|
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.If you have any questions, comments, concern, or feedback regarding the USPMT, then please use this thread: http://www.teamliquid.net/forum/website-feedback/510156-us-politics-thread |
On October 11 2020 14:03 Shingi11 wrote:Show nested quote +On October 11 2020 13:06 iPlaY.NettleS wrote:On October 11 2020 12:57 PhoenixVoid wrote:
Nate Silver said Biden's state polls were comparatively meh for one week after he just came off a dominating one where he got high quality polls that ranged from 7-11 points in Pennsylvania and similar numbers in Wisconsin, Michigan and Florida. I'd think Trump is polling slightly better in swing states compared to 2016 because there's fewer undecideds and third party voters and some have solidified on him rather than hovering on the fence like last time. I'm on RCP now and it shows Biden is polling worse in MI,PA,FL,WI than Clinton did at this stage four years ago.I'll note that he's only 0.1% worse in FL though, compared to -4% in MI and -1.2% in PA. He is however polling better in AZ than Clinton so maybe he can pick that one up. None of this changes Biden 400 EV votes as total delusion at this stage. I just going to post a couple of election forecasters here, all of them could be wrong but most have biden at 350+
the economist biden 350 trump 188 538 biden 352 trump 186 JHK forcast biden 356 trump 182
Our progress biden 389 trump 149 While 400 is a push it is not that unlikely. IT would require texas to flip and by 538 standers that has a higher chance of happening then trump winning. Texas has a 30 chance for biden the while trump is sitting at 14 to win. BIden is polling much better then Clinton was. SHe was never over 50% approve, she hovered around the mid 40s and biden is sitting at 52%. ALso trump was within margin of error with all polls in 2016. Even if you shift every poll this time by margin of error in trumps favor he still loses. I want some of that cool aid you are sipping but if you cant see that trump is way down right now. Trump is losing every swing state right now by large margins and is barely ahead in solid red states like georgia and iowa. IF anything they are toss ups right now. You have to get all the way to like missori and kansas before trump start having solid leads. That should tell you how dire things are for trump edit RCP no toss up map biden 358 trump 180
I feel like forecasts are gonna need to show me some reason I should believe them. After 2016, these "models" aren't doing a lot for me. I'm taking the perspective of "we'll see".
|
On October 11 2020 15:26 Mohdoo wrote:Show nested quote +On October 11 2020 14:03 Shingi11 wrote:On October 11 2020 13:06 iPlaY.NettleS wrote:On October 11 2020 12:57 PhoenixVoid wrote:
Nate Silver said Biden's state polls were comparatively meh for one week after he just came off a dominating one where he got high quality polls that ranged from 7-11 points in Pennsylvania and similar numbers in Wisconsin, Michigan and Florida. I'd think Trump is polling slightly better in swing states compared to 2016 because there's fewer undecideds and third party voters and some have solidified on him rather than hovering on the fence like last time. I'm on RCP now and it shows Biden is polling worse in MI,PA,FL,WI than Clinton did at this stage four years ago.I'll note that he's only 0.1% worse in FL though, compared to -4% in MI and -1.2% in PA. He is however polling better in AZ than Clinton so maybe he can pick that one up. None of this changes Biden 400 EV votes as total delusion at this stage. I just going to post a couple of election forecasters here, all of them could be wrong but most have biden at 350+
the economist biden 350 trump 188 538 biden 352 trump 186 JHK forcast biden 356 trump 182
Our progress biden 389 trump 149 While 400 is a push it is not that unlikely. IT would require texas to flip and by 538 standers that has a higher chance of happening then trump winning. Texas has a 30 chance for biden the while trump is sitting at 14 to win. BIden is polling much better then Clinton was. SHe was never over 50% approve, she hovered around the mid 40s and biden is sitting at 52%. ALso trump was within margin of error with all polls in 2016. Even if you shift every poll this time by margin of error in trumps favor he still loses. I want some of that cool aid you are sipping but if you cant see that trump is way down right now. Trump is losing every swing state right now by large margins and is barely ahead in solid red states like georgia and iowa. IF anything they are toss ups right now. You have to get all the way to like missori and kansas before trump start having solid leads. That should tell you how dire things are for trump edit RCP no toss up map biden 358 trump 180 I feel like forecasts are gonna need to show me some reason I should believe them. After 2016, these "models" aren't doing a lot for me. I'm taking the perspective of "we'll see".
