2012 Election Model Post-Mortem: Exactly right at 332 EV

07 Nov

2012 Election Model Post-Mortem: Exactly right at 332 EV

Richard Charnin
Nov.7, 2012

Obama won by 5 million recorded votes with 332 EV. But he did much better than that, just as he did much better than his recorded 9.5 million margin and 365 EV in 2008.

The recorded result was confirmed in the 2012 Election Forecast Model. Obama had 332 electoral votes based on the recorded vote total – not the True Vote. Obama was able once again to overcome the built-in fraud factor, which is reflected by a red-shift in the unadjusted exit polls.

But to paraphrase what Alec Baldwin told the real-estate salesmen in the famous opening scene of the classic film Glen Garry Glen Ross: “These are the unadjusted exit polls. They are gold – but you don’t get them. They’re for NEP only”.

The red-shift did not go away, even though the unadjusted exit polls (in just 31 states this time) have not been released. The recorded vote was projected based on the pre-election LV polls. The Monte Carlo Simulation derived Obama’s 99% win probability and exactly forecast his 332 (snapshot) EV.

But the True Vote Model forecast was 55-45% and 380 EV.

The expected theoretical 321 EV was calculated as the product sum: EV= ∑ P(i)* EV (i), i =1,51 states. The probability P(i) of winning the state’s electoral vote E(i) is based on the 2-party poll projection.

Obama had 51.9% of the two-party recorded vote. To match the recorded vote, Romney needed 17% of returning Obama 2008 voters and 50% of new voters. Romney needed a 2% turnout rate advantage of returning McCain voters over Obama voters. These are implausible vote shares, indicating that Obama did much better than the recorded vote. He won the True Vote by 56-44% (two-party).

Scroll down to row 375 in this spread sheet to view the sensitivity analysis.

Pre-election pollsters anticipate the red-shift without saying so. The Likely Voter Cutoff Model (LVCM) reduces Democratic turnout in LV polls. And the polling samples are based on previous election demographics in which recorded votes are always inflated for the Republicans.

As usual, the pollsters accurately projected the recorded vote. Also, as always, they avoided a True Vote analysis. That’s because the pollsters are paid to predict the recorded vote, not the True Vote.

Election Model Forecast; Post-election True Vote Model

2004 Election Model (2-party shares)
Kerry 51.8%, 337 EV (snapshot)
State exit poll aggregate: 51.7%, 337 EV
Recorded Vote: 48.3%, 255 EV
True Vote Model: 53.6%, 364 EV

2008 Election Model
Obama 53.1%, 365.3 EV (simulation mean);
Recorded: 52.9%, 365 EV
State exit poll aggregate: 58.0%, 420 EV
True Vote Model: 58.0%, 420 EV

2012 Election Model
Obama Projected: 51.6% (2-party), 332 EV snapshot; 320.7 expected; 321.6 mean
Adjusted National Exit Poll (recorded): 51.0-47.2%, 332 EV
True Vote Model 56.1%, 391 EV (snapshot); 385 EV (expected)
Unadjusted State Exit Polls: not released
Unadjusted National Exit Poll: not released


Posted by on November 7, 2012 in 2012 Election


Tags: ,

9 responses to “2012 Election Model Post-Mortem: Exactly right at 332 EV

  1. davidgmills

    November 7, 2012 at 12:10 pm

    Just the male/female exit poll should be enough to dispel any notion that the recorded vote was correct. Just curious, what did the male/female exit poll predict as the true vote?

  2. davidgmills

    November 7, 2012 at 1:27 pm

    Do they adjust male/female ratios and voting percentages to match the recorded vote as well? See:

    I am curious because there is a Huff Po blog that has these numbers differently:

    The numbers from CNN for women have come down from what they were reported earlier, if this blog post was correct when written.

  3. davidgmills

    November 9, 2012 at 10:14 pm

    How can they call it an exit poll if they just get to fudge any numbers they want? After they have their 26535 respondents and their given answers how do they change these answers and still call it an exit poll? But really, how could they change gender? If 54% of the respondents are women do they just get to decide it was 53%? And if 56% of the women said they voted for Obama do they get to say only 55% voted for him?

    • Richard Charnin

      December 7, 2012 at 1:34 am

      Changing the percentage from 53/47% to 54/46% has virtually no effect on the total vote share; they had to change the Romney and Obama vote shares of males and females.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: