Now that we have a new thread, in order to ensure that this thread continues to meet TL standards and follows the proper guidelines, we will be enforcing the rules in the OP more strictly. Be sure to give them a complete and thorough read before posting!
NOTE: When providing a source, please provide a very brief summary on what it's about and what purpose it adds to the discussion. The supporting statement should clearly explain why the subject is relevant and needs to be discussed. Please follow this rule especially for tweets.
Your supporting statement should always come BEFORE you provide the source.
On July 01 2024 01:14 Introvert wrote: And DBP, thr polls have been remarkably consistent all year. There's also less reason to think they will change because these are the two most well-known candidates in modern American history. They both have records people are evaluating, maybe with rose-tinted glasses but still. It's not like there's a magical date where polls become extra predictive, it's summer anyways and nothing has changed significantly in polling for 9 months.
My prediction has been that the election will end up being a coinflip, which would indeed be "remarkably consistent" with how the polls have shaped up so far, so there's a good chance that you and I end up being right! You seem to dismiss the idea that there may be any volatility between now and November though, such as Biden taking a dip due to his poor debate performance. I'm impressed by your confidence - that Biden's chances won't be affected and that he remains at basically a 50/50 chance with Trump - but I think it's more likely that we'll see at least a little movement in the polls (negative for Biden), especially over the next few weeks. Maybe his polling numbers will recover after that though. We'll have to see what happens in July and August, and then Biden's team will potentially need to deal with another problem around September (second debate), which could also lead to more volatility. But maybe you're correct that Biden's polling will end up being immune to negatives; that would certainly help to balance out the massive number of negative consequences that haven't affected Trump's polling!
GH has tried explaining this but Biden isn't tied, he's behind.
As of last week (before the debate), Biden was a negligible amount behind Trump. It was basically 50/50.
He has to win the popular vote by probably 2 points to be about tied.
That's false. Biden could lose the overall popular vote and still win the election. What matters is the outcome of a few key swing states. Whether he wins 70% or 51% of California is irrelevant for the election results, even though the former beefs up his overall popular vote.
On July 01 2024 02:21 GreenHorizons wrote:
On June 30 2024 22:12 DarkPlasmaBall wrote:
On June 30 2024 21:00 GreenHorizons wrote:
On June 30 2024 20:28 DarkPlasmaBall wrote: [quote]
Out of curiosity, why are you over Bernie?
[quote]
I never said that the next 4 months will be "so much better for Biden than the last 8 months". In fact, what I've been consistently saying is that the election will most likely come down to a coinflip. Spring polling (for whatever that's worth) agreed with me, and if summer/fall polling end up showing a post-debate dip for Biden that eventually starts to rebalance in 1-3 months, then it'll continue to be a coinflip. If the debate never happened, then staying the course of a coinflip would have also been a reasonable prediction. At no point had I suggested that Biden is going to start doing "so much better" in the polls.
For what it's worth, even if Biden is replaced, I think it'll still be a coinflip between the new candidate and Trump. Maybe the new candidate will have 55-45 odds instead of the reverse, but running an event like this only one time means that neither side can be super confident, even if it's 60-40 or 65-35 in favor of a candidate.
Bernie was a compromise to start with. The potential of him being the nominee/president was enough to keep a sliver of electoralism alive for me (back in 2016).
"Coinflip" (~50:50 and down by ~1-2% nationally) sounds a lot better than it is when you realize Biden was a 9:1 favorite with a 8.5% lead (4:1 with a 9%+ lead at this point in the race) going into an election he won by the skin of his teeth against the same guy in 2020 while citizens were dying by the hundreds or thousands per day from an ongoing pandemic Trump had clearly mismanaged.
Mind you this is up against a 2x impeached, 34x felony convicted, insurrectionist (who wasn't any of those things at this point last time, though he had been impeached once)
Biden/Democrats are in so much more trouble than you/his supporters/Democrats are prepared/willing to accept. This hubris could cost Dems the most important election of their lives and it's clear to anyone that isn't obsequiously supporting Biden imo.
Maybe I'll end up being wrong about my coinflip prediction - maybe in September and October, we'll see the polls tilted 70-30 or worse, in favor of Trump. Maybe things spiral out of control over the next few months. And if that happens, I'll have no problem saying that my prediction was wrong. But let's at least agree that my coinflip prediction is not the same as your claim that I "think the next 4 months will be so much better for Biden than the last 8 months", okay? I didn't say that, and I don't believe that.
With Biden's terrible debate performance, his chances of winning are surely worse now than they were a week ago. I said that I expect Biden to dip in the polls, and that I wonder if he'll recover from it; I'd hardly call that "hubris".
That's the fun of giving probabilities for presidential elections, as long as you give something a chance, you can't really be wrong.
We're presumably using the polls to inform the probabilities. For example, it would be inconsistent, based on the data, if someone said the current chance of Biden winning is 30% if the data says it's 50%.
One issue I'm trying to point out in that post is that it took a 8%+ lead in polls for Biden to squeak by Trump by thousands of votes. Biden consistently polling a couple points behind Trump is much worse than you/Democrats/Biden supporters are grasping. If the next 4 (3 before voting starts) months don't go much better than the last, then Biden is toast.
You're drawing a causal relationship that is unfounded. It's true that Biden was leading Trump by 8% in November 2020, and it's true that Biden just barely won the election, but it's incorrect to say that a polling lead of 8% is required for him to win again (especially in a different election).
The reason it's ~50:50 on 538 isn't the gap in the polls, it's the distance from the election (and the "fundamentals" favoring Biden).
The reason it's 50/50 is because a lot of people were asked who they'd vote for, and about half said Biden and about half said Trump (particularly for the key swing states they were polled in). That's it. 50/50 now means 50/50 now. It doesn't guarantee 50/50 in November, it doesn't guarantee 40/60 in November, and it doesn't guarantee anything else based on "the distance from the election". This is the problem with using early polling to extrapolate too far, rather than seeing a poll as a snapshot of a specific moment in time.
If you think people will gradually change their minds away from Biden, over the next few months, due to reasons like his poor debate performance, then that's fine, but those changes will still appear in the polls. For all the stock you've been putting into the polling data over the past few months, you're very quick to assert that those numbers really mean completely different numbers.
If Biden was polling like this in November it wouldn't be 50:50. To actually be 50:50, Biden has to do a lot better in polling than he is now and has been for 8 months.
If about half the people polled in November said they'd vote for Biden, and half said they've vote for Trump (particularly for the key swing states they were polled in), then the polls would say roughly 50/50.
So if you think it's going to be a 50:50 election you also believe that the next 4 months will be significantly better for Biden than the last 8 whether you say it explicitly or not.
I guess we'll just have to agree to disagree on what I believe. lol.
"inconsistent" isn't wrong.
I'm not saying he needs an 8% lead, I'm saying he needs better than to be losing as he has been for 8 months.
You're just misunderstanding how 538 is reaching its "50:50" probability.
When we have few polls or when it's early in the campaign, our model's predictions are mostly based on the fundamentals — with their standard deviation usually around 6 points or so. But when it's Election Day in a state with a lot of polls, our uncertainty about public opinion is a lot smaller, so we will put a lot more weight on the polling data (and subsequently, our polling averages that feed into the model) when generating our final prediction.
So 50:50 in the polls doesn't equate to a 50:50 chance. It only does now because "the fundamentals" in the model are more heavily weighted against polling and favor Biden. But those fundamentals are only going to become less weighted and if Biden's polling doesn't pick up (better than it has for 8 months) to counteract it, his odds will drop significantly even if his polling is marginally better.
I understand that Biden's strong economic and political fundamentals have been helping him in 538's early polling stages (I believe I've said about 10 times over the past few months that early polling is not necessarily a strong predictor of November outcomes, which is why I've been cautioning you and others to not take them as gospel). Introvert was the one who said that the next few months won't have fluctuations because the past few months have been steady. I didn't say that, and I don't think that one can safely make that causal claim.
I wouldn't be surprised at all if there's some volatility along the way to November (especially due to the debate), but I still predict that the November results will end up being a coinflip. And maybe I'm wrong. Maybe your prediction that Biden's polling numbers will consistently decline from now until November, as the weighing of fundamentals ease up in favor of other criteria, is correct. Maybe all three of us end up being wrong. We won't know until November, so I don't understand why you keep picking fights with me and others about these kinds of things.
I'm not saying (or conditioning) it not being a coinflip on Biden's polling getting worse, I'm saying that 538 wouldn't consider it a coinflip even if Biden was polling marginally better than he was before his debate disaster as we get closer to the election. He has to get higher in the polls than he's been in 8 months just to keep it near 50:50.
Biden's just in much more trouble than you appreciate currently.
Also you/Democrats/Biden supporters are advocating for a guy you iirc (and most of his 2020 voters) believe is aiding and abetting genocide, and could kill your mom in cold blood right in front of you and still not lose your vote or advocacy. I find that catastrophically problematic on several fronts, not the least of which, what it says of the ostensible "democracy" you're trying to "save" with that deplorable position in the first place.
You and I (and many others) have talked about this topic at great length. You disagree with our positions and explanations and rationalizations and justifications, and that's fine. I don't think I have anything else of substance to add to that particular conversation.
You asked why I keep challenging your narratives, so I gave you some of the reasons.
My hope from this exchange was that you'd at least recognize your error in assuming Biden polling about the same as he is now or slightly better when people start voting in September would net him 50:50 odds on election day. He has to do better for the next 4 months than he has for the last 8 months just to keep the odds near a coinflip. One of the big problems there being even his own staff says they can only get ~6 hours a day out of him and he's going to need to be sharp for far more than that to do the campaigning necessary just to salvage the debate, let alone actually take a lead. Based on letting/pushing him on the debate stage like that, he simply can't do what it will reasonably take to win.
Another aspect of Dem's problems is that replacing Biden is far from a magic bullet because they ignored the people saying Biden can't be the nominee for too long. So they're stuck unabashedly saying things like
[Biden] could have walked out, shit his pants, and then shuffled off and I'd still vote for him
like that isn't cult behavior
Question for you: If you were unilaterally in charge of the Democrats, and your single objective was to beat Trump in November, how would you play the (absolutely terrible) hand that you're currently being dealt? It's just about July now, you have less than 2 months before you formally appoint your best longshot candidate as the nominee to go up against the Republicans, and you have a little over 4 months total to magic your way into a victory on Election Day. However unlikely the win would be, how would GH approach things to maximize the Democrats' chances? Who's your presidential candidate? Who's your vice presidential candidate? What are the most important things they need to do, with the remaining time they have, and how do they do it?
If one had complete political control over everyone, then you could have Biden announce he steps back for medical reasons, on advice of his doctor and in consultation with his wife, and immediately have him endorse two new candidates. Have literally everyone immediately back them, Schiff, Pelosi, Obama, Newsom, AOC, Harris, Buttigieg, literally everyone.
And you'd have a good shot at beating Trump. Maybe even a better shot. Especially if that candidate can debate Trump again and crush him.
Problem is, that's not going to happen. If Biden is replaced, he is replaced in chaos. And that will be very bad.
If Biden believed he could step aside and endorse someone who had a better shot at beating Trump, he would have done so already. Unless some doctor tells him he can't run, or his wife begs him not to do it, he won't drop out. And if he does drop out, it will be Kamala Harris. That's literally the point of a VP. The backup if the president has a medical issue. Even if Biden wanted to drop out, there would be a lot of hurdles. Including the GOP blocking the new candidate from being on the ballot. The most practical way for Biden not to be the candidate is to keep both Biden and Harris on the ballot, but announce that Biden won't be president but Harris will. Even in that scenario, I believe you'd first have to inaugurate Biden as POTUS, to then immediately have Biden step down to Harris.
Then there's some crazy contested electoral college things aka West Wing aka House of Cards where electors & delegtes don't vote for the person they are pledged to vote for. If the scenario that Biden is replaced happens, those chaos scenarios are all injected into the whole process.
You didn't ask me, but GreenHorizons. But his answer won't be interesting, so don't bother.
On July 01 2024 01:14 Introvert wrote: And DBP, thr polls have been remarkably consistent all year. There's also less reason to think they will change because these are the two most well-known candidates in modern American history. They both have records people are evaluating, maybe with rose-tinted glasses but still. It's not like there's a magical date where polls become extra predictive, it's summer anyways and nothing has changed significantly in polling for 9 months.
My prediction has been that the election will end up being a coinflip, which would indeed be "remarkably consistent" with how the polls have shaped up so far, so there's a good chance that you and I end up being right! You seem to dismiss the idea that there may be any volatility between now and November though, such as Biden taking a dip due to his poor debate performance. I'm impressed by your confidence - that Biden's chances won't be affected and that he remains at basically a 50/50 chance with Trump - but I think it's more likely that we'll see at least a little movement in the polls (negative for Biden), especially over the next few weeks. Maybe his polling numbers will recover after that though. We'll have to see what happens in July and August, and then Biden's team will potentially need to deal with another problem around September (second debate), which could also lead to more volatility. But maybe you're correct that Biden's polling will end up being immune to negatives; that would certainly help to balance out the massive number of negative consequences that haven't affected Trump's polling!
GH has tried explaining this but Biden isn't tied, he's behind.
As of last week (before the debate), Biden was a negligible amount behind Trump. It was basically 50/50.
He has to win the popular vote by probably 2 points to be about tied.
That's false. Biden could lose the overall popular vote and still win the election. What matters is the outcome of a few key swing states. Whether he wins 70% or 51% of California is irrelevant for the election results, even though the former beefs up his overall popular vote.
On July 01 2024 02:21 GreenHorizons wrote:
On June 30 2024 22:12 DarkPlasmaBall wrote:
On June 30 2024 21:00 GreenHorizons wrote:
On June 30 2024 20:28 DarkPlasmaBall wrote:
On June 30 2024 20:00 GreenHorizons wrote: [quote]
I'm over Bernie personally
Out of curiosity, why are you over Bernie?
DPB's right that they aren't going to replace Biden unless they have to due to him just falling off a cliff medically. Why DPB and other Democrats/Biden supporters think the next 4 months will be so much better for Biden than the last 8 months I can't say though.
I never said that the next 4 months will be "so much better for Biden than the last 8 months". In fact, what I've been consistently saying is that the election will most likely come down to a coinflip. Spring polling (for whatever that's worth) agreed with me, and if summer/fall polling end up showing a post-debate dip for Biden that eventually starts to rebalance in 1-3 months, then it'll continue to be a coinflip. If the debate never happened, then staying the course of a coinflip would have also been a reasonable prediction. At no point had I suggested that Biden is going to start doing "so much better" in the polls.
For what it's worth, even if Biden is replaced, I think it'll still be a coinflip between the new candidate and Trump. Maybe the new candidate will have 55-45 odds instead of the reverse, but running an event like this only one time means that neither side can be super confident, even if it's 60-40 or 65-35 in favor of a candidate.
Bernie was a compromise to start with. The potential of him being the nominee/president was enough to keep a sliver of electoralism alive for me (back in 2016).
"Coinflip" (~50:50 and down by ~1-2% nationally) sounds a lot better than it is when you realize Biden was a 9:1 favorite with a 8.5% lead (4:1 with a 9%+ lead at this point in the race) going into an election he won by the skin of his teeth against the same guy in 2020 while citizens were dying by the hundreds or thousands per day from an ongoing pandemic Trump had clearly mismanaged.
Mind you this is up against a 2x impeached, 34x felony convicted, insurrectionist (who wasn't any of those things at this point last time, though he had been impeached once)
Biden/Democrats are in so much more trouble than you/his supporters/Democrats are prepared/willing to accept. This hubris could cost Dems the most important election of their lives and it's clear to anyone that isn't obsequiously supporting Biden imo.
Maybe I'll end up being wrong about my coinflip prediction - maybe in September and October, we'll see the polls tilted 70-30 or worse, in favor of Trump. Maybe things spiral out of control over the next few months. And if that happens, I'll have no problem saying that my prediction was wrong. But let's at least agree that my coinflip prediction is not the same as your claim that I "think the next 4 months will be so much better for Biden than the last 8 months", okay? I didn't say that, and I don't believe that.
With Biden's terrible debate performance, his chances of winning are surely worse now than they were a week ago. I said that I expect Biden to dip in the polls, and that I wonder if he'll recover from it; I'd hardly call that "hubris".
That's the fun of giving probabilities for presidential elections, as long as you give something a chance, you can't really be wrong.
We're presumably using the polls to inform the probabilities. For example, it would be inconsistent, based on the data, if someone said the current chance of Biden winning is 30% if the data says it's 50%.
One issue I'm trying to point out in that post is that it took a 8%+ lead in polls for Biden to squeak by Trump by thousands of votes. Biden consistently polling a couple points behind Trump is much worse than you/Democrats/Biden supporters are grasping. If the next 4 (3 before voting starts) months don't go much better than the last, then Biden is toast.
You're drawing a causal relationship that is unfounded. It's true that Biden was leading Trump by 8% in November 2020, and it's true that Biden just barely won the election, but it's incorrect to say that a polling lead of 8% is required for him to win again (especially in a different election).
The reason it's ~50:50 on 538 isn't the gap in the polls, it's the distance from the election (and the "fundamentals" favoring Biden).
The reason it's 50/50 is because a lot of people were asked who they'd vote for, and about half said Biden and about half said Trump (particularly for the key swing states they were polled in). That's it. 50/50 now means 50/50 now. It doesn't guarantee 50/50 in November, it doesn't guarantee 40/60 in November, and it doesn't guarantee anything else based on "the distance from the election". This is the problem with using early polling to extrapolate too far, rather than seeing a poll as a snapshot of a specific moment in time.
If you think people will gradually change their minds away from Biden, over the next few months, due to reasons like his poor debate performance, then that's fine, but those changes will still appear in the polls. For all the stock you've been putting into the polling data over the past few months, you're very quick to assert that those numbers really mean completely different numbers.
If Biden was polling like this in November it wouldn't be 50:50. To actually be 50:50, Biden has to do a lot better in polling than he is now and has been for 8 months.
If about half the people polled in November said they'd vote for Biden, and half said they've vote for Trump (particularly for the key swing states they were polled in), then the polls would say roughly 50/50.
So if you think it's going to be a 50:50 election you also believe that the next 4 months will be significantly better for Biden than the last 8 whether you say it explicitly or not.
I guess we'll just have to agree to disagree on what I believe. lol.
"inconsistent" isn't wrong.
I'm not saying he needs an 8% lead, I'm saying he needs better than to be losing as he has been for 8 months.
You're just misunderstanding how 538 is reaching its "50:50" probability.
When we have few polls or when it's early in the campaign, our model's predictions are mostly based on the fundamentals — with their standard deviation usually around 6 points or so. But when it's Election Day in a state with a lot of polls, our uncertainty about public opinion is a lot smaller, so we will put a lot more weight on the polling data (and subsequently, our polling averages that feed into the model) when generating our final prediction.
So 50:50 in the polls doesn't equate to a 50:50 chance. It only does now because "the fundamentals" in the model are more heavily weighted against polling and favor Biden. But those fundamentals are only going to become less weighted and if Biden's polling doesn't pick up (better than it has for 8 months) to counteract it, his odds will drop significantly even if his polling is marginally better.
I understand that Biden's strong economic and political fundamentals have been helping him in 538's early polling stages (I believe I've said about 10 times over the past few months that early polling is not necessarily a strong predictor of November outcomes, which is why I've been cautioning you and others to not take them as gospel). Introvert was the one who said that the next few months won't have fluctuations because the past few months have been steady. I didn't say that, and I don't think that one can safely make that causal claim.
I wouldn't be surprised at all if there's some volatility along the way to November (especially due to the debate), but I still predict that the November results will end up being a coinflip. And maybe I'm wrong. Maybe your prediction that Biden's polling numbers will consistently decline from now until November, as the weighing of fundamentals ease up in favor of other criteria, is correct. Maybe all three of us end up being wrong. We won't know until November, so I don't understand why you keep picking fights with me and others about these kinds of things.
I'm not saying (or conditioning) it not being a coinflip on Biden's polling getting worse, I'm saying that 538 wouldn't consider it a coinflip even if Biden was polling marginally better than he was before his debate disaster as we get closer to the election. He has to get higher in the polls than he's been in 8 months just to keep it near 50:50.
Biden's just in much more trouble than you appreciate currently.
Also you/Democrats/Biden supporters are advocating for a guy you iirc (and most of his 2020 voters) believe is aiding and abetting genocide, and could kill your mom in cold blood right in front of you and still not lose your vote or advocacy. I find that catastrophically problematic on several fronts, not the least of which, what it says of the ostensible "democracy" you're trying to "save" with that deplorable position in the first place.
