Exit polls show why the voters effectively shrugged

WASHINGTON, DC - NOVEMBER 08: House Minority Leader Rep. Kevin McCarthy (R-CA) delivers remarks to supporters alongside Ronna Romney McDaniel, Republican National Committee chair, and Rep. Tom Emmer (R-MN), at a watch party at the Westin Hotel on November 9, 2022 in Washington, DC. Republicans are hoping to take control of the House of Representatives away from Democrats. (Photo by Sarah Silbiger/Getty Images)

This week: The Briefing, Vol. X, Issue 47

  • What happened in the election? What the exit polls say
  • Abortion helped Democrats marginally, but the data call its importance into question
  • Among voters, Biden was unpopular; Trump was even more unpopular 


Republicans have clinched the U.S. House and will likely end up with about 222 seats out of 435. This will hold back the Biden administration’s worst impulses — it means that President Joe Biden cannot pack the Supreme Court or pass any other meaningful legislation without Republicans’ approval.

But Republican-leaning voters are not happy with this meager outcome. Ever since their disappointing 2022 election result, all the talk has been about what happened to the widely expected “Red wave.”

Perhaps one boring explanation is that it was always a bit too optimistic.

Republicans’ 3.3-point in the national popular vote for the U.S. House — 51% to 47.6% — was actually greater than the 2.5-point average win that generic congressional ballot polls predicted by election time, according to RealClearPolitics. 

There is also a tacit belief based on recent experience that when elections break, they break hard and they break in one direction. This has been true in many recent midterm elections, but it just wasn’t the case in 2022. Republicans will, in the end, lose governorships in Massachusetts, Maryland, and Arizona and state legislatures in Michigan and Pennsylvania, and gain perhaps 10 House seats.

In historical terms, this was only marginally worse than 2014, in which Republicans won a 5.5-point victory (51% to 45.5%) and elected a modern record 247 members to the House. So what happened? Were those 2.2 points really that important?

As Michael Barone has observed, there were a few key differences this time that probably diminished Republicans’ results beyond the proportionality of the overall popular vote result. 

One was that Republicans had less control over redistricting this year then they did after the 2010 election. Between states adopting non-partisan commissions and Democratic courts running interference for their party in the crucial states of Pennsylvania and North Carolina, Republicans simply did not have as much sway over the process this time as they had previously.

Republican improvements: Another reason is that a significant of the Republicans’ improvement in their vote totals came among non-white voters. This is a good thing, to be sure, but thanks to various court decisions and redistricting based on earlier partisan trends, many of those voters are drawn into seats that are numerically outside Republicans’ grasp. It is a hopeful sign that Republicans lost those districts by fewer points than usual, but it does not add to their House seats on the scoreboard this year.

For example, in Indiana’s 7th district, Republican Jennifer-Ruth Green lost by less than six points against Democrat Frank Mrvan, who won by 16 points and an almost identical district in 2020. That is an enormous improvement, but it does not affect the national outcome for House races this year. Likewise, the fact that Rep. Jim Clyburn (D-S.C.) won re-election in his majority-black district by only 24 points instead of the 37-point margin he had in 2020 or the 42-point margin of 2018 likewise doesn’t help Republicans in terms of the immediate outcome, even if it is a very positive sign for the future. After all, these incremental changes must occur first, and they don’t always bear fruit in terms of seats or clout right away.

Perhaps the most optimistic part of the 2022 outcome was Republican progress with Hispanic and Asian voters. These certainly did affect at least a few key House races.

In the 2018 midterm, Asian voters went Democratic by 54 points — a massive blowout. In 2022, however, Asians voted 58% to 40% Democrat, a margin of only 18 points. Part of this progress could be associated with the ongoing practice of discrimination against Asian university applicants — discrimination which Democrats and the Biden administration are actually defending. Republicans should be thinking of Asian voters as their next big project.

The current project started coming to fruition as Republicans won 39% of the Hispanic vote nationwide for House. This is a 10-point improvement over the last midterm and their best performance since 2004. This is very good, even if it only directly affected the outcome in a handful of House races in Florida and certain western states.

