I have my predictions about this election. They place me in opposition to just about every major forecasting agency in significant ways, and that makes me uncomfortable. That being said, I believe in my numbers and my process.
So I will be up front: my maps tonight represent a bull case for democrats, where they over-perform the assumptions of 538, Split-Ticket, Sabato, and Cook. I have my reasons for believing this will happen, but there is a ton of uncertainty!
Despite what disagreements with my maps you might have, understand that once the votes start coming in, all of my assumptions about the electorate are gone, and I do think my area of expertise far more lies on election night than before it.
• • •
Missing some Tweet in this thread? You can try to
force a refresh
Be on the lookout for blueshifts as the days drag on in NY, CA, and potentially AZ. NY-19 in particular might be a dem hold in the end, and the CA 3 (13/22/41) all will probably blueshift considerably over the next weeks.
I *believe* that all of the in-person votes have been counted in AZ, leaving around 300k mailers. If true, AZ races should be blueshifting from now on, generally - AZ1 and AZ6 are both doable for dems, though it's hard to say how likely.
NV I won't touch with a ten foot pole, but CO-8 has remaining ballots that point to Caraveo being advantaged; more adams votes are left than Weld+Larimer votes.
OR-5 is a big question; if I had to guess, dems will win it, but its very very close and hard to say.
I did pretty fucking good! National environment read looks spot-on, my overall house numbers is exactly what NYT predicts to be the average case, I was the only person to have Bobert as competitive...
To be sure, I was wrong in many races: PA-7, RI-2, between 3 and 5 NY races, *probably* two Colorado races, and potentially up to eight or nine races between OR, NV, AZ, and CA.
In statewide offices, I think I can claim victory: as of now, I have a 1.00 batting record.
When it comes down to it, my read on the national environment was more accurate than any of the pundits and most people who watch elections. I cleanly swept the Midwest house races, and I'm guessing that I'm going to do quite well in CA, once the votes are in.
Only substantial remaining votes out are from Mesa county - max 12kish. Bobert has a 5k vote deficit districtwide, so needs to win them 70/30. Mesa had a 5-point dem turnout advantage, compared to .5 in e.g. Montrose, Delta.
This poses a problem, because Montrose swung 14 points towards Ds from 2020, and virtually any shift in Mesa *overall* would mean that Bobert can't make her margins in the district. In fact, no county that is above 95% has swung for dems by little enough for Mesa to save her.
AND, this assumes that other ballots remaining out in the district are a net wash - not at all a guarantee, especially because the exact quantity of outvote is still a bit up in the air in places like deep-blue Pitkin and Eagle counties.
Adding on a few things that we talked about: 1. Pueblo is the last part of the district to have relatively low turnout, but state Dems are descending on the county like a flock of Vultures to a carcass; most of the red parts of the district really can't have that many more votes.
2. The race will be decided in four counties: Mesa, Garfield, Montrose, and Delta. All four are traditionally very red, and Mesa in particular is moving left rapidly. If Frisch can break 43 in Mesa, 36 in Delta/Montrose, or 55 in Garfield, it is ominous news for Bobert.
3. I, personally, think that Frisch wins NPAs in the district by four. The NPAs voting today are not necessarily the same NPAs that voted in 2020. If nothing else, the fact that they are voting earlier, rather than later, may be evidence of them being more democratic-leaning.
2022 MIDTERM PREDICTIONS:
This cycle is a hard one to be entering into the forecaster game, and at the end of the day, I have decided to stick with my guns. I believe that the Republican bull case is overstated, and that the final generic ballot looks more like D+0 than R+4.
I do not differ from other forecasters in what seats are competitive, for the most part - but my predictions are still to the left of the conventional wisdom by a pretty wide margin, especially in the house. This is in-keeping with my assumptions about the environment.
If I am wrong about the environment, and we are facing closer to an R+4 year, I am going to be wrong in significant ways, and inaccuracies in my predictions should be attributed to a misunderstanding on my part on the state of polls and pundits.
1.34 million votes have been cast, and the data points to the following:
Barbara Kirkmeyer will eke out a victory over Yadira Caraveo.
Brittany Pettersen will comfortably defeat Eric Aadland.
Lauren Bobert will underperform substantially.
There is a very clear separation of these three races when it comes to relative turnout: In both CO-07 and CO-03, Democrats have punched in the upper single digits above their weight, but in CO-08, Republicans have thus far over-performed.
I was looking for Republicans to pick up steam in CO-03 before making sure that this was not just a mirage, and once again, Adam Frisch leaves today in a better spot than he started it in, with Bobert having only a 14,134 vote lead with turnout in the stratosphere.