You and I (and many others) have talked about this topic at great length. You disagree with our positions and explanations and rationalizations and justifications, and that's fine. I don't think I have anything else of substance to add to that particular conversation.
On July 01 2024 01:14 Introvert wrote: And DBP, thr polls have been remarkably consistent all year. There's also less reason to think they will change because these are the two most well-known candidates in modern American history. They both have records people are evaluating, maybe with rose-tinted glasses but still. It's not like there's a magical date where polls become extra predictive, it's summer anyways and nothing has changed significantly in polling for 9 months.
My prediction has been that the election will end up being a coinflip, which would indeed be "remarkably consistent" with how the polls have shaped up so far, so there's a good chance that you and I end up being right! You seem to dismiss the idea that there may be any volatility between now and November though, such as Biden taking a dip due to his poor debate performance. I'm impressed by your confidence - that Biden's chances won't be affected and that he remains at basically a 50/50 chance with Trump - but I think it's more likely that we'll see at least a little movement in the polls (negative for Biden), especially over the next few weeks. Maybe his polling numbers will recover after that though. We'll have to see what happens in July and August, and then Biden's team will potentially need to deal with another problem around September (second debate), which could also lead to more volatility. But maybe you're correct that Biden's polling will end up being immune to negatives; that would certainly help to balance out the massive number of negative consequences that haven't affected Trump's polling!
GH has tried explaining this but Biden isn't tied, he's behind.
As of last week (before the debate), Biden was a negligible amount behind Trump. It was basically 50/50.
He has to win the popular vote by probably 2 points to be about tied.
That's false. Biden could lose the overall popular vote and still win the election. What matters is the outcome of a few key swing states. Whether he wins 70% or 51% of California is irrelevant for the election results, even though the former beefs up his overall popular vote.
On July 01 2024 02:21 GreenHorizons wrote:
On June 30 2024 22:12 DarkPlasmaBall wrote:
On June 30 2024 21:00 GreenHorizons wrote:
On June 30 2024 20:28 DarkPlasmaBall wrote:
On June 30 2024 20:00 GreenHorizons wrote: [quote]
I'm over Bernie personally
Out of curiosity, why are you over Bernie?
DPB's right that they aren't going to replace Biden unless they have to due to him just falling off a cliff medically. Why DPB and other Democrats/Biden supporters think the next 4 months will be so much better for Biden than the last 8 months I can't say though.
I never said that the next 4 months will be "so much better for Biden than the last 8 months". In fact, what I've been consistently saying is that the election will most likely come down to a coinflip. Spring polling (for whatever that's worth) agreed with me, and if summer/fall polling end up showing a post-debate dip for Biden that eventually starts to rebalance in 1-3 months, then it'll continue to be a coinflip. If the debate never happened, then staying the course of a coinflip would have also been a reasonable prediction. At no point had I suggested that Biden is going to start doing "so much better" in the polls.
For what it's worth, even if Biden is replaced, I think it'll still be a coinflip between the new candidate and Trump. Maybe the new candidate will have 55-45 odds instead of the reverse, but running an event like this only one time means that neither side can be super confident, even if it's 60-40 or 65-35 in favor of a candidate.
Bernie was a compromise to start with. The potential of him being the nominee/president was enough to keep a sliver of electoralism alive for me (back in 2016).
"Coinflip" (~50:50 and down by ~1-2% nationally) sounds a lot better than it is when you realize Biden was a 9:1 favorite with a 8.5% lead (4:1 with a 9%+ lead at this point in the race) going into an election he won by the skin of his teeth against the same guy in 2020 while citizens were dying by the hundreds or thousands per day from an ongoing pandemic Trump had clearly mismanaged.
Mind you this is up against a 2x impeached, 34x felony convicted, insurrectionist (who wasn't any of those things at this point last time, though he had been impeached once)
Biden/Democrats are in so much more trouble than you/his supporters/Democrats are prepared/willing to accept. This hubris could cost Dems the most important election of their lives and it's clear to anyone that isn't obsequiously supporting Biden imo.
Maybe I'll end up being wrong about my coinflip prediction - maybe in September and October, we'll see the polls tilted 70-30 or worse, in favor of Trump. Maybe things spiral out of control over the next few months. And if that happens, I'll have no problem saying that my prediction was wrong. But let's at least agree that my coinflip prediction is not the same as your claim that I "think the next 4 months will be so much better for Biden than the last 8 months", okay? I didn't say that, and I don't believe that.
With Biden's terrible debate performance, his chances of winning are surely worse now than they were a week ago. I said that I expect Biden to dip in the polls, and that I wonder if he'll recover from it; I'd hardly call that "hubris".
That's the fun of giving probabilities for presidential elections, as long as you give something a chance, you can't really be wrong.
We're presumably using the polls to inform the probabilities. For example, it would be inconsistent, based on the data, if someone said the current chance of Biden winning is 30% if the data says it's 50%.
One issue I'm trying to point out in that post is that it took a 8%+ lead in polls for Biden to squeak by Trump by thousands of votes. Biden consistently polling a couple points behind Trump is much worse than you/Democrats/Biden supporters are grasping. If the next 4 (3 before voting starts) months don't go much better than the last, then Biden is toast.
You're drawing a causal relationship that is unfounded. It's true that Biden was leading Trump by 8% in November 2020, and it's true that Biden just barely won the election, but it's incorrect to say that a polling lead of 8% is required for him to win again (especially in a different election).
The reason it's ~50:50 on 538 isn't the gap in the polls, it's the distance from the election (and the "fundamentals" favoring Biden).
The reason it's 50/50 is because a lot of people were asked who they'd vote for, and about half said Biden and about half said Trump (particularly for the key swing states they were polled in). That's it. 50/50 now means 50/50 now. It doesn't guarantee 50/50 in November, it doesn't guarantee 40/60 in November, and it doesn't guarantee anything else based on "the distance from the election". This is the problem with using early polling to extrapolate too far, rather than seeing a poll as a snapshot of a specific moment in time.
If you think people will gradually change their minds away from Biden, over the next few months, due to reasons like his poor debate performance, then that's fine, but those changes will still appear in the polls. For all the stock you've been putting into the polling data over the past few months, you're very quick to assert that those numbers really mean completely different numbers.
If Biden was polling like this in November it wouldn't be 50:50. To actually be 50:50, Biden has to do a lot better in polling than he is now and has been for 8 months.
If about half the people polled in November said they'd vote for Biden, and half said they've vote for Trump (particularly for the key swing states they were polled in), then the polls would say roughly 50/50.
So if you think it's going to be a 50:50 election you also believe that the next 4 months will be significantly better for Biden than the last 8 whether you say it explicitly or not.
I guess we'll just have to agree to disagree on what I believe. lol.
"inconsistent" isn't wrong.
I'm not saying he needs an 8% lead, I'm saying he needs better than to be losing as he has been for 8 months.
You're just misunderstanding how 538 is reaching its "50:50" probability.
When we have few polls or when it's early in the campaign, our model's predictions are mostly based on the fundamentals — with their standard deviation usually around 6 points or so. But when it's Election Day in a state with a lot of polls, our uncertainty about public opinion is a lot smaller, so we will put a lot more weight on the polling data (and subsequently, our polling averages that feed into the model) when generating our final prediction.
So 50:50 in the polls doesn't equate to a 50:50 chance. It only does now because "the fundamentals" in the model are more heavily weighted against polling and favor Biden. But those fundamentals are only going to become less weighted and if Biden's polling doesn't pick up (better than it has for 8 months) to counteract it, his odds will drop significantly even if his polling is marginally better.
I understand that Biden's strong economic and political fundamentals have been helping him in 538's early polling stages (I believe I've said about 10 times over the past few months that early polling is not necessarily a strong predictor of November outcomes, which is why I've been cautioning you and others to not take them as gospel). Introvert was the one who said that the next few months won't have fluctuations because the past few months have been steady. I didn't say that, and I don't think that one can safely make that causal claim.
I wouldn't be surprised at all if there's some volatility along the way to November (especially due to the debate), but I still predict that the November results will end up being a coinflip. And maybe I'm wrong. Maybe your prediction that Biden's polling numbers will consistently decline from now until November, as the weighing of fundamentals ease up in favor of other criteria, is correct. Maybe all three of us end up being wrong. We won't know until November, so I don't understand why you keep picking fights with me and others about these kinds of things.
The thing about all this is that you are the one hoping, against any evidence currently avaliable, that things are going to grt better for Biden when there's a good chance they get even worse instead. I'm looking at the history of how these things work. BTW, 538 is the only (or one of the only) model(s) that has Biden as good as a coinflip. Your assertion that things solidify later is true, but you asking it to do far too much work. Polls even now are generally predictive.
As to your point about the swing states... for all your focus on polling and models I'm surprised you don't know or understand the electoral college disadvantage dems have had and been complaining about for years. If Biden is even nationally he is almost certainly DOWN in the swing states, and indeed state level polling shows this. For decades the swing states vote right of the nation as a whole, and moreover states with similar demographics in the same region are correlated. If black turnout is down and rural turnout is better for Trump in WI it almost certainly is also true in MI, for example. The "Biden only has to win the swing states" is a problem for Biden, not a thing in his favor.
I don't know what will happen, but for months now you have been trying to pretend that he's in better shape, or that the state of the race is more ambigious than it actually is. Again, people can't handle that dems are actually losing.
I know that Democrats tend to win the popular vote, even if they lose the election. If the polling has him as 50/50 even chance to win the election, then that's talking about the electoral college, not the popular vote. The polls run scenarios for winning the election and individual state outcomes. If a poll announced something like "we asked a million American voters and 500,000 said they'll vote for Biden", that wouldn't be helpful at all without knowing where each of those 1 million voters live, but that's not what's happening here.
I am trying to figure out if I am misunderstanding whet you are saying here, because it seems wrong, and obviously so.
That's not what polls do, that's what modelers do, with polls as one input. Polls alone currently have Biden down slightly in the PV and down more in the swing states. Your one million person poll is precisely what they do...they weight their responses based on demographics and vote history but no poll is a poll of who is going to win the election, that doesn't even make sense. A national poll showing Trump 45% Biden 43% among all voters is literally a weighted tally of how people are going to vote, not Biden's chances of winning the election.
On July 01 2024 01:14 Introvert wrote: And DBP, thr polls have been remarkably consistent all year. There's also less reason to think they will change because these are the two most well-known candidates in modern American history. They both have records people are evaluating, maybe with rose-tinted glasses but still. It's not like there's a magical date where polls become extra predictive, it's summer anyways and nothing has changed significantly in polling for 9 months.
My prediction has been that the election will end up being a coinflip, which would indeed be "remarkably consistent" with how the polls have shaped up so far, so there's a good chance that you and I end up being right! You seem to dismiss the idea that there may be any volatility between now and November though, such as Biden taking a dip due to his poor debate performance. I'm impressed by your confidence - that Biden's chances won't be affected and that he remains at basically a 50/50 chance with Trump - but I think it's more likely that we'll see at least a little movement in the polls (negative for Biden), especially over the next few weeks. Maybe his polling numbers will recover after that though. We'll have to see what happens in July and August, and then Biden's team will potentially need to deal with another problem around September (second debate), which could also lead to more volatility. But maybe you're correct that Biden's polling will end up being immune to negatives; that would certainly help to balance out the massive number of negative consequences that haven't affected Trump's polling!
GH has tried explaining this but Biden isn't tied, he's behind.
As of last week (before the debate), Biden was a negligible amount behind Trump. It was basically 50/50.
He has to win the popular vote by probably 2 points to be about tied.
That's false. Biden could lose the overall popular vote and still win the election. What matters is the outcome of a few key swing states. Whether he wins 70% or 51% of California is irrelevant for the election results, even though the former beefs up his overall popular vote.
On July 01 2024 02:21 GreenHorizons wrote:
On June 30 2024 22:12 DarkPlasmaBall wrote:
On June 30 2024 21:00 GreenHorizons wrote:
On June 30 2024 20:28 DarkPlasmaBall wrote: [quote]
Out of curiosity, why are you over Bernie?
[quote]
I never said that the next 4 months will be "so much better for Biden than the last 8 months". In fact, what I've been consistently saying is that the election will most likely come down to a coinflip. Spring polling (for whatever that's worth) agreed with me, and if summer/fall polling end up showing a post-debate dip for Biden that eventually starts to rebalance in 1-3 months, then it'll continue to be a coinflip. If the debate never happened, then staying the course of a coinflip would have also been a reasonable prediction. At no point had I suggested that Biden is going to start doing "so much better" in the polls.
For what it's worth, even if Biden is replaced, I think it'll still be a coinflip between the new candidate and Trump. Maybe the new candidate will have 55-45 odds instead of the reverse, but running an event like this only one time means that neither side can be super confident, even if it's 60-40 or 65-35 in favor of a candidate.
Bernie was a compromise to start with. The potential of him being the nominee/president was enough to keep a sliver of electoralism alive for me (back in 2016).
"Coinflip" (~50:50 and down by ~1-2% nationally) sounds a lot better than it is when you realize Biden was a 9:1 favorite with a 8.5% lead (4:1 with a 9%+ lead at this point in the race) going into an election he won by the skin of his teeth against the same guy in 2020 while citizens were dying by the hundreds or thousands per day from an ongoing pandemic Trump had clearly mismanaged.
Mind you this is up against a 2x impeached, 34x felony convicted, insurrectionist (who wasn't any of those things at this point last time, though he had been impeached once)
Biden/Democrats are in so much more trouble than you/his supporters/Democrats are prepared/willing to accept. This hubris could cost Dems the most important election of their lives and it's clear to anyone that isn't obsequiously supporting Biden imo.
Maybe I'll end up being wrong about my coinflip prediction - maybe in September and October, we'll see the polls tilted 70-30 or worse, in favor of Trump. Maybe things spiral out of control over the next few months. And if that happens, I'll have no problem saying that my prediction was wrong. But let's at least agree that my coinflip prediction is not the same as your claim that I "think the next 4 months will be so much better for Biden than the last 8 months", okay? I didn't say that, and I don't believe that.
With Biden's terrible debate performance, his chances of winning are surely worse now than they were a week ago. I said that I expect Biden to dip in the polls, and that I wonder if he'll recover from it; I'd hardly call that "hubris".
That's the fun of giving probabilities for presidential elections, as long as you give something a chance, you can't really be wrong.
We're presumably using the polls to inform the probabilities. For example, it would be inconsistent, based on the data, if someone said the current chance of Biden winning is 30% if the data says it's 50%.
One issue I'm trying to point out in that post is that it took a 8%+ lead in polls for Biden to squeak by Trump by thousands of votes. Biden consistently polling a couple points behind Trump is much worse than you/Democrats/Biden supporters are grasping. If the next 4 (3 before voting starts) months don't go much better than the last, then Biden is toast.
You're drawing a causal relationship that is unfounded. It's true that Biden was leading Trump by 8% in November 2020, and it's true that Biden just barely won the election, but it's incorrect to say that a polling lead of 8% is required for him to win again (especially in a different election).
The reason it's ~50:50 on 538 isn't the gap in the polls, it's the distance from the election (and the "fundamentals" favoring Biden).
The reason it's 50/50 is because a lot of people were asked who they'd vote for, and about half said Biden and about half said Trump (particularly for the key swing states they were polled in). That's it. 50/50 now means 50/50 now. It doesn't guarantee 50/50 in November, it doesn't guarantee 40/60 in November, and it doesn't guarantee anything else based on "the distance from the election". This is the problem with using early polling to extrapolate too far, rather than seeing a poll as a snapshot of a specific moment in time.
If you think people will gradually change their minds away from Biden, over the next few months, due to reasons like his poor debate performance, then that's fine, but those changes will still appear in the polls. For all the stock you've been putting into the polling data over the past few months, you're very quick to assert that those numbers really mean completely different numbers.
If Biden was polling like this in November it wouldn't be 50:50. To actually be 50:50, Biden has to do a lot better in polling than he is now and has been for 8 months.
If about half the people polled in November said they'd vote for Biden, and half said they've vote for Trump (particularly for the key swing states they were polled in), then the polls would say roughly 50/50.
So if you think it's going to be a 50:50 election you also believe that the next 4 months will be significantly better for Biden than the last 8 whether you say it explicitly or not.
I guess we'll just have to agree to disagree on what I believe. lol.
"inconsistent" isn't wrong.
I'm not saying he needs an 8% lead, I'm saying he needs better than to be losing as he has been for 8 months.
You're just misunderstanding how 538 is reaching its "50:50" probability.
When we have few polls or when it's early in the campaign, our model's predictions are mostly based on the fundamentals — with their standard deviation usually around 6 points or so. But when it's Election Day in a state with a lot of polls, our uncertainty about public opinion is a lot smaller, so we will put a lot more weight on the polling data (and subsequently, our polling averages that feed into the model) when generating our final prediction.
So 50:50 in the polls doesn't equate to a 50:50 chance. It only does now because "the fundamentals" in the model are more heavily weighted against polling and favor Biden. But those fundamentals are only going to become less weighted and if Biden's polling doesn't pick up (better than it has for 8 months) to counteract it, his odds will drop significantly even if his polling is marginally better.
I understand that Biden's strong economic and political fundamentals have been helping him in 538's early polling stages (I believe I've said about 10 times over the past few months that early polling is not necessarily a strong predictor of November outcomes, which is why I've been cautioning you and others to not take them as gospel). Introvert was the one who said that the next few months won't have fluctuations because the past few months have been steady. I didn't say that, and I don't think that one can safely make that causal claim.
I wouldn't be surprised at all if there's some volatility along the way to November (especially due to the debate), but I still predict that the November results will end up being a coinflip. And maybe I'm wrong. Maybe your prediction that Biden's polling numbers will consistently decline from now until November, as the weighing of fundamentals ease up in favor of other criteria, is correct. Maybe all three of us end up being wrong. We won't know until November, so I don't understand why you keep picking fights with me and others about these kinds of things.
I'm not saying (or conditioning) it not being a coinflip on Biden's polling getting worse, I'm saying that 538 wouldn't consider it a coinflip even if Biden was polling marginally better than he was before his debate disaster as we get closer to the election. He has to get higher in the polls than he's been in 8 months just to keep it near 50:50.
Biden's just in much more trouble than you appreciate currently.
Also you/Democrats/Biden supporters are advocating for a guy you iirc (and most of his 2020 voters) believe is aiding and abetting genocide, and could kill your mom in cold blood right in front of you and still not lose your vote or advocacy. I find that catastrophically problematic on several fronts, not the least of which, what it says of the ostensible "democracy" you're trying to "save" with that deplorable position in the first place.
You and I (and many others) have talked about this topic at great length. You disagree with our positions and explanations and rationalizations and justifications, and that's fine. I don't think I have anything else of substance to add to that particular conversation.
On July 01 2024 01:14 Introvert wrote: And DBP, thr polls have been remarkably consistent all year. There's also less reason to think they will change because these are the two most well-known candidates in modern American history. They both have records people are evaluating, maybe with rose-tinted glasses but still. It's not like there's a magical date where polls become extra predictive, it's summer anyways and nothing has changed significantly in polling for 9 months.
My prediction has been that the election will end up being a coinflip, which would indeed be "remarkably consistent" with how the polls have shaped up so far, so there's a good chance that you and I end up being right! You seem to dismiss the idea that there may be any volatility between now and November though, such as Biden taking a dip due to his poor debate performance. I'm impressed by your confidence - that Biden's chances won't be affected and that he remains at basically a 50/50 chance with Trump - but I think it's more likely that we'll see at least a little movement in the polls (negative for Biden), especially over the next few weeks. Maybe his polling numbers will recover after that though. We'll have to see what happens in July and August, and then Biden's team will potentially need to deal with another problem around September (second debate), which could also lead to more volatility. But maybe you're correct that Biden's polling will end up being immune to negatives; that would certainly help to balance out the massive number of negative consequences that haven't affected Trump's polling!
GH has tried explaining this but Biden isn't tied, he's behind.
As of last week (before the debate), Biden was a negligible amount behind Trump. It was basically 50/50.
He has to win the popular vote by probably 2 points to be about tied.
That's false. Biden could lose the overall popular vote and still win the election. What matters is the outcome of a few key swing states. Whether he wins 70% or 51% of California is irrelevant for the election results, even though the former beefs up his overall popular vote.
On July 01 2024 02:21 GreenHorizons wrote:
On June 30 2024 22:12 DarkPlasmaBall wrote:
On June 30 2024 21:00 GreenHorizons wrote:
On June 30 2024 20:28 DarkPlasmaBall wrote: [quote]
Out of curiosity, why are you over Bernie?
