538 Model Megathread

[Rules for me, but not for thee]

RIP Nate Silver

You must be logged in to read this quote.

http://www.theonion.com/article/nate-silver-gunned-down-attempting-cross-mexican-b-54595

[The Other Castro]

RIP Nate Silver

You must be logged in to read this quote.

http://www.theonion.com/article/nate-silver-gunned-down-attempting-cross-mexican-b-54595

You must be logged in to read this quote.

Lol

[Mr. Morden]

What's wrong with what I said about 0.5% being an average?  That's around the average shift you'd expect when you run a second set of 10,000 simulations with the exact same polls.  Do you disagree?

IDk, it sounded to me as you were talking about mean when you say "the average shift". The mean of shift should be zero.
Or were you talking about abs(mean) or variance/std?

I assume he means root mean square, which in this case should work out to sqrt(2) * the usual standard deviation for sample proportion.  That usual standard deviation is sqrt(p * (1-p) / N), which for p=.643 in the polls-only and N = 10000, is 0.5 percentage points.  Multiplying that by the sqrt(2) factor gives 0.7 percentage points.

A bit more math if you like:

The difference between two independent runs of a normally-distributed random variable with some mean \mu and some standard deviation \sigma is itself a normally-distributed random variable with mean 0 and standard deviation \sqrt(2) * \sigma.

What this means in practice is that, at the moment, 68% of the time the (statistical error only) shift between simulation runs should be less than 0.7 percentage points, and 95% of the time it should be less than 1.4 percentage points.

If you are seeing shifts of 1.4 percentage points more than 5% of the time that means:

1) The poll updates are actually having an effect (which is why they are put into the model in the first place, of course).

and/or

2) You're cherry-picking (consciously or not) the large shifts.

Here’s what I was wondering about though Erc:

If you were talking about the forecast for 10,000 simulations of the outcome of a single state, then yes, your math as above is correct.  However, what the simulations actually do is simulate all 50 states (albeit with heavy correlations between them) and then sum up the electoral votes for each candidate in the simulation to project an overall national winner for that run of the simulation.  Since each state has its own run of random numbers, doesn’t that bring the statistical noise down at the national level?

Now, it probably doesn’t bring it down by that much, because 1) the states are heavily correlated with each other and 2) there are only a few swing states with any realistic chance of flipping the national winner from one candidate to the other.  But it should still bring the noise down at the national level a little, right?

[Devout Centrist]

RIP Nate Silver

You must be logged in to read this quote.

http://www.theonion.com/article/nate-silver-gunned-down-attempting-cross-mexican-b-54595

Sad News from Texas

[100% pro-life no matter what]

RIP Nate Silver

You must be logged in to read this quote.

http://www.theonion.com/article/nate-silver-gunned-down-attempting-cross-mexican-b-54595

Sad News from Texas

Where did the LA Times poll go?

[Erich Maria Remarque]

Before D-lean Marist showing Clinton+1

You must be logged in to read this quote.

So we can all agree now, that the only thing that was JUNK was Gustaf's and Co analysis of 538's model

[Meclazine for Israel]

RIP Nate Silver

You must be logged in to read this quote.

http://www.theonion.com/article/nate-silver-gunned-down-attempting-cross-mexican-b-54595

 RIP Nate.

[peterthlee]

RIP Nate Silver

You must be logged in to read this quote.

http://www.theonion.com/article/nate-silver-gunned-down-attempting-cross-mexican-b-54595

 RIP Nate.

How true it is!!!
At first glance I though it was real. And after reading the whole article--FAKE FAKE FAKE FAKE FAKE FAKE FAKE!

[ursulahx]

Nate is getting salty.

---------------------------------------------------------------------
Matthew Dowd @matthewjdowd
How do sketchy states polls have more impact on "data" odds sites (like 538) in this election than actual early vote results?  Bizarre.


Nate Silver @NateSilver538
32s
Next time we'll try the tried-and-true method of making sh**t up as we go along.

One of the enduring legacies of this election will be the increasing thinning of Nate Silver's skin.

Unless Trump wins, of course, then he'll have bragging rights until the day he dies.

[jaichind]

NV leaning Trump on all 3 models again.   Trump ahead in NV FL NC by low 50's percentages in all three models right now.

[Fmr President & Senator Polnut]

NV leaning Trump on all 3 models again.   Trump ahead in NV FL NC by low 50's percentages in all three models right now.

