Tuesday, February 02, 2016

After Iowa: Republicans Still in Trouble

The day after the Iowa Caucus, the conventional pundit wisdom is that the Republican position improved and that the Democrats are somehow (and here things get a little cloudy and ill-defined) in trouble. This is because the conventional wisdom is 1) relative, 2) obsessed with direction, and 3) amnesiac. So the Republican result gets spun as positive, because things are relatively better for the GOP and moving in the right direction, so that's "good." We forget all about the fact that last week -- just last week! -- various Republican heavyweights were actively trying to prevent the very result that is now being hailed as Good News for the GOP. And we measure everything by expectations, rather than by objective standards.

But even if the Republicans have taken a step toward climbing out of their hole, they are still in a deep hole, with a lot of climbing yet to do. They don't have a front-runner. They don't have a clear primary field. The Republican presidential campaign has gone from Completely Doomed to merely Basically Doomed. It could be even worse for them, and sometimes has been, but don't be fooled. "Could be worse" does not mean "good."

The big success story is that Trump came in second, so the Republicans can start the parade. Maybe this is better than Trump coming in first, unless of course you remember that last week the Grand Old Party was trying to throw Ted Cruz under the bus and openly rooting for Trump to beat him in Iowa. So Cruz beating Trump is a victory condition, and Trump beating Cruz is a victory condition, which must be convenient. Unless, of course, we remember that both Trump and Cruz are horrible general-election candidates and more than half of the Iowa Republicans voted for one of them.

Part of today's thinking is that Trump, having had a setback and badly underperformed his polls, will now collapse like a cheap tent. To this I say, maybe. It's certainly true that Der Trump has not behaved like a traditional candidate, proving immune to things that would end most campaigns. On the other hand, this scenario counts on Trump behaving unlike a traditional candidate. The conventional wisdom has never been that someone who comes in second in Iowa by a few percentage points and holds a hefty in lead in New Hampshire is no longer anyone to worry about. The idea that Trump is done because he came in second in Iowa, where several polls were showing him in second place over the last month, strikes me as wishful thinking.

The Republicans can be pleased and relieved that Trump didn't roll up 35% of the vote. On the other hand, when you step way from the endlessly-adjusted expectations, you find yourself facing the fact that a psychologically troubled amateur with no ground game still got 24% of the vote. That is not good at all. And that psychologically troubled amateur, who is still leading in New Hampshire, has enough money to stay in the race all the way until the convention if he's feeling stubborn, especially because he doesn't have to pay for a ground game or even much advertising, and still take a significant slice of the vote. Trump can certainly keep someone else from building a majority. And if he actually starts paying for commercials, who knows.

Meanwhile, Marco Rubio's 23% of the vote is allegedly cause for triumphant rejoicing, because he did better than expected and because one of the "electable" candidates got into the top three. So now the whole party can coalesce around Marco, right? Well, slow down. He managed to come in third. And he only got 23% of the vote. So his ability to expand to 50+% percent is still, ah, untested. So is Rubio's alleged ability to consolidate the "mainstream" or establishment vote, since most of his mainstream-y rivals basically abandoned Iowa and are waiting in New Hampshire, where three of them are polling at 10% or better. You can't say Rubio took the crown from Bush, Christie, or Kasich, since Bush, Christie, and Kasich conceded the round and are positioning themselves for the next one. Is Rubio the mainstream boy to beat? Maybe. But he still has to do it.

And then if Rubio emerged as the Sanest Remaining GOP Candidate, there's no evidence that he could beat Trump or Cruz. Rubio got 23% of the Iowa vote. Bush, Christie, and Kasich split less than 7% between them. That's barely 30% of the Iowa Republican electorate voting for sanity or electability. Meanwhile, if you add up the vote shares for Cruz, Trump, Carson, Rand Paul, Huckabee, and Santorum, you're seeing a very solid two-thirds preference, a supermajority, for a candidate who is bananapants crazy. Even if everybody gets behind Rubio, it's not clear that "everybody" is a majority of the party any more. The Bananapants Caucus is large and it's energized and it wants what it wants.