[quote]
I never said that the next 4 months will be "so much better for Biden than the last 8 months". In fact, what I've been consistently saying is that the election will most likely come down to a coinflip. Spring polling (for whatever that's worth) agreed with me, and if summer/fall polling end up showing a post-debate dip for Biden that eventually starts to rebalance in 1-3 months, then it'll continue to be a coinflip. If the debate never happened, then staying the course of a coinflip would have also been a reasonable prediction. At no point had I suggested that Biden is going to start doing "so much better" in the polls.
For what it's worth, even if Biden is replaced, I think it'll still be a coinflip between the new candidate and Trump. Maybe the new candidate will have 55-45 odds instead of the reverse, but running an event like this only one time means that neither side can be super confident, even if it's 60-40 or 65-35 in favor of a candidate.
Bernie was a compromise to start with. The potential of him being the nominee/president was enough to keep a sliver of electoralism alive for me (back in 2016).
"Coinflip" (~50:50 and down by ~1-2% nationally) sounds a lot better than it is when you realize Biden was a 9:1 favorite with a 8.5% lead (4:1 with a 9%+ lead at this point in the race) going into an election he won by the skin of his teeth against the same guy in 2020 while citizens were dying by the hundreds or thousands per day from an ongoing pandemic Trump had clearly mismanaged.
Mind you this is up against a 2x impeached, 34x felony convicted, insurrectionist (who wasn't any of those things at this point last time, though he had been impeached once)
Biden/Democrats are in so much more trouble than you/his supporters/Democrats are prepared/willing to accept. This hubris could cost Dems the most important election of their lives and it's clear to anyone that isn't obsequiously supporting Biden imo.
Maybe I'll end up being wrong about my coinflip prediction - maybe in September and October, we'll see the polls tilted 70-30 or worse, in favor of Trump. Maybe things spiral out of control over the next few months. And if that happens, I'll have no problem saying that my prediction was wrong. But let's at least agree that my coinflip prediction is not the same as your claim that I "think the next 4 months will be so much better for Biden than the last 8 months", okay? I didn't say that, and I don't believe that.
With Biden's terrible debate performance, his chances of winning are surely worse now than they were a week ago. I said that I expect Biden to dip in the polls, and that I wonder if he'll recover from it; I'd hardly call that "hubris".
That's the fun of giving probabilities for presidential elections, as long as you give something a chance, you can't really be wrong.
We're presumably using the polls to inform the probabilities. For example, it would be inconsistent, based on the data, if someone said the current chance of Biden winning is 30% if the data says it's 50%.
One issue I'm trying to point out in that post is that it took a 8%+ lead in polls for Biden to squeak by Trump by thousands of votes. Biden consistently polling a couple points behind Trump is much worse than you/Democrats/Biden supporters are grasping. If the next 4 (3 before voting starts) months don't go much better than the last, then Biden is toast.
You're drawing a causal relationship that is unfounded. It's true that Biden was leading Trump by 8% in November 2020, and it's true that Biden just barely won the election, but it's incorrect to say that a polling lead of 8% is required for him to win again (especially in a different election).
The reason it's ~50:50 on 538 isn't the gap in the polls, it's the distance from the election (and the "fundamentals" favoring Biden).
The reason it's 50/50 is because a lot of people were asked who they'd vote for, and about half said Biden and about half said Trump (particularly for the key swing states they were polled in). That's it. 50/50 now means 50/50 now. It doesn't guarantee 50/50 in November, it doesn't guarantee 40/60 in November, and it doesn't guarantee anything else based on "the distance from the election". This is the problem with using early polling to extrapolate too far, rather than seeing a poll as a snapshot of a specific moment in time.
If you think people will gradually change their minds away from Biden, over the next few months, due to reasons like his poor debate performance, then that's fine, but those changes will still appear in the polls. For all the stock you've been putting into the polling data over the past few months, you're very quick to assert that those numbers really mean completely different numbers.
If Biden was polling like this in November it wouldn't be 50:50. To actually be 50:50, Biden has to do a lot better in polling than he is now and has been for 8 months.
If about half the people polled in November said they'd vote for Biden, and half said they've vote for Trump (particularly for the key swing states they were polled in), then the polls would say roughly 50/50.
So if you think it's going to be a 50:50 election you also believe that the next 4 months will be significantly better for Biden than the last 8 whether you say it explicitly or not.
I guess we'll just have to agree to disagree on what I believe. lol.
"inconsistent" isn't wrong.
I'm not saying he needs an 8% lead, I'm saying he needs better than to be losing as he has been for 8 months.
You're just misunderstanding how 538 is reaching its "50:50" probability.
When we have few polls or when it's early in the campaign, our model's predictions are mostly based on the fundamentals — with their standard deviation usually around 6 points or so. But when it's Election Day in a state with a lot of polls, our uncertainty about public opinion is a lot smaller, so we will put a lot more weight on the polling data (and subsequently, our polling averages that feed into the model) when generating our final prediction.
So 50:50 in the polls doesn't equate to a 50:50 chance. It only does now because "the fundamentals" in the model are more heavily weighted against polling and favor Biden. But those fundamentals are only going to become less weighted and if Biden's polling doesn't pick up (better than it has for 8 months) to counteract it, his odds will drop significantly even if his polling is marginally better.
I understand that Biden's strong economic and political fundamentals have been helping him in 538's early polling stages (I believe I've said about 10 times over the past few months that early polling is not necessarily a strong predictor of November outcomes, which is why I've been cautioning you and others to not take them as gospel). Introvert was the one who said that the next few months won't have fluctuations because the past few months have been steady. I didn't say that, and I don't think that one can safely make that causal claim.
I wouldn't be surprised at all if there's some volatility along the way to November (especially due to the debate), but I still predict that the November results will end up being a coinflip. And maybe I'm wrong. Maybe your prediction that Biden's polling numbers will consistently decline from now until November, as the weighing of fundamentals ease up in favor of other criteria, is correct. Maybe all three of us end up being wrong. We won't know until November, so I don't understand why you keep picking fights with me and others about these kinds of things.
The thing about all this is that you are the one hoping, against any evidence currently avaliable, that things are going to grt better for Biden when there's a good chance they get even worse instead. I'm looking at the history of how these things work. BTW, 538 is the only (or one of the only) model(s) that has Biden as good as a coinflip. Your assertion that things solidify later is true, but you asking it to do far too much work. Polls even now are generally predictive.
As to your point about the swing states... for all your focus on polling and models I'm surprised you don't know or understand the electoral college disadvantage dems have had and been complaining about for years. If Biden is even nationally he is almost certainly DOWN in the swing states, and indeed state level polling shows this. For decades the swing states vote right of the nation as a whole, and moreover states with similar demographics in the same region are correlated. If black turnout is down and rural turnout is better for Trump in WI it almost certainly is also true in MI, for example. The "Biden only has to win the swing states" is a problem for Biden, not a thing in his favor.
I don't know what will happen, but for months now you have been trying to pretend that he's in better shape, or that the state of the race is more ambigious than it actually is. Again, people can't handle that dems are actually losing.
I know that Democrats tend to win the popular vote, even if they lose the election. If the polling has him as 50/50 even chance to win the election, then that's talking about the electoral college, not the popular vote. The polls run scenarios for winning the election and individual state outcomes. If a poll announced something like "we asked a million American voters and 500,000 said they'll vote for Biden", that wouldn't be helpful at all without knowing where each of those 1 million voters live, but that's not what's happening here.
I am trying to figure out if I am misunderstanding whet you are saying here, because it seems wrong, and obviously so.
That's not what polls do, that's what modelers do, with polls as one input. Polls alone currently have Biden down slightly in the PV and down more in the swing states. Your one million person poll is precisely what they do...they weight their responses based on demographics and vote history but no poll is a poll of who is going to win the election, that doesn't even make sense. A national poll showing Trump 45% Biden 43% among all voters is literally a weighted tally of how people are going to vote, not Biden's chances of winning the election.
Ah, I see what you're referring to. I was referring to the aggregate of the state polls - that's why I wrote "(particularly for the key swing states they were polled in)" twice in my earlier post, and I'm happy to call that a "model", as you're saying it's what "modelers" do. That's totally fair. It's definitely easier to use that different term than referring to every survey and meta-analysis of surveys as falling under the umbrella phrasing of what the polls say, so that we avoid semantics issues.
On July 01 2024 05:08 Conaing wrote: GreenHorizons, what are you even arguing for. It is 2024. We can't go back to 2016 and make Bernie Sanders beat Trump. That would have been great, but that didn't happen.
Biden is the candidate. Even if he shits his pants during the next debate. It isn't a 'cult' because the US is a two party system because of first past the post. And if the other guy wins, it may be the last election.
The entire US political establishment will let Netanyahu commit genocide, which he isn't doing, and cringe but not step in. That's baked into the cake since 20 years. Deal with it.
The cult is of the system. So much so that they'll willingly hand Trump, who they ostensibly believe will be a day 1 dictator destroying their system, control of the most dangerous military and nuclear arsenal in the world to do it with, and/or cast meaningless votes for someone that murders their mom in cold blood and/or shits themselves while speaking gibberish, and/or they believe is engaged in genocide, all in veneration of said system.
As far as Bernie goes, I just think he was Democrats last chance to salvage US democracy and they still can't accept that ship has sailed as they cling to Biden's husk with a blind hope that its not.
On July 01 2024 01:14 Introvert wrote: And DBP, thr polls have been remarkably consistent all year. There's also less reason to think they will change because these are the two most well-known candidates in modern American history. They both have records people are evaluating, maybe with rose-tinted glasses but still. It's not like there's a magical date where polls become extra predictive, it's summer anyways and nothing has changed significantly in polling for 9 months.
My prediction has been that the election will end up being a coinflip, which would indeed be "remarkably consistent" with how the polls have shaped up so far, so there's a good chance that you and I end up being right! You seem to dismiss the idea that there may be any volatility between now and November though, such as Biden taking a dip due to his poor debate performance. I'm impressed by your confidence - that Biden's chances won't be affected and that he remains at basically a 50/50 chance with Trump - but I think it's more likely that we'll see at least a little movement in the polls (negative for Biden), especially over the next few weeks. Maybe his polling numbers will recover after that though. We'll have to see what happens in July and August, and then Biden's team will potentially need to deal with another problem around September (second debate), which could also lead to more volatility. But maybe you're correct that Biden's polling will end up being immune to negatives; that would certainly help to balance out the massive number of negative consequences that haven't affected Trump's polling!
GH has tried explaining this but Biden isn't tied, he's behind.
As of last week (before the debate), Biden was a negligible amount behind Trump. It was basically 50/50.
He has to win the popular vote by probably 2 points to be about tied.
That's false. Biden could lose the overall popular vote and still win the election. What matters is the outcome of a few key swing states. Whether he wins 70% or 51% of California is irrelevant for the election results, even though the former beefs up his overall popular vote.
On July 01 2024 02:21 GreenHorizons wrote:
On June 30 2024 22:12 DarkPlasmaBall wrote:
On June 30 2024 21:00 GreenHorizons wrote: [quote] Bernie was a compromise to start with. The potential of him being the nominee/president was enough to keep a sliver of electoralism alive for me (back in 2016).
"Coinflip" (~50:50 and down by ~1-2% nationally) sounds a lot better than it is when you realize Biden was a 9:1 favorite with a 8.5% lead (4:1 with a 9%+ lead at this point in the race) going into an election he won by the skin of his teeth against the same guy in 2020 while citizens were dying by the hundreds or thousands per day from an ongoing pandemic Trump had clearly mismanaged.
Mind you this is up against a 2x impeached, 34x felony convicted, insurrectionist (who wasn't any of those things at this point last time, though he had been impeached once)
Biden/Democrats are in so much more trouble than you/his supporters/Democrats are prepared/willing to accept. This hubris could cost Dems the most important election of their lives and it's clear to anyone that isn't obsequiously supporting Biden imo.
Maybe I'll end up being wrong about my coinflip prediction - maybe in September and October, we'll see the polls tilted 70-30 or worse, in favor of Trump. Maybe things spiral out of control over the next few months. And if that happens, I'll have no problem saying that my prediction was wrong. But let's at least agree that my coinflip prediction is not the same as your claim that I "think the next 4 months will be so much better for Biden than the last 8 months", okay? I didn't say that, and I don't believe that.
With Biden's terrible debate performance, his chances of winning are surely worse now than they were a week ago. I said that I expect Biden to dip in the polls, and that I wonder if he'll recover from it; I'd hardly call that "hubris".
That's the fun of giving probabilities for presidential elections, as long as you give something a chance, you can't really be wrong.
We're presumably using the polls to inform the probabilities. For example, it would be inconsistent, based on the data, if someone said the current chance of Biden winning is 30% if the data says it's 50%.
One issue I'm trying to point out in that post is that it took a 8%+ lead in polls for Biden to squeak by Trump by thousands of votes. Biden consistently polling a couple points behind Trump is much worse than you/Democrats/Biden supporters are grasping. If the next 4 (3 before voting starts) months don't go much better than the last, then Biden is toast.
You're drawing a causal relationship that is unfounded. It's true that Biden was leading Trump by 8% in November 2020, and it's true that Biden just barely won the election, but it's incorrect to say that a polling lead of 8% is required for him to win again (especially in a different election).
The reason it's ~50:50 on 538 isn't the gap in the polls, it's the distance from the election (and the "fundamentals" favoring Biden).
The reason it's 50/50 is because a lot of people were asked who they'd vote for, and about half said Biden and about half said Trump (particularly for the key swing states they were polled in). That's it. 50/50 now means 50/50 now. It doesn't guarantee 50/50 in November, it doesn't guarantee 40/60 in November, and it doesn't guarantee anything else based on "the distance from the election". This is the problem with using early polling to extrapolate too far, rather than seeing a poll as a snapshot of a specific moment in time.
If you think people will gradually change their minds away from Biden, over the next few months, due to reasons like his poor debate performance, then that's fine, but those changes will still appear in the polls. For all the stock you've been putting into the polling data over the past few months, you're very quick to assert that those numbers really mean completely different numbers.
If Biden was polling like this in November it wouldn't be 50:50. To actually be 50:50, Biden has to do a lot better in polling than he is now and has been for 8 months.
If about half the people polled in November said they'd vote for Biden, and half said they've vote for Trump (particularly for the key swing states they were polled in), then the polls would say roughly 50/50.
So if you think it's going to be a 50:50 election you also believe that the next 4 months will be significantly better for Biden than the last 8 whether you say it explicitly or not.
I guess we'll just have to agree to disagree on what I believe. lol.
"inconsistent" isn't wrong.
I'm not saying he needs an 8% lead, I'm saying he needs better than to be losing as he has been for 8 months.
You're just misunderstanding how 538 is reaching its "50:50" probability.
When we have few polls or when it's early in the campaign, our model's predictions are mostly based on the fundamentals — with their standard deviation usually around 6 points or so. But when it's Election Day in a state with a lot of polls, our uncertainty about public opinion is a lot smaller, so we will put a lot more weight on the polling data (and subsequently, our polling averages that feed into the model) when generating our final prediction.
So 50:50 in the polls doesn't equate to a 50:50 chance. It only does now because "the fundamentals" in the model are more heavily weighted against polling and favor Biden. But those fundamentals are only going to become less weighted and if Biden's polling doesn't pick up (better than it has for 8 months) to counteract it, his odds will drop significantly even if his polling is marginally better.
I understand that Biden's strong economic and political fundamentals have been helping him in 538's early polling stages (I believe I've said about 10 times over the past few months that early polling is not necessarily a strong predictor of November outcomes, which is why I've been cautioning you and others to not take them as gospel). Introvert was the one who said that the next few months won't have fluctuations because the past few months have been steady. I didn't say that, and I don't think that one can safely make that causal claim.
I wouldn't be surprised at all if there's some volatility along the way to November (especially due to the debate), but I still predict that the November results will end up being a coinflip. And maybe I'm wrong. Maybe your prediction that Biden's polling numbers will consistently decline from now until November, as the weighing of fundamentals ease up in favor of other criteria, is correct. Maybe all three of us end up being wrong. We won't know until November, so I don't understand why you keep picking fights with me and others about these kinds of things.
I'm not saying (or conditioning) it not being a coinflip on Biden's polling getting worse, I'm saying that 538 wouldn't consider it a coinflip even if Biden was polling marginally better than he was before his debate disaster as we get closer to the election. He has to get higher in the polls than he's been in 8 months just to keep it near 50:50.
Biden's just in much more trouble than you appreciate currently.
Also you/Democrats/Biden supporters are advocating for a guy you iirc (and most of his 2020 voters) believe is aiding and abetting genocide, and could kill your mom in cold blood right in front of you and still not lose your vote or advocacy. I find that catastrophically problematic on several fronts, not the least of which, what it says of the ostensible "democracy" you're trying to "save" with that deplorable position in the first place.
You and I (and many others) have talked about this topic at great length. You disagree with our positions and explanations and rationalizations and justifications, and that's fine. I don't think I have anything else of substance to add to that particular conversation.
You asked why I keep challenging your narratives, so I gave you some of the reasons.
My hope from this exchange was that you'd at least recognize your error in assuming Biden polling about the same as he is now or slightly better when people start voting in September would net him 50:50 odds on election day. He has to do better for the next 4 months than he has for the last 8 months just to keep the odds near a coinflip. One of the big problems there being even his own staff says they can only get ~6 hours a day out of him and he's going to need to be sharp for far more than that to do the campaigning necessary just to salvage the debate, let alone actually take a lead. Based on letting/pushing him on the debate stage like that, he simply can't do what it will reasonably take to win.
Another aspect of Dem's problems is that replacing Biden is far from a magic bullet because they ignored the people saying Biden can't be the nominee for too long. So they're stuck unabashedly saying things like
[Biden] could have walked out, shit his pants, and then shuffled off and I'd still vote for him
like that isn't cult behavior
Question for you: If you were unilaterally in charge of the Democrats, and your single objective was to beat Trump in November, how would you play the (absolutely terrible) hand that you're currently being dealt? It's just about July now, you have less than 2 months before you formally appoint your best longshot candidate as the nominee to go up against the Republicans, and you have a little over 4 months total to magic your way into a victory on Election Day. However unlikely the win would be, how would GH approach things to maximize the Democrats' chances? Who's your presidential candidate? Who's your vice presidential candidate? What are the most important things they need to do, with the remaining time they have, and how do they do it?
Pretty much this:
If one had complete political control over everyone, then you could have Biden announce he steps back for medical reasons, on advice of his doctor and in consultation with his wife, and immediately have him endorse two new candidates. Have literally everyone immediately back them, Schiff, Pelosi, Obama, Newsom, AOC, Harris, Buttigieg, literally everyone.
And you'd have a good shot at beating Trump. Maybe even a better shot. Especially if that candidate can debate Trump again and crush him.
But I agree with Cona in that it aint happenin. Which is one of several reasons Democrats are in so much worse shape than you and Democrats/their supporters recognize currently.
@Cona I'd consider Barack Obama for VP if that makes it any more interesting for you?
On July 01 2024 01:14 Introvert wrote: And DBP, thr polls have been remarkably consistent all year. There's also less reason to think they will change because these are the two most well-known candidates in modern American history. They both have records people are evaluating, maybe with rose-tinted glasses but still. It's not like there's a magical date where polls become extra predictive, it's summer anyways and nothing has changed significantly in polling for 9 months.
My prediction has been that the election will end up being a coinflip, which would indeed be "remarkably consistent" with how the polls have shaped up so far, so there's a good chance that you and I end up being right! You seem to dismiss the idea that there may be any volatility between now and November though, such as Biden taking a dip due to his poor debate performance. I'm impressed by your confidence - that Biden's chances won't be affected and that he remains at basically a 50/50 chance with Trump - but I think it's more likely that we'll see at least a little movement in the polls (negative for Biden), especially over the next few weeks. Maybe his polling numbers will recover after that though. We'll have to see what happens in July and August, and then Biden's team will potentially need to deal with another problem around September (second debate), which could also lead to more volatility. But maybe you're correct that Biden's polling will end up being immune to negatives; that would certainly help to balance out the massive number of negative consequences that haven't affected Trump's polling!
GH has tried explaining this but Biden isn't tied, he's behind.
As of last week (before the debate), Biden was a negligible amount behind Trump. It was basically 50/50.
He has to win the popular vote by probably 2 points to be about tied.
That's false. Biden could lose the overall popular vote and still win the election. What matters is the outcome of a few key swing states. Whether he wins 70% or 51% of California is irrelevant for the election results, even though the former beefs up his overall popular vote.
