|
If this thread turns into a USPMT 2.0, we will not hesitate to shut it down. Do not even bother posting if all you're going to do is shit on the Democratic candidates while adding nothing of value.
Rules: - Don't post meaningless one-liners. - Don't turn this into a X doesn't stand a chance against Trump debate. - Sources MUST have a supporting comment that summarizes the source beforehand. - Do NOT turn this thread into a Republicans vs. Democrats shit-storm.
This thread will be heavily moderated. Expect the same kind of strictness as the USPMT. |
Having a hard time buying these 538 forecasts. Bernie losing in Hawaii, New York, Arizona makes no sense to me.
To me, Idaho, Michigan and Washington will be good indicators for how much of a chance Bernie has.
If Bernie does as bad as 538 predicts he'll do, I'll find myself a lot more accepting of Warren endorsing Biden. If Bernie is even remotely viable, she should endorse Bernie. But if she needs to be the executioner, I accept it.
|
538 seems heavily influenced by latest events. Before South Carolina Biden was said to have 0.8% chance of getting a majority, or something like that. Now that he has won Super Tuesday convincingly, it changes. If Bernie does decent tuesday it will change again.
When Bernie did well in the first three states the model predicted that Bernie would win South Carolina for a while despite not a single poll saying he was going to be less than 5 points behind in the state.
|
On March 10 2020 01:52 Nebuchad wrote: 538 seems heavily influenced by latest events. Before South Carolina Biden was said to have 0.8% chance of getting a majority, or something like that. Now that he has won Super Tuesday convincingly, it changes. If Bernie does decent tuesday it will change again.
When Bernie did well in the first three states the model predicted that Bernie would win South Carolina for a while despite not a single poll saying he was going to be less than 5 points behind in the state.
While true, models of this nature tend to improve with age and more data. Especially as more people drop.
But TBH as long as old people are dying from Corona, we should probably keep people around lol. Tulsi as the last standing candidate = LOL
|
On March 10 2020 01:57 Mohdoo wrote:Show nested quote +On March 10 2020 01:52 Nebuchad wrote: 538 seems heavily influenced by latest events. Before South Carolina Biden was said to have 0.8% chance of getting a majority, or something like that. Now that he has won Super Tuesday convincingly, it changes. If Bernie does decent tuesday it will change again.
When Bernie did well in the first three states the model predicted that Bernie would win South Carolina for a while despite not a single poll saying he was going to be less than 5 points behind in the state. While true, models of this nature tend to improve with age and more data. Especially as more people drop. But TBH as long as old people are dying from Corona, we should probably keep people around lol. Tulsi as the last standing candidate = LOL
I don't disagree, Biden looks like a clear favourite right now. Just talking about how it seems the model can come to make predictions that appear weird for specific states by overadjusting for new data in other states.
|
On March 10 2020 02:02 Nebuchad wrote:Show nested quote +On March 10 2020 01:57 Mohdoo wrote:On March 10 2020 01:52 Nebuchad wrote: 538 seems heavily influenced by latest events. Before South Carolina Biden was said to have 0.8% chance of getting a majority, or something like that. Now that he has won Super Tuesday convincingly, it changes. If Bernie does decent tuesday it will change again.
When Bernie did well in the first three states the model predicted that Bernie would win South Carolina for a while despite not a single poll saying he was going to be less than 5 points behind in the state. While true, models of this nature tend to improve with age and more data. Especially as more people drop. But TBH as long as old people are dying from Corona, we should probably keep people around lol. Tulsi as the last standing candidate = LOL I don't disagree, Biden looks like a clear favourite right now. Just talking about how it seems the model can come to make predictions that appear weird for specific states by overadjusting for new data in other states.
Absolutely, and it isn't a mistake that 538's model has enormous swings that generate headlines and page views. 538 describes their model as nimble and impressionable, but I also kinda think that is done to make it change often.
|
On March 10 2020 01:49 Mohdoo wrote: Having a hard time buying these 538 forecasts. Bernie losing in Hawaii, New York, Arizona makes no sense to me.
To me, Idaho, Michigan and Washington will be good indicators for how much of a chance Bernie has.
If Bernie does as bad as 538 predicts he'll do, I'll find myself a lot more accepting of Warren endorsing Biden. If Bernie is even remotely viable, she should endorse Bernie. But if she needs to be the executioner, I accept it.
