538 Update Thread

[Lief 🗽]

Nate Silver tweets:

You must be logged in to read this quote.

Sounds about right.

[Joe Biden 2020]

Nate Silver tweets:

You must be logged in to read this quote.

Sounds about right.

I don't always agree with Nate Silver, but in this instance I do.

[Jake]

At this point, I'd put a lot less stock in Nate's numbers than I would've several months ago. His model is likely to be "right" if the turnout numbers are where the aggregate of all the polls place them. His model's problem is it doesn't know anything about the early voting data that's coming in. Meaning that if the early voting partisan turnout numbers (and the resultant overall partisan turnout numbers) are different than what the aggregate of the polls predicted, his model will be "wrong". I'm not at all sure whether that will be the case, but that's an extremely important factor that he's counting on polling companies to be correct about. Now polling companies could be adjusting their turnout projections based off early voting numbers, but they also might not be, especially the less savvy ones.

All in all, I'm much more worried that the pollsters are overestimating Obama's turnout than Romney's, if only because their main data point will be the crazy turnout in 2008.

[Skill and Chance]

At this point, I'd put a lot less stock in Nate's numbers than I would've several months ago. His model is likely to be "right" if the turnout numbers are where the aggregate of all the polls place them. His model's problem is it doesn't know anything about the early voting data that's coming in. Meaning that if the early voting partisan turnout numbers (and the resultant overall partisan turnout numbers) are different than what the aggregate of the polls predicted, his model will be "wrong". I'm not at all sure whether that will be the case, but that's an extremely important factor that he's counting on polling companies to be correct about. Now polling companies could be adjusting their turnout projections based off early voting numbers, but they also might not be, especially the less savvy ones.

All in all, I'm much more worried that the pollsters are overestimating Obama's turnout than Romney's, if only because their main data point will be the crazy turnout in 2008.

Well I'm not sure that D or R in a phone poll = D or R in registration or even D or R in an exit poll.  In 2004, Democrats set up unskewing blogs because they couldn't believe turnout would be even between the parties.

[Sbane]

At this point, I'd put a lot less stock in Nate's numbers than I would've several months ago. His model is likely to be "right" if the turnout numbers are where the aggregate of all the polls place them. His model's problem is it doesn't know anything about the early voting data that's coming in. Meaning that if the early voting partisan turnout numbers (and the resultant overall partisan turnout numbers) are different than what the aggregate of the polls predicted, his model will be "wrong". I'm not at all sure whether that will be the case, but that's an extremely important factor that he's counting on polling companies to be correct about. Now polling companies could be adjusting their turnout projections based off early voting numbers, but they also might not be, especially the less savvy ones.

All in all, I'm much more worried that the pollsters are overestimating Obama's turnout than Romney's, if only because their main data point will be the crazy turnout in 2008.

Was there crazy turnout in 2008 besides Black voters? What evidence is there of that?

I think Republicans think there are more of them in the country than there actually are. Democrats do have a turnout problem in some elections, which causes this perception. The fact of the matter is that if all registered voters were forced to vote, Obama would win easily with perhaps a margin close to 2008. Polls are already taking into account that a lot of soft Obama supporters will likely not show up to the polls. This is why national numbers are around even and even the state polls don't point to a race more favorable than D+2-3. Lower turnout for Democratic groups are already baked into the polls. I actually would argue Hispanics and Asians are being under sampled by the polls currently.

I vehemently disagree that there was historic turnout amongst the youth and Hispanics in 2008. You really have to show me some evidence to convince me otherwise. I am looking at the exit poll numbers and for the youth it only increased from 17 to 18% in 2008 from 2004. Maybe it goes down to 17% this year, but the main thing to watch will be the margin of victory. For Hispanics and Asians, their share of the vote barely went up in 2008 even though they are the fastest growing groups in America. This does not lead one to believe there was historic turnout amongst them. Looking at the Democratic primary results and how those groups voted, this should make sense.

[Zanas]

The figures stand now as follows :

307.2-230.8 EV (Obama +0.7, Romney -0.7)
86.3%-13.7% chance of EC victory (Obama +1.2%, Romney -1.2%)
50.6%-48.5% PV margin (Obama +0.0%, Romney +0.2%)

Nowcast

308.4-229.6 EV
88%-12% chance of EC victory
50.6%-48.5% PV margin

The chance of EC victory is second best for Obama only to that of October 4th when it was 87.1%.
The popular vote is now the same in the nowcast and the Nov 6th.
Florida flipped a bit to Romney at 55.5% chance.
NC 77.2% chance Romney, but it may decrease a bit following the last polls.
CO is 69.7% chance Obama, I wouldn't bet my money on this particular one, I don't feel it. It's Obama's most fragile state as of now.
VA seems a bit solider, at 72.6%.
NH is at 80.2%, and I finally am confident that Obama will keep it.

Smelling good.

[Yank2133]

EC-314.4-223.6
CW-91.4%-8.6%
PV-50.9%-48.2%

Obama is peaking at the right time.........

