538 Model Megathread

[IceSpear]

Nothing to see here. The email news will have no impact whatsoever. The Baghdad Bobs told me so.

[Mr. Morden]

It doesn't make any sense to only rate the accuracy of the various models on the basis of the predictions they're making on election day itself.  The models also make predictions a week before election day, a month before election day, four months before election day, etc.  The model that's most accurate on election day itself might not be the most accurate a month out.  And I think it's more interesting to look at accuracy from a month out or a week out.  After all, if you've already reached election day, then how accurate your model is is no longer a very interesting question, as you're going to know the results in a few hours anyway.

[Slander and/or Libel]

It doesn't make any sense to only rate the accuracy of the various models on the basis of the predictions they're making on election day itself.  The models also make predictions a week before election day, a month before election day, four months before election day, etc.  The model that's most accurate on election day itself might not be the most accurate a month out.  And I think it's more interesting to look at accuracy from a month out or a week out.  After all, if you've already reached election day, then how accurate your model is is no longer a very interesting question, as you're going to know the results in a few hours anyway.

This is one thing I've been wondering a lot about. I think somebody needs to develop a methodology for measuring time-based performance of these models. I've thought that maybe one metric would be showing how long before the election the model decided a correct outcome was more likely than not and didn't shift back. That has the disadvantage of treating anything over 50% as a "call," but it would be illustrative, I think. If one model wound up saying Florida was 50.3% for Obama, but half a day earlier had said it was 50.2% for Romney, then that's not a very valuable call. But if it had said for two weeks that it was likelier than not to go for Obama, we could have some more confidence in it as a predictive model.

[Mr. Morden]

It doesn't make any sense to only rate the accuracy of the various models on the basis of the predictions they're making on election day itself.  The models also make predictions a week before election day, a month before election day, four months before election day, etc.  The model that's most accurate on election day itself might not be the most accurate a month out.  And I think it's more interesting to look at accuracy from a month out or a week out.  After all, if you've already reached election day, then how accurate your model is is no longer a very interesting question, as you're going to know the results in a few hours anyway.

This is one thing I've been wondering a lot about. I think somebody needs to develop a methodology for measuring time-based performance of these models. I've thought that maybe one metric would be showing how long before the election the model decided a correct outcome was more likely than not and didn't shift back. That has the disadvantage of treating anything over 50% as a "call," but it would be illustrative, I think. If one model wound up saying Florida was 50.3% for Obama, but half a day earlier had said it was 50.2% for Romney, then that's not a very valuable call. But if it had said for two weeks that it was likelier than not to go for Obama, we could have some more confidence in it as a predictive model.

The obvious way to measure the accuracy of the models as a function of time is to take, at each time interval, all of the races that the model thinks one candidate is favored in 50-60% of the time and check if the favorite does indeed win 50-60% of the time, then the same where the model has someone as a 60-70% favorite, a 70-80% favorite, etc.  Do the probabilities at a week out match the outcomes?  Or a month out?  Or whatever time interval you want.

[Donnie]

NC just flipped to Trump on polls-plus.

[Ozymandias]

This is a pretty cool animated gif:

Josh Katz ‏@jshkatz  17m17 minutes ago
Clinton win chances over the past three months, forecast by forecast

https://twitter.com/jshkatz/status/793553664724140032

[Erich Maria Remarque]

Florida is back to Daddy in Now-Cast!

[Slander and/or Libel]

Seeing a Trump fan call him Daddy is the creepiest thing I've seen all week, so thanks for that.

[Dr. Arch]

Seeing a Trump fan call him Daddy is the creepiest thing I've seen all week, so thanks for that.

Pretty much.

[Speed of Sound]

Seeing a Trump fan call him Daddy is the creepiest thing I've seen all week, so thanks for that.

Pretty much.

Young men in desperate need of a "cool" father figure makes up 90% of Trump's online support.

[Meclazine for Israel]

Choo choo...trump trump trump trump trump trump....choo choo.

Here comes the Atlas Forum's worst nightmare.

31% and rising.

[Erich Maria Remarque]

Seeing a Trump fan call him Daddy is the creepiest thing I've seen all week, so thanks for that.

Pretty much.

Young angry white men in desperate need of a "cool" father figure makes up 90% of Trump's online support.

Fixed for you.

Choo choo...trump trump trump trump trump trump....choo choo.
Here comes the Atlas Forum's worst nightmare.
31% and rising.

But they'll rather discuss something else. Bubbbbbbble... Boom

[elcorazon]

I'm ashamed of this country

[Speed of Sound]

Seeing a Trump fan call him Daddy is the creepiest thing I've seen all week, so thanks for that.

Pretty much.

Young angry white men in desperate need of a "cool" father figure makes up 90% of Trump's online support.

Fixed for you.

We're on the internet, so I didn't think that I really needed to qualify those obvious parts.

[StatesPoll]

few weeks ago,
One of Red Avatar said, "Nate Silver is a GOP Operative."

I couldn't stop laughing.

Nate Silver is a liberal man and pro-HRC!!!!

[100% pro-life no matter what]

ME-02 flips back to Trump in all three models.  In polls-plus and nowcast, he is nearing having every state in the 266 back in his column.  His chances are nearing 30 in his best freiwal states (Colorado and New Hampshire), and in the 20s in several more.  A good PA or WI poll today would do wonders.

[jaichind]

Nowcast has NV FL NC OH IA all leaning Trump.  Some by the narrowest of margins of course.

[Speed of Sound]

Nowcast has NV FL NC OH IA all leaning Trump.  Some by the narrowest of margins of course.

Input garbage, output garbage. Only Nate could have his model claim Trump's leading NV and then release an article about how Clinton is likely to exceed her already positive NV polls.

[Devout Centrist]

Nowcast has NV FL NC OH IA all leaning Trump.  Some by the narrowest of margins of course.

Input garbage, output garbage. Only Nate could have his model claim Trump's leading NV and then release an article about how Clinton is likely to exceed her already positive NV polls.

Sabato may be more accurate this time.

[Yank2133]

Nowcast has NV FL NC OH IA all leaning Trump.  Some by the narrowest of margins of course.

Input garbage, output garbage. Only Nate could have his model claim Trump's leading NV and then release an article about how Clinton is likely to exceed her already positive NV polls.

The primary broke Nate.

[Slander and/or Libel]

Nowcast has NV FL NC OH IA all leaning Trump.  Some by the narrowest of margins of course.

Input garbage, output garbage. Only Nate could have his model claim Trump's leading NV and then release an article about how Clinton is likely to exceed her already positive NV polls.

The primary broke Nate.

The fact that he's resorting to quoting political futures markets to back his models is telling. Also the fact that he's basing his uncertainty on polling data going back to 1972 betrays that he really still wants the lesson of the primary to have been that nobody can know a thing, rather than that he just got it wrong.

[Figueira]

Clinton's odds are now lower than the Senate Democrats' odds.

[Erich Maria Remarque]

Lol. Model is basically the same as 2012..

Bad polls? I recall Nate mentioned what would be the diff live-polls (often A rating) vs others. The difference was not big. Probably. It would show Trump 25% vs now 30%

[Speed of Sound]

Lol. Model is basically the same as 2012...

Good thing the quality of polls and the nature of the electorate are the same as 2012.

[Erich Maria Remarque]

Sigh... So far other model eventually followed 538. In balance Upshot gives 5-10% less chances to Trump. But averages number nationally/states are ~the same.

But I give up. Othervise, mods would ban for repetitive posts (it was the same displeasure among Atlas last time Trump surged in polls)