Why budge on this. The whole reason people want Warren to endorse is to change or help Bernie's odds.
The only thing Biden is being elected on is "electability", so high profile endorsements of Bernie that can bolster his opinion of being electable would go a long way. Warren withholding an endorsement doesn't make her pragmatic, just makes her either not a fighter or not a progressive.
|
I thought their model would be better at predicting the relative strengths from past polls instead of reacting to changes in relative strengths from polls I can read myself.
But this race has been so unpredictable. After Biden fell 20 points in Iowa in a span of two weeks, I never saw him bouncing back. Buttigieg and Klobuchar were just so much better than him in debates and occupying a relatively moderate position compared to Bernie. I thought their own interests would propel their campaigns through to Super Tuesday to see if one would be the clear frontrunner AND THEM pressure the other two to drop out and support the leader.
Instead, the guy that is having trouble stringing sentences together in each new week of video got their endorsement prior to Super Tuesday which changes the path of everything. Bernie’s finally directly making the case that he’s profoundly different than Biden, but it all looks like too little too late. Rhetoric on social security cuts and stuff like that should’ve been done every day from South Carolina too present day, not just after Super Tuesday.
|
On March 10 2020 02:14 Danglars wrote: I thought their model would be better at predicting the relative strengths from past polls instead of reacting to changes in relative strengths from polls I can read myself.
But this race has been so unpredictable. After Biden fell 20 points in Iowa in a span of two weeks, I never saw him bouncing back. Buttigieg and Klobuchar were just so much better than him in debates and occupying a relatively moderate position compared to Bernie. I thought their own interests would propel their campaigns through to Super Tuesday to see if one would be the clear frontrunner AND THEM pressure the other two to drop out and support the leader.
Instead, the guy that is having trouble stringing sentences together in each new week of video got their endorsement prior to Super Tuesday which changes the path of everything. Bernie’s finally directly making the case that he’s profoundly different than Biden, but it all looks like too little too late. Rhetoric on social security cuts and stuff like that should’ve been done every day from South Carolina too present day, not just after Super Tuesday.
I wonder if Bernie will join in the Trump campaign line of attack that Biden is simply losing it.
|
Norway28558 Posts
|
On March 10 2020 02:08 Logo wrote:Show nested quote +On March 10 2020 01:49 Mohdoo wrote: Having a hard time buying these 538 forecasts. Bernie losing in Hawaii, New York, Arizona makes no sense to me.
To me, Idaho, Michigan and Washington will be good indicators for how much of a chance Bernie has.
If Bernie does as bad as 538 predicts he'll do, I'll find myself a lot more accepting of Warren endorsing Biden. If Bernie is even remotely viable, she should endorse Bernie. But if she needs to be the executioner, I accept it. Why budge on this. The whole reason people want Warren to endorse is to change or help Bernie's odds. The only thing Biden is being elected on is "electability", so high profile endorsements of Bernie that can bolster his opinion of being electable would go a long way. Warren withholding an endorsement doesn't make her pragmatic, just makes her either not a fighter or not a progressive.
You might be right that it is too early, but I do think there is a time when it makes sense. And while I think there is a very practical reason to say Bernie should stay in, I am just saying there may be a point when it is lights out. Biden is doing a lot better than Clinton in a lot of states.
The only thing that makes me comfortable with drawing this out is the fact that Corona kills old people and everyone involved is old.
|
On March 10 2020 02:34 GoTuNk! wrote:Show nested quote +On March 10 2020 02:14 Danglars wrote: I thought their model would be better at predicting the relative strengths from past polls instead of reacting to changes in relative strengths from polls I can read myself.
But this race has been so unpredictable. After Biden fell 20 points in Iowa in a span of two weeks, I never saw him bouncing back. Buttigieg and Klobuchar were just so much better than him in debates and occupying a relatively moderate position compared to Bernie. I thought their own interests would propel their campaigns through to Super Tuesday to see if one would be the clear frontrunner AND THEM pressure the other two to drop out and support the leader.
