Within the Margin

Stability was the theme in the 36 elections for governor in 2022, with the incumbent party winning 35 of them. Only Nevada changed hands, with Republican Joe Lombardo narrowly defeating Democrat Steve Sisolak. For our part, we called 35 of the races correctly, including that Nevada flip. The lone miss was Arizona, which we had at Tilt Republican on the eve of the election. It’s easy to dismiss Kari Lake now, in the aftermath of her refusal to accept the result of the election she lost and the ever-rightward and increasingly unhinged drift in her rhetoric. But in the fall of 2022, she led in almost every poll – though the final edition of two of the only higher-quality surveys (Siena and Marist) showed a tie and a one-point Katie Hobbs lead, respectively.

Most of those calls were pretty straightforward, of course. Even two of the flips (Maryland and Massachusetts moving from popular but term-limited moderate Republican governors to Dems) were essentially written in stone from the start of the cycle as Republicans lacked another candidate with the political skills or moderation of Larry Hogan or Charlie Baker. In fact, the GOP nominated hardcore Trumpers in both states, further ensuring Dem pickups in two of America’s bluest states.

Let’s review the other notable races, including a few where we were a shade too confident or not confident enough:

• Georgia: It never quite felt like Stacey Abrams was recapturing the magic of her near-miss in the 2018 race for governor. In part that stemmed from the nature of the two election cycles, each involving a president with declining popularity from opposite parties. It also had something to do with incumbent Republican Brian Kemp managing to thread the needle of backing Donald Trump without supporting the latter’s attempts to overturn the results of the 2020 election. Predictably enraged, Trump endorsed Kemp’s primary opponent, former Senator David Perdue…and Kemp absolutely smoked him in the primary, 74%-22%. Creating that mostly one-sided separation in voters’ minds helped Kemp thrive even as the state’s high-profile race for U.S. Senate showcased Republican extremism and their tendency to nominate oddballs. We kept the race at Lean Republican throughout the cycle, and Kemp’s margin of victory of 7.5 points is indeed at the outer edge of that Lean band.
• Kansas: Democratic incumbent Laura Kelly was defending her 2018 win over far-right former KS Secretary of State Kris Kobach. It was a massive win in a Dem-friendly year and 2022 stood to be tougher – after all, it’s not like someone would ever nominate Kris Kobach again, right? This time the GOP nominated KS Attorney General Derek Schmidt. But just as Kobach’s extremism created room for Greg Orman’s independent candidacy in 2018, Schmidt’s relative moderation (by today’s GOP standards) allowed Dennis Pyle to run as the self-proclaimed “conservative” choice. Kelly’s 2.2% margin of victory was greater than Pyle’s vote share, though. Her popularity – along with the continued drift of pro-choice GOP moderates in eastern KS into the Dem column following the Dobbs ruling – bumped her vote share upward from 2018’s 48% to 49.5%. There wasn’t much polling here throughout the cycle, but the non-partisan polls showed her ahead. With a state like Kansas, you’re rarely if ever going to see a Safe or Likely Dem rating – but Kelly’s strengths allowed us to keep it in the Tilt Dem column throughout the cycle.
• New Mexico: Michelle Lujan Grisham was another Democratic incumbent who flipped a seat in 2018, in a much blue seat than Kansas. Grisham’s approval ratings were not quite as strong as Kelly’s, and she was not facing a fellow statewide officeholder this time around. But Republican meteorologist Mark Ronchetti had run a surprisingly strong U.S. Senate race in 2020 and was viewed as a solid recruit for this race. Most polling, including every nonpartisan poll, showed Grisham ahead…but the margin was narrowing as the election approached. As a result, we moved this from Lean Dem to Tilt Dem in our final Election Eve forecast. We could have left it where it was: that late movement in the polls turned out to be ephemeral if it was real at all, and Grisham won by nearly six and a half points.
• New York: Closer to home, the Empire State’s gubernatorial race turned out to be a tricky one for us to get our arms around. From the start of the race, our position was that Lee Zeldin was too conservative to thrive in a statewide race: an unapologetic supporter of Trump’s efforts to “investigate” and overturn the 2020 election, Zeldin was also staunchly anti-choice and had voted against marriage equality during his time in the state senate back in 2011. On both social issues and the democracy issues stemming from the January 6 insurrection, Zeldin was well to the right of the national electorate, let alone blue New York’s. And yet: he consistently dismissed those attacks, on the rare occasion incumbent Kathy Hochul or the state Democratic party attempted to make them. He was teflon. Hochul had risen to the position after the resignation of three-term governor Andrew Cuomo in the summer of 2021; she was probably damaged by the brief re-imposition of some COVID lockdown measures the following winter but much more so by the perception that crime was rampant – particularly in New York City. Dem vote shares had already slipped a bit in NYC’s Latino, Asian-American and Haredi communities in the 2020 presidential election, and that slide continued in 2022. It was