Obamacare moderate emerges from the Louisiana jungle, but why?

Sam Knight, a smart guy who occasionally holds down the Washington Monthly's  Political Animal Blog for the weekend, notes some interesting election results in a Louisiana Congressional election on Saturday. Republican candidate Vance McAllister easily defeated fellow Republican Ned Riser, who was backed by tea-party groups, 59.6 percent to 40.4 percent

Knight argues that the election shows that the tea party is losing sway in its scorched-earth war against the Affordable Care Act, as McAllister suggested that he wanted to work with Democrats to improve the law, and hinted that these stances may show a recognition among GOP voters of the reality of Obamacare. Riser had the backing of other establishment Republican groups as well, while McAllister self-financed his campaign and earned a celebrity endorsement from Duck Dynasty reality-TV star Phil Robertson.

I hope Knight is right, but I suspect that there are some funky electoral dynamics at work here that may explain McAllister's stances.

Louisiana uses a two-stage "jungle primary" system in which a wide number of candidates jump into the first round, and if none win a majority, the top two candidates advance to a run-off. Riser finished first in the primary with 32 percent of the vote, while McAllister had 17.9 percent. (primary results here)

McAllister had more ground to make-up, but he had several avenues to get it:

1. He could try to pick up the votes from the other GOP candidates who got knocked out of the run off, whose support made up 18.2 percent of the votes.

2. He could try to pick up support from voters who supported the Democratic candidates -- none of whom made the run-off, who collectively collected 29.9 percent of the votes.

The math looks pretty obvious here. McAllister can probably keep his primary supporters, pick up at least a sliver of other supporters from eliminated GOP candidates and grab almost all the Democratic voters by running conservative on most issues (God and guns in this case) while speaking some sense on the side about health care.

He was able to increase his vote total by 36,000, while Riser only increased his by 3,800. Turnout only decreased from 21.2 percent to 18.3 percent of registered voters, so my theory at least seems plausible. (Since the results from the run-off aren't official, we don't know official turn out figures from partisans in either party. In any case, in this part of Louisiana those numbers are only marginally useful for our analysis, as Democrats still outnumber the GOP, but often tend to vote for Republicans in large numbers.)

I read this as a short-term tactical move to attract moderate and a few liberal voters in a two-way general election who liked the opportunity to keep a pure tea-partier out of office. Notice how as a federal office holder, he'll have no say in whether Louisiana will expand Medicaid, essentially making his election position a no-cost proposition. He'll have plenty of time to establish his hard-right credentials over a year in Congress during which he'll get to vote "no" on lots of things. Next year as an incumbent, he can run on that record and discard the Democratic part of his coalition.

Finally, as a self-funder, he's not as dependent on either tea-party or other GOP party organizations to put together a campaign, so he might be able afford to drift a bit from party orthodoxy on one issue. 

Incidentally, the Christian Science Monitor article that Knight cites hints at this dynamic as well, though Knight doesn't go into it:

Riser and McAllister largely agreed on many issues. Both opposed abortion, favor strong gun rights and criticize the levels of federal spending and debt.
Their sharpest distinction rested with President Barack Obama's signature health care law.

Both opposed the health overhaul, but Riser wanted only repeal, saying the law will harm businesses and families and can't be fixed.

McAllister said repeal had no chance with Democrats leading the Senate and White House, so he said Congress should work to improve the law. He also wants Louisiana to expand its Medicaid program to give insurance to the working poor, an expansion that Riser opposed.

The positions put McAllister at odds with some tea party supporters but generated support from Democrats who had no candidate of their own in the runoff.

(Italics mine)

Again, I'd love to be wrong here, but I think my theory is just as plausible as Knight's with the data we have.

The more general question is what sort of impact the "jungle primary" has on results, but I don't know too much of the political science on it -- though France has had essentially the same system for years with some odd results from time to time.