Where I was wrong and where I was right

I predicted Republican Carl DeMaio would win the CA-52 race and he lost. I was wrong. I missed on a few others too. I didn't think Republican Catherine Baker had a shot in AD-16, but she's likely to win. I predicted Tony Strickland was going to win and he lost. In fairness to me, and I like being fair to me, I got more districts right. I called CA-16 as possibly going to Johnny Tacherra back in August. No one, I mean no one, even thought this district was in play. Every congressional district except one fell into my range. It appears that I may have almost exactly hit the final results in my high turnout number, even though they happened in a low turn-out election. When the dust settles I may end up hitting the margins on party on party contests in CA-4 and CA-17 exactly. I got every state senate contest right. I 3 of the 4 assembly districts flipping to the GOP right and got the 1 that flipped to the Democrats right.

But I'll let you pat me on the back for the ones I called correctly and I'll focus on two ones I got pretty wrong. I'd been touting Carl DeMaio for CA-52 as far back as March 2013, citing DeMaio winning the district in the San Diego mayoral race. He was unable to translate that. It's possible that Democrats concentrated on the areas of San Diego outside CA-52 in that race and that resulted in DeMaio doing better.

DeMaio and the Republicans romped in the primary, giving the GOP their largest margin in a competitive district. It was much larger than in 2012 and if Peters only closed the gap by how much he did in 2012, he would've lost. The margin was so big, it was unlikely Democrats would gain that much ground. I think when the final numbers are in you'll see that Democrats didn't gain 18% in any other district from the primary to the general.

There were the VBM numbers, which showed more Republicans this time than last. Most other districts correlated fairly well between VBM numbers and the final vote. When the VBMs were better in 2014, Republicans did better. When they were worse, as they were in CA-36, Republicans did worse. Yet Republicans improved in a number of districts where the VBM numbers weren't nearly as positive as CA-52.

Finally, there was that last CA-52 SurveyUSA poll. It had DeMaio leading by 12% with those already voted. Had he led by 12%, he would've won. When the votes came in, he led by 2%. I assumed that one thing SurveyUSA couldn't get wrong would be those who already voted telling people who they voted for.

Every ounce of data I had told me DeMaio would win. And that data proved to be correct in almost every other race. It was just wrong here.

It was also wrong in AD-16, a match-up between Republican Catherine Baker and Democrat Tim Sbranti. Joan Buchanan won the district by nearly 20% in 2012 and that would appear to be too large a margin to close. In fact, no Democrats had more than 55% in 2012 in any of the other districts Republicans flipped in 2014. The district was also one that Jerry Brown won in 2010. Republicans don't hold any assembly district he won then.

The primary numbers were eye popping for the Democrats. They finished with 63% of the vote. From 2008-2012 Republicans did improve in a small percentage of the districts, but never enough to close a 26% gap. I did know that one party's primary total can be inflated if that party has two competitive candidates and other has a candidate or candidates who don't invest heavily in the primary. It just usually isn't this much. When the financial numbers showed that Sbranti was outspending Baker significantly, Baker's fate seemed sealed. One thing I didn't count on was the value of Democrat Steve Glazer's endorsement of the Republican Baker. I think the primary's bitterness caused a number of Democrats to cross over for her.