The models were not off though. they predicted a hillary win by +3% and i believe she ended at like +1.9%. There is a thing called margin of error that a lot of people forget about when they look at polls and models. This is the national average but the same can be some with battleground. She was leading by about 3 to 4% in state polls and guess what the margin of error is. About 3 to 4% in most polls. DO the same thing with biden now. Shift the polls and models in trumps favor and he is still losing.
edit
For reference at this day in 2016
Clinton was at 44
Trump was at 40
As of today
biden 52.2
trump 41.9
|
Nettles, you are only convincing yourself. Although you are technically right, you forget that what matters is the number of undecided voters. It's easy to fill a 5 point bridge when you have 20% voters who are undecided. It's almost impossible when that number is 2%.
I think that reading in polls in a non partisan way is EXTREMELY hard. 538 are still the best aggregator and still have by far the best and most complete model. They give Trump a 14% chance to win as of today. That is low but non negligible. It's like having to hit a 6 on a single dice roll. That happens.
For reference he had 35% chances of winning on election day last time.
One thing to add is that unless something happens, this 14% is going to go down because the model takes time into account. So unless he starts going up in the poll soon, he is probably going to end up in one digit territory.
There is also the possibility that the race is close in which case he will almost certainly try to steal the election. That's a real possibility that is not taken into consideration in the model.
|
For reference he had 35% chances of winning on election day last time.
NYT had him at 15% and 538 had him at 28.6% chance of winning on election day.
538 had him at about 13-14% ~this week of October
NYT had him as low as 7% in mid-late October
www.nytimes.com
projects.fivethirtyeight.com
EDIT: It makes me wonder how 538 is coming to essentially the same conclusion despite this difference in polling/favorability between 2016 and 2020.
|
On October 11 2020 18:09 GreenHorizons wrote:NYT had him at 15% and 538 had him at 28.6% chance of winning on election day. 538 had him at about 13-14% ~this week of October NYT had him as low as 7% in mid-late October www.nytimes.comprojects.fivethirtyeight.comEDIT: It makes me wonder how 538 is coming to essentially the same conclusion despite this difference in polling/favorability between 2016 and 2020.
It's just math, and in the last four weeks of the race A LOT happened in 2016. Clinton got one bad blow after another. It's totally possible that Biden gets one news cycle after another and that 538 models ends up at 29% chances for Trump.
And it's also possible Trump stays around the 10-15% mark and wins.
|
On October 11 2020 19:36 Biff The Understudy wrote:Show nested quote +On October 11 2020 18:09 GreenHorizons wrote:For reference he had 35% chances of winning on election day last time. NYT had him at 15% and 538 had him at 28.6% chance of winning on election day. 538 had him at about 13-14% ~this week of October NYT had him as low as 7% in mid-late October www.nytimes.comprojects.fivethirtyeight.comEDIT: It makes me wonder how 538 is coming to essentially the same conclusion despite this difference in polling/favorability between 2016 and 2020. It's just math, and in the last four weeks of the race A LOT happened in 2016. Clinton got one bad blow after another. It's totally possible that Biden gets one news cycle after another and that 538 models ends up at 29% chances for Trump. And it's also possible Trump stays around the 10-15% mark and wins.
I am curious if the Trump campaign have some aces up their sleeve to try to gain momentum in the final weeks. The democrats and other anti-Trumpers might have saved some gun powder as well.
The situation is very different to 2016, though. Trump's main hurdles are the economy and the pandemic, and neither are going anywhere the next month.
|
On October 11 2020 07:20 Danglars wrote:Show nested quote +On October 11 2020 06:25 Nouar wrote:What the actual fuck Lindsey Graham ? On a question about careers in SC, he said he believed his opponent would lose because he is a democrat, not because he is black, and he just said that african-americans in South Carolina can go/become where/what they want, they just have to be conservative, not liberal. Seriously ?? “Do I believe our cops are systemically racist? No,” Graham said. “Do I believe South Carolina is a racist state? No. Let me tell you why. To young people out there, young people of color, young immigrants, this is a great state, but one thing I can say without any doubt, you can be an African American and go to the Senate but you just have to share our values.”
He went on to say: “If you’re a young, African American or an immigrant, you can go anywhere in this state, you just need to be conservative, not liberal”. https://www.theguardian.com/us-news/live/2020/oct/10/trump-coronavirus-live-white-house-event-debate-biden-latest-updates?page=with:block-5f821e478f088cf1f1a6489a#block-5f821e478f088cf1f1a6489a It's a pretty inarticulate statement, but follows along with mainstream conservative thought. If you're taught all your life from a liberal perspective, you can come away thinking that the country is too systemically racist for any minority to achieve anything of consequence. The only way to rise up is to campaign for anti-racist programs in companies, reparations, affirmative action for universities, racial quotas in management positions, and all the rest. Maybe that's a biased oversimplification of the ideology, but it's a debate and you'll hear the usual smears about Republicans and the poor and elderly or whatnot. Contrast that with the conservative view that America is the land of opportunity, and the most important value to inculcate in the young is that you can rise with hard work and determination. You can be president one day like Barack Obama, or on the Supreme Court like Clarence Thomas, or chair your own congressional committee like Elijah Cummings. Don't make mountains out of molehills in obstacles, and take personal ownership of your track in life. The more obvious point is that voters identify with the ideas and platform of candidates, so if you're proposing dumb ideas and backwards stances, then that should be what stops you. Not the color of your skin, but the quality of your directives. The lefties on this forum should be well aware at how often they criticize the bulk of the Republican platform, so should understand Republicans thinking the same same thing about the left-wing platform (IE give a true/false to "It shouldn't matter if you're a minority representing the Republican party in an election, because their platform and leadership is too extreme for consideration" and compare/contrast)
I agree that many create obstacles for themselves, and some times use them as excuses. The conservative platform you describe is equally flawed, though, as the US has shockingly little social mobility for being "the land of opportunity." In that regard, having your access to college depend on your grades and dedication instead of your ability to pay obscene tuition fees should be considered "conservative" values.
It is also sad how tribal the US has become, in Europe, ethnicity has little to do with the conservative/liberal/left/right axis, and it is much healthier that way.
|
On October 11 2020 19:36 Biff The Understudy wrote:Show nested quote +On October 11 2020 18:09 GreenHorizons wrote:For reference he had 35% chances of winning on election day last time. NYT had him at 15% and 538 had him at 28.6% chance of winning on election day. 538 had him at about 13-14% ~this week of October NYT had him as low as 7% in mid-late October www.nytimes.comprojects.fivethirtyeight.comEDIT: It makes me wonder how 538 is coming to essentially the same conclusion despite this difference in polling/favorability between 2016 and 2020. It's just math, and in the last four weeks of the race A LOT happened in 2016. Clinton got one bad blow after another. It's totally possible that Biden gets one news cycle after another and that 538 models ends up at 29% chances for Trump. And it's also possible Trump stays around the 10-15% mark and wins.
There's not really any way to know whether a probability estimate of a presidential election by 538 is anything more than a mathematically derived guess with no more accuracy than assigning a random chance to win to both candidates.
My main point was just that the "35% chance on election day" was flat wrong and that Trump's not significantly less likely to win today than he was at this time in 2016 according to 538 (NYT doesn't seem to be doing it this cycle).
So whatever you and others are seeing in the data (besides the mistaken 35% figure) isn't changing the likeliness Trump wins in 538's estimate compared to 2016 at this point in time.
|
On October 11 2020 22:21 GreenHorizons wrote:Show nested quote +On October 11 2020 19:36 Biff The Understudy wrote:On October 11 2020 18:09 GreenHorizons wrote:For reference he had 35% chances of winning on election day last time. NYT had him at 15% and 538 had him at 28.6% chance of winning on election day. 538 had him at about 13-14% ~this week of October NYT had him as low as 7% in mid-late October www.nytimes.comprojects.fivethirtyeight.comEDIT: It makes me wonder how 538 is coming to essentially the same conclusion despite this difference in polling/favorability between 2016 and 2020. It's just math, and in the last four weeks of the race A LOT happened in 2016. Clinton got one bad blow after another. It's totally possible that Biden gets one news cycle after another and that 538 models ends up at 29% chances for Trump. And it's also possible Trump stays around the 10-15% mark and wins. There's not really any way to know whether a probability estimate of a presidential election by 538 is anything more than a mathematically derived guess with no more accuracy than assigning a random chance to win to both candidates. My main point was just that the "35% chance on election day" was flat wrong and that Trump's not actually less likely to win today than he was this time in 2016 according to 538 (NYT doesn't seem to be doing it this cycle). So whatever you and others are seeing in the data (besides the mistaken 35% figure) isn't changing the likeliness Trump wins in 538's estimate compared to 2016 at this point in time. How was it flat wrong in 2016? Trump won by a tiny margin in several states. Just because Trump won doesn't mean the probability was wrong.