On July 01 2024 02:21 GreenHorizons wrote:
On June 30 2024 22:12 DarkPlasmaBall wrote: [quote]
Maybe I'll end up being wrong about my coinflip prediction - maybe in September and October, we'll see the polls tilted 70-30 or worse, in favor of Trump. Maybe things spiral out of control over the next few months. And if that happens, I'll have no problem saying that my prediction was wrong. But let's at least agree that my coinflip prediction is not the same as your claim that I "think the next 4 months will be so much better for Biden than the last 8 months", okay? I didn't say that, and I don't believe that.
With Biden's terrible debate performance, his chances of winning are surely worse now than they were a week ago. I said that I expect Biden to dip in the polls, and that I wonder if he'll recover from it; I'd hardly call that "hubris".
That's the fun of giving probabilities for presidential elections, as long as you give something a chance, you can't really be wrong.
We're presumably using the polls to inform the probabilities. For example, it would be inconsistent, based on the data, if someone said the current chance of Biden winning is 30% if the data says it's 50%.
One issue I'm trying to point out in that post is that it took a 8%+ lead in polls for Biden to squeak by Trump by thousands of votes. Biden consistently polling a couple points behind Trump is much worse than you/Democrats/Biden supporters are grasping. If the next 4 (3 before voting starts) months don't go much better than the last, then Biden is toast.
You're drawing a causal relationship that is unfounded. It's true that Biden was leading Trump by 8% in November 2020, and it's true that Biden just barely won the election, but it's incorrect to say that a polling lead of 8% is required for him to win again (especially in a different election).
The reason it's ~50:50 on 538 isn't the gap in the polls, it's the distance from the election (and the "fundamentals" favoring Biden).
The reason it's 50/50 is because a lot of people were asked who they'd vote for, and about half said Biden and about half said Trump (particularly for the key swing states they were polled in). That's it. 50/50 now means 50/50 now. It doesn't guarantee 50/50 in November, it doesn't guarantee 40/60 in November, and it doesn't guarantee anything else based on "the distance from the election". This is the problem with using early polling to extrapolate too far, rather than seeing a poll as a snapshot of a specific moment in time.
If you think people will gradually change their minds away from Biden, over the next few months, due to reasons like his poor debate performance, then that's fine, but those changes will still appear in the polls. For all the stock you've been putting into the polling data over the past few months, you're very quick to assert that those numbers really mean completely different numbers.
If Biden was polling like this in November it wouldn't be 50:50. To actually be 50:50, Biden has to do a lot better in polling than he is now and has been for 8 months.
If about half the people polled in November said they'd vote for Biden, and half said they've vote for Trump (particularly for the key swing states they were polled in), then the polls would say roughly 50/50.
So if you think it's going to be a 50:50 election you also believe that the next 4 months will be significantly better for Biden than the last 8 whether you say it explicitly or not.
I guess we'll just have to agree to disagree on what I believe. lol.
"inconsistent" isn't wrong.
I'm not saying he needs an 8% lead, I'm saying he needs better than to be losing as he has been for 8 months.
You're just misunderstanding how 538 is reaching its "50:50" probability.
When we have few polls or when it's early in the campaign, our model's predictions are mostly based on the fundamentals — with their standard deviation usually around 6 points or so. But when it's Election Day in a state with a lot of polls, our uncertainty about public opinion is a lot smaller, so we will put a lot more weight on the polling data (and subsequently, our polling averages that feed into the model) when generating our final prediction.
So 50:50 in the polls doesn't equate to a 50:50 chance. It only does now because "the fundamentals" in the model are more heavily weighted against polling and favor Biden. But those fundamentals are only going to become less weighted and if Biden's polling doesn't pick up (better than it has for 8 months) to counteract it, his odds will drop significantly even if his polling is marginally better.
I understand that Biden's strong economic and political fundamentals have been helping him in 538's early polling stages (I believe I've said about 10 times over the past few months that early polling is not necessarily a strong predictor of November outcomes, which is why I've been cautioning you and others to not take them as gospel). Introvert was the one who said that the next few months won't have fluctuations because the past few months have been steady. I didn't say that, and I don't think that one can safely make that causal claim.
I wouldn't be surprised at all if there's some volatility along the way to November (especially due to the debate), but I still predict that the November results will end up being a coinflip. And maybe I'm wrong. Maybe your prediction that Biden's polling numbers will consistently decline from now until November, as the weighing of fundamentals ease up in favor of other criteria, is correct. Maybe all three of us end up being wrong. We won't know until November, so I don't understand why you keep picking fights with me and others about these kinds of things.
I'm not saying (or conditioning) it not being a coinflip on Biden's polling getting worse, I'm saying that 538 wouldn't consider it a coinflip even if Biden was polling marginally better than he was before his debate disaster as we get closer to the election. He has to get higher in the polls than he's been in 8 months just to keep it near 50:50.
Biden's just in much more trouble than you appreciate currently.
Also you/Democrats/Biden supporters are advocating for a guy you iirc (and most of his 2020 voters) believe is aiding and abetting genocide, and could kill your mom in cold blood right in front of you and still not lose your vote or advocacy. I find that catastrophically problematic on several fronts, not the least of which, what it says of the ostensible "democracy" you're trying to "save" with that deplorable position in the first place.
You and I (and many others) have talked about this topic at great length. You disagree with our positions and explanations and rationalizations and justifications, and that's fine. I don't think I have anything else of substance to add to that particular conversation.
You asked why I keep challenging your narratives, so I gave you some of the reasons.
My hope from this exchange was that you'd at least recognize your error in assuming Biden polling about the same as he is now or slightly better when people start voting in September would net him 50:50 odds on election day. He has to do better for the next 4 months than he has for the last 8 months just to keep the odds near a coinflip. One of the big problems there being even his own staff says they can only get ~6 hours a day out of him and he's going to need to be sharp for far more than that to do the campaigning necessary just to salvage the debate, let alone actually take a lead. Based on letting/pushing him on the debate stage like that, he simply can't do what it will reasonably take to win.
Another aspect of Dem's problems is that replacing Biden is far from a magic bullet because they ignored the people saying Biden can't be the nominee for too long. So they're stuck unabashedly saying things like
[Biden] could have walked out, shit his pants, and then shuffled off and I'd still vote for him
like that isn't cult behavior
Question for you: If you were unilaterally in charge of the Democrats, and your single objective was to beat Trump in November, how would you play the (absolutely terrible) hand that you're currently being dealt? It's just about July now, you have less than 2 months before you formally appoint your best longshot candidate as the nominee to go up against the Republicans, and you have a little over 4 months total to magic your way into a victory on Election Day. However unlikely the win would be, how would GH approach things to maximize the Democrats' chances? Who's your presidential candidate? Who's your vice presidential candidate? What are the most important things they need to do, with the remaining time they have, and how do they do it?
If one had complete political control over everyone, then you could have Biden announce he steps back for medical reasons, on advice of his doctor and in consultation with his wife, and immediately have him endorse two new candidates. Have literally everyone immediately back them, Schiff, Pelosi, Obama, Newsom, AOC, Harris, Buttigieg, literally everyone.
And you'd have a good shot at beating Trump. Maybe even a better shot. Especially if that candidate can debate Trump again and crush him.
But I agree with Cona in that it aint happenin. Which is one of several reasons Democrats are in so much worse shape than you and Democrats/their supporters recognize currently.
But which two candidates would you put on the Democratic ticket?
On July 01 2024 04:52 Conaing wrote: @RenSC2 If you platform Trump, then it is your responsibility to fact check him. You can't say that you can't do that because it is a debate.
Honestly. How can you even say this. What do you want? A debate where both sides just yell "you are lying." and "No, you are lying". On top of that, the format made any attempt to even fact check the other side impossible. You can put the world champ real time fact checker on the other stage vs Trump. And he can't do much because their microphone is muted.
This is 100% on CNN. Saying anything else is completely normalizing Trump. Trump is not normal. If you are not willing to fact check Trump because you don't want the moderator to debate Trump, then just say you can't do the debate because Trump is unmoderatable and it would be irresponsible to have him lie on your news channel. It is as simple as that.
Trump actually wants to be fact checked. Even fact checking Trump wouldn't have worked because a CNN vs Trump debate would be bad for Biden. That's why Trump shouldn't even be in this race. You can't suddenly begin to fact check Trump. You should have done that starting in 2016. And every American voter should have Trump's lies, plus almost 10 years of fact checking already burned into their memories, before a 2024 Trump vs Biden debate even starts.
You could even have Biden debate RFK (who is actually at 10% in popular vote poll, which is crazy). And then Trump can do his lying spree alone on Fox News or OneAmericanNews. And if you think that's unfair to the voters. Guess what. Trump shouldn't morally have been on the ballot in 2016. And Trump shouldn't legally have been on the ballot in 2024. You, CNN, already made this unfair to the voter. Because they don't know wtf is true anyone because of post-Truth politics.
BTW people here don't seem to care what I say. This is a good political commentator, mostly on Russia, analysing the debate. Way better than any of the cookie cutter 'The rest is politics' or David Pacman or Destiny or Meidas Touch or The Bullwark stuff:
good points in the video but i don't know about the arrogance part; to me it looked more like a desire/want to overcome a weakness(of his, Bidens') regardless of costs(optic or otherwise) and means needed to accomplish it.
but, Destiny would teach Biden to meme himself into a new term, fo sho; people like comedians.
An Al Gore & Barack Obama ticket sounds like maybe a good idea for the establishment to try to beat Trump, but Obama cannot be vice president according to the 12th amendment. Obama can't be vice president because if anything were to happen to his president, he would automatically become president. Which he cannot be, because of his two previous terms. So he can't be a vice presidential candidate.
Would Obama actually be that popular nowadays, though? Politicians burn through a lot of their political credit during their terms. Especially if they do not deliver on their promises. It would also play into Trump's conspiracy theories that Obama is actually president right now. Obama being VP is so Putin-esque. The GW Bush -> B Clinton -> W Bush -> B Obama -> H Clinton -> M Obama is so bad. Why not just elect a royal family then? People talk about Michelle Obama now. Which seems crazy to me, as if she had ambitions she could have become a senator or something. That's what Hillary did.
Delegates hand-picked by the Biden campaign would need to vote for someone else during a contested convention. Biden himself has no control over this process either. He can endorse someone else, but they are pledged to him. Not to the person he endorses. It will likely be chaos.
On July 01 2024 01:14 Introvert wrote: And DBP, thr polls have been remarkably consistent all year. There's also less reason to think they will change because these are the two most well-known candidates in modern American history. They both have records people are evaluating, maybe with rose-tinted glasses but still. It's not like there's a magical date where polls become extra predictive, it's summer anyways and nothing has changed significantly in polling for 9 months.
My prediction has been that the election will end up being a coinflip, which would indeed be "remarkably consistent" with how the polls have shaped up so far, so there's a good chance that you and I end up being right! You seem to dismiss the idea that there may be any volatility between now and November though, such as Biden taking a dip due to his poor debate performance. I'm impressed by your confidence - that Biden's chances won't be affected and that he remains at basically a 50/50 chance with Trump - but I think it's more likely that we'll see at least a little movement in the polls (negative for Biden), especially over the next few weeks. Maybe his polling numbers will recover after that though. We'll have to see what happens in July and August, and then Biden's team will potentially need to deal with another problem around September (second debate), which could also lead to more volatility. But maybe you're correct that Biden's polling will end up being immune to negatives; that would certainly help to balance out the massive number of negative consequences that haven't affected Trump's polling!
GH has tried explaining this but Biden isn't tied, he's behind.
As of last week (before the debate), Biden was a negligible amount behind Trump. It was basically 50/50.
He has to win the popular vote by probably 2 points to be about tied.
That's false. Biden could lose the overall popular vote and still win the election. What matters is the outcome of a few key swing states. Whether he wins 70% or 51% of California is irrelevant for the election results, even though the former beefs up his overall popular vote.
On July 01 2024 02:21 GreenHorizons wrote: [quote]
That's the fun of giving probabilities for presidential elections, as long as you give something a chance, you can't really be wrong.
We're presumably using the polls to inform the probabilities. For example, it would be inconsistent, based on the data, if someone said the current chance of Biden winning is 30% if the data says it's 50%.
One issue I'm trying to point out in that post is that it took a 8%+ lead in polls for Biden to squeak by Trump by thousands of votes. Biden consistently polling a couple points behind Trump is much worse than you/Democrats/Biden supporters are grasping. If the next 4 (3 before voting starts) months don't go much better than the last, then Biden is toast.
You're drawing a causal relationship that is unfounded. It's true that Biden was leading Trump by 8% in November 2020, and it's true that Biden just barely won the election, but it's incorrect to say that a polling lead of 8% is required for him to win again (especially in a different election).
The reason it's ~50:50 on 538 isn't the gap in the polls, it's the distance from the election (and the "fundamentals" favoring Biden).
The reason it's 50/50 is because a lot of people were asked who they'd vote for, and about half said Biden and about half said Trump (particularly for the key swing states they were polled in). That's it. 50/50 now means 50/50 now. It doesn't guarantee 50/50 in November, it doesn't guarantee 40/60 in November, and it doesn't guarantee anything else based on "the distance from the election". This is the problem with using early polling to extrapolate too far, rather than seeing a poll as a snapshot of a specific moment in time.
If you think people will gradually change their minds away from Biden, over the next few months, due to reasons like his poor debate performance, then that's fine, but those changes will still appear in the polls. For all the stock you've been putting into the polling data over the past few months, you're very quick to assert that those numbers really mean completely different numbers.
If Biden was polling like this in November it wouldn't be 50:50. To actually be 50:50, Biden has to do a lot better in polling than he is now and has been for 8 months.
If about half the people polled in November said they'd vote for Biden, and half said they've vote for Trump (particularly for the key swing states they were polled in), then the polls would say roughly 50/50.
So if you think it's going to be a 50:50 election you also believe that the next 4 months will be significantly better for Biden than the last 8 whether you say it explicitly or not.
I guess we'll just have to agree to disagree on what I believe. lol.
"inconsistent" isn't wrong.
I'm not saying he needs an 8% lead, I'm saying he needs better than to be losing as he has been for 8 months.
You're just misunderstanding how 538 is reaching its "50:50" probability.
When we have few polls or when it's early in the campaign, our model's predictions are mostly based on the fundamentals — with their standard deviation usually around 6 points or so. But when it's Election Day in a state with a lot of polls, our uncertainty about public opinion is a lot smaller, so we will put a lot more weight on the polling data (and subsequently, our polling averages that feed into the model) when generating our final prediction.
So 50:50 in the polls doesn't equate to a 50:50 chance. It only does now because "the fundamentals" in the model are more heavily weighted against polling and favor Biden. But those fundamentals are only going to become less weighted and if Biden's polling doesn't pick up (better than it has for 8 months) to counteract it, his odds will drop significantly even if his polling is marginally better.
I understand that Biden's strong economic and political fundamentals have been helping him in 538's early polling stages (I believe I've said about 10 times over the past few months that early polling is not necessarily a strong predictor of November outcomes, which is why I've been cautioning you and others to not take them as gospel). Introvert was the one who said that the next few months won't have fluctuations because the past few months have been steady. I didn't say that, and I don't think that one can safely make that causal claim.
I wouldn't be surprised at all if there's some volatility along the way to November (especially due to the debate), but I still predict that the November results will end up being a coinflip. And maybe I'm wrong. Maybe your prediction that Biden's polling numbers will consistently decline from now until November, as the weighing of fundamentals ease up in favor of other criteria, is correct. Maybe all three of us end up being wrong. We won't know until November, so I don't understand why you keep picking fights with me and others about these kinds of things.
I'm not saying (or conditioning) it not being a coinflip on Biden's polling getting worse, I'm saying that 538 wouldn't consider it a coinflip even if Biden was polling marginally better than he was before his debate disaster as we get closer to the election. He has to get higher in the polls than he's been in 8 months just to keep it near 50:50.
Biden's just in much more trouble than you appreciate currently.
Also you/Democrats/Biden supporters are advocating for a guy you iirc (and most of his 2020 voters) believe is aiding and abetting genocide, and could kill your mom in cold blood right in front of you and still not lose your vote or advocacy. I find that catastrophically problematic on several fronts, not the least of which, what it says of the ostensible "democracy" you're trying to "save" with that deplorable position in the first place.
You and I (and many others) have talked about this topic at great length. You disagree with our positions and explanations and rationalizations and justifications, and that's fine. I don't think I have anything else of substance to add to that particular conversation.
You asked why I keep challenging your narratives, so I gave you some of the reasons.
My hope from this exchange was that you'd at least recognize your error in assuming Biden polling about the same as he is now or slightly better when people start voting in September would net him 50:50 odds on election day. He has to do better for the next 4 months than he has for the last 8 months just to keep the odds near a coinflip. One of the big problems there being even his own staff says they can only get ~6 hours a day out of him and he's going to need to be sharp for far more than that to do the campaigning necessary just to salvage the debate, let alone actually take a lead. Based on letting/pushing him on the debate stage like that, he simply can't do what it will reasonably take to win.
Another aspect of Dem's problems is that replacing Biden is far from a magic bullet because they ignored the people saying Biden can't be the nominee for too long. So they're stuck unabashedly saying things like
[Biden] could have walked out, shit his pants, and then shuffled off and I'd still vote for him
like that isn't cult behavior
Question for you: If you were unilaterally in charge of the Democrats, and your single objective was to beat Trump in November, how would you play the (absolutely terrible) hand that you're currently being dealt? It's just about July now, you have less than 2 months before you formally appoint your best longshot candidate as the nominee to go up against the Republicans, and you have a little over 4 months total to magic your way into a victory on Election Day. However unlikely the win would be, how would GH approach things to maximize the Democrats' chances? Who's your presidential candidate? Who's your vice presidential candidate? What are the most important things they need to do, with the remaining time they have, and how do they do it?
Pretty much this:
If one had complete political control over everyone, then you could have Biden announce he steps back for medical reasons, on advice of his doctor and in consultation with his wife, and immediately have him endorse two new candidates. Have literally everyone immediately back them, Schiff, Pelosi, Obama, Newsom, AOC, Harris, Buttigieg, literally everyone.
And you'd have a good shot at beating Trump. Maybe even a better shot. Especially if that candidate can debate Trump again and crush him.
But I agree with Cona in that it aint happenin. Which is one of several reasons Democrats are in so much worse shape than you and Democrats/their supporters recognize currently.
But which two candidates would you put on the Democratic ticket?
I'd have been polling this question months ago, had a real primary, etc. and would never be in this hole to dig out of, but assuming I have to start from today I'd have to poll it now in this hypothetical. Also I'm not exactly clear if I get unanimity of "vote blue no matter who" support from Democrats regardless of whom I put on the ticket.
I don't have an exact duo to give you regardless, but some of the short list would be Barack Obama (the most popular/favorable person Democrats have), Michelle Obama (probably more popular, but less of a politician), Gavin Newsom, JB Pritzker, (generic white guys to assuage swing voting bigots) and I guess Whitmer specifically for MI (but don't think she could carry Dems nationally).
Jesus the pickins are slim among Democrats. Hard to appreciate just how badly Hillary and Joe's ambitions, hubris and presence in the party fucked everyone over for generations.
On Obama being VP that would go to SCOTUS, but possibly not in time for the election. It's basically whether you interpret the 12th amendment to include "being elected president" instead of the literal "No person constitutionally ineligible to the office of President shall be eligible to that of Vice-President of the United States," which he wouldn't be. He's ineligible to be "elected" president, but he's literally eligible to be + Show Spoiler +
"President must be at least 35 years of age, be a natural born citizen, and must have lived in the United States for at least 14 years."
On July 01 2024 01:14 Introvert wrote: And DBP, thr polls have been remarkably consistent all year. There's also less reason to think they will change because these are the two most well-known candidates in modern American history. They both have records people are evaluating, maybe with rose-tinted glasses but still. It's not like there's a magical date where polls become extra predictive, it's summer anyways and nothing has changed significantly in polling for 9 months.
My prediction has been that the election will end up being a coinflip, which would indeed be "remarkably consistent" with how the polls have shaped up so far, so there's a good chance that you and I end up being right! You seem to dismiss the idea that there may be any volatility between now and November though, such as Biden taking a dip due to his poor debate performance. I'm impressed by your confidence - that Biden's chances won't be affected and that he remains at basically a 50/50 chance with Trump - but I think it's more likely that we'll see at least a little movement in the polls (negative for Biden), especially over the next few weeks. Maybe his polling numbers will recover after that though. We'll have to see what happens in July and August, and then Biden's team will potentially need to deal with another problem around September (second debate), which could also lead to more volatility. But maybe you're correct that Biden's polling will end up being immune to negatives; that would certainly help to balance out the massive number of negative consequences that haven't affected Trump's polling!
