538 Model Megathread

[Spark]

Of course she will..

[heatcharger]

> Kansas

In fairness, he gave that Zogby poll a C-, so once there's a real poll that could change a lot. But it really wouldn't surprise me if Kansas is a single digit win for Trump given how awful Brownback is.

[ElectionsGuy]

Kansas and Mississippi are some of the most competitive states? I like Nate Silver's forecasts, but just like his primary prognostications, I think these will end up totally wrong (Gary Johnson probably won't get anywhere near the 9% this forecast shows).

[President Johnson]

Let's dispel once and for all the fiction that Nate Silver makes correct predictions.

[Illiniwek]

Just a bit surprised, because I have felt Nate has been a bit bearish on Hillary of late. Still checks out though.

[DS0816]

His model of Hillary Clinton winning by +7.3, over Donald Trump, yields a few states which have Democratic tilts underperforming: Colorado and Iowa, to name two, but also New Hampshire and Nevada. None of those four were less than +1.5 above for Barack Obama in both 2008 and 2012.

The state of Washington is at +14 when I would expect it to be +16 or +17.

Details like that.

But, the most noticeable thing is that Donald Trump would apparently carry no state by greater than the +23 as indicated with Wyoming. That state, in 2012, was +41.

I think people are underestimating how an electoral map can really move. If these numbers really shifted Democratic to such levels, we're not going to see only one or two states flip from 2012 Republican to 2016 Democratic. We're going to see at least a half-dozen states do that.

[Illiniwek]

Very cool to look at. It will be fun to watch this throughout the rest of the year.

[john cage bubblegum]

Features I like from this year's site:

- It shows you the "adjusted" margin from an individual poll after the model takes into account the pollster's house effect.  (I don't think they did that last year?  I might be wrong.)

- It gives you the numbers from a state before and after the model takes into account things like demographics and fundamentals.

[Statilius the Epicurean]

80/20 is about right.

[Beefalow and the Consumer]

Clinton's closest states (polls only):

Arizona
North Carolina
Colorado
Ohio
Iowa
Florida <-- Tipping point
Virginia
New Hampshire
Nevada
Pennsylvania

Clinton's closest states (polls-plus):

Ohio
Colorado
Florida <-- Tipping Point
Iowa
New Hampshire
Virginia
Pennsylvania
Nevada

Ohio I expect to see past the tipping point.  Colorado I don't.  Iowa I especially don't.  And Pennsylvania being that far behind the tipping point is weird.  What's up with PA?

Oh, it's the Marist College and Evolving Strategies outlier polls.  WTF are they doing in there, completely jacking up his model?

[Gass3268]

Nate Silver may suck when it comes to primaries, but he is so far 100/100 in predicting how each state will go in a general election.

Here is the electoral map based on the currently projected odds:

As of right now Arizona, Georgia, Missouri, and North Carolina are the pure toss up states.

[Gass3268]

Kansas and Mississippi are some of the most competitive states? I like Nate Silver's forecasts, but just like his primary prognostications, I think these will end up totally wrong (Gary Johnson probably won't get anywhere near the 9% this forecast shows).

To be fair this will adjust with new polling (both statewide and nationally). Obviously those two states should move towards Trump.

[Beefalow and the Consumer]

Kansas and Mississippi are some of the most competitive states? I like Nate Silver's forecasts, but just like his primary prognostications, I think these will end up totally wrong (Gary Johnson probably won't get anywhere near the 9% this forecast shows).

To be fair this will adjust with new polling (both statewide and nationally). Obviously those two states should move towards Trump.

I believe his models don't just take polls in isolation.  They take polls in the context of polls in other states and form a holistic model.

For instance, say you have several polls in Mississippi, when averaged, saying Trump is only ahead by five points.  Is that statistical noise, or is it meaningful?  Well, take a look at surrounding states, or states with similar demographics.  Do their polls corroborate this?  If so, weigh that data heavily.  If not, weigh it less.  Similar states move in concert.

If his model says Mississippi is competitive, Mississippi is competitive.  I have good reasons of my own to believe it should be, so this does not surprise me.  Kansas genuinely does.  There's a massive revolt against the Brownback regime going on, but that shouldn't blow up-ticket.

Silver's data analysis is very, very good.  His rating of polls is very, very good.  His political punditry is crap.