And while we're talking about the 30% threshold, how is it that none of the Republican candidates got to 30% of their own party's vote? Even the "winner," Cruz, did not get to 28%. You can say that this is because the vote is split so many ways. But actually, the vote is split so many ways because none of the candidates are very popular. None of them can even get one third of the vote. By contrast, in 2008 all three of the leading Democrats got a larger vote share than Cruz did this time. Part of that is because the Iowa Democrats shunt caucusers to second choices, but the third-place Democratic finisher in 2008, one Senator Hillary Clinton, got 29% of her party's vote, while 2016's glorious victor Ted Cruz doesn't quite have 28% of his party backing him.

(Oh, yes and because of expectations, Hillary's whisper-thin victory, or statistical tie, is supposed to be a huge comedown. On the other hand, she got nearly 50% of the vote in a state where last time she got 29%. Disaster!)

And while we're comparing Democratic and Republican vote tallies, we can ask ourselves how many people voted for the top two Democrats. We can't say in detail, because the Democratic Caucus only releases state-delegate counts, but we can use registration data, turnout history, and the published results to estimate a ballpark figure. If it were a close comparison we wouldn't have enough data, but in this case the ballpark estimate will do because Ted Cruz is not actually in Sanders's or Clinton's ballpark. Even using the most conservative estimates, both Sanders and Clinton each collected at least 10,000 more individual votes last night than Ted Cruz did.

Just under 187,000 Iowa Republicans caucused last night, out of 650,000 registered Republicans: better GOP turnout than the last two presidential cycles, where Iowa turnout was around 20%. (Here I'm using Dennis J. Goldford's turnout date from The Iowa Caucus Project, which is very much worth a look.) Of those 187,000, Cruz garnered just over 51,000 votes.

Now, there are 700,000 registered Democrats in Iowa (and 750,000 registered independents). Goldford's data shows that the last two contested Democratic caucuses, 2004 and 2008, had a turnout of more than 23% in 2004 and just under 40%, a whopping amount by caucus standards, during the Obama/Clinton/Edwards showdown of 2008. We don't know this year's turnout, except for meaningless media anecdotes, so lets err on the small-c conservative side and say that the minimum Democratic turnout is 20%, lower than in 2004. (We could get super-conservative and set it to 18% and it wouldn't matter much.) That would mean 140,000 Democratic caucusgoers, with much lower turnout than the Republicans this year, and that sounds very low, but let's stick with it.

If 140,000 people caucused for the Democrats and split nearly in half, that's just shy of 70,000 votes for the winner AND 70,000 for the runner-up. That means the second-place Democrat had to pull at least 18,000 or 19,000 votes more than the Republican winner.

(Don't like that number? Let's set Democratic turnout to a bottom-falling-out 18%, much worse than 2004. Now we have a mere 126,000 Dem voters, giving Clinton and Sanders a mere 62,000 votes and change. Still a five-figure advantage over Cruz.)

And of course, if the Democratic turnout was actually higher than that 20% (or 18%) minimum, the difference between Sanders's vote and Cruz's vote only expands. If we find that 25% of Democrats turned out, a fairly middling number by recent standards, that would mean that Sanders and Clinton each had around 35,000 more supporters than Cruz did last night. And if the Democratic turnout was even 30% (on the high side, but far below 2008's high water mark of 39-40%), then both Clinton and Sanders collected one hundred thousand votes apiece, basically doubling Ted Cruz's total.

These aren't exact figures, because we don't have the exact figures. They are only estimates of general scale. But the difference between Cruz's support (or Trump's or Rubio's) on one hand and Sanders's or Clinton's on the other, is not a matter of exact figures. You don't need most of the decimal places. It is a comparison of scale. Even if you low-ball Clinton and Sanders, they had to swamp Cruz in the raw vote count.

So obviously, as everyone on TV concludes, things are looking pretty tough for the Democrats.

cross-posted from, and all comments welcome at, Dagblog

No comments: