AMAC Exclusive – By Daniel Berman
After leading by four points in February on the RealClearPolitics (RCP) average, Republicans are now down to just a 0.2 percent lead on the generic congressional ballot. This apparent tightening of the polls has prompted narratives about a “Democratic recovery,” which are likely to reach a new crescendo in coming weeks as Democrats remain hopeful that they may yet defy the odds and retain control of Congress this November.
Careful optimism for Democrats’ chances this fall has been reported throughout the media landscape. Newsweek typified the narrative, declaring on August 5, “With fewer than 100 days until Election Day, talk of massive Republican gains in Congress has faded substantially amid a series of victories for the president and his fellow Democrats.”
But as David Drucker of the Washington Examiner laid out in a tweet, we have seen this movie before. “Midterm cycles since ‘06 have certain rhythm,” he wrote. “1) Maybe POTUS’ party’ll avoid losses. 2) Things look good for out [of power] party. 3) Things look REALLY good for out party. 4) Hold on, maybe POTUS’ party won’t lose as many seats as thought. 5) Could POTUS’ party avoid wipeout? 6) WIPEOUT.”
Quoting this tweet in an article for the Cook Political Report, Amy Walter argued that we are now in “phase 4” of that cycle, when the evidence for a “red wave” has been so overwhelming for so long that any further data in that direction is no longer newsworthy. Conversely, any evidence to the contrary, no matter how circumstantial, is deemed worthy of in-depth analysis.
But the truth is that predictions of a Democrat resurgence often ignore the deep deficit Democrats need to make up, run against historical precedent for midterms, and defy recent precedent in which individual candidates and campaigns matter less than partisanship.
To determine if reports of a “Democratic recovery” are to be trusted, we first need to examine the evidence. There are three sources to which we can look. The first is polling. Here we have three different data points: Biden’s approval ratings, the generic congressional ballot, and specific state polls.
The first is cold comfort for Democrats. Biden’s job approval rating stands at negative 15% on RCP, with an average of 40.9% approving and 55.8% disapproving. This is an improvement from his lowest point, when he was more than 20% underwater in late July, but it is still lower than at any point between January 2021 and June 2022. Whether Biden’s numbers are improving depends on context. If the drop in June and July was the result of temporary factors such as $5 a gallon gasoline, then the fall in gas prices would be expected to cause his numbers to revert to their “normal” trajectory. Rather than providing comfort to Democrats, this should be a cause for worry, as implies that Biden has already received whatever “bounce” he can expect from the easing of gas prices, and it still leaves him at a lower point than any time prior to May 2022.
The second category is the generic ballot. The GOP’s current lead of just .2% is the smallest since Democrats last led on November 10, 2021. This does seem to indicate a degree of movement, and at 44%, support for Democrats is at the highest level since October of 2021. However, as with Biden’s numbers, that 44% is at best a glass half full. Prior to October 2021, Democrats had not dropped below 48% average support for more than three years. Hence, while these numbers may indicate that bleeding has recovered somewhat, they still point to a GOP landslide. Factoring in the 41%/56% approval numbers for Joe Biden, not only do almost all currently undecided voters disapprove of Biden, but even some Democrats. Democrats need to win over voters who do not support Biden to even get to a tie, much less make any further recovery, whereas the GOP can gain further support simply by winning over those who think Biden has failed.
The third category is polling in individual races. Here a multitude of Democrat candidates are outperforming their state’s leans by leading in races even when Joe Biden is underwater by as much as 30%. Pundits can point to polls not just in Pennsylvania, Arizona, Nevada, Georgia, and New Hampshire, but even Ohio for examples of Democratic candidates leading in races while Biden sits at woeful approval levels.
For a mainstream media obsessed with the mechanics of individual races, state-level polling is a godsend. There is a strong assumption that if data points exist, they must be important. However, history provides ample reason to be skeptical of state-level polling for lower-level races. The problem is that in races with a very high disparity in name recognition between candidates, it is relatively easy for incumbents or better-known candidates to lead. One of the best examples was the Tennessee Senate race in 2018, when former Democratic Governor Phil Bredesen led for virtually the entire campaign until the final month, before losing by 10% to then-Congresswoman Marsha Blackburn. Democratic incumbents Claire McCaskill and Joe Donnelly also led in Missouri and Indiana until August and Election Day, respectively, that year. If part of the reason Democrat candidates are leading is higher name identification, as is clearly the case in New Hampshire where there is no Republican nominee, and Arizona where the Republican candidate was only recently nominated, then we should expect the numbers for the incumbents to matter more.
There’s also an emerging narrative that Republicans have several weak nominees in key states. That is the argument fielded by pundits when it comes to Blake Masters in Arizona, Herschel Walker in Georgia, Dr. Mehmet Oz in Pennsylvania, Adam Laxalt in Nevada, and J.D. Vance in Ohio. It is worth noting all of these candidates were endorsed by Donald Trump, and much of the presumption of their weakness rests on assumptions about both the quality of the former president’s political judgment and the impact of his image on their candidacy. Many of the same charges were made in 2018 against Josh Hawley in Missouri and Marsha Blackburn in Tennessee (too extreme) and Rick Scott in Florida (“unpopular”) only to see all three win their races.
Yet even if we were to concede that this year’s slate of Republicans may be “weaker” than a mythical “generic” Republican, historical evidence is that this matters increasingly less. In 2016 and 2020 combined, only a single Senate race went to a candidate of a party different from the way the state voted for president. In 2018, a Democratic wave year, Democrats did win Senate races in 7 states which went for Donald Trump in 2016, but only two of those cast their electoral votes for Donald Trump in 2020. Both those races, Ohio and West Virginia, were all but written off by the press and featured Republican candidates who were badly outspent but nevertheless lost by single digits.
Why does this matter? Because the playing field for Democrats is difficult. Arizona, Georgia, Pennsylvania, and Nevada all officially went for Joe Biden by well under his national margin of 4%. In a generic ballot where Democrats win by even 1%, odds are every one of the aforementioned races would go Republican. This is not to say that the Republican candidates there are perfect, but the question is whether they are closer to Josh Hawley, Rick Scott, and Mike Braun, unknown candidates seen as polarizing but who nevertheless had money and profile, or whether they are closer to candidates like Patrick Morrisey in West Virginia or Jim Renacci in Ohio who were outspent several times over and all but written off. It does not seem that Masters, Walker, or Oz will lack for money or attention.
The clear conclusion is that arguments that Democrats have in any way turned the corner are being overblown. There does seem to be some movement in the generic ballot, largely made up of Democratic-leaning voters who disapprove of Biden but nevertheless prefer Democrats on policy. That may be a result of Dobbs, along with falling gas prices, pushing otherwise left-leaning voters back into the voter pool. But there is almost zero evidence voters are warming to Joe Biden, and this recovery is precarious. It is built on voters who still lack faith in Biden.
Furthermore, it is almost certainly not enough for Democrats to have a serious prospect of holding the House, or even the Senate, unless one ignores the historical trends of polling several months out. To do that, Democrats would need to actually win over enough voters to lead Republicans by several percentage points, not merely tie or hold their loss to 1-2%. It is hard to see how that can happen without Biden turning himself around.
The best that can be said for Democrats is that for the last few weeks, things have stopped getting worse. But absent them getting a whole lot better, Democrats should follow Walter’s advice and get ready for the final phase of midterm election years: Acceptance.
Daniel Berman is a frequent commentator and lecturer on foreign policy and political affairs, both nationally and internationally. He holds a Ph.D. in International Relations from the London School of Economics. He also writes as Daniel Roman.