Lots of buzzing political junkies about the release of fivethirtyeight’s forecast of the U.S. House races, which gives Dems about a three in four chance of taking the majority. The forecast is bearish on Dems’ chances in Arkansas, however, predicting low odds of success for Clarke Tucker and other Democratic challengers.

I guess we might as well get it out of the way first that Nate Silver’s website gave Hillary Clinton a 71 percent chance to win the presidency in November of 2016, versus a 29 percent chance for Donald Trump. We know how that turned out. But these are probabilistic models — if they work, the underdog with the 30 percent chance should win some of the time (and fivethirtyeight was higher on Trump’s chances than most). And keep in mind that the national polls were actually reasonably accurate — Trump just won where it counted and took the electoral college. All that said, predictions are hard, especially about the future! Douse the following with however many grains of salt you think appropriate:


* The forecast gives Democrats a 75 percent chance to take the House, which requires flipping 23 seats. But the range of plausible outcomes is quite wide. That big part in the middle of the graphic above— covering 80 precent of possible outcomes — goes from 14 pickups (short of a majority) all the way to a historic wave of 57 pickups. Some analysts have argued that with so many tossups, things will swing one way or the other en masse, so the most likely outcomes are huge gains for Dems or the GOP holding on to the House.


Silver concurred, noting that the fivethirtyeight model finds that there’s a 60 percent chance that either Democrats win 40+ seats or fail to take the House at all.

* One finding that remains kind of astonishing is that Democrats would need to win the national popular vote by 5-6 points to be favored to take over the majority of seats in the House.


* While Silver’s model likes Democrats’ chances nationally, the forecasts are gloomier for the Ds in Arkansas:

The Democrat with the best shot to pull off a win, state Rep. Clarke Tucker, is given just a 22 percent chance to win versus a 78 percent chance for incumbent U.S. Rep. French Hill. The model predicts a close race, with the estimated voteshare 51 percent for Hill, 46 percent for Tucker and 3 percent for Libertarian Joe Ryne Swafford.

Meanwhile, incumbent Republicans Bruce Westerman, Rick Crawford, and Steve Womack are all considered essentially locks by the fivethirtyeight forecast, with predictions that they’ll trounce their opponents nearly 2-1 at the ballot. 

Westerman is given a 99.9 percent chance of winning. The model estimates that Westerman will win with 62.5 percent of the vote, topping Democrat Hayden Shamel with 33.8.


Womack is likewise given a 99.9 percent chance of winning, forecasting the voteshare as a 63.2-to-33.1 win over Democrat Josh Mahony.

Crawford is given 99.8 percent chance of winning, with the forecasted voteshare 62.5 percent for Crawford, 34.1 percent for Democrat Chintan Desai, and 3.3. percent for Libertarian Elvis Presley. Never underestimate Elvis.

* One more time with feeling: Even if this model is constructed perfectly, it is merely meant to express probabilities. The Dems are the favorites to retake the House and Hill is the favorite to top Tucker … but Hillary Clinton was the favorite, by about the same odds, to win in 2016. Underdogs pull off upsets. Stuff happens. We’ll see.

* Oh, and what about the Senate? Even with a massive blue wave, Republicans are likely to hold — and add to — their majority in the upper chamber, writes David Wasserman of the Cook Political Report in the New York Times:

It’s hard to believe, but true: If every state’s and district’s election results on Nov. 6 were a uniform eight-point swing in the Democrats’ direction from the 2016 presidential result, Democrats would gain 44 House seats — almost twice the 23 they need to control the chamber. But with that same eight-point swing, the party would lose four Senate seats, leaving them six seats short of a majority.