Here's How Big the GOP's Midterm Hole Is

Republicans have a problem, and pretending otherwise helps nobody. The generic congressional ballot, the poll question that asks voters whether they would rather see a Republican or a Democrat in their House district, has moved against the GOP this fall. The election is Nov. 3. Here is what the polls say, how much weight to give them, and what Republicans can still do.

What the polls show

Every number below is a poll or a polling average, which is an estimate of opinion on the day the poll was taken, not a prediction of the outcome.

A CBS News/YouGov survey, as reported secondhand in a political.org roundup dated Oct. 8 and in social media posts quoting CBS, put Democrats at 54% and Republicans at 45% on the generic ballot, a nine-point gap. We have not seen CBS's own release, so treat that figure with that caveat.

A PBS News/NPR/Marist poll, in a broadcast dated Sept. 23, found registered voters preferring Democrats by 12 points. The broadcast noted that this was the same gap seen at this point in Trump's first midterm in 2018. We could not confirm the exact percentages behind that margin, so we are not stating them.

Poll averages tell a similar story. According to that same political.org post, Nate Silver's tracker had Democrats ahead by 8.9 points as of Oct. 7. We did not find a current RealClearPolitics number.

Not every poll agrees. A Rasmussen survey from January had Democrats up only 47% to 41%, and Morning Consult had Democrats up three points in July. Pollsters differ in who they count, and registered-voter and likely-voter polls often produce different leads.

Why a generic ballot is not a seat count

The national popular vote does not elect a House. 435 separate districts do, and the map matters as much as the mood. A nine-point national lead would be dangerous for Republicans in the many competitive seats, but a poll cannot tell you how those seats break. Polls can also miss, in either direction, and turnout in a midterm is different from turnout in a presidential year.

That said, history suggests the president's party usually struggles in midterms, and Marist's comparison to 2018 is not a comfortable one for Republicans. In that year the party lost control of the House.

What the GOP can still do

The party has about three weeks. Our view is that Republicans have more control over their fate than the headlines suggest, though not unlimited control.

First, talk about what voters feel. The September jobs report, with only 29,000 jobs added and unemployment at 4.2%, and inflation that remains above the Fed's target, are kitchen-table problems. A party that sounds as if it is changing the subject will lose. A party that offers specific answers on prices, energy and wages can still win a hearing.

Second, tighten the immigration message. As we have reported, some Republican senators are urging a sharper focus on criminals. A Fox News poll in September found 53% saying enforcement has gone too far. Border security remains a strength, and voters have noticed that crossings are down. Enforcement that looks precise protects that strength.

Third, turn out the base. Midterms are decided by who shows up. Republican voters who feel taken for granted tend to stay home. Campaigns that give them a reason to vote, rather than a reason to be angry, usually do better.

The take

A conservative who looks at these numbers and sees only media bias is making a mistake. The polls are not destiny, but they are a warning, and they deserve to be read as one. The same discipline applies to Democrats, who have seen leads like this evaporate before and who should not pick out furniture for the majority.

The only poll that counts is the one on Nov. 3. Until then, the job for Republicans is to earn votes, not to argue with the people who count them.