GH has tried explaining this but Biden isn't tied, he's behind.
As of last week (before the debate), Biden was a negligible amount behind Trump. It was basically 50/50.
He has to win the popular vote by probably 2 points to be about tied.
That's false. Biden could lose the overall popular vote and still win the election. What matters is the outcome of a few key swing states. Whether he wins 70% or 51% of California is irrelevant for the election results, even though the former beefs up his overall popular vote.
[quote]
We're presumably using the polls to inform the probabilities. For example, it would be inconsistent, based on the data, if someone said the current chance of Biden winning is 30% if the data says it's 50%.
[quote]
You're drawing a causal relationship that is unfounded. It's true that Biden was leading Trump by 8% in November 2020, and it's true that Biden just barely won the election, but it's incorrect to say that a polling lead of 8% is required for him to win again (especially in a different election).
[quote]
The reason it's 50/50 is because a lot of people were asked who they'd vote for, and about half said Biden and about half said Trump (particularly for the key swing states they were polled in). That's it. 50/50 now means 50/50 now. It doesn't guarantee 50/50 in November, it doesn't guarantee 40/60 in November, and it doesn't guarantee anything else based on "the distance from the election". This is the problem with using early polling to extrapolate too far, rather than seeing a poll as a snapshot of a specific moment in time.
If you think people will gradually change their minds away from Biden, over the next few months, due to reasons like his poor debate performance, then that's fine, but those changes will still appear in the polls. For all the stock you've been putting into the polling data over the past few months, you're very quick to assert that those numbers really mean completely different numbers.
[quote]
If about half the people polled in November said they'd vote for Biden, and half said they've vote for Trump (particularly for the key swing states they were polled in), then the polls would say roughly 50/50.
[quote]
I guess we'll just have to agree to disagree on what I believe. lol.
"inconsistent" isn't wrong.
I'm not saying he needs an 8% lead, I'm saying he needs better than to be losing as he has been for 8 months.
You're just misunderstanding how 538 is reaching its "50:50" probability.
When we have few polls or when it's early in the campaign, our model's predictions are mostly based on the fundamentals — with their standard deviation usually around 6 points or so. But when it's Election Day in a state with a lot of polls, our uncertainty about public opinion is a lot smaller, so we will put a lot more weight on the polling data (and subsequently, our polling averages that feed into the model) when generating our final prediction.
So 50:50 in the polls doesn't equate to a 50:50 chance. It only does now because "the fundamentals" in the model are more heavily weighted against polling and favor Biden. But those fundamentals are only going to become less weighted and if Biden's polling doesn't pick up (better than it has for 8 months) to counteract it, his odds will drop significantly even if his polling is marginally better.
I understand that Biden's strong economic and political fundamentals have been helping him in 538's early polling stages (I believe I've said about 10 times over the past few months that early polling is not necessarily a strong predictor of November outcomes, which is why I've been cautioning you and others to not take them as gospel). Introvert was the one who said that the next few months won't have fluctuations because the past few months have been steady. I didn't say that, and I don't think that one can safely make that causal claim.
I wouldn't be surprised at all if there's some volatility along the way to November (especially due to the debate), but I still predict that the November results will end up being a coinflip. And maybe I'm wrong. Maybe your prediction that Biden's polling numbers will consistently decline from now until November, as the weighing of fundamentals ease up in favor of other criteria, is correct. Maybe all three of us end up being wrong. We won't know until November, so I don't understand why you keep picking fights with me and others about these kinds of things.
I'm not saying (or conditioning) it not being a coinflip on Biden's polling getting worse, I'm saying that 538 wouldn't consider it a coinflip even if Biden was polling marginally better than he was before his debate disaster as we get closer to the election. He has to get higher in the polls than he's been in 8 months just to keep it near 50:50.
Biden's just in much more trouble than you appreciate currently.
Also you/Democrats/Biden supporters are advocating for a guy you iirc (and most of his 2020 voters) believe is aiding and abetting genocide, and could kill your mom in cold blood right in front of you and still not lose your vote or advocacy. I find that catastrophically problematic on several fronts, not the least of which, what it says of the ostensible "democracy" you're trying to "save" with that deplorable position in the first place.
You and I (and many others) have talked about this topic at great length. You disagree with our positions and explanations and rationalizations and justifications, and that's fine. I don't think I have anything else of substance to add to that particular conversation.
You asked why I keep challenging your narratives, so I gave you some of the reasons.
My hope from this exchange was that you'd at least recognize your error in assuming Biden polling about the same as he is now or slightly better when people start voting in September would net him 50:50 odds on election day. He has to do better for the next 4 months than he has for the last 8 months just to keep the odds near a coinflip. One of the big problems there being even his own staff says they can only get ~6 hours a day out of him and he's going to need to be sharp for far more than that to do the campaigning necessary just to salvage the debate, let alone actually take a lead. Based on letting/pushing him on the debate stage like that, he simply can't do what it will reasonably take to win.
Another aspect of Dem's problems is that replacing Biden is far from a magic bullet because they ignored the people saying Biden can't be the nominee for too long. So they're stuck unabashedly saying things like
[Biden] could have walked out, shit his pants, and then shuffled off and I'd still vote for him
like that isn't cult behavior
Question for you: If you were unilaterally in charge of the Democrats, and your single objective was to beat Trump in November, how would you play the (absolutely terrible) hand that you're currently being dealt? It's just about July now, you have less than 2 months before you formally appoint your best longshot candidate as the nominee to go up against the Republicans, and you have a little over 4 months total to magic your way into a victory on Election Day. However unlikely the win would be, how would GH approach things to maximize the Democrats' chances? Who's your presidential candidate? Who's your vice presidential candidate? What are the most important things they need to do, with the remaining time they have, and how do they do it?
Pretty much this:
If one had complete political control over everyone, then you could have Biden announce he steps back for medical reasons, on advice of his doctor and in consultation with his wife, and immediately have him endorse two new candidates. Have literally everyone immediately back them, Schiff, Pelosi, Obama, Newsom, AOC, Harris, Buttigieg, literally everyone.
And you'd have a good shot at beating Trump. Maybe even a better shot. Especially if that candidate can debate Trump again and crush him.
But I agree with Cona in that it aint happenin. Which is one of several reasons Democrats are in so much worse shape than you and Democrats/their supporters recognize currently.
But which two candidates would you put on the Democratic ticket?
I'd have been polling this question months ago, had a real primary, etc. and would never be in this hole to dig out of, but assuming I have to start from today I'd have to poll it now in this hypothetical. Also I'm not exactly clear if I get unanimity of "vote blue no matter who" support from Democrats regardless of whom I put on the ticket.
I don't have an exact duo to give you regardless, but some of the short list would be Barack Obama (the most popular/favorable person Democrats have), Michelle Obama (probably more popular, but less of a politician), Gavin Newsom, JB Pritzker, (generic white guys to assuage swing voting bigots) and I guess Whitmer specifically for MI (but don't think she could carry Dems nationally).
Jesus the pickins are slim among Democrats. Hard to appreciate just how badly Hillary and Joe's ambitions, hubris and presence in the party fucked everyone over for generations.
On Obama being VP that would go to SCOTUS, but possibly not in time for the election. It's basically whether you interpret the 12th amendment to include "being elected president" instead of the literal "No person constitutionally ineligible to the office of President shall be eligible to that of Vice-President of the United States," which he wouldn't be. He's ineligible to be "elected" president, but he's literally eligible to be + Show Spoiler +
"President must be at least 35 years of age, be a natural born citizen, and must have lived in the United States for at least 14 years."
So 50:50 in the polls doesn't equate to a 50:50 chance. It only does now because "the fundamentals" in the model are more heavily weighted against polling and favor Biden. But those fundamentals are only going to become less weighted and if Biden's polling doesn't pick up (better than it has for 8 months) to counteract it, his odds will drop significantly even if his polling is marginally better.
I understand that Biden's strong economic and political fundamentals have been helping him in 538's early polling stages (I believe I've said about 10 times over the past few months that early polling is not necessarily a strong predictor of November outcomes, which is why I've been cautioning you and others to not take them as gospel). Introvert was the one who said that the next few months won't have fluctuations because the past few months have been steady. I didn't say that, and I don't think that one can safely make that causal claim.
I wouldn't be surprised at all if there's some volatility along the way to November (especially due to the debate), but I still predict that the November results will end up being a coinflip. And maybe I'm wrong. Maybe your prediction that Biden's polling numbers will consistently decline from now until November, as the weighing of fundamentals ease up in favor of other criteria, is correct. Maybe all three of us end up being wrong. We won't know until November, so I don't understand why you keep picking fights with me and others about these kinds of things.
I'm not saying (or conditioning) it not being a coinflip on Biden's polling getting worse, I'm saying that 538 wouldn't consider it a coinflip even if Biden was polling marginally better than he was before his debate disaster as we get closer to the election. He has to get higher in the polls than he's been in 8 months just to keep it near 50:50.
Biden's just in much more trouble than you appreciate currently.
Also you/Democrats/Biden supporters are advocating for a guy you iirc (and most of his 2020 voters) believe is aiding and abetting genocide, and could kill your mom in cold blood right in front of you and still not lose your vote or advocacy. I find that catastrophically problematic on several fronts, not the least of which, what it says of the ostensible "democracy" you're trying to "save" with that deplorable position in the first place.
You and I (and many others) have talked about this topic at great length. You disagree with our positions and explanations and rationalizations and justifications, and that's fine. I don't think I have anything else of substance to add to that particular conversation.
You asked why I keep challenging your narratives, so I gave you some of the reasons.
My hope from this exchange was that you'd at least recognize your error in assuming Biden polling about the same as he is now or slightly better when people start voting in September would net him 50:50 odds on election day. He has to do better for the next 4 months than he has for the last 8 months just to keep the odds near a coinflip. One of the big problems there being even his own staff says they can only get ~6 hours a day out of him and he's going to need to be sharp for far more than that to do the campaigning necessary just to salvage the debate, let alone actually take a lead. Based on letting/pushing him on the debate stage like that, he simply can't do what it will reasonably take to win.
Another aspect of Dem's problems is that replacing Biden is far from a magic bullet because they ignored the people saying Biden can't be the nominee for too long. So they're stuck unabashedly saying things like
[Biden] could have walked out, shit his pants, and then shuffled off and I'd still vote for him
like that isn't cult behavior
Question for you: If you were unilaterally in charge of the Democrats, and your single objective was to beat Trump in November, how would you play the (absolutely terrible) hand that you're currently being dealt? It's just about July now, you have less than 2 months before you formally appoint your best longshot candidate as the nominee to go up against the Republicans, and you have a little over 4 months total to magic your way into a victory on Election Day. However unlikely the win would be, how would GH approach things to maximize the Democrats' chances? Who's your presidential candidate? Who's your vice presidential candidate? What are the most important things they need to do, with the remaining time they have, and how do they do it?
Pretty much this:
If one had complete political control over everyone, then you could have Biden announce he steps back for medical reasons, on advice of his doctor and in consultation with his wife, and immediately have him endorse two new candidates. Have literally everyone immediately back them, Schiff, Pelosi, Obama, Newsom, AOC, Harris, Buttigieg, literally everyone.
And you'd have a good shot at beating Trump. Maybe even a better shot. Especially if that candidate can debate Trump again and crush him.
But I agree with Cona in that it aint happenin. Which is one of several reasons Democrats are in so much worse shape than you and Democrats/their supporters recognize currently.
But which two candidates would you put on the Democratic ticket?
I'd have been polling this question months ago, had a real primary, etc. and would never be in this hole to dig out of, but assuming I have to start from today I'd have to poll it now in this hypothetical. Also I'm not exactly clear if I get unanimity of "vote blue no matter who" support from Democrats regardless of whom I put on the ticket.
I don't have an exact duo to give you regardless, but some of the short list would be Barack Obama (the most popular/favorable person Democrats have), Michelle Obama (probably more popular, but less of a politician), Gavin Newsom, JB Pritzker, (generic white guys to assuage swing voting bigots) and I guess Whitmer specifically for MI (but don't think she could carry Dems nationally).
Jesus the pickins are slim among Democrats. Hard to appreciate just how badly Hillary and Joe's ambitions, hubris and presence in the party fucked everyone over for generations.
On Obama being VP that would go to SCOTUS, but possibly not in time for the election. It's basically whether you interpret the 12th amendment to include "being elected president" instead of the literal "No person constitutionally ineligible to the office of President shall be eligible to that of Vice-President of the United States," which he wouldn't be. He's ineligible to be "elected" president, but he's literally eligible to be + Show Spoiler +
"President must be at least 35 years of age, be a natural born citizen, and must have lived in the United States for at least 14 years."
the president.
Okay, cool. Thank you!
Might want to clarify if it's with the assumption that Democrats will uniformly fall in line behind whoever before you to try to apply that information to anything.
Because if they're not (and I don't think they would), I might be calling the CIA/FBI and asking if they have any magic bullets and good stories lying around, because that might be the only hope lol.
Biden had a real primary and he won with North Korean level percentages. If the people you are thinking about thought they could beat Biden in a primary, they would have entered. If it had been closer because more serious people had entered, there would have been a tv debate, Biden would have fumbled that one too, but Biden would still have won that primary. The only option was for Biden, knowing his age, to announce from the start that he would be a one term president. Put Trump behind bars. And have an open democratic field run vs DJ Vance, Nicki Haley or Ted Cruz or Rubio or MTG or whoever then would have been the GOP nomenee.
Asking SCOTUS to decide who should be your ticket vs Trump is quite ironic.
BTW, the 'vote blue no matter who' crowd would vote for your hypothetical ticket, if it had been the ticket. It is not about if people would support them vs Trump. The issue is if you should replace Biden right now with said hypothetical ticket.
It is more about you not supporting Biden vs Trump given that it already is Biden vs Trump because you want to go back in time.
On July 01 2024 01:14 Introvert wrote: And DBP, thr polls have been remarkably consistent all year. There's also less reason to think they will change because these are the two most well-known candidates in modern American history. They both have records people are evaluating, maybe with rose-tinted glasses but still. It's not like there's a magical date where polls become extra predictive, it's summer anyways and nothing has changed significantly in polling for 9 months.
My prediction has been that the election will end up being a coinflip, which would indeed be "remarkably consistent" with how the polls have shaped up so far, so there's a good chance that you and I end up being right! You seem to dismiss the idea that there may be any volatility between now and November though, such as Biden taking a dip due to his poor debate performance. I'm impressed by your confidence - that Biden's chances won't be affected and that he remains at basically a 50/50 chance with Trump - but I think it's more likely that we'll see at least a little movement in the polls (negative for Biden), especially over the next few weeks. Maybe his polling numbers will recover after that though. We'll have to see what happens in July and August, and then Biden's team will potentially need to deal with another problem around September (second debate), which could also lead to more volatility. But maybe you're correct that Biden's polling will end up being immune to negatives; that would certainly help to balance out the massive number of negative consequences that haven't affected Trump's polling!
GH has tried explaining this but Biden isn't tied, he's behind.
As of last week (before the debate), Biden was a negligible amount behind Trump. It was basically 50/50.
He has to win the popular vote by probably 2 points to be about tied.
That's false. Biden could lose the overall popular vote and still win the election. What matters is the outcome of a few key swing states. Whether he wins 70% or 51% of California is irrelevant for the election results, even though the former beefs up his overall popular vote.
On July 01 2024 02:21 GreenHorizons wrote:
On June 30 2024 22:12 DarkPlasmaBall wrote:
On June 30 2024 21:00 GreenHorizons wrote: [quote] Bernie was a compromise to start with. The potential of him being the nominee/president was enough to keep a sliver of electoralism alive for me (back in 2016).
"Coinflip" (~50:50 and down by ~1-2% nationally) sounds a lot better than it is when you realize Biden was a 9:1 favorite with a 8.5% lead (4:1 with a 9%+ lead at this point in the race) going into an election he won by the skin of his teeth against the same guy in 2020 while citizens were dying by the hundreds or thousands per day from an ongoing pandemic Trump had clearly mismanaged.
Mind you this is up against a 2x impeached, 34x felony convicted, insurrectionist (who wasn't any of those things at this point last time, though he had been impeached once)
Biden/Democrats are in so much more trouble than you/his supporters/Democrats are prepared/willing to accept. This hubris could cost Dems the most important election of their lives and it's clear to anyone that isn't obsequiously supporting Biden imo.
Maybe I'll end up being wrong about my coinflip prediction - maybe in September and October, we'll see the polls tilted 70-30 or worse, in favor of Trump. Maybe things spiral out of control over the next few months. And if that happens, I'll have no problem saying that my prediction was wrong. But let's at least agree that my coinflip prediction is not the same as your claim that I "think the next 4 months will be so much better for Biden than the last 8 months", okay? I didn't say that, and I don't believe that.
With Biden's terrible debate performance, his chances of winning are surely worse now than they were a week ago. I said that I expect Biden to dip in the polls, and that I wonder if he'll recover from it; I'd hardly call that "hubris".
That's the fun of giving probabilities for presidential elections, as long as you give something a chance, you can't really be wrong.
We're presumably using the polls to inform the probabilities. For example, it would be inconsistent, based on the data, if someone said the current chance of Biden winning is 30% if the data says it's 50%.
One issue I'm trying to point out in that post is that it took a 8%+ lead in polls for Biden to squeak by Trump by thousands of votes. Biden consistently polling a couple points behind Trump is much worse than you/Democrats/Biden supporters are grasping. If the next 4 (3 before voting starts) months don't go much better than the last, then Biden is toast.
You're drawing a causal relationship that is unfounded. It's true that Biden was leading Trump by 8% in November 2020, and it's true that Biden just barely won the election, but it's incorrect to say that a polling lead of 8% is required for him to win again (especially in a different election).
The reason it's ~50:50 on 538 isn't the gap in the polls, it's the distance from the election (and the "fundamentals" favoring Biden).
The reason it's 50/50 is because a lot of people were asked who they'd vote for, and about half said Biden and about half said Trump (particularly for the key swing states they were polled in). That's it. 50/50 now means 50/50 now. It doesn't guarantee 50/50 in November, it doesn't guarantee 40/60 in November, and it doesn't guarantee anything else based on "the distance from the election". This is the problem with using early polling to extrapolate too far, rather than seeing a poll as a snapshot of a specific moment in time.
If you think people will gradually change their minds away from Biden, over the next few months, due to reasons like his poor debate performance, then that's fine, but those changes will still appear in the polls. For all the stock you've been putting into the polling data over the past few months, you're very quick to assert that those numbers really mean completely different numbers.
If Biden was polling like this in November it wouldn't be 50:50. To actually be 50:50, Biden has to do a lot better in polling than he is now and has been for 8 months.
If about half the people polled in November said they'd vote for Biden, and half said they've vote for Trump (particularly for the key swing states they were polled in), then the polls would say roughly 50/50.
So if you think it's going to be a 50:50 election you also believe that the next 4 months will be significantly better for Biden than the last 8 whether you say it explicitly or not.
I guess we'll just have to agree to disagree on what I believe. lol.
"inconsistent" isn't wrong.
I'm not saying he needs an 8% lead, I'm saying he needs better than to be losing as he has been for 8 months.
You're just misunderstanding how 538 is reaching its "50:50" probability.
When we have few polls or when it's early in the campaign, our model's predictions are mostly based on the fundamentals — with their standard deviation usually around 6 points or so. But when it's Election Day in a state with a lot of polls, our uncertainty about public opinion is a lot smaller, so we will put a lot more weight on the polling data (and subsequently, our polling averages that feed into the model) when generating our final prediction.
So 50:50 in the polls doesn't equate to a 50:50 chance. It only does now because "the fundamentals" in the model are more heavily weighted against polling and favor Biden. But those fundamentals are only going to become less weighted and if Biden's polling doesn't pick up (better than it has for 8 months) to counteract it, his odds will drop significantly even if his polling is marginally better.
I understand that Biden's strong economic and political fundamentals have been helping him in 538's early polling stages (I believe I've said about 10 times over the past few months that early polling is not necessarily a strong predictor of November outcomes, which is why I've been cautioning you and others to not take them as gospel). Introvert was the one who said that the next few months won't have fluctuations because the past few months have been steady. I didn't say that, and I don't think that one can safely make that causal claim.
