Thursday, November 8, 2012

Marking My (Electoral) Beliefs To Market

In the interests of self-accountability, while I'd live to toot my own horn about getting the Presidential Election exactly right, it's worth reflecting on what I got wrong. Here are a few things I thought I knew about how elections work, but where reality has smacked me in the face to show me who's the real boss.
  • I underestimated the ability of Democratic candidates to overcome a large Romney victory in their state. Mitt Romney beat Barack Obama by almost 20 percentage points in North Dakota, yet somehow Heidi Heitkamp squeaked out a victory. Jon Tester survived in Montana. Richard Mourdock in Indiana is something of a special case, so we'll set that one aside. I had thought that the 2010 marked a permanent decline in ticket splitting. That appears to be incorrect.

    I think this means Ben Nelson should have been encouraged not to wuss out the way Kent Conrad did.
  • I overestimated the staying power of moderate Republicans incumbents who look good in early polling. I didn't think Elizabeth Warren would beat Scott Brown, let alone win handily by seven percentage points. Susan Collins may be in real trouble in 2014 if she doesn't retire or become an independent.

    Note that these two beliefs aren't contradictory, but they are definitely asymmetrical. I basically thought that a Republican moderate who survived the primary in a Blue State would be close to invincible, while a Democratic moderate in a Red State is doomed, or at least well behind the eight ball.
  • I overestimated Todd Akin's ability to overcome his crazy comments about rape/pregnancy/abortion. In Indiana, Richard Mourdock didn't have time to recover, but the polls made it look like Akin might have made it close.  Instead he got walloped.

    In general, Patty Murray deserves a lot of credit for getting recruiting viable Senate candidates in what look like a tough cycle for Democrats, all at a time (2011) when people thought Obama might be a one-term President. And no one in either party should think of any Senate seat as safe. There's a median voter in every state; the right candidate can find a way to reach them.
  • I expected turnout, even in swing states, to decline much further than it really did. Nationwide, raw turnout is down by probably about 7.5% turnout as a share of estimated eligible voters is down about 6% (there are lots of uncounted mail-in ballots on the West Coast so we don't have an exact figure), putting the vote total roughly on par with the 2004 election. I thought we'd be well below that. And in swing states, turnout sometimes went up. Update these percentages now use estimates of Voter Eligible Population in each state.
    • Iowa +0.4%
    • Nevada +0.2%
    • North Carolina -0.5%
    • Virginia -2.9%
    • Florida -2.7%
    • Colorado -4.5% (?!?!)
    • Ohio -4.6%

      In the major swing states, turnout is up or declined more slowly than the national average. Note that the voter ID laws and general election maladministraion do seem to be having an effect. The rules are most burdensome in Ohio, and then in Florida, and then in Virginia.

      (technically, the proper thing to do here is compare turnout to Voter Eligible Population at the time of the election in both states, since Nevada and North Carolina are growing while Ohio's population is flat or in decline, but I had a hard time tracking down that data).
  • I expected the demographic shape of the electorate to "revert to normal" after Obama's historic 2008 election. I thought the share of the electorate under the age of 30 would decline significantly, and the non-white share of the electorate would stay the same or decline slightly, as the lack of an opportunity to make history would mean fewer people of color would vote. Instead exit polls suggest the share of non-white voters increased again, from 26% of the electorate to 28%; similarly, the share of the electorate under 30 dropped by one percentage point while the share age 30-45 grew by one point. That's a big deal, and if it's permanent, it's something that will help Democrats unless they start losing young white voters by larger margins than they currently lose old white voters. In swing states minority turnout is up substantially--evidence that the Obama ground game was indeed vastly superior.
I'm not sure what these last two points mean going forward. Is high swing state and non-white turnout unique to Barack Obama being the President? Can the next Presidential candidate capture the same enthusiasm? Can it be replicated in midterm elections? Answering those questions will prove crucial.

1 comment:

Neil Sinhababu said...

Hey, why aren't you marking up the good ones? You got all the electoral votes right, beating me and Sam Wang.