We have got used to polls being so reliable when it comes to predicting election results in the US that actual elections seem to be just a pro forma exercise to validate the polls. So last evening’s result in Virginia, where Republican House majority leader Eric Cantor lost his primary race was shocking.
I had heard some minor rumblings about a possible upset during the past week but did not pay it much attention since the media often like to make races appear closer than they are in order to drum up interest and readership. The classic case was the 2012 presidential election when the polling for some time predicted a convincing win for Barack Obama but the media talking heads and pundits were breathlessly treating it as a competitive race and some even had Mitt Romney winning easily.
So what happened in Virginia? Cantor had good name recognition and outspent his rival by a huge margin, $5.2 million to a measly $120,000, two things that usually lead to victory. Cantor had no scandals to deal with or major political missteps. And yet while his internal polls had shown him with a lead of over 30 points, he ended up losing 56-44%, a massive swing.
I don’t think this signals a major shift in the Republican party. One likely mundane reason for this outcome is that turnout in primary elections, especially in years that do not have a presidential election, tend to be very low and thus far more volatile. In yesterday’s primary, just 65,000 people voted. In the 2012 general election, 382,000 people voted in that same district of whom 223,000 voted for Cantor. This can be compared to just 29,000 votes that he got yesterday, while his opponent David Brat got 36,000 votes. In the 2012 primary election, Cantor won easily with 37,000 votes out of just 47,000 cast. The fact that more people voted in the 2014 primary than in the 2012 primary (confounding the usual off-presidential year downturn) suggests more that Brat managed to get some traction among the voters than that Cantor lost support.
I am not sorry to see Cantor lose. He seemed to be pretty much a snake-oil salesman. But I don’t expect much change in the party as a result of this. Brat is supported by the Tea Party but the Republican party establishment, of which Cantor is a member, has embraced most of the Tea Party agenda and tactics.
But this result will give huge boost to both the Tea Party and the media.
It may also show, to quote Mark Twain, that rumors of the death of the teabagger movement are somewhat exaggerated.
It may be that an exciting primary here has attracted a lot of non-republicans to vote on the Republican Party ballot for just this one day, which Virginia allows. Thus, for example, maybe Brat got only 8,000 votes from tea party Republicans, but also got 28,000 votes from Democrats and independents who would have skipped the primary if not for their opposition to Cantor. There was an effort in the district to encourage this. So it might not mean anything for tea party support, but just that Dems there saw a way to oust Cantor, and did so.
Here’s a hint of Brat’s libertarian-propagandist career:
http://www.vox.com/2014/6/10/5798904/the-bank-funded-libertarian-curriculum-david-brat-taught-at-randolph
Another “professor” running a propaganda organ funded by big business. This should raise some serious questions about his “outsider” and “anti-establishment” schtick.
The most parsimonious explanation is that the polls of Republican voter preference and turnout were completely correct, and that Cantor won by a margin of 10-30% among Republicans, just as predicted. But if nobody polled non-Republicans about any plans to vote by crossing party lines for one day, to make their vote count in the most significant way, then we would see the observed result and still fit the polling predictions.
In this scenario, Brat only needed an extra 10,000 Cantor-hating non-Republicans to cross over, out of 300,000 voters who otherwise would not have voted in this Republican primary. That’s only asking 3% of non-R’s to do this to stop Cantor. That seems most plausible to me.
Anecdotal of course, but I do have non-Republican FB friends in VA who did just this and crossed party lines in the primary to oust Cantor on the assumption that this guy Brat is easier to defeat in the general. We’ll see if it works.
