If the election were held today, I'm not sure who would win the House. First time I've been able to say that since summer of last year.
Right now my mentions is a war between "SPECIAL ELECTIONS!!!!" and "PRIMARY TURNOUT!!!!" Of course I'm cognizant of both, and think both tell us something, especially the former. 2/
Indeed, the former is why I'd still put a thumb on the scale for Ds. At the same time, while I picked up on the Ds dropoff issue in 2013, and saw the increased primary turnout in 2010, I consider these more qualitative predictors than quantitative. 3/
That is to say, we haven't used them to try to predict elections directly in the past, but rather have used them as heuristics. Until we get a few elections under our belts with models using these (which tend to be overfit/not account for covariances correctly) 4/
I'm still going to regard them as heuristics. In other words, I put a *lot* more weight on traditional metrics like presidential job approval and the generic ballot. 5/
I've had 44% job approval and a 5-point lead on the generic ballot as the point where my expectations would shift from a D win to an R win. We're getting pretty close to that. 6/
Especially when you narrow down to Registered/Likely voter polls. So, my expectation remains that Ds will take the House, but it is now a VERY weakly held one. I would not be surprised if it did not flip. And if trends continue (IF), it probably won't. 7/7