Instead, the guy that is having trouble stringing sentences together in each new week of video got their endorsement prior to Super Tuesday which changes the path of everything. Bernie’s finally directly making the case that he’s profoundly different than Biden, but it all looks like too little too late. Rhetoric on social security cuts and stuff like that should’ve been done every day from South Carolina too present day, not just after Super Tuesday. I wonder if Bernie will join in the Trump campaign line of attack that Biden is simply losing it.
To be fair Trump picked that up from Castro, Booker, and a bunch of people on MSNBC. I think even Andrea Mitchell of all people pointed out his apparent decline months ago.
|
I don't think Bernie will go after Biden's awareness, but I do think Bernie is keeping in mind it is a way he can distinguish himself in this next debate kinda thing. And it gives him reason to stick around just in case Biden crashes hard. Like if Biden actually did forget where he was or something, it would likely end it.
|
On March 10 2020 03:06 Mohdoo wrote: I don't think Bernie will go after Biden's awareness, but I do think Bernie is keeping in mind it is a way he can distinguish himself in this next debate kinda thing. And it gives him reason to stick around just in case Biden crashes hard. Like if Biden actually did forget where he was or something, it would likely end it.
Bernie is way too nice (maybe slightly empathetic?) to Biden to go after him for it. I think it was on healthcare when Castro pointed it out that Bernie actually reached over to help him remember what he said (in what under most other circumstances would be an understandable confusion).
|
Canada8988 Posts
On March 10 2020 03:10 GreenHorizons wrote:Show nested quote +On March 10 2020 03:06 Mohdoo wrote: I don't think Bernie will go after Biden's awareness, but I do think Bernie is keeping in mind it is a way he can distinguish himself in this next debate kinda thing. And it gives him reason to stick around just in case Biden crashes hard. Like if Biden actually did forget where he was or something, it would likely end it. Bernie is way too nice (maybe slightly empathetic?) to Biden to go after him for it. I think it was on healthcare when Castro pointed it out that Bernie actually reached over to help him remember what he said (in what under most other circumstances would be an understandable confusion).
He also probably don't want to hurt Biden, in case he end up the nominee, 2016 has to loom over his conscience a bit. I'm not saying he saying he did anything wrong, but give Hillary a free coronation at the convention instead of months of debate and she probably enter the general election much stronger. Again, it's suppose to be a democracy, I'm not arguing he shouldn't have run, but on a personnal level it's probably a bit daunting.
Also, if Biden actually physicly or mentaly break down, I don't know what happen. We will know more tomorrow night, but right now it's looking like Sander will underperform quite hard compared to 2016, I don't know if he would have the nomination anyway. Others can always "reactivate" their campain... In any case it would be a bloody mess
|
On March 10 2020 02:34 GoTuNk! wrote:Show nested quote +On March 10 2020 02:14 Danglars wrote: I thought their model would be better at predicting the relative strengths from past polls instead of reacting to changes in relative strengths from polls I can read myself.
But this race has been so unpredictable. After Biden fell 20 points in Iowa in a span of two weeks, I never saw him bouncing back. Buttigieg and Klobuchar were just so much better than him in debates and occupying a relatively moderate position compared to Bernie. I thought their own interests would propel their campaigns through to Super Tuesday to see if one would be the clear frontrunner AND THEM pressure the other two to drop out and support the leader.
Instead, the guy that is having trouble stringing sentences together in each new week of video got their endorsement prior to Super Tuesday which changes the path of everything. Bernie’s finally directly making the case that he’s profoundly different than Biden, but it all looks like too little too late. Rhetoric on social security cuts and stuff like that should’ve been done every day from South Carolina too present day, not just after Super Tuesday. I wonder if Bernie will join in the Trump campaign line of attack that Biden is simply losing it. I think he knows it’s unwise to criticize someone born a year after him for age-related senility. A repartee of style “if I’m losing it, you’re right around the corner grandpa” would be deserved, and hard-hitting.
Secondly, they have another debate soon with fewer people on stage. Bernie just has to hit him on issues, and voters will hear if Biden’s responses are at all coherent.
|
On March 10 2020 04:21 Danglars wrote:Show nested quote +On March 10 2020 02:34 GoTuNk! wrote:On March 10 2020 02:14 Danglars wrote: I thought their model would be better at predicting the relative strengths from past polls instead of reacting to changes in relative strengths from polls I can read myself.