Speaking of which, it is important to point out that in Florida, not only did Gov. Ron DeSantis (R) win Cuban (69%) and overall Hispanic voters (58%), but he also won traditionally democratic Puerto Rican voters, 56% to 43%. He also won 46% of non-white voters. Despite liberal complaints, this is not something that anyone could have done through trickery or voter suppression or through gerrymandering. Florida is once again a model for Republican parties across the country who want to expand and broaden their appeal, beginning with the issue of education and school choice.

Also, speaking of possible presidential contenders, Texas Gov. Greg Abbott (R) won the votes of 40% of Hispanic voters — not excellent for his state, but a pretty good showing that points the way to the future. 

By the way, fun fact: in Texas, one of the best predictors of how you’re going to vote is whether someone in your household owns a gun. It turns out that 62% of households of Texans who voted in this election include someone who owns a gun.

Trump omnipresent: Looking at the exit polling, a few additional issues jump out. 

One pertains to former President Donald Trump and the role that his continued involvement in politics might have had in turning the race against Republican candidates.

All along, it was assumed that President Joe Biden’s extremely high negative ratings would drag Democrats down. But although President Biden’s favorables were low among voters nationwide at 41% favorable to 56% unfavorable, former President Trump’s numbers were even worse at 39% favorable and 58% unfavorable. Only 20% within that anti-Trump majority voted for Republican candidates. That definitely limited Republicans’ ability to succeed in this election.

This is something Republican voters are going to have to think about. In an election where they actually did win the national popular vote for House, they did so in spite of their de facto party leader’s extremely negative ratings. 

Trump remains popular among Republican voters, and will not be easy to dislodge in the 2024 primaries. However, this is precisely the sort of data that is convincing some Republican leaders and donors to back other candidates already for 2024. Trump loomed over the 2022 election In the background, and for a large majority of voters, his presence diminished the odds they would vote Republican.

In Arizona, Kari Lake won only 16% out of the 57% majority of voters who said they viewed Trump unfavorably. If you do the math, this means she would have had to win nearly 100% of the 42% of Trump-favorable voters in order to get to 50%. She won only 94% of them, and that wasn’t quite enough to win her race.

Trump-style candidates: speaking of Lake, we certainly enjoyed her candidacy and made no secret of it. But you can’t argue with the results: the voters in Arizona clearly did not enjoy it. The exit polls reflect the voting tallies. 

Lake underperformed Republican House candidates in her state. She also won only 50% of white voters in Arizona, according to exit polls. She did not do too poorly among non-white voters, earning 46% of their votes, but a Republican who cannot break 50% among whites simply cannot win. 

Tudor Dixon (R) In Michigan likewise won only 50% of the white vote in her loss to Gov. Gretchen Whitmer (D). Doug Mastrano (R) in Pennsylvania actually lost the white vote, 50% to 48%, on his way to a blowout loss. 

Contrast those Arizona, Michigan, and Pennsylvania results with governor’s races that Republicans won in Nevada (where they won 58% of white voters), Florida (65% of whites), Texas (66% of whites) Ohio (67% of whites), and Georgia (74% of whites). It is excellent that Republicans generally improved among non-white voters, and something to build upon. But you cannot win elections without running up the score in your best voter demographic. When you nominate eccentric candidates who are not acceptable to most voters, you cannot expect good results and in some states, Republicans did just that in 2022.

It may also be that candidates who profess that Trump actually won in 2020 or otherwise attempt to adopt Trump’s style of politics are unattractive to a vast number of voters who might otherwise be more open to voting Republican.

Again, Republicans have to rethink the sort of candidates they want to nominate if they intend to win and to govern in the future. There is no indication that they need to sacrifice the extent of the candidates’ conservatism, but the style of the candidate and also perhaps his or her blind loyalty to Trump really does matter, or so the exit polling suggests. 

Also, when Democrats start advertising on behalf of a Republican candidate in a primary, maybe it’s time to reconsider supporting that candidate.

The ‘our democracy’ shibboleth: So what about January 6?The Democrats’ theatrical and long-running attempts to confuse in the public eye the fate of democracy with support for the Democratic Party does not appear to have succeeded, and that remains true even though Democrats outperformed expectations in the final election results. 