[Gass3268]

Obama has gone up by 30% since mid-October

[Lemmiwinks]

The model has Obama winning Florida, too!

[Speed of Sound]

Said on twitter that this isn't his last run of the numbers tonight, so stay tuned as they add the last few polls, etc.

[Starbucks Union Thug HokeyPuck]

Your 90% club (271)

and your 80% club (303)

[Speed of Sound]

AWall, man, if you will have nailed this election as long ago as you started publishing that map of yours, you will have surely earned top-shelf kudos.

[DemPGH]

Man, I'm just trying to be realistic because there's so much at stake (like, I actually care who sits on the Supreme Court!), but I think 303 is a very, very real possibility for Obama. FL would be beyond sweet - I hear that ground op is kicking butt. Still, waves and waves of rural folks tomorrow to offset will be tough.

[exopolitician]

The model has Obama winning Florida, too!

If he grabs Florida by a hair, I'll be a happy camper. Shame that North Carolina seems completely out of reach now though, I hoped Obama would try and pull out a win there again.

[Starbucks Union Thug HokeyPuck]

AWall, man, if you will have nailed this election as long ago as you started publishing that map of yours, you will have surely earned top-shelf kudos.

I won't say a word until the last state is called, but I do feel like I've had a good grip on this election for quite a while.  We'll see if my confidence is rewarded tomorrow.  

[Jake]

<>

Crazy turnout meaning very heavily a Democratic electorate. Turnout was 55/45 D/R among those who display a partisan preference in 2008,  50/50 D/R in 2004, 53/47 D/R in 2000, and 53/47 D/R in 1996. Any partisan turnout model based primarily on 2008 (and note I'm not claiming this is actually the case) will over-estimate D turnout. It's pretty likely that we get turnout much closer to '96 and '00.

[Fmr President & Senator Polnut]

Obama now at 92.2%

[Thomas D]

Obama now at 92.2%

So if Obama doesn't win, Nate is right up there with "Dewey Defeats Truman"

[Badger]

The model has Obama winning Florida, too!

Yep, with a mere 53% certainty.

Gonna be damn close....

[Antonio the Sixth]

This late Obama bump is awesome news, but I don't exactly get what prompted it. Did we get a streak of great polls for Obama today?

[Lief 🗽]

Obama now at 92.2%

So if Obama doesn't win, Nate is right up there with "Dewey Defeats Truman"

Oh god, I hope if Romney wins he does the Truman picture pose but with an iPhone showing 538.

[Speed of Sound]

This late Obama bump is awesome news, but I don't exactly get what prompted it. Did we get a streak of great polls for Obama today?

More confirmation in OH+Rassy shows Obama up in NH+Angus(I think?) poll out of WI+1.5% bounce in national polls+no more time for any surprises/last minute shifts?


I agree the shift seems large even so, but especially with those national numbers, he must be sensing the momentum and the calculation must be taking that into account for this to be the number, really.

[Adam Griffin]

(WHAT APPEARS TO BE) FINAL 538 RESULTS:


EV
Obama: 315.2
Romney: 222.8

PV
Obama: 50.9%
Romney: 48.2%

% OF VICTORY
Obama: 92.0%
Romney: 8.0%

STATE-BY-STATE CHANCES:

WA: 100.0%
CA: 100.0%
IL: 100.0%
NJ: 100.0%
NY: 100.0%
CT: 100.0%
RI: 100.0%
MA: 100.0%
VT: 100.0%
HI: 100.0%
MD: 100.0%
DC: 100.0%
ME: 99.9%
DE: 99.9%
MN: 99.8%
OR: 99.6%
NM: 99.5%
MI: 99.5%
PA: 98.8%
WI: 97.2%
NV: 94.4%
OH: 91.8%
NH: 86.2%
IA: 86.0%
CO: 81.8%
VA: 81.4%
FL: 53.1%
NC: 72.5%
MT: 97.7%
AZ: 97.9%
SC: 99.5%
IN: 99.6%
MO: 99.6%
GA: 99.8%
WV: 99.9%
MS: 99.9%
KS: 99.9%
SD: 99.9%
AK: 99.9%
UT: 100.0%
WY: 100.0%
ID: 100.0%
ND: 100.0%
NE: 100.0%
OK: 100.0%
TX: 100.0%
LA: 100.0%
AR: 100.0%
KY: 100.0%
AL: 100.0%
TN: 100.0%

[Lief 🗽]

Well the shift in the national polls has reduced a lot of the uncertainty. Before today (and arguably yesterday as well, IMO), there was a clear gap between the national polls (showing about a tie) and the state polls (which were more representative of Obama +2 or so). Now the national polls have shifted to match the state polls and that contradiction no longer exists.

That's my explanation anyway.

[Antonio the Sixth]

Yeah, the closing of the national-State poll gap (in the latter's favor) is a pretty big deal at this point. Anyways, that's good to take. If this map is the one we'll see on Election Night, things look really awesome.