Skip to main content
ABC News
Holding Pattern in House Forecast

There is little change in the House forecast from last week. Republicans are assigned a 65 percent chance of taking over the House by our forecasting model, up slightly from 62 percent last week. They are projected to control 224 House seats in the new Congress, up a single seat from 223 last week; this would imply a gain of 45 seats from the 179 they have now (counting one vacant seat most recently held by a Republican member). Substantially larger — or smaller — gains are possible, however.

The Republicans’ lead among likely voters on the generic ballot is down slightly to 6.5 points this week from 7.5 points last week, according to our estimate. However, any potential penalty from this was offset by gains made by Republican candidates in the characterizations of individual races made by some expert forecasters, which are also used in our model. Cook Political, for instance, moved 7 races in the Republicans’ direction, while CQ Politics — which had been more conservative about projecting Republican gains — did so in more than a dozen races.

Polls of local districts, another indicator that our model uses, are becoming much more plentiful, but continue to show ambiguous results. In general, Republicans are doing very well in polling in Pennsylvania, Ohio and Michigan and are poised to win a large number of seats there and in other parts of the Midwest.

Elsewhere in the country, however, the polling tends to be more equivocal. Local polls tend to become considerably more accurate in the final weeks of a campaign; Republicans would need to emerge with somewhat better results in these other regions to make a gain as large as 50 to 60 seats nationwide in the most prudent forecast.

The phrase “holding pattern,” in the headline of this article, implies that we are awaiting the next significant development in the race: either a further break toward Republicans, which could make their taking over the House all but inevitable and would put scenarios like a 60-plus seat gain in squarely into play, or, a shift toward the Democrats, which would make them more likely than not to hold the House.

While either of those things are possible, it is equally possible that there will not be much momentum for either party between now and Nov. 2. This would nevertheless imply an uncertain outcome because of the very large number of seats in play: currently, there are 89 seats that each party has at least a 10 percent chance of winning, according to the model.

We are a bit overdue in providing additional detail on our House forecasts, available for our Senate and gubernatorial forecasts, but we should be rolling that out for you over the course of the next week or so.

Nate Silver founded and was the editor in chief of FiveThirtyEight.