NATIONAL LANDSCAPE
My topline prediction of a R +2 environment will be right on the money or off by a point, depending on your method of adjusting for uncontested races. But the traditional national environment this year was extremely unpredictive of the eventual results. Blue-leaning state environments took precedent over the national wave in states like Michigan and Pennsylvania, while Red Wave/Tsunami environments took precedent in important states for House control such as California, Florida, and New York. While some polling indicated this was coming, most prognosticators including myself believed it was an artifact of the typical polling error of “safe state underestimation”, or that the dominant party is underestimated in state polls.
THE WRONGS: THE INCORRECT CALLS
US Senate: Lumping call my incorrect Senate calls (AZ + NV + PA) because I basically played a parlay bet for this one. I thought that the probability of all or nothing was much more likely in the grand scheme of things than the 51-49R consensus outcome people seemed to believe would happen, and the mirage of the R +2 environment and the polarization of politics in recent years led me to pick the right-tailed outcome of a GOP sweep. Well, the nation depolarized much to my surprise, and so despite the nominal Republican-leaning environment, it was the left-tailed outcome, the Democrats sweep, which occurred. The Georgia runoff will determine if Team Donkey has a lifeline or if they’re out of the game for 2024.
Arizona Governor: The turnout and Congressional results show that the Arizona electorate this year was somewhere in the ballpark of R +5 or even higher, but that wasn’t enough to stop Kari Lake’s campaign from crashing down last moment. The idea before the election was that Lake should outrun Masters by a decent 3-4 points and surely Masters won’t lose by more than 4 right? Whoops.
Wisconsin Governor: This was always the purest tossup Governorship on my board and I’m not that surprised to see it go wrong. Evers winning by 3.4% is quite impressive though, and the long standing tradition of Wisconsin polls underestimating Republicans have come to an end. (While it has not for their best friend, Ohio polls)
WA-03: The biggest shock on the house side, nominally anyways, was WA-03 which I rated Likely Republican. I expected Tiffany Smiley to win this district by a lot (she did) and thought the district was downballot Republican enough for Kent to overcome his struggles as a candidate. But candidate quality was quite a big factor this year, and Kent was worse than one could’ve guessed, leading to his dramatic loss to complete the Democratic pacific coast sweep.
CO-08: State Senator Barbara Kirkmeyer’s strengths as a candidate was not enough to counter the unexpected complete collapse of the Colorado Republican ticket on the top of the ballot. O’Dea lost by more than Trump did, which basically no one expected, and outrunning him by more than ~6-7% proved too difficult of a task for the Kirkmeyer. State Rep. Yadira Caraveo is the Representative-elect.
ME-02: If you told me this was going to be a state environment above-else environment, I’d definitely have chosen Golden to win. But alas, the national environment was deceptive. Another underestimation for Team Blue.
MI-07: Similar patterns playing out all over, with the state environment taking precedent over national patterns. My entire case for Barrett was that he was mediocre enough to take a Biden +1 seat in a R +2 national environment. Well, the environment in Michigan was certainty not R +2, as Democrats are still in celebration of their newly formed trifecta.
NC-13: This was a candidate quality issue, as all statewide Rs carried this district. Bo Hines’ frequent carpetbagging and questionable political stances did him in at the end, despite largely favorable internal numbers and a good Republican environment in North Carolina.
NV-03: Even Ralston got this one wrong, and there wasn’t much ticket splitting between this race and the Senate race as one would’ve assumed beforehand. The NV GOP is now in a frantic search for a credible 2024 Senate nominee after their heir to the nomination was defeated by a 52-48 margin.
NM-02: One of the closest races in the country, Gabe Vasquez barely knocked off the incumbent in this Biden +6 seat even as Ronchetti was narrowly carrying it on the top of the ballot (I think). Herrell’s residual weakness as a candidate ended up hurting her at the barest but most significant of margins.
OH-09: Another victim of the national environment theory, JR Majewski proved to be so bad that he lost by double digits. Turns out faking your military record isn’t a great idea.
OH-13: I probably underestimated the Skyes name in the Akron area, and Tim Ryan’s strong showing here got Democrats the checkmark in this district.
PA-07/PA-08: I was truly shocked by PA-07, since as a national environment truther it was difficult to digest how Wild won a district she would have lost in 2020. The moral here was that Mastriano and Oz dragged down Scheller and Bognet just enough for Democrats to hold onto both of their vulnerable districts in PA.
RI-02: The truly shocking thing here was how a generally unpopular Governor basically matched Biden’s margins (never underestimate the machines!) Rhode Island wasn’t expected to be a great area for Democrats this cycle and the DGA was even getting McKee last minute support, but McKee’s surprising showing helped out Magaziner in his bid to succeed Jim Langevin. Fung’s tradition of significantly outrunning fundamentals but falling up short continues.
TX-34: The hype for this seat was real but it was well short of what it needed to be for the GOP. While Flores was the only of the three Congressional Republican candidates to significantly outperform Trump in the heavily Hispanic south Texas counties, the partisan lean proved too much to overcome.
WA-08: The theory here was that the Washington primaries took place in August, the height of the Democratic momentum, and Tiffany Smiley looked poised to outrun the primary and it’d probably translate downballot. Smiley ended up matching the primary, and Schrier ended up matching Biden’s margins in a truly impressive showing.
NY-04: The races where I underestimated the GOP all came from New York. A Long Island shocker, Kevin McCarthy can thank Lee Zeldin for this extra vote in his bid for the Speakership.
NY-17: This was genuinely kinda shocking, as I figured the big GOP spending onslaught against was more of a statement and the New Square Orthodox endorsement sealed the deal. But Zeldin’s win here along with Lawler being a strong candidate in his own right narrowly carried him through in the tipping race for house control.
NY-19: I thought Ithaca would’ve proved too much for Molinaro to overcome, but Riley’s 47 point win in Tompkins was not enough to keep Molinaro at bay in the rest of the district. A GOP pickup and likely a perennial battleground in the rest of the decade.
LESSONS LEARNED
Guessing the correct national environment is not that hard if you use the correct indicators (it was nail-able in the past 3 cycles), but the state-by-state predictions probably need variation with candidate quality and other factors. My 2024 prediction will probably still be quite nationalized (it’s a presidential cycle after all), but the “neutral midterms are local” phenomena of 1998/2002/2022 will probably continue to be a thing if one occurs again.
Oh yeah, trends are real, dread it or live it.
