538 model & poll tracker thread

[ProudModerate2]

Don't talk to me about the race "tightening" until this changes:

Oh yes.
I forgot there was a grade slider (top-left corner area), to weed-out the lower-graded pollsters.
Nice. Thank you for this image.

[Landslide Lyndon]

Oh yes.
I forgot there was a grade slider (top-left corner area), to weed-out the lower-graded pollsters.
Nice. Thank you for this image.

I used it a week ago and found that out of all the polls released since the beginning of June by A+ pollsters only one or two didn't show a double digit Biden lead.

[Beefalow and the Consumer]

Oh yes.
I forgot there was a grade slider (top-left corner area), to weed-out the lower-graded pollsters.
Nice. Thank you for this image.

I used it a week ago and found that out of all the polls released since the beginning of June by A+ pollsters only one or two didn't show a double digit Biden lead.

Trump has not led in any poll graded B or higher since Biden clinched the nomination. Not a one. I count two polls where Trump and Biden are even, 19 where Biden is up 3 points or less, and 55 where Biden led by double digits.

If the roles were reversed, we'd all be resigned to another four years of Trump.

[Gass3268]

https://twitter.com/gelliottmorris/status/1298797791523868673?s=20

[GeorgiaModerate]

I have to admit that I'm enjoying the Morris-Silver feud.

[ProgressiveModerate]

Trump's chances went up by 1.5% in my model too because of that stupid Rasmussen poll! However Trump's chances are still at 14.32%

[Admiral Stockdale]

https://twitter.com/gelliottmorris/status/1298797791523868673?s=20

Isn't the fact that the model was "wrong" in 2016 mean that they adjusted it in some way - maybe it was too "confident" based on early polls and now it is more conservative?





 

[TiltsAreUnderrated]

https://twitter.com/gelliottmorris/status/1298797791523868673?s=20

Isn't the fact that the model was "wrong" in 2016 mean that they adjusted it in some way - maybe it was too "confident" based on early polls and now it is more conservative?
 

Somewhat wider confidence intervals are appropriate because of shifts in VBM and COVID-19’s potential effects on differential turnout.

[Non Swing Voter]

https://twitter.com/gelliottmorris/status/1298797791523868673?s=20

Isn't the fact that the model was "wrong" in 2016 mean that they adjusted it in some way - maybe it was too "confident" based on early polls and now it is more conservative?





 

Maybe but didn't they already adjust the underlying polls themselves to adjust for 2016?

[Annatar]

https://twitter.com/gelliottmorris/status/1298797791523868673?s=20

The model isn't based off national polls, Nate Silver has pointed this out, the model works off state polling, the national polls are only used to apply a trend adjustment so it doesn't matter what the national poll say, the 538 model just operates based off state polls.

For example on September 7 2016, Clinton was projected to win PA by 4% and had a 72% chance of doing so, today Biden is projected to win PA by 4% and has a 71% chance of doing so, if you go state by state, the margins and probability from 2016 are the same in 2020, the only difference is compared to this time in 2016 Trump is doing much worse nationally but not worse in a lot of states.

[Fmr. Gov. NickG]

https://twitter.com/gelliottmorris/status/1298797791523868673?s=20

Certainly it’s not entirely reconcilable, but one factor to remember is that the conventions were a full month earlier in 2016, which would probably imply less uncertainty about the state of the race on this date in 2016.  But it does seem like Silver is just arbitrarily building more uncertainty into his model this year.

[Ferguson97]

https://twitter.com/gelliottmorris/status/1298812940871766017

https://twitter.com/NateSilver538/status/1298816716500336641

[Crumpets]

https://twitter.com/gelliottmorris/status/1298812940871766017

https://twitter.com/NateSilver538/status/1298816716500336641

Lol. Nate can't stand that there's someone else at his level who can challenge him on his own terms.

[Landslide Lyndon]

Silver had the monopoly on this kind of electoral forecasting and of course he is annoyed now that a newcomer not only is competing for a piece of the pie but also challenges him directly about the flaws of his model.

[Alben Barkley]

https://twitter.com/gelliottmorris/status/1298797791523868673?s=20

Certainly it’s not entirely reconcilable, but one factor to remember is that the conventions were a full month earlier in 2016, which would probably imply less uncertainty about the state of the race on this date in 2016.  But it does seem like Silver is just arbitrarily building more uncertainty into his model this year.