This is beyond stupid...

[muon2]

If you use HuffPo 3-way and limit the polls to nonpartisan pollsters and likely voters the current numbers from their trend lines are Clinton 45.1%, Trump 40.7%, Johnson 5.1%. For the 2-way it gives 45.8% to 42.4% (moderate smoothing)

Why would you do any of this?

This in response to posts on both sides complaining about inclusion of partisan pollsters. HuffPo lets me filter them out so I did this as a point of comparison.

By further comparison that same exercise today has Clinton 45.2%, Trump 42.5%, Johnson 4.9% in the 3-way and Clinton 45.4% to 42.8%  in the 2-way. Clinton's share in both is essentially unchanged from my post 4 days ago.

The Trump line in my exercise is still rising but more slowly than it did last week.

3-way: Clinton 44.5%, Trump 43.6%, Johnson 4.4%
2-way: Clinton 45.3%, Trump 43.6%

The Clinton data is still very steady as it has been over the last month. Note that when Johnson is not included Hillary gets a quarter of his vote, but the rest goes to undecided. To eliminate recent noise and movement between undecided/third party I can put on more smoothing on the two-way results and she is at 46.0%. However, over time Trump continues to gain at the expense of Johnson, but the trend lines don't show him catching her by Tuesday.

[Mr. Morden]

At the top of their Friday daily podcast, the 538ers said they were getting a lot of tweet and emails about "Why did this poll with this rating push the win probabilities this much?", and in his answer Harry didn't mention statistical noise at all.  Nate himself wasn't on the podcast though.  I'm assuming that Nate has thought about statistical noise in the model, but not sure if anyone else there has.

[Ozymandias]

Nick Gourevitch ‏@nickgourevitch  1h1 hour ago
There has not been a large (2 points+) polling error that has gone against Democrats since 1996. Chart via @ForecasterEnten:

[Person Man]

Nick Gourevitch ‏@nickgourevitch  1h1 hour ago
There has not been a large (2 points+) polling error that has gone against Democrats since 1996. Chart via @ForecasterEnten:

The polls have never actually been wrong in 50..maybe 70 years.

[Mr. Morden]

Nick Gourevitch ‏@nickgourevitch  1h1 hour ago
There has not been a large (2 points+) polling error that has gone against Democrats since 1996. Chart via @ForecasterEnten:

But that's only five elections ago, so not sure it really means much one way or the other.  Small number statistics on the basis of five events isn't that compelling.

[Ozymandias]

The polls have never actually been wrong in 50..maybe 70 years.

Well, given that the goal of national polls is to predict the winner of the popular vote, they actually were wrong in 2000.

[Ricky1121]

http://projects.fivethirtyeight.com/2016-election-forecast/ohio/#now

[Lyin' Steve]

Good for him.  What are his odds of winning Nevada, Florida, North Carolina and one of New Hampshire/Colorado?  If he can't do that he's dead.

[Ricky1121]

Nevada: 51.0%
Florida: 52.5%
North Carolina: 51.9%
Colorado: 27.3%
New Hampshire: 39.5%

[Wisconsin+17]

Well, taking into account error margins and sample sizes, that's not a great confidence that the election can distinguish between a margin of less than two points. It can do that about half the time.

That would mean that half the time - Clinton would lose half the time, but the other half has Clinton winning. Would estimate 75-25 based on that measure.

[Xing]

And yet, according to the same model, he still only has a 35% chance of winning at all.

[Erich Maria Remarque]

But Trump doesn't need 2+ error. Besided that this year we don't have so much good polls this year. So there is a larger probability that polling error will be larger as well. In both sides.

I f wonly look into last "A" live polls.

FOX News                   11/1 - 11/3            Clinton +2
McClatchy/Marist         11/1 - 11/3            Clinton +1
IBD/TIPP Tracking       11/1 - 11/4            Tie
ABC/Wash Post            10/31 - 11/3         Clinton +4
CBS News/NY Times    10/28 - 11/1         Clinton +3

(2+1+0+4+3)/5 = 2.0%

Trump might win EC losing PV by 0.0-1.5 (if he is extremely lucky).

So 2% error will do! Even 1% error might do. Maybe. At least 15%, that Trump win due to polling error

[pppolitics]

There's nothing like having the FBI trying to manipulate the election in your favor.

[Lief 🗽]

Please don't sully this fine forum with the trash numbers spat out by discredited fraud Nate Silver.