[Crumpets]

It looks like they fixed the Trump-Obama state and Clinton-Romney state numbers. Now its:

Clinton wins at least one state Mitt Romney won in 2012 87.4%
Trump wins at least one state President Obama won in 2012 63.3%

[Xing]

Not flawless by any means, but certainly interesting.

[john cage bubblegum]

Kansas and Mississippi are some of the most competitive states? I like Nate Silver's forecasts, but just like his primary prognostications, I think these will end up totally wrong (Gary Johnson probably won't get anywhere near the 9% this forecast shows).

Kansas and Mississippi are some of the closest states in the "polls only" model because that's what the very limited/questionable polling indicates.  Especially this far out, it's better to stick to the "polls plus" model, which is the model that factors in the economy and various state fundamentals, and it's what he's used in the past.  

The "polls only" model is for the benefit of people who might have issues with those other factors FiveThirtyEight takes into account besides polls, and so it's a lot closer to a simple average of the polls.  But there's still some adjustments for things like pollster house effect and for demographics in states with little-to-no polling.

The general election is where FiveThirtyEight shines.  Happy to see that the forecast is up and running.

[Beefalow and the Consumer]

It looks like they fixed the Trump-Obama state and Clinton-Romney state numbers. Now its:

Clinton wins at least one state Mitt Romney won in 2012 87.4%
Trump wins at least one state President Obama won in 2012 63.3%

Seems about right.

It's hard to model correlations between regions.  OH and CO are mostly independent variables, but CO and NV are closely linked.  Assuming OH and CO are completely independent (they aren't), Clinton has about a 50% chance of winning them both.  If she does win them both, there's a good chance she'll retain the rest of Obama's 2012 states, but a small chance Trump could pick off a FL or VA.

[pbrower2a]

Other than AZ, it looks ok ...

I'd give AZ to Trump in a close race, because the state's laws that will block Latinos from voting.

Not with the Obama Administration doing everything possible to protect the voting rights of blacks, Latinos, and First Peoples in Arizona. Nobody is going to get a chance to do anything squirrely with the vote in Arizona.

More troubling to Donald Trump could be that many white Anglo Arizonans have a Mexican-American in the family... and there could be many non-Latinos deciding that Trump's ugly statements about Mexican-Americans badly fit the reality of an in-law, co-worker, etc.    

[Brittain33]

It's funny that Polls-Plus shows that the probability of matching the 2012 map is 0.4% when that's exactly what is predicted now. I understand why that is—all those states which could theoretically go either Clinton or Trump—but it's still funny.

[Beefalow and the Consumer]

Most likely states to flip (polls only):

NC   59.00%
AZ   53.90%
MO   47.30%
GA   41.90%
SC   38.50%
KS   32.60%
CO   31.50%
MS   31.10%
SD   30.00%
IN   29.60%

Most likely states to flip (polls-plus):

NC   46.00%
OH   40.10%
CO   38.80%
AZ   37.70%
IA   37.10%
FL   36.70%
NH   32.90%
MO   31.80%
VA   27.70%
NV   27.00%

[RJEvans]

Most likely states to flip (polls only):

NC   59.00%
AZ   53.90%
MO   47.30%
GA   41.90%
SC   38.50%
KS   32.60%
CO   31.50%
MS   31.10%
SD   30.00%
IN   29.60%

Most likely states to flip (polls-plus):

NC   46.00%
OH   40.10%
CO   38.80%
AZ   37.70%
IA   37.10%
FL   36.70%
NH   32.90%
MO   31.80%
VA   27.70%
NV   27.00%

Just eyeballing it, it looks like the most likely Electoral College outcome in the model is Clinton 368-170, with AZ, NC and MO flipping. The model is also bullish on Clinton 400+ with a relatively high probability of 483-55.

[DS0816]

The people who underestimated Donald Trump in the 2016 Republican presidential primaries, and yet they make their living in politics (like as pundits), should not be forgiven.

ABC News's Matthew Dowd was an exception. How many more? (Kind of rhetorical. I'm not that interested.)

[Free Bird]

Nate Silver predicts Drumpf will never win the Republican nomination!

[Free Bird]

Making any definitive prediction before the conventions and debates is foolish

[Wells]

People appear to be confusing Nate Silver himself and Nate Silver's model. During the primaries, Nate's models were generally bullish on Trump, while Silver himself. . .