538 Model Megathread

[Person Man]

I would say his chances lie between 15 and 35%.

[Ricky1121]

I would say his chances lie between 15 and 35%.

Where?

[Ashley Biden's Diary]

More like 60%.

[Rand]

Cool, he can have it. Give him Florida, North Carolina, Iowa, and Nevada as well and he still comes up short. That's pretty sad.

[Speed of Sound]

Nate is NOT AMUSED!

You must be logged in to read this quote.

You must be logged in to read this quote.

[Ebsy]

DOGSWEAT.

[The Other Castro]

Ryan GrimVerified account
‏@ryangrim
@NateSilver538 Sick burn, non-bro.

AWWWWWSHEEEEEEEEEET it's on.


NERD FIGHT!

[The Other Castro]

Tapping out due to drowning in dogsweat, Nate Silver calls in his wingman Harry Enten:

(((Harry Enten))) ‏@ForecasterEnten  12s13 seconds ago
You can agree or disagree with the model. (I have some qualms myself.) But it's empirical & tested. To say otherwise is idiotic.

[Erich Maria Remarque]

Lol, negative campaigning is on even among poll aggregators


S**tshow

[The Other Castro]

Now Nate is trying to pin the Great Recession on Ryan Grim. Truly a sight to behold.

Nate Silver Verified account
‏@NateSilver538
Not just an issue in elections models. Failure to understand how risks are correlated is part of what led to the 2007/8 financial crisis.

[Ebsy]

Now Nate is trying to pin the Great Recession on Ryan Grim. Truly a sight to behold.

Nate Silver Verified account
‏@NateSilver538
Not just an issue in elections models. Failure to understand how risks are correlated is part of what led to the 2007/8 financial crisis.

lol, that's one way to frame it.

[The Other Castro]

Ok yeah Ryan was kind of wrong on some things, though I still don't really agree with Nate's adjusted polling methods.

[heatcharger]

That HuffPo writer is out of his league on this one, but wow Nate is sweating so badly -- honesty it's disgusting. My real qualm is that he's including so many junk polls that it's skewing the averages, which makes quality pollsters seem like they have a huge D-house effect.

[Speed of Sound]

The best part of this isn't that Nate's defense of his model is wrong- cause he's right on that- it's how obvious it is that he knows it won't be right.

[Erich Maria Remarque]

Ok yeah Ryan was kind of wrong on some things, though I still don't really agree with Nate's adjusted polling methods.

Explain?

Do you not like their trend line adjustment? Or house effect adjustment?

That HuffPo writer is out of his league on this one, but wow Nate is sweating so badly -- honesty it's disgusting. My real qualm is that he's including so many junk polls that it's skewing the averages, which makes quality pollsters seem like they have a huge D-house effect.

Trend line adjustement is mostly responcible for that, not D-house effect.

But it would be better, if one could see, how adjustment works for each poll by itself.

[Speed of Sound]

You must be logged in to read this quote.

sweatin'dog.gif

[Erich Maria Remarque]

You must be logged in to read this quote.

sweatin'dog.gif

I_want_confirmation_so_i_seek_approval.mpg

[Speed of Sound]

You must be logged in to read this quote.

sweatin'dog.gif

I_want_confirmation_so_i_seek_approval.mpg

IenjoythismememorethanIshould.xml

[The Other Castro]

And Matthew Dowd enters the mix! We got a ballgame folks:

Nate SilverVerified account
‏@NateSilver538
The problem is that we're doing this in a world where people—like @ryangrim—don't actually give a sh**t about evidence and proof.

To Nate:
Ryan Grim ‏@ryangrim  1m1 minute ago
@NateSilver538 ur trend line adjustment was certainly theoretically defensible 18 mos. ago, but you can't ignore the wild effect it has had

To Nate:
Matthew Dowd ‏@matthewjdowd  3m3 minutes ago
@NateSilver538 @goodnewsgoddess @ryangrim couldn't that be said of you Nate in the GOP primary? Data said trump would win then.

To Ryan:
Nate Silver ‏@NateSilver538  2m2 minutes ago
@ryangrim: Very professional of you not bother asking me for comment ahead of time. So unbelievably lazy.

[Speed of Sound]

Oh Nate.....

[heatcharger]

It's not his fault there have been so many garbage polls released recently -- but his model is supposed to weigh them accordingly based on their rating. That hasn't happened the way it's supposed to because the sheer volume of junk polls is flooding the model in a way that didn't happen in 2012.

At least in pollster.com you can modify which polls are included.

[Maxwell]

It's not his fault there have been so many garbage polls released recently -- but his model is supposed to weigh them accordingly based on their rating. That hasn't happened the way it's supposed to because the sheer volume of junk polls is flooding the model in a way that didn't happen in 2012.

not to mention how highly he's weighed such obvious junk like GOOGLE CONSUMER SURVEYS or Gravis

[Blackacre]

Yeah, his model could use some updating next time to be more selective with its polls. But I still like him and the model (otherwise I would have abandoned it) and I understand his concerns

[The Other Castro]

I love a good statistician nerd fight. Meanwhile Harry Enten is just chillin:

(((Harry Enten))) ‏@ForecasterEnten  13m13 minutes ago
So many subtweets to be had.

[Erich Maria Remarque]

It's not his fault there have been so many garbage polls released recently -- but his model is supposed to weigh them accordingly based on their rating. That hasn't happened the way it's supposed to because the sheer volume of junk polls is flooding the model in a way that didn't happen in 2012.

not to mention how highly he's weighed such obvious junk like GOOGLE CONSUMER SURVEYS or Gravis

It hasn't. The main "issue" here is that 538 adjusts for trend line. Do you think it is stupid?