I wouldn't be surprised at all if there's some volatility along the way to November (especially due to the debate), but I still predict that the November results will end up being a coinflip. And maybe I'm wrong. Maybe your prediction that Biden's polling numbers will consistently decline from now until November, as the weighing of fundamentals ease up in favor of other criteria, is correct. Maybe all three of us end up being wrong. We won't know until November, so I don't understand why you keep picking fights with me and others about these kinds of things.
I'm not saying (or conditioning) it not being a coinflip on Biden's polling getting worse, I'm saying that 538 wouldn't consider it a coinflip even if Biden was polling marginally better than he was before his debate disaster as we get closer to the election. He has to get higher in the polls than he's been in 8 months just to keep it near 50:50.
Biden's just in much more trouble than you appreciate currently.
Also you/Democrats/Biden supporters are advocating for a guy you iirc (and most of his 2020 voters) believe is aiding and abetting genocide, and could kill your mom in cold blood right in front of you and still not lose your vote or advocacy. I find that catastrophically problematic on several fronts, not the least of which, what it says of the ostensible "democracy" you're trying to "save" with that deplorable position in the first place.
You and I (and many others) have talked about this topic at great length. You disagree with our positions and explanations and rationalizations and justifications, and that's fine. I don't think I have anything else of substance to add to that particular conversation.
On July 01 2024 01:14 Introvert wrote: And DBP, thr polls have been remarkably consistent all year. There's also less reason to think they will change because these are the two most well-known candidates in modern American history. They both have records people are evaluating, maybe with rose-tinted glasses but still. It's not like there's a magical date where polls become extra predictive, it's summer anyways and nothing has changed significantly in polling for 9 months.
My prediction has been that the election will end up being a coinflip, which would indeed be "remarkably consistent" with how the polls have shaped up so far, so there's a good chance that you and I end up being right! You seem to dismiss the idea that there may be any volatility between now and November though, such as Biden taking a dip due to his poor debate performance. I'm impressed by your confidence - that Biden's chances won't be affected and that he remains at basically a 50/50 chance with Trump - but I think it's more likely that we'll see at least a little movement in the polls (negative for Biden), especially over the next few weeks. Maybe his polling numbers will recover after that though. We'll have to see what happens in July and August, and then Biden's team will potentially need to deal with another problem around September (second debate), which could also lead to more volatility. But maybe you're correct that Biden's polling will end up being immune to negatives; that would certainly help to balance out the massive number of negative consequences that haven't affected Trump's polling!
GH has tried explaining this but Biden isn't tied, he's behind.
As of last week (before the debate), Biden was a negligible amount behind Trump. It was basically 50/50.
He has to win the popular vote by probably 2 points to be about tied.
That's false. Biden could lose the overall popular vote and still win the election. What matters is the outcome of a few key swing states. Whether he wins 70% or 51% of California is irrelevant for the election results, even though the former beefs up his overall popular vote.
On July 01 2024 02:21 GreenHorizons wrote:
On June 30 2024 22:12 DarkPlasmaBall wrote:
On June 30 2024 21:00 GreenHorizons wrote: [quote] Bernie was a compromise to start with. The potential of him being the nominee/president was enough to keep a sliver of electoralism alive for me (back in 2016).
"Coinflip" (~50:50 and down by ~1-2% nationally) sounds a lot better than it is when you realize Biden was a 9:1 favorite with a 8.5% lead (4:1 with a 9%+ lead at this point in the race) going into an election he won by the skin of his teeth against the same guy in 2020 while citizens were dying by the hundreds or thousands per day from an ongoing pandemic Trump had clearly mismanaged.
Mind you this is up against a 2x impeached, 34x felony convicted, insurrectionist (who wasn't any of those things at this point last time, though he had been impeached once)
Biden/Democrats are in so much more trouble than you/his supporters/Democrats are prepared/willing to accept. This hubris could cost Dems the most important election of their lives and it's clear to anyone that isn't obsequiously supporting Biden imo.
Maybe I'll end up being wrong about my coinflip prediction - maybe in September and October, we'll see the polls tilted 70-30 or worse, in favor of Trump. Maybe things spiral out of control over the next few months. And if that happens, I'll have no problem saying that my prediction was wrong. But let's at least agree that my coinflip prediction is not the same as your claim that I "think the next 4 months will be so much better for Biden than the last 8 months", okay? I didn't say that, and I don't believe that.
With Biden's terrible debate performance, his chances of winning are surely worse now than they were a week ago. I said that I expect Biden to dip in the polls, and that I wonder if he'll recover from it; I'd hardly call that "hubris".
That's the fun of giving probabilities for presidential elections, as long as you give something a chance, you can't really be wrong.
We're presumably using the polls to inform the probabilities. For example, it would be inconsistent, based on the data, if someone said the current chance of Biden winning is 30% if the data says it's 50%.
One issue I'm trying to point out in that post is that it took a 8%+ lead in polls for Biden to squeak by Trump by thousands of votes. Biden consistently polling a couple points behind Trump is much worse than you/Democrats/Biden supporters are grasping. If the next 4 (3 before voting starts) months don't go much better than the last, then Biden is toast.
You're drawing a causal relationship that is unfounded. It's true that Biden was leading Trump by 8% in November 2020, and it's true that Biden just barely won the election, but it's incorrect to say that a polling lead of 8% is required for him to win again (especially in a different election).
The reason it's ~50:50 on 538 isn't the gap in the polls, it's the distance from the election (and the "fundamentals" favoring Biden).
The reason it's 50/50 is because a lot of people were asked who they'd vote for, and about half said Biden and about half said Trump (particularly for the key swing states they were polled in). That's it. 50/50 now means 50/50 now. It doesn't guarantee 50/50 in November, it doesn't guarantee 40/60 in November, and it doesn't guarantee anything else based on "the distance from the election". This is the problem with using early polling to extrapolate too far, rather than seeing a poll as a snapshot of a specific moment in time.
If you think people will gradually change their minds away from Biden, over the next few months, due to reasons like his poor debate performance, then that's fine, but those changes will still appear in the polls. For all the stock you've been putting into the polling data over the past few months, you're very quick to assert that those numbers really mean completely different numbers.
If Biden was polling like this in November it wouldn't be 50:50. To actually be 50:50, Biden has to do a lot better in polling than he is now and has been for 8 months.
If about half the people polled in November said they'd vote for Biden, and half said they've vote for Trump (particularly for the key swing states they were polled in), then the polls would say roughly 50/50.
So if you think it's going to be a 50:50 election you also believe that the next 4 months will be significantly better for Biden than the last 8 whether you say it explicitly or not.
I guess we'll just have to agree to disagree on what I believe. lol.
"inconsistent" isn't wrong.
I'm not saying he needs an 8% lead, I'm saying he needs better than to be losing as he has been for 8 months.
You're just misunderstanding how 538 is reaching its "50:50" probability.
When we have few polls or when it's early in the campaign, our model's predictions are mostly based on the fundamentals — with their standard deviation usually around 6 points or so. But when it's Election Day in a state with a lot of polls, our uncertainty about public opinion is a lot smaller, so we will put a lot more weight on the polling data (and subsequently, our polling averages that feed into the model) when generating our final prediction.
So 50:50 in the polls doesn't equate to a 50:50 chance. It only does now because "the fundamentals" in the model are more heavily weighted against polling and favor Biden. But those fundamentals are only going to become less weighted and if Biden's polling doesn't pick up (better than it has for 8 months) to counteract it, his odds will drop significantly even if his polling is marginally better.
I understand that Biden's strong economic and political fundamentals have been helping him in 538's early polling stages (I believe I've said about 10 times over the past few months that early polling is not necessarily a strong predictor of November outcomes, which is why I've been cautioning you and others to not take them as gospel). Introvert was the one who said that the next few months won't have fluctuations because the past few months have been steady. I didn't say that, and I don't think that one can safely make that causal claim.
I wouldn't be surprised at all if there's some volatility along the way to November (especially due to the debate), but I still predict that the November results will end up being a coinflip. And maybe I'm wrong. Maybe your prediction that Biden's polling numbers will consistently decline from now until November, as the weighing of fundamentals ease up in favor of other criteria, is correct. Maybe all three of us end up being wrong. We won't know until November, so I don't understand why you keep picking fights with me and others about these kinds of things.
The thing about all this is that you are the one hoping, against any evidence currently avaliable, that things are going to grt better for Biden when there's a good chance they get even worse instead. I'm looking at the history of how these things work. BTW, 538 is the only (or one of the only) model(s) that has Biden as good as a coinflip. Your assertion that things solidify later is true, but you asking it to do far too much work. Polls even now are generally predictive.
As to your point about the swing states... for all your focus on polling and models I'm surprised you don't know or understand the electoral college disadvantage dems have had and been complaining about for years. If Biden is even nationally he is almost certainly DOWN in the swing states, and indeed state level polling shows this. For decades the swing states vote right of the nation as a whole, and moreover states with similar demographics in the same region are correlated. If black turnout is down and rural turnout is better for Trump in WI it almost certainly is also true in MI, for example. The "Biden only has to win the swing states" is a problem for Biden, not a thing in his favor.
I don't know what will happen, but for months now you have been trying to pretend that he's in better shape, or that the state of the race is more ambigious than it actually is. Again, people can't handle that dems are actually losing.
I know that Democrats tend to win the popular vote, even if they lose the election. If the polling has him as 50/50 even chance to win the election, then that's talking about the electoral college, not the popular vote. The polls run scenarios for winning the election and individual state outcomes. If a poll announced something like "we asked a million American voters and 500,000 said they'll vote for Biden", that wouldn't be helpful at all without knowing where each of those 1 million voters live, but that's not what's happening here.
I am trying to figure out if I am misunderstanding whet you are saying here, because it seems wrong, and obviously so.
That's not what polls do, that's what modelers do, with polls as one input. Polls alone currently have Biden down slightly in the PV and down more in the swing states. Your one million person poll is precisely what they do...they weight their responses based on demographics and vote history but no poll is a poll of who is going to win the election, that doesn't even make sense. A national poll showing Trump 45% Biden 43% among all voters is literally a weighted tally of how people are going to vote, not Biden's chances of winning the election.
Ah, I see what you're referring to. I was referring to the aggregate of the state polls - that's why I wrote "(particularly for the key swing states they were polled in)" twice in my earlier post, and I'm happy to call that a "model", as you're saying it's what "modelers" do. That's totally fair. It's definitely easier to use that different term than referring to every survey and meta-analysis of surveys as falling under the umbrella phrasing of what the polls say, so that we avoid semantics issues.
I think part of the confusion is that the polling aggregates have Trump with a PV and swing state lead too, there's almost no polling that says the race is a tie, only by modeling can get to a tie. And of the three major modelers (538, Nate Silver, and The Economist) the latter two have Trump at ~60% to win. You can hang your hat on 538 I guess but you can't expect everyone to go along with the more rosy outlook you are trying to will into existence. Calling it "meta-analysis" gives it a scientific rigor that imo it does not deserve. It's not a meta-analysis of polls anyways, it's polls AND "fundamentals." "The polls" are pretty clear, espeically when you look at the bias they had in 2020.
I'm sorry to everyone else I know this conversation seems pointless lol
On July 01 2024 01:14 Introvert wrote: And DBP, thr polls have been remarkably consistent all year. There's also less reason to think they will change because these are the two most well-known candidates in modern American history. They both have records people are evaluating, maybe with rose-tinted glasses but still. It's not like there's a magical date where polls become extra predictive, it's summer anyways and nothing has changed significantly in polling for 9 months.
My prediction has been that the election will end up being a coinflip, which would indeed be "remarkably consistent" with how the polls have shaped up so far, so there's a good chance that you and I end up being right! You seem to dismiss the idea that there may be any volatility between now and November though, such as Biden taking a dip due to his poor debate performance. I'm impressed by your confidence - that Biden's chances won't be affected and that he remains at basically a 50/50 chance with Trump - but I think it's more likely that we'll see at least a little movement in the polls (negative for Biden), especially over the next few weeks. Maybe his polling numbers will recover after that though. We'll have to see what happens in July and August, and then Biden's team will potentially need to deal with another problem around September (second debate), which could also lead to more volatility. But maybe you're correct that Biden's polling will end up being immune to negatives; that would certainly help to balance out the massive number of negative consequences that haven't affected Trump's polling!
GH has tried explaining this but Biden isn't tied, he's behind.
As of last week (before the debate), Biden was a negligible amount behind Trump. It was basically 50/50.
He has to win the popular vote by probably 2 points to be about tied.
That's false. Biden could lose the overall popular vote and still win the election. What matters is the outcome of a few key swing states. Whether he wins 70% or 51% of California is irrelevant for the election results, even though the former beefs up his overall popular vote.
On July 01 2024 02:21 GreenHorizons wrote:
On June 30 2024 22:12 DarkPlasmaBall wrote: [quote]
Maybe I'll end up being wrong about my coinflip prediction - maybe in September and October, we'll see the polls tilted 70-30 or worse, in favor of Trump. Maybe things spiral out of control over the next few months. And if that happens, I'll have no problem saying that my prediction was wrong. But let's at least agree that my coinflip prediction is not the same as your claim that I "think the next 4 months will be so much better for Biden than the last 8 months", okay? I didn't say that, and I don't believe that.
With Biden's terrible debate performance, his chances of winning are surely worse now than they were a week ago. I said that I expect Biden to dip in the polls, and that I wonder if he'll recover from it; I'd hardly call that "hubris".
That's the fun of giving probabilities for presidential elections, as long as you give something a chance, you can't really be wrong.
We're presumably using the polls to inform the probabilities. For example, it would be inconsistent, based on the data, if someone said the current chance of Biden winning is 30% if the data says it's 50%.
One issue I'm trying to point out in that post is that it took a 8%+ lead in polls for Biden to squeak by Trump by thousands of votes. Biden consistently polling a couple points behind Trump is much worse than you/Democrats/Biden supporters are grasping. If the next 4 (3 before voting starts) months don't go much better than the last, then Biden is toast.
You're drawing a causal relationship that is unfounded. It's true that Biden was leading Trump by 8% in November 2020, and it's true that Biden just barely won the election, but it's incorrect to say that a polling lead of 8% is required for him to win again (especially in a different election).
The reason it's ~50:50 on 538 isn't the gap in the polls, it's the distance from the election (and the "fundamentals" favoring Biden).
The reason it's 50/50 is because a lot of people were asked who they'd vote for, and about half said Biden and about half said Trump (particularly for the key swing states they were polled in). That's it. 50/50 now means 50/50 now. It doesn't guarantee 50/50 in November, it doesn't guarantee 40/60 in November, and it doesn't guarantee anything else based on "the distance from the election". This is the problem with using early polling to extrapolate too far, rather than seeing a poll as a snapshot of a specific moment in time.
If you think people will gradually change their minds away from Biden, over the next few months, due to reasons like his poor debate performance, then that's fine, but those changes will still appear in the polls. For all the stock you've been putting into the polling data over the past few months, you're very quick to assert that those numbers really mean completely different numbers.
If Biden was polling like this in November it wouldn't be 50:50. To actually be 50:50, Biden has to do a lot better in polling than he is now and has been for 8 months.
If about half the people polled in November said they'd vote for Biden, and half said they've vote for Trump (particularly for the key swing states they were polled in), then the polls would say roughly 50/50.
So if you think it's going to be a 50:50 election you also believe that the next 4 months will be significantly better for Biden than the last 8 whether you say it explicitly or not.
I guess we'll just have to agree to disagree on what I believe. lol.
"inconsistent" isn't wrong.
I'm not saying he needs an 8% lead, I'm saying he needs better than to be losing as he has been for 8 months.
You're just misunderstanding how 538 is reaching its "50:50" probability.
When we have few polls or when it's early in the campaign, our model's predictions are mostly based on the fundamentals — with their standard deviation usually around 6 points or so. But when it's Election Day in a state with a lot of polls, our uncertainty about public opinion is a lot smaller, so we will put a lot more weight on the polling data (and subsequently, our polling averages that feed into the model) when generating our final prediction.
So 50:50 in the polls doesn't equate to a 50:50 chance. It only does now because "the fundamentals" in the model are more heavily weighted against polling and favor Biden. But those fundamentals are only going to become less weighted and if Biden's polling doesn't pick up (better than it has for 8 months) to counteract it, his odds will drop significantly even if his polling is marginally better.
I understand that Biden's strong economic and political fundamentals have been helping him in 538's early polling stages (I believe I've said about 10 times over the past few months that early polling is not necessarily a strong predictor of November outcomes, which is why I've been cautioning you and others to not take them as gospel). Introvert was the one who said that the next few months won't have fluctuations because the past few months have been steady. I didn't say that, and I don't think that one can safely make that causal claim.
I wouldn't be surprised at all if there's some volatility along the way to November (especially due to the debate), but I still predict that the November results will end up being a coinflip. And maybe I'm wrong. Maybe your prediction that Biden's polling numbers will consistently decline from now until November, as the weighing of fundamentals ease up in favor of other criteria, is correct. Maybe all three of us end up being wrong. We won't know until November, so I don't understand why you keep picking fights with me and others about these kinds of things.
I'm not saying (or conditioning) it not being a coinflip on Biden's polling getting worse, I'm saying that 538 wouldn't consider it a coinflip even if Biden was polling marginally better than he was before his debate disaster as we get closer to the election. He has to get higher in the polls than he's been in 8 months just to keep it near 50:50.
Biden's just in much more trouble than you appreciate currently.
Also you/Democrats/Biden supporters are advocating for a guy you iirc (and most of his 2020 voters) believe is aiding and abetting genocide, and could kill your mom in cold blood right in front of you and still not lose your vote or advocacy. I find that catastrophically problematic on several fronts, not the least of which, what it says of the ostensible "democracy" you're trying to "save" with that deplorable position in the first place.
You and I (and many others) have talked about this topic at great length. You disagree with our positions and explanations and rationalizations and justifications, and that's fine. I don't think I have anything else of substance to add to that particular conversation.
On July 01 2024 01:14 Introvert wrote: And DBP, thr polls have been remarkably consistent all year. There's also less reason to think they will change because these are the two most well-known candidates in modern American history. They both have records people are evaluating, maybe with rose-tinted glasses but still. It's not like there's a magical date where polls become extra predictive, it's summer anyways and nothing has changed significantly in polling for 9 months.
My prediction has been that the election will end up being a coinflip, which would indeed be "remarkably consistent" with how the polls have shaped up so far, so there's a good chance that you and I end up being right! You seem to dismiss the idea that there may be any volatility between now and November though, such as Biden taking a dip due to his poor debate performance. I'm impressed by your confidence - that Biden's chances won't be affected and that he remains at basically a 50/50 chance with Trump - but I think it's more likely that we'll see at least a little movement in the polls (negative for Biden), especially over the next few weeks. Maybe his polling numbers will recover after that though. We'll have to see what happens in July and August, and then Biden's team will potentially need to deal with another problem around September (second debate), which could also lead to more volatility. But maybe you're correct that Biden's polling will end up being immune to negatives; that would certainly help to balance out the massive number of negative consequences that haven't affected Trump's polling!
GH has tried explaining this but Biden isn't tied, he's behind.
As of last week (before the debate), Biden was a negligible amount behind Trump. It was basically 50/50.
He has to win the popular vote by probably 2 points to be about tied.
That's false. Biden could lose the overall popular vote and still win the election. What matters is the outcome of a few key swing states. Whether he wins 70% or 51% of California is irrelevant for the election results, even though the former beefs up his overall popular vote.
On July 01 2024 02:21 GreenHorizons wrote:
On June 30 2024 22:12 DarkPlasmaBall wrote: [quote]
Maybe I'll end up being wrong about my coinflip prediction - maybe in September and October, we'll see the polls tilted 70-30 or worse, in favor of Trump. Maybe things spiral out of control over the next few months. And if that happens, I'll have no problem saying that my prediction was wrong. But let's at least agree that my coinflip prediction is not the same as your claim that I "think the next 4 months will be so much better for Biden than the last 8 months", okay? I didn't say that, and I don't believe that.
With Biden's terrible debate performance, his chances of winning are surely worse now than they were a week ago. I said that I expect Biden to dip in the polls, and that I wonder if he'll recover from it; I'd hardly call that "hubris".
That's the fun of giving probabilities for presidential elections, as long as you give something a chance, you can't really be wrong.
We're presumably using the polls to inform the probabilities. For example, it would be inconsistent, based on the data, if someone said the current chance of Biden winning is 30% if the data says it's 50%.
