Thursday, August 30, 2018

Yet More House Forecasting Models: Democratic Wins Favored and Big Democratic Wins in Reach

As we approach Labor Day, the forecasts are coming thick and fast. Sabato's Crystal Ball reports two more models, the first merely an update of Alan Abramowitz' model using current data (it predicts a 30 seat Democratic gain) and the second an elaborate 3000 simulation exercise by political scientists Joseph Bafumi, Robert S. Erikson, and Christopher Wlezien combining national polling data with district-level voting patterns (their average prediction is for a 26 seat Democratic gain).
But they also note the following:
"[Democrats] have a 54% chance of regaining the majority. A Republican win is only a little less likely (46%) but notice that there is the possibility of a big Democratic win.
The forecast requires an important note of caution. It assumes in effect that parties apply their normal level of effort across all districts. The reports of Democratic enthusiasm about newly winnable races suggest that this assumption may not be correct. For that reason the forecast here may be best considered a lower bound of Democratic prospects."
The serious possibility that Democrats could win big is stressed by Nathaniel Rakich in his latest piece on 538 about their forecasting model. These additional gains, according to their model, would likely occur in Sunbelt states like Florida, Arizona and Oklahoma (!)
"[T]he model does hint at the possibility that Democrats could make a lot more gains, both generally and in these states specifically. To win the House, Democrats need only capture all the seats where they’re at least moderate favorites, plus some of the ones where it’s close to 50-50. But our forecast has a long “tail,” which means that the model envisions plenty of scenarios in which the midterms result in an extreme outcome — say, a Democratic gain of 60 seats or more. If that happens, our forecast suggests that many of those gains could come in Arizonan, Floridian and Oklahoman districts with little Democratic pedigree. There are eight districts in the three states that the Classic version of our model rates as “lean R” or “likely R” in the average outcome, which means they’d be the first to become toss-ups or Democrat-favored in the event of a Democratic best-case scenario."
Interesting. I shall have more to say about the Democrats' "Sunbelt path" in future posts.
The Bafumi-Erikson-Wlezien team first applied our model in 2006 and were among the first to predict the return of the Democratic majority. That forecast was off by only two seats. Like most others, our model predicted a Republican takeover in 2010, underestimating the GOP swing by about 10 seats. In...

No comments:

Post a Comment

Note: Only a member of this blog may post a comment.