But this race has been so unpredictable. After Biden fell 20 points in Iowa in a span of two weeks, I never saw him bouncing back. Buttigieg and Klobuchar were just so much better than him in debates and occupying a relatively moderate position compared to Bernie. I thought their own interests would propel their campaigns through to Super Tuesday to see if one would be the clear frontrunner AND THEM pressure the other two to drop out and support the leader.
Instead, the guy that is having trouble stringing sentences together in each new week of video got their endorsement prior to Super Tuesday which changes the path of everything. Bernie’s finally directly making the case that he’s profoundly different than Biden, but it all looks like too little too late. Rhetoric on social security cuts and stuff like that should’ve been done every day from South Carolina too present day, not just after Super Tuesday. I wonder if Bernie will join in the Trump campaign line of attack that Biden is simply losing it. I think he knows it’s unwise to criticize someone born a year after him for age-related senility. A repartee of style “if I’m losing it, you’re right around the corner grandpa” would be deserved, and hard-hitting. Secondly, they have another debate soon with fewer people on stage. Bernie just has to hit him on issues, and voters will hear if Biden’s responses are at all coherent. + he said time and time again that he wont attack Joe personally, he's going after his record and his stances
|
but give Hillary a free coronation at the convention instead of months of debate and she probably enter the general election much stronger
This line of argument to me is just bullshit meant to keep you in line the same as "Vote blue no matter who". If Sanders drops out early there's no reason for the party to compromise with him, he will have no leverage.
A candidate only looks worse entering the general after a long primary if she makes herself look bad in the primary. Otherwise everything they will get in the primary is the same as what they'll get in the general.
Obama-Hillary had a really drawn out primary (with Hillary running one of the darkest primaries ever) and it that turned out pretty well for him.
|
On March 10 2020 04:32 Logo wrote:Show nested quote +but give Hillary a free coronation at the convention instead of months of debate and she probably enter the general election much stronger This line of argument to me is just bullshit meant to keep you in line the same as "Vote blue no matter who". If Sanders drops out early there's no reason for the party to compromise with him, he will have no leverage. A candidate only looks worse entering the general after a long primary if she makes herself look bad in the primary. Otherwise everything they will get in the primary is the same as what they'll get in the general. Obama-Hillary had a really drawn out primary (with Hillary running one of the darkest primaries ever) and it that turned out pretty well for him.
Republicans will never find less things to complain about than the primary. If anything, the primary gave the Clinton campaign a trove of data indicating extreme anti-elite sentiment across the country. The primary should have been what saved us from Trump. Instead, the most arrogant campaign of all time managed to lose.
|
|
On March 10 2020 02:14 Danglars wrote: I thought their model would be better at predicting the relative strengths from past polls instead of reacting to changes in relative strengths from polls I can read myself.
But this race has been so unpredictable. After Biden fell 20 points in Iowa in a span of two weeks, I never saw him bouncing back. Buttigieg and Klobuchar were just so much better than him in debates and occupying a relatively moderate position compared to Bernie. I thought their own interests would propel their campaigns through to Super Tuesday to see if one would be the clear frontrunner AND THEM pressure the other two to drop out and support the leader.
Instead, the guy that is having trouble stringing sentences together in each new week of video got their endorsement prior to Super Tuesday which changes the path of everything. Bernie’s finally directly making the case that he’s profoundly different than Biden, but it all looks like too little too late. Rhetoric on social security cuts and stuff like that should’ve been done every day from South Carolina too present day, not just after Super Tuesday.
Their model over-corrects if someone over-performs their projected result, or vice-versa. Like, it gives an extra benefit of the doubt regarding of current polling aggregate, if that candidate overperformed their polls in the previous round (i think).
it also takes endorsements into account, very much, and it's not wrong to do so. If you scroll back a few pages, people were calling out their model for swinging way the f over to Joe just before ST, saying how it cannot be taken seriously, if it swings like that in 48 hours. But those 48 hours introduced a lot of new data so it reacted to that according to plan. But lo and behold, Joe even over-performed their adjusted, very favorable ST predictions, cuz indeed those dropouts and endorsements + the boost from over-performing in SC really did propel Joe forward. Likewise, if for some reason Sanders beats his pollnumbers significantly in the next round, the model will overcorrect (probably) in his favor.
|
|
|
|