The Washington PostDemocracy Dies in Darkness

Opinion ‘Candidate quality’ won’t be enough to save the Senate for Democrats

Pennsylvania Lt. Gov. John Fetterman, the Democratic nominee for the state's U.S. Senate seat poses for a photograph with supporters after speaking at a rally in Erie, Pa., on Aug. 12. (Gene J. Puskar/AP)

There’s a growing belief, based on state polling and a perception that Republicans have nominated an unusually poor slate of candidates in swing states, that Democratic Senate candidates can significantly outrun President Biden’s feeble job approval ratings and thereby keep control of the chamber. It’s not impossible, but the bulk of polling evidence from the last four cycles shows that’s it’s very unlikely.

It’s true that polls currently show Democrats ahead in many swing or Republican-leaning states. In fact, polling averages compiled by FiveThirtyEight show Democratic candidates tied or ahead in every state that is considered up for grabs this year.

Those leads include double digits in Pennsylvania, a state where Republicans nominated a Trump-backed political neophyte. This gives Democrats ammunition to argue that the Republicans’ purported lurch to the right is making GOP nominees unelectable in any state that’s not safely red.

This argument, however, runs against the massive weight of evidence from recent history. Our elections are increasingly partisan, with voters first choosing which party they back and then voting for its candidates up and down the ticket. I went back to 2014 and compared the exit poll results for the incumbent president’s job approval in those four elections to the vote share received by incumbent senators from the president’s party.

Follow Henry Olsen's opinionsFollow

The results should make Democrats temper their expectations.

In 2014, longtime Democratic senators ran as much as nine points ahead of President Obama’s job approval in their state, but most still lost despite significantly outrunning the president. No Democrat in a contested race ran more than five points ahead of Obama’s job approval rating in 2016.

Partisanship increased even more in the Trump era. Nine Republican candidates in the 16 most contested Senate races in 2018 and 2020 ran within just two points of Trump’s job approval. Another four ran within three or four points. Only two GOP nominees ran 10 or more points ahead or behind Trump’s job approval: Maine’s Susan Collins (+10) and West Virginia’s Patrick Morrisey, who ran 17 points behind Trump’s job approval in his state against Sen. Joe Manchin III.

These data strongly suggest that the fate of this year’s Democratic Senate nominees, like those in years past, is tied to the president’s job approval. The party’s Senate candidates likely will run a couple of points ahead of that figure. They may even run five points ahead of it. But that’s of scant comfort when Biden’s national job approval is languishing around 41 percent.

The pro-Democratic argument also ignores how dramatically state polls have exaggerated Democratic strength in recent elections. The FiveThirtyEight final projections in 2020 for close states all overestimated the Democrats’ performance when compared with the actual results. Most of the errors were large, including by 2.3 points in Arizona and a whopping 7.7 points in Wisconsin. Similar errors in this election would erase Democrats’ current polling leads in most states.

This year’s bout of Democratic optimism should sound familiar because it is similar to a case liberal analysts made in the summer of 2014. Back then, Democrats argued that their moderate incumbents could win reelection despite President Obama’s declining job approval. Polls in mid-July of that year showed Democrats ahead or less than a point behind in states like Louisiana, Arkansas, Georgia and Iowa. Those advantages disappeared as the partisan fundamentals asserted themselves in the fall, and Democrats lost those races by large margins.

This happened again to Democrat Steve Bullock in his campaign for Montana’s Senate seat in 2020. Bullock had long won election statewide as a moderate Democratic governor, regularly winning votes from Republicans. Summertime polls had Bullock ahead, and even polls right before Election Day had him ahead or within the margin of error. But he lost by 10 points as Republican Steve Daines ran only four points behind Trump’s job approval.

There are clear clues that this reassertion of partisanship might happen again in 2022. The most recent Politico/Morning Consult poll has Democrats up by four points in the generic congressional ballot, better than they did in 2020’s House election. That’s mainly because partisans are backing their parties. A narrow plurality of independents (36 percent) didn’t choose between the parties, while 35 percent are backing Democrats.

The same poll found that only 34 percent of independents approved of Biden’s job performance. So nearly all of the undecided voters are independents who don’t approve of Biden. What direction do you think those voters will break when they vote in November?

Maybe this time really is different, and Democrats can swim against the extreme partisan tides of our recent elections. But if I were a betting man, I wouldn’t wager on it.