One issue I'm trying to point out in that post is that it took a 8%+ lead in polls for Biden to squeak by Trump by thousands of votes. Biden consistently polling a couple points behind Trump is much worse than you/Democrats/Biden supporters are grasping. If the next 4 (3 before voting starts) months don't go much better than the last, then Biden is toast.
You're drawing a causal relationship that is unfounded. It's true that Biden was leading Trump by 8% in November 2020, and it's true that Biden just barely won the election, but it's incorrect to say that a polling lead of 8% is required for him to win again (especially in a different election).
The reason it's ~50:50 on 538 isn't the gap in the polls, it's the distance from the election (and the "fundamentals" favoring Biden).
The reason it's 50/50 is because a lot of people were asked who they'd vote for, and about half said Biden and about half said Trump (particularly for the key swing states they were polled in). That's it. 50/50 now means 50/50 now. It doesn't guarantee 50/50 in November, it doesn't guarantee 40/60 in November, and it doesn't guarantee anything else based on "the distance from the election". This is the problem with using early polling to extrapolate too far, rather than seeing a poll as a snapshot of a specific moment in time.
If you think people will gradually change their minds away from Biden, over the next few months, due to reasons like his poor debate performance, then that's fine, but those changes will still appear in the polls. For all the stock you've been putting into the polling data over the past few months, you're very quick to assert that those numbers really mean completely different numbers.
If Biden was polling like this in November it wouldn't be 50:50. To actually be 50:50, Biden has to do a lot better in polling than he is now and has been for 8 months.
If about half the people polled in November said they'd vote for Biden, and half said they've vote for Trump (particularly for the key swing states they were polled in), then the polls would say roughly 50/50.
So if you think it's going to be a 50:50 election you also believe that the next 4 months will be significantly better for Biden than the last 8 whether you say it explicitly or not.
I guess we'll just have to agree to disagree on what I believe. lol.
"inconsistent" isn't wrong.
I'm not saying he needs an 8% lead, I'm saying he needs better than to be losing as he has been for 8 months.
You're just misunderstanding how 538 is reaching its "50:50" probability.
When we have few polls or when it's early in the campaign, our model's predictions are mostly based on the fundamentals — with their standard deviation usually around 6 points or so. But when it's Election Day in a state with a lot of polls, our uncertainty about public opinion is a lot smaller, so we will put a lot more weight on the polling data (and subsequently, our polling averages that feed into the model) when generating our final prediction.
So 50:50 in the polls doesn't equate to a 50:50 chance. It only does now because "the fundamentals" in the model are more heavily weighted against polling and favor Biden. But those fundamentals are only going to become less weighted and if Biden's polling doesn't pick up (better than it has for 8 months) to counteract it, his odds will drop significantly even if his polling is marginally better.
I understand that Biden's strong economic and political fundamentals have been helping him in 538's early polling stages (I believe I've said about 10 times over the past few months that early polling is not necessarily a strong predictor of November outcomes, which is why I've been cautioning you and others to not take them as gospel). Introvert was the one who said that the next few months won't have fluctuations because the past few months have been steady. I didn't say that, and I don't think that one can safely make that causal claim.
I wouldn't be surprised at all if there's some volatility along the way to November (especially due to the debate), but I still predict that the November results will end up being a coinflip. And maybe I'm wrong. Maybe your prediction that Biden's polling numbers will consistently decline from now until November, as the weighing of fundamentals ease up in favor of other criteria, is correct. Maybe all three of us end up being wrong. We won't know until November, so I don't understand why you keep picking fights with me and others about these kinds of things.
The thing about all this is that you are the one hoping, against any evidence currently avaliable, that things are going to grt better for Biden when there's a good chance they get even worse instead. I'm looking at the history of how these things work. BTW, 538 is the only (or one of the only) model(s) that has Biden as good as a coinflip. Your assertion that things solidify later is true, but you asking it to do far too much work. Polls even now are generally predictive.
As to your point about the swing states... for all your focus on polling and models I'm surprised you don't know or understand the electoral college disadvantage dems have had and been complaining about for years. If Biden is even nationally he is almost certainly DOWN in the swing states, and indeed state level polling shows this. For decades the swing states vote right of the nation as a whole, and moreover states with similar demographics in the same region are correlated. If black turnout is down and rural turnout is better for Trump in WI it almost certainly is also true in MI, for example. The "Biden only has to win the swing states" is a problem for Biden, not a thing in his favor.
I don't know what will happen, but for months now you have been trying to pretend that he's in better shape, or that the state of the race is more ambigious than it actually is. Again, people can't handle that dems are actually losing.
I know that Democrats tend to win the popular vote, even if they lose the election. If the polling has him as 50/50 even chance to win the election, then that's talking about the electoral college, not the popular vote. The polls run scenarios for winning the election and individual state outcomes. If a poll announced something like "we asked a million American voters and 500,000 said they'll vote for Biden", that wouldn't be helpful at all without knowing where each of those 1 million voters live, but that's not what's happening here.
I am trying to figure out if I am misunderstanding whet you are saying here, because it seems wrong, and obviously so.
That's not what polls do, that's what modelers do, with polls as one input. Polls alone currently have Biden down slightly in the PV and down more in the swing states. Your one million person poll is precisely what they do...they weight their responses based on demographics and vote history but no poll is a poll of who is going to win the election, that doesn't even make sense. A national poll showing Trump 45% Biden 43% among all voters is literally a weighted tally of how people are going to vote, not Biden's chances of winning the election.
Ah, I see what you're referring to. I was referring to the aggregate of the state polls - that's why I wrote "(particularly for the key swing states they were polled in)" twice in my earlier post, and I'm happy to call that a "model", as you're saying it's what "modelers" do. That's totally fair. It's definitely easier to use that different term than referring to every survey and meta-analysis of surveys as falling under the umbrella phrasing of what the polls say, so that we avoid semantics issues.
I think part of the confusion is that the polling aggregates have Trump with a PV and swing state lead too, there's almost no polling that says the race is a tie, only by modeling can get to a tie. And of the three major modelers (538, Nate Silver, and The Economist) the latter two have Trump at ~60% to win. You can hang your hat on 538 I guess but you can't expect everyone to go along with the more rosy outlook you are trying to will into existence. Calling it "meta-analysis" gives it a scientific rigor that imo it does not deserve. It's not a meta-analysis of polls anyways, it's polls AND "fundamentals." "The polls" are pretty clear, espeically when you look at the bias they had in 2020.
I'm sorry to everyone else I know this conversation seems pointless lol
I appreciate the elaboration/conversation, however pointless it may be!
On July 01 2024 01:14 Introvert wrote: And DBP, thr polls have been remarkably consistent all year. There's also less reason to think they will change because these are the two most well-known candidates in modern American history. They both have records people are evaluating, maybe with rose-tinted glasses but still. It's not like there's a magical date where polls become extra predictive, it's summer anyways and nothing has changed significantly in polling for 9 months.
My prediction has been that the election will end up being a coinflip, which would indeed be "remarkably consistent" with how the polls have shaped up so far, so there's a good chance that you and I end up being right! You seem to dismiss the idea that there may be any volatility between now and November though, such as Biden taking a dip due to his poor debate performance. I'm impressed by your confidence - that Biden's chances won't be affected and that he remains at basically a 50/50 chance with Trump - but I think it's more likely that we'll see at least a little movement in the polls (negative for Biden), especially over the next few weeks. Maybe his polling numbers will recover after that though. We'll have to see what happens in July and August, and then Biden's team will potentially need to deal with another problem around September (second debate), which could also lead to more volatility. But maybe you're correct that Biden's polling will end up being immune to negatives; that would certainly help to balance out the massive number of negative consequences that haven't affected Trump's polling!
GH has tried explaining this but Biden isn't tied, he's behind.
As of last week (before the debate), Biden was a negligible amount behind Trump. It was basically 50/50.
He has to win the popular vote by probably 2 points to be about tied.
That's false. Biden could lose the overall popular vote and still win the election. What matters is the outcome of a few key swing states. Whether he wins 70% or 51% of California is irrelevant for the election results, even though the former beefs up his overall popular vote.
On July 01 2024 02:21 GreenHorizons wrote:
On June 30 2024 22:12 DarkPlasmaBall wrote: [quote]
Maybe I'll end up being wrong about my coinflip prediction - maybe in September and October, we'll see the polls tilted 70-30 or worse, in favor of Trump. Maybe things spiral out of control over the next few months. And if that happens, I'll have no problem saying that my prediction was wrong. But let's at least agree that my coinflip prediction is not the same as your claim that I "think the next 4 months will be so much better for Biden than the last 8 months", okay? I didn't say that, and I don't believe that.
With Biden's terrible debate performance, his chances of winning are surely worse now than they were a week ago. I said that I expect Biden to dip in the polls, and that I wonder if he'll recover from it; I'd hardly call that "hubris".
That's the fun of giving probabilities for presidential elections, as long as you give something a chance, you can't really be wrong.
We're presumably using the polls to inform the probabilities. For example, it would be inconsistent, based on the data, if someone said the current chance of Biden winning is 30% if the data says it's 50%.
One issue I'm trying to point out in that post is that it took a 8%+ lead in polls for Biden to squeak by Trump by thousands of votes. Biden consistently polling a couple points behind Trump is much worse than you/Democrats/Biden supporters are grasping. If the next 4 (3 before voting starts) months don't go much better than the last, then Biden is toast.
You're drawing a causal relationship that is unfounded. It's true that Biden was leading Trump by 8% in November 2020, and it's true that Biden just barely won the election, but it's incorrect to say that a polling lead of 8% is required for him to win again (especially in a different election).
The reason it's ~50:50 on 538 isn't the gap in the polls, it's the distance from the election (and the "fundamentals" favoring Biden).
The reason it's 50/50 is because a lot of people were asked who they'd vote for, and about half said Biden and about half said Trump (particularly for the key swing states they were polled in). That's it. 50/50 now means 50/50 now. It doesn't guarantee 50/50 in November, it doesn't guarantee 40/60 in November, and it doesn't guarantee anything else based on "the distance from the election". This is the problem with using early polling to extrapolate too far, rather than seeing a poll as a snapshot of a specific moment in time.
If you think people will gradually change their minds away from Biden, over the next few months, due to reasons like his poor debate performance, then that's fine, but those changes will still appear in the polls. For all the stock you've been putting into the polling data over the past few months, you're very quick to assert that those numbers really mean completely different numbers.
If Biden was polling like this in November it wouldn't be 50:50. To actually be 50:50, Biden has to do a lot better in polling than he is now and has been for 8 months.
If about half the people polled in November said they'd vote for Biden, and half said they've vote for Trump (particularly for the key swing states they were polled in), then the polls would say roughly 50/50.
So if you think it's going to be a 50:50 election you also believe that the next 4 months will be significantly better for Biden than the last 8 whether you say it explicitly or not.
I guess we'll just have to agree to disagree on what I believe. lol.
"inconsistent" isn't wrong.
I'm not saying he needs an 8% lead, I'm saying he needs better than to be losing as he has been for 8 months.
You're just misunderstanding how 538 is reaching its "50:50" probability.
When we have few polls or when it's early in the campaign, our model's predictions are mostly based on the fundamentals — with their standard deviation usually around 6 points or so. But when it's Election Day in a state with a lot of polls, our uncertainty about public opinion is a lot smaller, so we will put a lot more weight on the polling data (and subsequently, our polling averages that feed into the model) when generating our final prediction.
So 50:50 in the polls doesn't equate to a 50:50 chance. It only does now because "the fundamentals" in the model are more heavily weighted against polling and favor Biden. But those fundamentals are only going to become less weighted and if Biden's polling doesn't pick up (better than it has for 8 months) to counteract it, his odds will drop significantly even if his polling is marginally better.
I understand that Biden's strong economic and political fundamentals have been helping him in 538's early polling stages (I believe I've said about 10 times over the past few months that early polling is not necessarily a strong predictor of November outcomes, which is why I've been cautioning you and others to not take them as gospel). Introvert was the one who said that the next few months won't have fluctuations because the past few months have been steady. I didn't say that, and I don't think that one can safely make that causal claim.
I wouldn't be surprised at all if there's some volatility along the way to November (especially due to the debate), but I still predict that the November results will end up being a coinflip. And maybe I'm wrong. Maybe your prediction that Biden's polling numbers will consistently decline from now until November, as the weighing of fundamentals ease up in favor of other criteria, is correct. Maybe all three of us end up being wrong. We won't know until November, so I don't understand why you keep picking fights with me and others about these kinds of things.
I'm not saying (or conditioning) it not being a coinflip on Biden's polling getting worse, I'm saying that 538 wouldn't consider it a coinflip even if Biden was polling marginally better than he was before his debate disaster as we get closer to the election. He has to get higher in the polls than he's been in 8 months just to keep it near 50:50.
Biden's just in much more trouble than you appreciate currently.
Also you/Democrats/Biden supporters are advocating for a guy you iirc (and most of his 2020 voters) believe is aiding and abetting genocide, and could kill your mom in cold blood right in front of you and still not lose your vote or advocacy. I find that catastrophically problematic on several fronts, not the least of which, what it says of the ostensible "democracy" you're trying to "save" with that deplorable position in the first place.
You and I (and many others) have talked about this topic at great length. You disagree with our positions and explanations and rationalizations and justifications, and that's fine. I don't think I have anything else of substance to add to that particular conversation.
On July 01 2024 01:14 Introvert wrote: And DBP, thr polls have been remarkably consistent all year. There's also less reason to think they will change because these are the two most well-known candidates in modern American history. They both have records people are evaluating, maybe with rose-tinted glasses but still. It's not like there's a magical date where polls become extra predictive, it's summer anyways and nothing has changed significantly in polling for 9 months.
My prediction has been that the election will end up being a coinflip, which would indeed be "remarkably consistent" with how the polls have shaped up so far, so there's a good chance that you and I end up being right! You seem to dismiss the idea that there may be any volatility between now and November though, such as Biden taking a dip due to his poor debate performance. I'm impressed by your confidence - that Biden's chances won't be affected and that he remains at basically a 50/50 chance with Trump - but I think it's more likely that we'll see at least a little movement in the polls (negative for Biden), especially over the next few weeks. Maybe his polling numbers will recover after that though. We'll have to see what happens in July and August, and then Biden's team will potentially need to deal with another problem around September (second debate), which could also lead to more volatility. But maybe you're correct that Biden's polling will end up being immune to negatives; that would certainly help to balance out the massive number of negative consequences that haven't affected Trump's polling!
GH has tried explaining this but Biden isn't tied, he's behind.
As of last week (before the debate), Biden was a negligible amount behind Trump. It was basically 50/50.
He has to win the popular vote by probably 2 points to be about tied.
That's false. Biden could lose the overall popular vote and still win the election. What matters is the outcome of a few key swing states. Whether he wins 70% or 51% of California is irrelevant for the election results, even though the former beefs up his overall popular vote.
On July 01 2024 02:21 GreenHorizons wrote:
On June 30 2024 22:12 DarkPlasmaBall wrote: [quote]
Maybe I'll end up being wrong about my coinflip prediction - maybe in September and October, we'll see the polls tilted 70-30 or worse, in favor of Trump. Maybe things spiral out of control over the next few months. And if that happens, I'll have no problem saying that my prediction was wrong. But let's at least agree that my coinflip prediction is not the same as your claim that I "think the next 4 months will be so much better for Biden than the last 8 months", okay? I didn't say that, and I don't believe that.
With Biden's terrible debate performance, his chances of winning are surely worse now than they were a week ago. I said that I expect Biden to dip in the polls, and that I wonder if he'll recover from it; I'd hardly call that "hubris".
That's the fun of giving probabilities for presidential elections, as long as you give something a chance, you can't really be wrong.
We're presumably using the polls to inform the probabilities. For example, it would be inconsistent, based on the data, if someone said the current chance of Biden winning is 30% if the data says it's 50%.
One issue I'm trying to point out in that post is that it took a 8%+ lead in polls for Biden to squeak by Trump by thousands of votes. Biden consistently polling a couple points behind Trump is much worse than you/Democrats/Biden supporters are grasping. If the next 4 (3 before voting starts) months don't go much better than the last, then Biden is toast.
You're drawing a causal relationship that is unfounded. It's true that Biden was leading Trump by 8% in November 2020, and it's true that Biden just barely won the election, but it's incorrect to say that a polling lead of 8% is required for him to win again (especially in a different election).
The reason it's ~50:50 on 538 isn't the gap in the polls, it's the distance from the election (and the "fundamentals" favoring Biden).
The reason it's 50/50 is because a lot of people were asked who they'd vote for, and about half said Biden and about half said Trump (particularly for the key swing states they were polled in). That's it. 50/50 now means 50/50 now. It doesn't guarantee 50/50 in November, it doesn't guarantee 40/60 in November, and it doesn't guarantee anything else based on "the distance from the election". This is the problem with using early polling to extrapolate too far, rather than seeing a poll as a snapshot of a specific moment in time.
If you think people will gradually change their minds away from Biden, over the next few months, due to reasons like his poor debate performance, then that's fine, but those changes will still appear in the polls. For all the stock you've been putting into the polling data over the past few months, you're very quick to assert that those numbers really mean completely different numbers.
If Biden was polling like this in November it wouldn't be 50:50. To actually be 50:50, Biden has to do a lot better in polling than he is now and has been for 8 months.
If about half the people polled in November said they'd vote for Biden, and half said they've vote for Trump (particularly for the key swing states they were polled in), then the polls would say roughly 50/50.
So if you think it's going to be a 50:50 election you also believe that the next 4 months will be significantly better for Biden than the last 8 whether you say it explicitly or not.
I guess we'll just have to agree to disagree on what I believe. lol.
"inconsistent" isn't wrong.
I'm not saying he needs an 8% lead, I'm saying he needs better than to be losing as he has been for 8 months.
You're just misunderstanding how 538 is reaching its "50:50" probability.
When we have few polls or when it's early in the campaign, our model's predictions are mostly based on the fundamentals — with their standard deviation usually around 6 points or so. But when it's Election Day in a state with a lot of polls, our uncertainty about public opinion is a lot smaller, so we will put a lot more weight on the polling data (and subsequently, our polling averages that feed into the model) when generating our final prediction.
So 50:50 in the polls doesn't equate to a 50:50 chance. It only does now because "the fundamentals" in the model are more heavily weighted against polling and favor Biden. But those fundamentals are only going to become less weighted and if Biden's polling doesn't pick up (better than it has for 8 months) to counteract it, his odds will drop significantly even if his polling is marginally better.
I understand that Biden's strong economic and political fundamentals have been helping him in 538's early polling stages (I believe I've said about 10 times over the past few months that early polling is not necessarily a strong predictor of November outcomes, which is why I've been cautioning you and others to not take them as gospel). Introvert was the one who said that the next few months won't have fluctuations because the past few months have been steady. I didn't say that, and I don't think that one can safely make that causal claim.
I wouldn't be surprised at all if there's some volatility along the way to November (especially due to the debate), but I still predict that the November results will end up being a coinflip. And maybe I'm wrong. Maybe your prediction that Biden's polling numbers will consistently decline from now until November, as the weighing of fundamentals ease up in favor of other criteria, is correct. Maybe all three of us end up being wrong. We won't know until November, so I don't understand why you keep picking fights with me and others about these kinds of things.
The thing about all this is that you are the one hoping, against any evidence currently avaliable, that things are going to grt better for Biden when there's a good chance they get even worse instead. I'm looking at the history of how these things work. BTW, 538 is the only (or one of the only) model(s) that has Biden as good as a coinflip. Your assertion that things solidify later is true, but you asking it to do far too much work. Polls even now are generally predictive.
As to your point about the swing states... for all your focus on polling and models I'm surprised you don't know or understand the electoral college disadvantage dems have had and been complaining about for years. If Biden is even nationally he is almost certainly DOWN in the swing states, and indeed state level polling shows this. For decades the swing states vote right of the nation as a whole, and moreover states with similar demographics in the same region are correlated. If black turnout is down and rural turnout is better for Trump in WI it almost certainly is also true in MI, for example. The "Biden only has to win the swing states" is a problem for Biden, not a thing in his favor.
I don't know what will happen, but for months now you have been trying to pretend that he's in better shape, or that the state of the race is more ambigious than it actually is. Again, people can't handle that dems are actually losing.
I know that Democrats tend to win the popular vote, even if they lose the election. If the polling has him as 50/50 even chance to win the election, then that's talking about the electoral college, not the popular vote. The polls run scenarios for winning the election and individual state outcomes. If a poll announced something like "we asked a million American voters and 500,000 said they'll vote for Biden", that wouldn't be helpful at all without knowing where each of those 1 million voters live, but that's not what's happening here.