|
|
|
On October 11 2020 22:21 GreenHorizons wrote:Show nested quote +On October 11 2020 19:36 Biff The Understudy wrote:On October 11 2020 18:09 GreenHorizons wrote:For reference he had 35% chances of winning on election day last time. NYT had him at 15% and 538 had him at 28.6% chance of winning on election day. 538 had him at about 13-14% ~this week of October NYT had him as low as 7% in mid-late October www.nytimes.comprojects.fivethirtyeight.comEDIT: It makes me wonder how 538 is coming to essentially the same conclusion despite this difference in polling/favorability between 2016 and 2020. It's just math, and in the last four weeks of the race A LOT happened in 2016. Clinton got one bad blow after another. It's totally possible that Biden gets one news cycle after another and that 538 models ends up at 29% chances for Trump. And it's also possible Trump stays around the 10-15% mark and wins. There's not really any way to know whether a probability estimate of a presidential election by 538 is anything more than a mathematically derived guess with no more accuracy than assigning a random chance to win to both candidates. My main point was just that the "35% chance on election day" was flat wrong and that Trump's not actually less likely to win today than he was this time in 2016 according to 538 (NYT doesn't seem to be doing it this cycle). So whatever you and others are seeing in the data (besides the mistaken 35% figure) isn't changing the likeliness Trump wins in 538's estimate compared to 2016 at this point in time.
That is not how probabilities work. Just because something turned out one way doesn't mean that the probability for that effect is higher than 50%.
If i roll a die, the probability of rolling a 3 is 1/6. If i now roll that die, and actually do get a 3, that does not mean that that probability was incorrect.
Of course, in real world situations, stuff gets more complicated. What a number like "Trump has a 35% chance of winning" means is "given our mathematical model, fed with the data that we currently use, Trump wins in 35% of the cases". Of course, if you had perfect information (and assumed that the universe is deterministic in some way), that 35% chance is incorrect, and the chance is either 100% or 0%.
The problem with these chances is that they are hard to actually test, because you can not have 1000 elections from the starting position.
|
On October 11 2020 22:53 Gorsameth wrote:Show nested quote +On October 11 2020 22:21 GreenHorizons wrote:On October 11 2020 19:36 Biff The Understudy wrote:On October 11 2020 18:09 GreenHorizons wrote:For reference he had 35% chances of winning on election day last time. NYT had him at 15% and 538 had him at 28.6% chance of winning on election day. 538 had him at about 13-14% ~this week of October NYT had him as low as 7% in mid-late October www.nytimes.comprojects.fivethirtyeight.comEDIT: It makes me wonder how 538 is coming to essentially the same conclusion despite this difference in polling/favorability between 2016 and 2020. It's just math, and in the last four weeks of the race A LOT happened in 2016. Clinton got one bad blow after another. It's totally possible that Biden gets one news cycle after another and that 538 models ends up at 29% chances for Trump. And it's also possible Trump stays around the 10-15% mark and wins. There's not really any way to know whether a probability estimate of a presidential election by 538 is anything more than a mathematically derived guess with no more accuracy than assigning a random chance to win to both candidates. My main point was just that the "35% chance on election day" was flat wrong and that Trump's not actually less likely to win today than he was this time in 2016 according to 538 (NYT doesn't seem to be doing it this cycle). So whatever you and others are seeing in the data (besides the mistaken 35% figure) isn't changing the likeliness Trump wins in 538's estimate compared to 2016 at this point in time. How was it flat wrong in 2016? Trump won by a tiny margin in several states. Just because Trump won doesn't mean the probability was wrong.
28.6% or 15% ≠ 35%? Are you guys messin with me? I know (granted not nearly as well as Simberto) how probability works.
I was clarifying that saying "Trump's chance to win was 35% on election day" was flat wrong because that's not what 538 (where the 2020 probability was referenced from) actually said. It said 28.6% (or 28.2% for "polls plus" since it's unclear which, is being used in 2020) and NYT had him at 15% on election day.
The problem with these chances is that they are hard to actually test, because you can not have 1000 elections from the starting position.
Also this ^
and:Trump's not significantly less likely to win today than he was this time in 2016 according to 538 since the "35% chance on election day" was being used to support an argument that his chances today in 2020 are significantly worse than this time in 2016.
|
ok I didn't realise you merely meant the number itself. It helps if you make it clearer what your arguing against.
|
On October 11 2020 23:14 Gorsameth wrote:
ok I didn't realise you merely meant the number itself. It helps if you make it clearer what your arguing against.
... I said:
On October 11 2020 18:09 GreenHorizons wrote:NYT had him at 15% and 538 had him at 28.6% chance of winning on election day. 538 had him at about 13-14% ~this week of October NYT had him as low as 7% in mid-late October www.nytimes.comprojects.fivethirtyeight.comEDIT: It makes me wonder how 538 is coming to essentially the same conclusion despite this difference in polling/favorability between 2016 and 2020.
|
On October 11 2020 22:21 GreenHorizons wrote: any way to know whether a probability estimate... ... is anything more than a mathematically derived guess That's what probability is GH.