The conventions don’t f—king matter this year. No one’s even watching them. Expecting them to have a real significant impact on the election outcome is dubious even in a normal year, let alone this year. Silver is an idiot if he thinks otherwise. It’s obvious all this “UnCeRtAiNty” BS was just his way of forcing the model to not show what the data clearly shows: Biden is the clear and overwhelming favorite to win this election, far more than Hillary ever was. Period. End of story. The discrepancy between the models then and now is absolutely irreconcilable. Nate may act like a robot sometimes, but he is very human, and therefore very much prone to both flaws and emotional bias. Which it is clear, sadly, that he has allowed to influence how he interprets the data: He doesn’t want to be accused of being “wrong” again if another 2016 takes him by surprise, so he’s hedging his bets against the data. Downright cowardly, if you ask me. And the ongoing petty Twitter feud with a man who has made much more logical, less personal arguments and has a much sounder model is not helping his case at all.

As Charlie Cook said earlier this week, the emperor has no clothes:

https://cookpolitical.com/analysis/national/national-politics/many-are-afraid-say-it-not-close-race

But sadly, Silver is among those pundits too afraid to admit it. And yeah, that’s pretty clearly what he is now: Full-time pundit. The objective statistician is long dead. Silver’s  reputation has been immensely and irreparably damaged in my eyes at least over the course of just a few weeks. I will never take anything Silver says or any projection he makes as seriously as I once did again.

[Alben Barkley]

https://twitter.com/gelliottmorris/status/1298797791523868673?s=20

The model isn't based off national polls, Nate Silver has pointed this out, the model works off state polling, the national polls are only used to apply a trend adjustment so it doesn't matter what the national poll say, the 538 model just operates based off state polls.

For example on September 7 2016, Clinton was projected to win PA by 4% and had a 72% chance of doing so, today Biden is projected to win PA by 4% and has a 71% chance of doing so, if you go state by state, the margins and probability from 2016 are the same in 2020, the only difference is compared to this time in 2016 Trump is doing much worse nationally but not worse in a lot of states.

Biden is polling better than Clinton was at this time in most swing states as well.

Also, even if there wasn’t a big difference in the state polls and even if the national polls don’t affect the model, why would there be such a large discrepancy between the overall odds for who is projected to win?

[Landslide Lyndon]

I was astonished to learn that Silver's model doesn't take into account Trump's approval rating.
This is literally the most salient number in every contest where an incumbent ran for reelection.

[wbrocks67]

I was astonished to learn that Silver's model doesn't take into account Trump's approval rating.
This is literally the most salient number in every contest where an incumbent ran for reelection.

Yeah, the fact that Silver's doesn't incorp approval rating or even rely really on national polls is really problematic to me, especially when there has been a serious lack of good state polling, on which apparently the model relies on heavily.

[Skye]

The most cursed thing about these model wars is that Biden is going to win so the discussion of which model was correct will never be answered and they'll keep fighting over it.

[Fmr. Gov. NickG]

I think the model must incorporate national polls somewhat, because Biden’s chances in every swing state dropped a point or two yesterday afternoon when the Rasmussen poll came out, even those that haven’t been polled in quite a while.

[Gustaf]

IIRC, it includes national polling by creating a trend line. So if Biden loses a point from July nationally they adjust the state polls by a point (not exactly, because there is elasticity etc but basically)

[wbrocks67]

The economist one was too far down but really, this is just getting ridiculous. Trying to factor in how 1M+ job losses this week is somehow good for Trump just seems as ridiculous as the 538 forecast incorporating NYT headlines.

https://twitter.com/gelliottmorris/status/1298991572210479104

[Del Tachi]

On national vs state polling:  National polls are of limited relevance because there isn't a national election.  The winner will be decided in the Electoral College, which is mostly awarded at the statewide level.  If you're letting national polls color your perception of statewide races, then you're fooling yourself.  Biden being +8 nationally doesn't "shift" bad statewide polls toward him, as the two are determined simultaneously.     

[Fmr. Gov. NickG]

IIRC, it includes national polling by creating a trend line. So if Biden loses a point from July nationally they adjust the state polls by a point (not exactly, because there is elasticity etc but basically)

This seems like basically the right way to incorporate national polls to me.

I don't fault them for not incorporating Trump's approval rating.  I -do- fault them for incorporating a number of other "fundamentals" that are really just there to artificially generate more uncertainty.

538 really should have just stood by their model from 2016.

[Gass3268]

So they just added a bunch of those USC polls. Biden's polling lead jumped back over 9 points and Trump's at his lowest support in over a month. Yet the model didn't move in this mornings update. I really don't understand how it works anymore.

Update: The new "social desirability bias" Trafalgar poll of Michigan affected the model more than weeks worth of national polls showing Biden up in the double digits.