Between now and election day, we’ll be doing a series of weekly updates on the state of the race in the House, Senate, and Presidency. Our goal is to keep you informed on what our models say on a weekly basis about the evolving state of the race as the election nears.
President
At the presidential level, Kamala Harris remains a very slight favorite over Donald Trump for the presidency, with our latest model update giving her a 57% chance of winning the presidency. This race remains exceptionally tight, and a lower-than-average polling miss in either direction could see Donald Trump win the presidency or lose all 7 swing states. But we would probably rather be the Vice President right now if forced to pick, if only by a whisker.
Although Harris trails in Georgia and Arizona by just over a point, she continues to hold steady in Michigan, Wisconsin, and Pennsylvania, with polls showing her either holding ground or gaining on Joe Biden’s narrow 2020 margins in each of these states. Winning these three states, plus Nebraska’s 2nd Congressional District (where she leads by over 5 points) would net her the necessary 270 electoral votes to win.
In addition to her healthier rust belt margins, Harris also maintains a slim lead in Nevada and is virtually tied in North Carolina, the latter of which opens up a few other pathways to the presidency for her (for example, carrying North Carolina and Nevada would allow her the buffer to lose one of the rust belt states and still win the election).
Our model remains slightly skeptical of Harris’ polling strength in North Carolina, but it is worth noting that there are factors that may really move the needle — for instance, Mark Robinson’s toxicity amidst his collapsing campaign has injected fresh uncertainty into the state’s political landscape, and Democrats continue to maintain an advertising advantage on the airwaves.
The movement of states is very random, and while North Carolina voted to the right of all the other swing states in 2020, it was also to the left of Georgia in both 2016 and 2022. To us, while we may not have expected this heading into the election, it is eminently plausible that North Carolina votes to the left of Georgia and Arizona alike at the moment.
Senate
In the Senate, Republicans remain favorites, with the model giving them a 79% chance of retaking the chamber. Currently, Republicans are favored in 51 seats to the Democrats’ 48. Ohio’s Senate race is a tossup but our model considers Republican Bernie Moreno a favorite to win in November.
Otherwise, the playing field has evolved significantly since our last update. In Montana, Republican Tim Sheehy continues to broaden his lead in the polls against incumbent Democrat Jon Tester. Our model now gives Republicans an 83% chance of flipping the seat, pushing Democrats to allocate resources elsewhere.
In Florida and Texas, where Democrats are hoping to score surprise pickups, Republicans remain modest favorites. Our model gives Rick Scott and Ted Cruz 73% and 71% chances of winning reelection, respectively. Still, Democrats now have a better chance of winning both of those states than they do Montana.
Aside from Montana and Ohio (where the GOP is narrowly favored), the Republican offense is still at a disadvantage. Out west, Democrat Ruben Gallego has a 77% chance of defeating Republican Kari Lake — who lost the 2022 gubernatorial election. Similarly, in Nevada, Senator Jacky Rosen has retained a significant polling lead despite the close presidential matchup, and her race sits at Likely Democratic.
In the Rust Belt, Republican nominees in the key Senate races are also at a disadvantage, as none of the Republican nominees can claim to have anything more than an underdog’s shot at the moment. The fundamentals would suggest their best bet is in Michigan’s open seat, but it is a bit curious that the GOP is spending so much more in trying to unseat Bob Casey in Pennsylvania right now.
House
Lastly, Democrats remain slight favorites in the House, with a 61% chance to retake the chamber. Our model shows Democrats favored in 215 seats, while Republicans are favored in 208 seats; 12 seats are tossups.
Perhaps the most tenuous hold for Democrats is Jared Golden’s seat in Maine’s 2nd District — though Golden has won in two tough cycles for Democrats, he is perhaps the most imperiled House incumbent this time around, with our model actually finding him a slight underdog against state senator Austin Theriault. This, along with Elissa Slotkin’s vacant seat in Michigan’s 7th district, are probably two of the most at-risk seats that House Democrats stand to lose.
However, a wave of decent polling in California has served as a valuable counterweight, with non-partisan polls finding Rudy Salas (CA-22), Adam Gray (CA-13), George Whitesides (CA-27), Derek Tran (CA-45), and Will Rollins (CA-41) all either tied or leading Republican incumbents. It goes without saying that California and New York are ground zero for the House in this cycle, and stronger-than-expected showings here from either party in these two states would put them in prime position to take the House back.
Here, it’s worth noting that Democrats have seen a welcome boost to their chances in New York, as news of an ethics violation committed by Anthony D’Esposito (NY-04) has thrown his already-slim chances into turmoil, with the former congressman’s extramarital affair coming to light in the process.
D’Esposito represented the bluest district won by Republicans in 2022 in any district, and in a Biden +14 seat, he was already fighting against the odds. Our model already placed his district at Leans Democratic, giving him just a 26% chance of retaining his seat. This new scandal may doom him.
D’Esposito’s potential collapse would be a major break for Democrats, especially considering that things in Mike Lawler’s seat appear to be drifting significantly in his favor, and against his old challenger Mondaire Jones — this formerly Leans Democratic seat is now squarely in the tossup category, and it actually would not surprise us to see it move more to the right as the election nears and more polls and fundraising data come out. (In fact, this update was finished and locked right before an Emerson College survey of New York found the Empire State at D+14 this year, which would be a 9 point swing right from 2023).
I’m a computer scientist who has an interest in machine learning, politics, and electoral data. I’m a cofounder and partner at Split Ticket and make many kinds of election models. I graduated from UC Berkeley and work as a software & AI engineer. You can contact me at lakshya@splitticket.org
My name is Harrison Lavelle and I am a co-founder and partner at Split Ticket. I write about a variety of electoral topics and handle our Datawrapper visuals.
Contact me at @HWLavelleMaps or harrison@splitticket.org