Off the Kuff Rotating Header Image

How many votes did we think we’d need?

We’ve already established that this election saw an historically high level of turnout in the county, and an historically high level of vote support for Republican candidates. It’s clear that Republican candidates rode a wave of voters who don’t normally participate in the off year elections to across the board victory. But would Democratic levels of support that candidates got this year have been sufficient to win in a more normal year?

I did a little number crunching to try to figure that out. Here’s the first pass at it:

TO% Voters Win Total # Wins =============================== 35 671,137 308,660 64 36 690,312 317,543 64 37 709,488 326,364 50 38 728,663 335,185 28 39 747,838 344,005 6 40 767,014 352,826 0

“TO%” is turnout percent; according to the results page there were 1,917,534 registered voters for this election, up about 42,000 from 2002. “Voters”, therefore, is just 1,917,534 multiplied by the turnout number. “Win total” is the number of votes needed in a countywide election to get a majority. You will note that they are less than 50% of the number os voters. This is because not everybody actually casts a vote in these races. If you look at the countywide races from 2002 through 2008, the undervote rate for them was about 8%. It varied a bit from year to year and race to race, but overall it was pretty consistently within about a point of that. As such, “Win total” represents 46% of the number of voters, which is to say half of the 92% of voters who actually picked a candidate in these races. “# Wins” is how many Democratic candidates out of the 65 countywide races cleared that bar.

I started at 35% because turnout in 2002 was 35.1% As you can see, for turnout levels of 37% and less the Democrats did quite well. At 38%, there’s a sizeable dropoff in the number of winners, but it’s still nearly half the slate. Note that that level of turnout would have been an increase of 10.9% in absolute terms over what we experienced in 2002. By contrast, the 2008 turnout level was 8.5% higher than it was in 2004. All things being equal, that level of surge was still manageable for Democrats. At the 39% level, Republicans would have dominated but there would still have been six survivors: Loren Jackson, Diane Trautman, Katy Kennedy, Kathy Stone, David Longoria, and Bruce Kessler. At 40%, it was all under water.

I said “all things being equal” for a reason, because there was one additional difference this year: The undervote rate was lower. In 2010, about 94.5% of all voters cast a ballot in the county races. That changes the calculus as follows:

TO% Voters Win Total # Wins =============================== 35 671,137 317,112 64 36 690,312 326,172 50 37 709,488 335,234 28 38 728,663 344,293 6 39 747,838 353,353 0 40 767,014 362,414 0

Basically, it moves everything down a level, so that the slate won big at 36%, won some at 37%, held on to a few at 38%, and was wiped out above that. While we could have seen the higher turnout coming, I don’t know how anyone could have seen this. You might attribute it to the higher rate of straight ticket voting, and I think that is the best explanation, but note that in Presidential years, when the straight ticket rate is about what it was this year (62% in 2004, 64% in 2008, 66% in 2010; it was 48% in 2006 and 55% in 2002), the undervote rate was not significantly different than in the other off years.

So the question then becomes, how much turnout did the party and the candidates plan for? I had thought we’d get about 700,000 voters, at which level the slate would have done reasonably well. Beverly Kaufman predicted 750,000 voters after Early Voting was over. That would have necessitated the 8% undervote rate to avoid a complete wipeout, and still would have been a lousy showing even with it. In reality, with turnout just shy of 800,000, the district court races had about 756,000 votes in them, and the county court races had about 753,000, meaning you needed at least 377,000 to 378,000 votes to win. We sure didn’t get that. I can understand not expecting 800,000 people to show up, but it would help to know what number we did expect. I’m sure plenty of people reading this have an answer to that, so I trust we’ll get some feedback on it. For now, at least you know how the actual performance of Democratic candidates would have fared under this range of scenarios.

Related Posts:

3 Comments

  1. […] discussed before, the undervote rate in the judicial elections was about 5.5%, so clearly there was a higher rate of […]

  2. […] at work next year, Democrats ought to be in pretty good shape countywide – as I’ve noted before, Democratic turnout was pretty decent in 2010 despite the butt-kicking – but if there’s […]

  3. […] and white people are shrinking as a share of population in Harris County. Remember, Democrats had decent turnout in 2010, and despite a perceived advantage in enthusiasm last year, Republicans still lost most of the […]

Bookmark and Share