I would really like to believe that the Dems were successful in causing grief for Cantor. However, per the Washington Post, turnout was highest in the strongest Republican precincts and lowest in the Democratic ones.
http://www.washingtonpost.com/blogs/the-fix/wp/2014/06/11/did-democratic-votes-doom-eric-cantor/
One of the commenters had this to say:
fburgdave
7:36 AM PDT
“As a Chief Election official in one of the Cantor’s precincts, I didn’t see higher turnout from democratic voters. And included in the 4 localities that Cantor did win is Richmond city, which is heavily Democratic. No, Cantor lost because he forgot the basic rule of ANY election, which is get YOUR supporters out by reminding them there IS an election. There were no signs, no advertising, no robo-calls, or anything else that most candidates do when they are in an election. He took it for granted he would win, and didn’t expect teh other guy to be able to get out HIS supporters.”
Next question: Will Brat’s Tea Party taint be sufficient to motivate Dem voters to go the polls when the time comes? And how likely is it that the half of Republicans who don’t watch Fox or listen to AM radio will turn out to vote for Brat?
If Cantor is Right, then Brat is Even Righter (sic). He’s a nightmare of theocratic idiocy in his policies, Randian Christofascist, basically.
According to Washington Post’s ‘The Fix’, http://www.washingtonpost.com/blogs/the-fix/wp/2014/06/11/did-democratic-votes-doom-eric-cantor/, his loss was not due to Democrats voting for his opponent in the primary.
“Cantor received nearly 8,500 fewer votes this year than he did in the 2012 Republican primary, a drop that was larger than Brat’s 7,200-vote margin of victory. “
Following on from the (convincing) rebuttal of the hypothesis that crossover voters were responsible for this result, does anyone know of any clear cases of such a thing happening? I’m sceptical about whether many voters are ever that strategic.
Seems to me the simplest explanation is that the pollster badly miscalculated likely/motivated voters.
“We have elections to prove that the polls were right.”
-- old political joke
Cantor paid a pollster $75,000 during the elections. But was it to make the numbers the message, or to massage the numbers?
http://gawker.com/eric-cantors-pollster-got-paid-75-000-to-be-wrong-by-4-1589269225
In the 1984 US presidential election, the early returns changed a close race into a landslide. Voters thought, “Reagan’s going to win anyway, so I might as well vote for him.”
Odds are, Cantor knew he was behind and his campaign produced phony numbers to entice people to vote for him, to make peope think he was ahead. Printing phony poll numbers won’t result in fraud charges, so there was no reason to tell the truth and no punishment for lying. He can simply claim error.
Give it a couple more years and Brat will be out too for being too liberal -- no one is pure enough for the ‘baggers.
I don’t think that there has been any real evidence that the strategy of asking the voters of one party to vote strategically in the primary of the other has ever succeeded, except marginally. If anyone has such an example, please let me know. There have been times in the past when media heavyweights like Rush Limbaugh who command a big following of loyal listeners have tried it but as far as I know, those efforts proved to be a bust.
Of course, in off-year primaries with low voter turnout like in the Cantor race that could be different. But you still would have had to have a significant public effort to get that message out and I have not read reports of any such thing.
Seems to me that it’s the more zealous voters who voted, the sort who likely consider Eric Cantor a RINO sellout.
During the 2008 campaign, I remember checking on caucuses vs. primaries proper. The former gets the voters who are more involved in politics than the latter. It was evident in how Ron Paul did: much better in the caucuses than in the primaries proper. Ron Paul’s followers have had a LOT of zeal, but they have been unable or unwilling to get more broad support for their hero, and often unwilling to recognize that that’s a problem.
I now think comments 6, 13, 14 have the key insights here. Lower turnout usually means lower turnout for moderates, not the motivated. Cantor did not have effort invested to identify his own voters. He needed to hire or inspire people to phone or knock on doors of Republican primary voters and ask them who they supported. Then he needed to send reminders to them to vote.
In his local district party convention, Cantor didn’t even succeed in identifying which of the precinct committee people there in the room were on his side, as his party chair candidate’s humiliation showed a couple of months ago.
So he spent $5 million but couldn’t ID his own voters. That made him afraid to remind anyone to vote, because of his lack of knowledge. He assumed being the incumbent would mean he’d win in a low turnout race. But it was only low turnout for him, not for the tea challenger.