clear to us as the campaign moved along that this was not a Safe Dem race like mot of NY’s recent gubernatorial elections, and we toyed with moving it from Likely Dem to Lean Dem in our final forecast. We stuck with Likely Dem, but should not have: Hochul only won by six points as the “red wave” that missed most of the country squarely struck New York – particularly on Long Island, but also neighborhoods of NYC that Republicans had not won in decades, like Brooklyn’s Chinatown(s). (note that while Zeldin claims victory in Manhattan’s Chinatown, he did not win a single precinct there. Zeldin is lying when he says that. He did reduce Dem margins there, though). With New York politics in a more dynamic state than we’ve seen in quite a while, forecasting elections here will again be a real challenge in 2024 and 2026, despite our proximity.
• Oklahoma: This race looked increasingly interesting as it unfolded. Incumbent Republican Kevin Stitt was less popular than one would expect in a deep-red state, and faced a competitive challenger in Rep-turned-Dem Joy Hofmeister, previously elected statewide as Superintendent of Public Instruction. Polling tends to be scarce in Oklahoma, with the better-rated pollsters rarely making a foray into the Sooner State. A series of Republican-affiliated pollsters descended up on the state in October, initially finding Hofmeister ahead – a directional difference from many swing state polls in ’22 where the Rep-affiliated companies were finding GOP leads that ended up far removed from the final result. With the better-rated Emerson showing a single-digit Stitt lead in late October, and a competitive high-profile race for Superintendent of Public Instruction involving the controversial Ryan Walters, our final forecast moved our rating from Likely Rep to Lean Dem, just in case something was really happening. In the end, Stitt won comfortably – albeit behind most of the Oklahoma GOP ticket, and far behind recent GOP presidential showings in the state. A 55%-42% margin of defeat passes for respectable these days in Oklahoma, but with Stitt and Walters continuing to stir up controversy one wonders if 2026 might be closer. We’ll have to be disciplined, though: the 2022 example (and some before that) indicates that pollsters understate GOP support in OK regardless of whether it’s a presidential year or not.
• Oregon: Throughout the 2022 cycle, Oregon looked like one of the GOP’s best opportunities to capture a governorship – something they had not done in Oregon since Victor Atiyeh was term-limited in 1986. Democrat Kate Brown was not running, having reached her term limit; the Dem nominee was former state House Speaker Tina Kotek. The GOP had a reasonably strong nominee in former state House Minority Leader Christine Drazan. Brown left behind abysmal approval ratings, while Democratic state senator Betsy Johnson was running as an independent candidate likely to eat into the usual Dem margins in Oregon. And sure enough, polling throughout the summer and early fall showed Drazan ahead, with Johnson polling in double-digits – occasionally passing the 20% mark. But as tends to happen, the autumn saw the independent/third party support drop off, and by late October Johnson had fallen into single-digits. Both major-party candidates saw their vote shares rising, but Kotek was gaining faster and leading more often than not down the stretch. We shifted our rating from Tilt Rep on October 21 and 27 to Tilt Dem as midnight struck on November 8, and that proved to be correct. Kotek prevailed 47%-43.5%, with Johnson trailing at 8.6%. The GOP drought in the Oregon governor’s mansion (I assume they have a mansion?) continues for another term.
• Rhode Island: Man, I don’t even remember precisely why we thought this was Likely Dem instead of Safe Dem. It probably owed to a mix of factors – starting with the lack of any public polling after the first week of October, which did show a closer-than-usual contest in the Ocean State, and the RI-2 open seat race was understood to be quite competitive. And Dems were tearing each other up in various primaries up and down the ballot. But Dan McKee won by 19 points, running only slightly behind the baseline Dem performance here. We hedged the appropriate amount, but no serious person would have looked askance at a Safe Dem call here, either.
• South Carolina: The Palmetto State is really the inverse of Rhode Island: with former one-term Congressman Joe Cunningham carrying the torch for Dems, there was a chance this race could pop. Summertime polling showed a single digit lead for incumbent Republican Henry McMaster, and then pollsters disappeared from the scene entirely a week into September. That necessitated a bit of hedging to Likely Rep, but no one on the ground was saying Dems really had a shot here down the stretch. Sure enough, McMaster won by a little over 17 points. Our hedging had some logic but as with RI, few would quibble with a Safe rating in a state with such a strong partisan lean.

If we wanted, we could probably add Michigan as a state where we were too conservative at Lean Dem, as incumbent Gretchen Whitmer ended up with a comfortable double-digit win over GOP challenger Tudor Dixon. All told, though, pretty solid work. Wish we had nailed Arizona, obviously, to get all of these correct.

That wraps up our look back at the ’22 elections. Next up, it’s time to put together some assessments of the present state of affairs in what has rather suddenly turned into a dynamic 2024 election cycle.