Dem
47
GOP
53
Although we don't normally put House races near the top, this polling item and the one above may be related. Presidential approval certainly is a major factor in the midterms, since voters often see midterms as a way to send the president a message. An even more direct indicator is to ask people: "Will you vote for the Democrat or the Republican for the House?" This factors in presidential approval, but also other factors such as "I love/hate my congresscritter," "the economy is wonderful/terrible," and other things. It is far from perfect because candidate quality, money, and other factors can play a role, but it is still an interesting indicator.
The latest generic poll is from Emerson College and has the generic Democrat at 50% and the generic Republican at 40%. This is the biggest gap this cycle. It is also consistent with the trend in Emerson polls, so factors like "polling house bias" can't explain this. Throughout 2025, the Democrats were ahead by under 4 points. Since then it has exploded, as shown below.
Again, one poll does not a House majority make. Here are earlier House generic polls for comparison. What is interesting is that Emerson is a reputable pollster and there is a clear trend.
If the 10-point gap holds, there are 80 House Republicans in districts R+9 or bluer. All of them would potentially be in danger. Some incumbents are popular enough in their district to survive and some might luck out by drawing a weak opponent, etc. Still, it would be a bloodbath.
The director of the Emerson poll, Spencer Kimball, says that Democrats' gains are due to Latinos (D+35), women (D+21), and independents (D+19). The poll also puts Donald Trump's approval rating at 40%, much better than the 34% in the Ipsos poll, so Emerson is definitely not oversampling Democrats. Kimball also noted that Trump is underwater by 41 points among Latinos, 29% approve to 70% disapprove, vs. an almost even split a year ago. This could be the story of 2026—Republicans lost the elections by infuriating Latinos beyond belief. We shall see. (V)