I don't think you know what probability is, nevermind how it works. Inherent in probability is the recognition of imperfect data, otherwise there will be certainty of a prediction. Is probability not taught in US schools? This is super basic "high school" level stuff in the UK. People shouldn't be so proud to display their ignorance. This thread have been through this since the previous election. How is it possible not to learn anything since?
On October 11 2020 22:21 GreenHorizons wrote:
no more accuracy than assigning a random chance to win to both candidates.
The only way to know whether a probability is a good prediction of chance is to go through the data and the modelling, but that needs both the data and the methodologies and is an impossibility for the mathematically illiterate anyways.
|
+ Show Spoiler +On October 11 2020 23:18 Dangermousecatdog wrote:Show nested quote +On October 11 2020 22:21 GreenHorizons wrote: any way to know whether a probability estimate... ... is anything more than a mathematically derived guess That's what probability is GH. I don't think you know what probability is, nevermind how it works. Inherent in probability is the recognition of imperfect data, otherwise there will be certainty of a prediction. Is probability not taught in US schools? This is super basic "high school" level stuff in the UK. People shouldn't be so proud to display their ignorance. This thread have been through this since the previous election. How is it possible not to learn anything since? Show nested quote +On October 11 2020 22:21 GreenHorizons wrote:
no more accuracy than assigning a random chance to win to both candidates. The only way to know whether a probability is a good prediction of chance is to go through the data and the modelling, but that needs both the data and the methodologies and is an impossibility for the mathematically illiterate anyways.
Holy shit guys. I know my grammar isn't great, but this is an epic reading comprehension failure on your parts.
|
On October 11 2020 23:20 GreenHorizons wrote:+ Show Spoiler +On October 11 2020 23:18 Dangermousecatdog wrote:Show nested quote +On October 11 2020 22:21 GreenHorizons wrote: any way to know whether a probability estimate... ... is anything more than a mathematically derived guess That's what probability is GH. I don't think you know what probability is, nevermind how it works. Inherent in probability is the recognition of imperfect data, otherwise there will be certainty of a prediction. Is probability not taught in US schools? This is super basic "high school" level stuff in the UK. People shouldn't be so proud to display their ignorance. This thread have been through this since the previous election. How is it possible not to learn anything since? Show nested quote +On October 11 2020 22:21 GreenHorizons wrote:
no more accuracy than assigning a random chance to win to both candidates. The only way to know whether a probability is a good prediction of chance is to go through the data and the modelling, but that needs both the data and the methodologies and is an impossibility for the mathematically illiterate anyways. Holy shit guys. I know my grammar isn't great, but this is an epic reading comprehension failure on your parts.
When a bunch of people read something and don't get it is that their fault or your fault? 538 posts extensively about how their model works so saying
any way to know whether a probability estimate... ... is anything more than a mathematically derived guess
Makes you seem like you have no idea what you're talking about.
I'd also add that most people attacking predictive models do it in a way that outs them as not understanding what they're talking about. Sam Wang is most fun example. His math wasn't wrong. It was the assumption in the model he used that each state election would be an independent event. All probability estimates are a mathematically derived guess. You must attack the model that got it.
|
|
|
On October 11 2020 23:35 Blitzkrieg0 wrote:Show nested quote +On October 11 2020 23:20 GreenHorizons wrote:+ Show Spoiler +On October 11 2020 23:18 Dangermousecatdog wrote:Show nested quote +On October 11 2020 22:21 GreenHorizons wrote: any way to know whether a probability estimate... ... is anything more than a mathematically derived guess That's what probability is GH. I don't think you know what probability is, nevermind how it works. Inherent in probability is the recognition of imperfect data, otherwise there will be certainty of a prediction. Is probability not taught in US schools? This is super basic "high school" level stuff in the UK. People shouldn't be so proud to display their ignorance. This thread have been through this since the previous election. How is it possible not to learn anything since? Show nested quote +On October 11 2020 22:21 GreenHorizons wrote:
no more accuracy than assigning a random chance to win to both candidates. The only way to know whether a probability is a good prediction of chance is to go through the data and the modelling, but that needs both the data and the methodologies and is an impossibility for the mathematically illiterate anyways. Holy shit guys. I know my grammar isn't great, but this is an epic reading comprehension failure on your parts. When a bunch of people read something and don't get it is that their fault or your fault?
It can be either. This time it was definitely theirs.
|
There is really no need to turn this into a major mudfight. GH said something, people responded, GH clarified, stuff is clear now. No need to fight about who won that exchange.
|
|
|
|
|
|
EPT