The key to understanding this is in the exit polls. Among voters who believed democracy was under threat (68% of the electorate), Democrats and Republicans split the vote almost evenly, 50% to 48%. In fact, the ones who considered democracy to be “very threatened” leaned Republican (50% to 48%). We surmise that this is because people have vastly different ideas about what it means for democracy to be under threat, depending upon their partisan affiliation. Does it mean that there is widespread Democratic voter fraud changing the outcome of big elections, or that Republican voters are “semi-fascists” preparing a coup d’etat? Does it mean that leftists are stifling free speech and attempting to assassinate Supreme Court justices in order to change the court’s decisions? Or does it mean that January 6 rioters supposedly almost toppled the fragile republic?

The voting public seems evenly divided on such questions, If the exit polls are any indication. Perhaps this campaign of show-trials and shrill rhetoric motivated base Democratic voters, but that’s probably the extent of it.

Abortion: It’s no big surprise, but abortion was a lightning rod issue in this election. Of those who ranked it as their top issue in the election (27% of voters), 76% voted Democrat. But that number could be somewhat misleading, because Republican voters may simply have been loath to describe abortion as their most important issue in the year that Roe v. Wade was overturned.

Fortunately, the exit polls asked a question that is probably more relevant and illuminating on the question of abortion. In general, regardless of which issue was considered most important to any given voter, 53% of voters trusted Democrats more on the issue of abortion compared to 42% who trusted Republicans more. That represents a decisive but not overwhelming advantage for Democrats. This was reflected in the failure of pro-life measures even in such states as Montana and Kentucky, and the success of Michigan’s anything-goes abortion referendum.

As the numbers go, voters’ feelings about Trump had a much bigger effect on the outcome than voter feelings on abortion.

It also bears noting that voters were almost as brand-loyal on some of the other issues as they were on abortion. By margins greater than 80 points in most cases, people voted for the party they trusted more on immigration (51% to 45% in favor of Republicans), crime (52% to 43% Republican), and the most important issue to voters overall, inflation (54% to 42% Republican).

This all suggests that abortion played some role, but it was arguably a very limited role. It also bears noting that several of the governors who were easily reelected — including Georgia’s Brian Kemp, South Dakota’s Kristi Noem, Texas’s Greg Abbott, Idaho’s Brad Little, Tennessee’s Bill Lee, and Oklahoma’s Kevin Stitt, among others — suffered no repercussions at all after signing relatively restrictive abortion bans into law. Kemp and Noem, in particular, went from nearly lossing their races in 2018 to broad victories this time around — Noem went from a four-point win in 2018 to a nearly 30-point win this time.

Likewise, the pro-life Sarah Huckabee Sanders won her election in Arkansas with no controversy after that state’s abortion ban took effect in June

Yes, some of these officials govern very Red states. But if there is as much backlash to Dobbs as the media predicted, then you would still expect some signs in the electorate even of the deepest Red states — perhaps the loss of at least a few state legislative seats or points in their election — something. Instead, all of these governors won easily, in some cases despite being massively outspent by their Democratic opponents. 

A similar thing could be said of the legislators who passed new abortion restrictions in Missouri (passed in May), Mississippi (took effect in June), West Virginia (passed in September), Kentucky (passed in April), and Indiana (took effect in September) — states which did not have governor’s races this year. As it turned out, Republicans in these states suffered no losses at all. They appear to have gained several seats in West Virginia’s legislature, and none of the others changed significantly.

Another argument against pro-abortion sentiment as the driving force in 2022: timing. People who decided how they were going to vote during the month of October — that is, 19% of the electorate — voted Democratic, 55% to 42%. The 6% who decided in November and the 68% who decided before October leaned Republican, according to the exit polls. The Dobbs decision overturning Roe v. Wade was leaked early and then officially released in June. For voters whose main focus was to preserve legal abortion, October would have been a strange time to decide they’d vote Democratic.

Next week: We will briefly preview the race for House Speaker, then take a preliminary look at presidential states and Senate races in 2024.