I am trying to figure out if I am misunderstanding whet you are saying here, because it seems wrong, and obviously so.
That's not what polls do, that's what modelers do, with polls as one input. Polls alone currently have Biden down slightly in the PV and down more in the swing states. Your one million person poll is precisely what they do...they weight their responses based on demographics and vote history but no poll is a poll of who is going to win the election, that doesn't even make sense. A national poll showing Trump 45% Biden 43% among all voters is literally a weighted tally of how people are going to vote, not Biden's chances of winning the election.
Ah, I see what you're referring to. I was referring to the aggregate of the state polls - that's why I wrote "(particularly for the key swing states they were polled in)" twice in my earlier post, and I'm happy to call that a "model", as you're saying it's what "modelers" do. That's totally fair. It's definitely easier to use that different term than referring to every survey and meta-analysis of surveys as falling under the umbrella phrasing of what the polls say, so that we avoid semantics issues.
I think part of the confusion is that the polling aggregates have Trump with a PV and swing state lead too, there's almost no polling that says the race is a tie, only by modeling can get to a tie. And of the three major modelers (538, Nate Silver, and The Economist) the latter two have Trump at ~60% to win. You can hang your hat on 538 I guess but you can't expect everyone to go along with the more rosy outlook you are trying to will into existence. Calling it "meta-analysis" gives it a scientific rigor that imo it does not deserve. It's not a meta-analysis of polls anyways, it's polls AND "fundamentals." "The polls" are pretty clear, espeically when you look at the bias they had in 2020.
I'm sorry to everyone else I know this conversation seems pointless lol
Havent the Democrats been outperforming the polls in recent smaller elections?
On July 01 2024 01:14 Introvert wrote: And DBP, thr polls have been remarkably consistent all year. There's also less reason to think they will change because these are the two most well-known candidates in modern American history. They both have records people are evaluating, maybe with rose-tinted glasses but still. It's not like there's a magical date where polls become extra predictive, it's summer anyways and nothing has changed significantly in polling for 9 months.
My prediction has been that the election will end up being a coinflip, which would indeed be "remarkably consistent" with how the polls have shaped up so far, so there's a good chance that you and I end up being right! You seem to dismiss the idea that there may be any volatility between now and November though, such as Biden taking a dip due to his poor debate performance. I'm impressed by your confidence - that Biden's chances won't be affected and that he remains at basically a 50/50 chance with Trump - but I think it's more likely that we'll see at least a little movement in the polls (negative for Biden), especially over the next few weeks. Maybe his polling numbers will recover after that though. We'll have to see what happens in July and August, and then Biden's team will potentially need to deal with another problem around September (second debate), which could also lead to more volatility. But maybe you're correct that Biden's polling will end up being immune to negatives; that would certainly help to balance out the massive number of negative consequences that haven't affected Trump's polling!
GH has tried explaining this but Biden isn't tied, he's behind.
As of last week (before the debate), Biden was a negligible amount behind Trump. It was basically 50/50.
He has to win the popular vote by probably 2 points to be about tied.
That's false. Biden could lose the overall popular vote and still win the election. What matters is the outcome of a few key swing states. Whether he wins 70% or 51% of California is irrelevant for the election results, even though the former beefs up his overall popular vote.
On July 01 2024 02:21 GreenHorizons wrote: [quote]
That's the fun of giving probabilities for presidential elections, as long as you give something a chance, you can't really be wrong.
We're presumably using the polls to inform the probabilities. For example, it would be inconsistent, based on the data, if someone said the current chance of Biden winning is 30% if the data says it's 50%.
One issue I'm trying to point out in that post is that it took a 8%+ lead in polls for Biden to squeak by Trump by thousands of votes. Biden consistently polling a couple points behind Trump is much worse than you/Democrats/Biden supporters are grasping. If the next 4 (3 before voting starts) months don't go much better than the last, then Biden is toast.
You're drawing a causal relationship that is unfounded. It's true that Biden was leading Trump by 8% in November 2020, and it's true that Biden just barely won the election, but it's incorrect to say that a polling lead of 8% is required for him to win again (especially in a different election).
The reason it's ~50:50 on 538 isn't the gap in the polls, it's the distance from the election (and the "fundamentals" favoring Biden).
The reason it's 50/50 is because a lot of people were asked who they'd vote for, and about half said Biden and about half said Trump (particularly for the key swing states they were polled in). That's it. 50/50 now means 50/50 now. It doesn't guarantee 50/50 in November, it doesn't guarantee 40/60 in November, and it doesn't guarantee anything else based on "the distance from the election". This is the problem with using early polling to extrapolate too far, rather than seeing a poll as a snapshot of a specific moment in time.
If you think people will gradually change their minds away from Biden, over the next few months, due to reasons like his poor debate performance, then that's fine, but those changes will still appear in the polls. For all the stock you've been putting into the polling data over the past few months, you're very quick to assert that those numbers really mean completely different numbers.
If Biden was polling like this in November it wouldn't be 50:50. To actually be 50:50, Biden has to do a lot better in polling than he is now and has been for 8 months.
If about half the people polled in November said they'd vote for Biden, and half said they've vote for Trump (particularly for the key swing states they were polled in), then the polls would say roughly 50/50.
So if you think it's going to be a 50:50 election you also believe that the next 4 months will be significantly better for Biden than the last 8 whether you say it explicitly or not.
I guess we'll just have to agree to disagree on what I believe. lol.
"inconsistent" isn't wrong.
I'm not saying he needs an 8% lead, I'm saying he needs better than to be losing as he has been for 8 months.
You're just misunderstanding how 538 is reaching its "50:50" probability.
When we have few polls or when it's early in the campaign, our model's predictions are mostly based on the fundamentals — with their standard deviation usually around 6 points or so. But when it's Election Day in a state with a lot of polls, our uncertainty about public opinion is a lot smaller, so we will put a lot more weight on the polling data (and subsequently, our polling averages that feed into the model) when generating our final prediction.
So 50:50 in the polls doesn't equate to a 50:50 chance. It only does now because "the fundamentals" in the model are more heavily weighted against polling and favor Biden. But those fundamentals are only going to become less weighted and if Biden's polling doesn't pick up (better than it has for 8 months) to counteract it, his odds will drop significantly even if his polling is marginally better.
I understand that Biden's strong economic and political fundamentals have been helping him in 538's early polling stages (I believe I've said about 10 times over the past few months that early polling is not necessarily a strong predictor of November outcomes, which is why I've been cautioning you and others to not take them as gospel). Introvert was the one who said that the next few months won't have fluctuations because the past few months have been steady. I didn't say that, and I don't think that one can safely make that causal claim.
I wouldn't be surprised at all if there's some volatility along the way to November (especially due to the debate), but I still predict that the November results will end up being a coinflip. And maybe I'm wrong. Maybe your prediction that Biden's polling numbers will consistently decline from now until November, as the weighing of fundamentals ease up in favor of other criteria, is correct. Maybe all three of us end up being wrong. We won't know until November, so I don't understand why you keep picking fights with me and others about these kinds of things.
I'm not saying (or conditioning) it not being a coinflip on Biden's polling getting worse, I'm saying that 538 wouldn't consider it a coinflip even if Biden was polling marginally better than he was before his debate disaster as we get closer to the election. He has to get higher in the polls than he's been in 8 months just to keep it near 50:50.
Biden's just in much more trouble than you appreciate currently.
Also you/Democrats/Biden supporters are advocating for a guy you iirc (and most of his 2020 voters) believe is aiding and abetting genocide, and could kill your mom in cold blood right in front of you and still not lose your vote or advocacy. I find that catastrophically problematic on several fronts, not the least of which, what it says of the ostensible "democracy" you're trying to "save" with that deplorable position in the first place.
You and I (and many others) have talked about this topic at great length. You disagree with our positions and explanations and rationalizations and justifications, and that's fine. I don't think I have anything else of substance to add to that particular conversation.
On July 01 2024 01:14 Introvert wrote: And DBP, thr polls have been remarkably consistent all year. There's also less reason to think they will change because these are the two most well-known candidates in modern American history. They both have records people are evaluating, maybe with rose-tinted glasses but still. It's not like there's a magical date where polls become extra predictive, it's summer anyways and nothing has changed significantly in polling for 9 months.
My prediction has been that the election will end up being a coinflip, which would indeed be "remarkably consistent" with how the polls have shaped up so far, so there's a good chance that you and I end up being right! You seem to dismiss the idea that there may be any volatility between now and November though, such as Biden taking a dip due to his poor debate performance. I'm impressed by your confidence - that Biden's chances won't be affected and that he remains at basically a 50/50 chance with Trump - but I think it's more likely that we'll see at least a little movement in the polls (negative for Biden), especially over the next few weeks. Maybe his polling numbers will recover after that though. We'll have to see what happens in July and August, and then Biden's team will potentially need to deal with another problem around September (second debate), which could also lead to more volatility. But maybe you're correct that Biden's polling will end up being immune to negatives; that would certainly help to balance out the massive number of negative consequences that haven't affected Trump's polling!
GH has tried explaining this but Biden isn't tied, he's behind.
As of last week (before the debate), Biden was a negligible amount behind Trump. It was basically 50/50.
He has to win the popular vote by probably 2 points to be about tied.
That's false. Biden could lose the overall popular vote and still win the election. What matters is the outcome of a few key swing states. Whether he wins 70% or 51% of California is irrelevant for the election results, even though the former beefs up his overall popular vote.
On July 01 2024 02:21 GreenHorizons wrote: [quote]
That's the fun of giving probabilities for presidential elections, as long as you give something a chance, you can't really be wrong.
We're presumably using the polls to inform the probabilities. For example, it would be inconsistent, based on the data, if someone said the current chance of Biden winning is 30% if the data says it's 50%.
One issue I'm trying to point out in that post is that it took a 8%+ lead in polls for Biden to squeak by Trump by thousands of votes. Biden consistently polling a couple points behind Trump is much worse than you/Democrats/Biden supporters are grasping. If the next 4 (3 before voting starts) months don't go much better than the last, then Biden is toast.
You're drawing a causal relationship that is unfounded. It's true that Biden was leading Trump by 8% in November 2020, and it's true that Biden just barely won the election, but it's incorrect to say that a polling lead of 8% is required for him to win again (especially in a different election).
The reason it's ~50:50 on 538 isn't the gap in the polls, it's the distance from the election (and the "fundamentals" favoring Biden).
The reason it's 50/50 is because a lot of people were asked who they'd vote for, and about half said Biden and about half said Trump (particularly for the key swing states they were polled in). That's it. 50/50 now means 50/50 now. It doesn't guarantee 50/50 in November, it doesn't guarantee 40/60 in November, and it doesn't guarantee anything else based on "the distance from the election". This is the problem with using early polling to extrapolate too far, rather than seeing a poll as a snapshot of a specific moment in time.
If you think people will gradually change their minds away from Biden, over the next few months, due to reasons like his poor debate performance, then that's fine, but those changes will still appear in the polls. For all the stock you've been putting into the polling data over the past few months, you're very quick to assert that those numbers really mean completely different numbers.
If Biden was polling like this in November it wouldn't be 50:50. To actually be 50:50, Biden has to do a lot better in polling than he is now and has been for 8 months.
If about half the people polled in November said they'd vote for Biden, and half said they've vote for Trump (particularly for the key swing states they were polled in), then the polls would say roughly 50/50.
So if you think it's going to be a 50:50 election you also believe that the next 4 months will be significantly better for Biden than the last 8 whether you say it explicitly or not.
I guess we'll just have to agree to disagree on what I believe. lol.
"inconsistent" isn't wrong.
I'm not saying he needs an 8% lead, I'm saying he needs better than to be losing as he has been for 8 months.
You're just misunderstanding how 538 is reaching its "50:50" probability.
When we have few polls or when it's early in the campaign, our model's predictions are mostly based on the fundamentals — with their standard deviation usually around 6 points or so. But when it's Election Day in a state with a lot of polls, our uncertainty about public opinion is a lot smaller, so we will put a lot more weight on the polling data (and subsequently, our polling averages that feed into the model) when generating our final prediction.
So 50:50 in the polls doesn't equate to a 50:50 chance. It only does now because "the fundamentals" in the model are more heavily weighted against polling and favor Biden. But those fundamentals are only going to become less weighted and if Biden's polling doesn't pick up (better than it has for 8 months) to counteract it, his odds will drop significantly even if his polling is marginally better.
I understand that Biden's strong economic and political fundamentals have been helping him in 538's early polling stages (I believe I've said about 10 times over the past few months that early polling is not necessarily a strong predictor of November outcomes, which is why I've been cautioning you and others to not take them as gospel). Introvert was the one who said that the next few months won't have fluctuations because the past few months have been steady. I didn't say that, and I don't think that one can safely make that causal claim.
I wouldn't be surprised at all if there's some volatility along the way to November (especially due to the debate), but I still predict that the November results will end up being a coinflip. And maybe I'm wrong. Maybe your prediction that Biden's polling numbers will consistently decline from now until November, as the weighing of fundamentals ease up in favor of other criteria, is correct. Maybe all three of us end up being wrong. We won't know until November, so I don't understand why you keep picking fights with me and others about these kinds of things.
The thing about all this is that you are the one hoping, against any evidence currently avaliable, that things are going to grt better for Biden when there's a good chance they get even worse instead. I'm looking at the history of how these things work. BTW, 538 is the only (or one of the only) model(s) that has Biden as good as a coinflip. Your assertion that things solidify later is true, but you asking it to do far too much work. Polls even now are generally predictive.
As to your point about the swing states... for all your focus on polling and models I'm surprised you don't know or understand the electoral college disadvantage dems have had and been complaining about for years. If Biden is even nationally he is almost certainly DOWN in the swing states, and indeed state level polling shows this. For decades the swing states vote right of the nation as a whole, and moreover states with similar demographics in the same region are correlated. If black turnout is down and rural turnout is better for Trump in WI it almost certainly is also true in MI, for example. The "Biden only has to win the swing states" is a problem for Biden, not a thing in his favor.
I don't know what will happen, but for months now you have been trying to pretend that he's in better shape, or that the state of the race is more ambigious than it actually is. Again, people can't handle that dems are actually losing.
I know that Democrats tend to win the popular vote, even if they lose the election. If the polling has him as 50/50 even chance to win the election, then that's talking about the electoral college, not the popular vote. The polls run scenarios for winning the election and individual state outcomes. If a poll announced something like "we asked a million American voters and 500,000 said they'll vote for Biden", that wouldn't be helpful at all without knowing where each of those 1 million voters live, but that's not what's happening here.
I am trying to figure out if I am misunderstanding whet you are saying here, because it seems wrong, and obviously so.
That's not what polls do, that's what modelers do, with polls as one input. Polls alone currently have Biden down slightly in the PV and down more in the swing states. Your one million person poll is precisely what they do...they weight their responses based on demographics and vote history but no poll is a poll of who is going to win the election, that doesn't even make sense. A national poll showing Trump 45% Biden 43% among all voters is literally a weighted tally of how people are going to vote, not Biden's chances of winning the election.
Ah, I see what you're referring to. I was referring to the aggregate of the state polls - that's why I wrote "(particularly for the key swing states they were polled in)" twice in my earlier post, and I'm happy to call that a "model", as you're saying it's what "modelers" do. That's totally fair. It's definitely easier to use that different term than referring to every survey and meta-analysis of surveys as falling under the umbrella phrasing of what the polls say, so that we avoid semantics issues.
I think part of the confusion is that the polling aggregates have Trump with a PV and swing state lead too, there's almost no polling that says the race is a tie, only by modeling can get to a tie. And of the three major modelers (538, Nate Silver, and The Economist) the latter two have Trump at ~60% to win. You can hang your hat on 538 I guess but you can't expect everyone to go along with the more rosy outlook you are trying to will into existence. Calling it "meta-analysis" gives it a scientific rigor that imo it does not deserve. It's not a meta-analysis of polls anyways, it's polls AND "fundamentals." "The polls" are pretty clear, espeically when you look at the bias they had in 2020.
I'm sorry to everyone else I know this conversation seems pointless lol
Havent the Democrats been outperforming the polls in recent smaller elections?
I think Democrats have been generally doing quite well in elections, ever since Trump took office, from success in smaller elections to weathering potential Red Waves, even throughout Biden's presidency.
As predicted, SCOTUS wastes everyones time, punts the immunity issue, and is allowing courts to get jammed up well past 2025 arguing what is considered official and what is considered unofficial. Gee thanks SCOTUS, I really wonder if this is going to end up having disastrous consequences unless you believe the United States of America should be ruled by 1 President and 9 judges.
Biden, as the current President, could just drone strike Trump and argue that he was doing an official act to protect the country from insurrectionists. Funny how the American president technically has more power than every monarch in the modern world if they find the right judges.
On June 28 2024 23:32 frontgarden2222 wrote: SCOTUS just overruled Chevron what the fuck. Ruling appears to be based on the Administrative Procedure Act but still this literally fucks every government agency.
Can you please elaborate on what this is and what this means?
Basically Chevron deference argues that government agencies (like the EPA) are best positioned to interpret federal statutes if they're ambigious, within reason of course. The argument against it is that the courts should handle interpretation of law.
Like everything with this SCOTUS, they're absolutely obsessed with the minutiae and completely willing to nuke common sense laws becaue of it.
So the problem they have appears to be with the Administrative Procedure Act. Fine, legislators can resolve the impending quagmire of Art III judges making decisions they have literally no real idea about by drafting new legislation. But lol lmao have you seen Congress operate? Half this Congress takes open "brides" in some way or form and want to dismantle specific parts of the administrative state that get in the way of the people handing the "bribes".
In the meantime there's probably going to be a huge amount of circuit splits and different regions are going to see differing levels of things people take for granted like water quality or workplace protections.
Which already happens I guess because the administrative state keeps getting kneecapped by every subsequent government. But this is rolling out the welcome mat for some real bad actors to do some real bad actor things when combined with some other recent rulings regarding things like gratuities being kinda OK if they're done after the fact.
I like the maybe intentional, maybe Freudian slip "The argument against it is that the courts should handle interpretation of law" because yes, that is exactly the issue, but when phrased that way, it's blindingly obvious that it is the job of thr courts to interpret laws!
Theoretically this should force congress to be more clear, and stop administrations of both parties from engaging in policy ping-pong using novel statutory readings.
But of course since the bureaucracy at places like the EPA overwhelming tilt in one direction, it's clear why the big-government left was fearing this day.
Judges do not have the expertise or time to waste time on things like scientific or urban design minutiae. This is why we have government agencies. No, don't tell me they can get expert help. They still do not understand things created through a lifetime of experience.
Theoretically this should force Congress to legislate. They will not, they will use this ruling to their advantage. You are not stupid, you know exactly what is going to happen.
If you think the EPA is idelogically opposed to conservative viewpoints, that's up to the Republicans to fix and not in a way that involves judge shopping. Agencies, guided by Congress, should be able to produce regulations created through compromises. When it comes to this sort of garbage no one cares about, Congress is still pretty effective. I'm still confused how you think the EPA is overwhelmingly filled to the gills with leftists or whatever you're suggesting.
An administrative state must still exist to ensure regulations are consistently applied. The solution is not for bad actors to judge shop for a judge, who is technically legally able to accept brides after the fact after the recent SCOTUS ruling, and for every region in the United States to have haphazard standards. This is what a failing country does.
So, since the President is immune while doing official constitutional acts, and the constitution declares the President is the Commander in chief, SCOTUS has just ruled that Biden in fact can order seal team 6 to assassinate Trump and the conservative supreme court justices while being immune to prosecution.
Apparently Mike Johnson said this with regards to the immunity case:
Today’s ruling by the Court is a victory for former President Trump and all future presidents, and another defeat for President Biden’s weaponized Department of Justice and Jack Smith. The Court clearly stated that presidents are entitled to immunity for their official acts. This decision is based on the obviously unique power and position of the presidency, and comports with the Constitution and common sense. As President Trump has repeatedly said, the American people, not President Biden’s bureaucrats, will decide the November 5th election.
If Biden were truly trying to weaponize something against Trump, hasn't the SC just given him cart blanche to weaponize... well.. actual weapons against Trump now? Why bother with the judiciary route when they have literally given you the freedom to take the extrajudiciary one?
It's incredibly dangerous, because we trust Biden not to take that route, but there's the very real possibility that Trump will. He has stated as much himself. So is it justified to throw Trump in Guantanamo in order to prevent him from becoming the first King of the USA, thereby effectively becoming the first King of the USA yourself? How on earth did the SC not consider the horrors they have just given blanket immunity for?
EPT
