Five questions I thought of, anyway. With my own answers, some of which are admittedly on the weaselish side. Feel free to discuss/disagree/ask your own questions/etc.
1. What kind of turnout are we going to have?
The short answer is “a lot”. Texas doesn’t always get to be a part of a contested Presidential primary, but when we are, we go to the polls. Dems in 2008 and Republicans in 2016 both topped 2.8 million voters – hell, more Dems voted in the 2008 primary than in the 2004 general election. I think the bidding on the Dem side starts at 3 million, with at least 500K in Harris County (we had 410K in 2008). I think 3.5 million is in play, which means a lot of first-time Dem voters. It’s going to be really interesting to see people’s voting histories in VAN after this.
2. What does this mean for all of the other races on the ballot?
It’s really hard to say. I feel like when turnout is super low, it levels the field a bit for those who are challenging incumbents or maybe haven’t raised a ton of money because the electorate is limited to the hardcore faithful, who probably know more about the candidates, or at least pay attention to endorsements and stuff like that. In a normal high turnout environment, I figure incumbents and candidates who have raised more money have the edge, since they’re better positioned to be known to the voters. In a super high turnout election, where a significant number of people won’t be all that familiar with the many names before them, who knows? I still think incumbents will be better off, but even the high-money candidates will have to fight for attention as most voters are tuned into the Presidential race. I really don’t feel comfortable making any predictions. At least the number of goofball candidates is pretty low, so even with the likelihood of some random results, there don’t appear to be any Gene Kellys or Jim Hogans out there.
That said, some number of people who vote will just be voting in the Presidential race, so the topline turnout number will be higher, maybe a lot higher, than the size of the electorate downballot. I went and looked at primary turnout in recent elections to see what this factor looks like:
Year President Next Most % Pres ====================================== 2004 D 839,231 605,789 72.2% 2004 R 687,615 567,835 82.6% 2008 D 2,874,986 2,177,252 75.8% 2008 R 1,362,322 1,223,865 89.8% 2012 D 590,164 497,487 84.3% 2012 R 1,449,477 1,406,648 97.0% 2016 D 1,435,895 1,087,976 75.8% 2016 R 2,836,488 2,167,838 76.4%
“President” is the number of votes cast in that Presidential primary race, “Next Most” is the next highest vote total, which was in the Senate primaries in 2008 and 2012 and in either the Railroad Commissioner or a Supreme Court race otherwise, and “% Pres” is the share of the highest non-Presidential total. Some people could have voted for President and then skipped to a Congressional race or some other non-statewide contest, but this is a reasonable enough approximation of the dropoff. Bear in mind that context matters as well. In 2004, none of the Dem statewide primaries were contested, which likely meant more people skipped those races. The infamous Senate primary between Ted Cruz and David Dewhurst and other lesser candidates was in 2012, which is why nearly everyone also voted in that race. All but one of the Dem statewide races are contested, though none are as high profile as 2012 R Senate – we may never see a race like that again.
So my best guess would be that if 3 million people vote in the Dem Presidential primary, somewhere between 2.2 million and 2.4 million people will then vote in the Senate and other statewide primaries. That’s still a lot, but the downballot races will have a slightly more engaged electorate as a result.
3. What about that Presidential primary?
Again, who knows? The polling evidence we have is mixed. Before the UT/Trib poll, the evidence we had said that Joe Biden was the leading candidate, though whether he has a big lead or a small lead over Bernie Sanders depends on which poll you’re looking at. Throw that UT/Trib poll in there, and maybe he doesn’t have a lead at all. Who knows?
The primaries that take place between now and March 3 will have an effect as well – candidates may gain or lose momentum before March 3. Bear in mind, though, that a whole lot of Texas primary voting will happen before either the Nevada caucus or the South Carolina primary happen, so the effect from those states will be limited. And Texas is one of many states voting on Super Tuesday, so candidates can’t just camp here, they have other states to worry about as well. They do all have campaign presences, however, with some of them having been here for months. Finally, quite a few candidates who have already dropped out, including Kamala Harris, Cory Booker, and Julian Castro will still be on the ballot and will get some number of votes. That UT/Trib poll still had Andrew Yang in it, and he polled at six percent, higher than Amy Klobuchar. There are a lot of moving parts here.
To me, the X factor in all this is Michael Bloomberg, who has been carpet-bombing the airwaves (seriously, where do I go to surrender?) and has been ramping up his field presence in a way that other candidates may have a hard time matching. He was basically tied for third or just behind third but still super close in the UT-Tyler poll, and fourth in the UT/Trib poll, in double digits in each case. I won’t be surprised if these polls underestimate his strength. I mean, he sure seems like a candidate positioned to do quite well among those less-frequent Dem voters, and if your top priority is beating Trump, he did quite well on that score in the UT-Tyler poll, too. He’s now getting some establishment support, too. To say the least, Bloomberg is a problematic candidate, and the inevitable round of scrutiny of his baggage may drag him back down, but if you’re not prepared for the possibility that Bloomberg could do quite well in Texas in March, you’re not paying attention.
4. What about the runoffs?
Three statewide races – Senate, RRC, and Court of Criminal Appeals, Place 3 – as well as 15 Congressional races have at least three candidates and could go to runoffs, plus who knows how many other downballot contests. Runoffs generally get far less attention and participation than the main event, but this could be a year where a reasonable share of the initial vote turns out again in May.
Because that’s the kind of person I am, I looked at the recent history of primary runoff turnout. Here you go:
Year President Runoff % Pres ====================================== 2004 R 687,615 223,769 32.5% 2008 D 2,874,986 187,708 6.5% 2012 D 590,164 236,305 40.0% 2012 R 1,449,477 1,111,938 76.7% 2016 D 1,435,477 188,592 13.1% 2016 R 2,836,488 376,387 13.3%
There were no statewide runoffs in 2004 for Dems (those races were all uncontested) or in 2008 for Republicans. We already know that the 2012 GOP Senate race is a unicorn, and you can see another dimension of that here. There was a Senate runoff in 2012 on the Dem side as well, and that’s the high water mark for turnout in the modern era. This Senate race isn’t that high profile, but I think there will be some money in it, and there will be some Congressional races of interest, so maybe 300K or 400K in May for Dems? I’m totally guessing, but it wouldn’t shock me if we hit a new height this year. The bar to clear is not at all high.
5. What about the Republicans?
What about them? This is basically a 2004 year for them – incumbent President, a super low-key Senate race, no other statewide races of interest, with a few hot Congressional races being the main driver of turnout. They’ll have several of those to finish up in May as well, but my guess is they top out at about a million in March, and don’t reach 200K in May. There just isn’t that much to push them to the polls at this time.