2024 Senate Race Rundown: Part 1
2024 Senate Race Rundown: Part 1
Could Democrats win a Senate seat in Texas? Florida? Nebraska??? It’s not as crazy as it sounds. A politics update by Jacob Cohen (@Conflux).
In a midterm year like 2018 or 2022, there’s frequent talk about Congress. But whenever there’s a presidential year, I feel like the presidential election always sucks up the oxygen.
Today, I’ll right that wrong! In this post is a comprehensive overview of Manifold’s forecasts for half of the Senate elections this year. (The other half will be in Part 2.)
But first: if the presidency is a tossup, and the Democrats currently have a 51-49 advantage in the Senate (including independents like Bernie Sanders who “caucus with Democrats”), why do Republicans have an approximately 80% chance to take the chamber?1
It’s because I lied above. This post will not talk about half of the 34 Senate elections which are up this year. It’ll talk about less than a third: the 11/34 seats currently held by Republicans. So Republicans have only 11 places to lose, but 23 to gain.
Additionally, there’s stark contrast in the amount of competitive races. Joe Manchin is retiring, so Democrats can write off West Virginia. (This takes Democrats from 51-49 down to 50-50.) They also have an uphill battle in Montana, and are defending competitive races in Ohio, Arizona, Michigan, Pennsylvania, Wisconsin … well, that’s for next week, but the list goes on.
If Democrats lost WV but held on in all those states — a tall order — and won the presidency, they’d have a winning 50-50 majority.
But where can they make gains? Their main avenues are Texas, Florida, and Nebraska.
That’s right, Nebraska! Even though it’s a ruby-red state, there’s a 19% chance that we’ll have a Democratic senator from Nebraska!
I’ll explain later, but I should get down to business: in addition to those three states, I’ll be covering Indiana, Mississippi, Missouri, Nebraska (special), North Dakota, Tennessee, Utah, and Wyoming. Let’s get started!
But first: did you know that Manifold is hosting a Election Day Watch Party in Berkeley? Here are ten reasons you should come:
It’s free!
There will be food!
It’s at Lighthaven which is a magical place!
It will last from 2pm-1am, and you can come for as short or as long as you want!
It will have special guests! Including Maxim Lott (who’s behind electionbettingodds.com).
There will be fun sessions! The schedule isn’t totally finalized, but it’s planned to include Election Pastcasting (I tell an anonymized story of a historical election, and you bet as I go!), a trivia contest, and perhaps an election-based game with secret roles that I’ve been concocting. (If you have ideas for activities, get in touch!)
Community! You can also just hang out with other Manifolders. Or your friends who you bring. Lighthaven has so many great nooks to hang out and chat.
You can ask me questions! In addition to attempting to be really on top of the election news, I’ll be helping manage the Manifold Twitter (aka X), where I’m planning to issue live decision desk-style projections of races.
You can bug the Manifold team with questions! For example, “The website is down, pls fix???”
You can look at physical live election needles! Like the New York Times needle, which graphically displays a live prediction of the election as results come in. But they’ll be physical devices hooked up via robotics to our own markets!!! They will display results for the presidency, major swing states, and Congress. They’re currently being crafted by the multi-talented Verda Korzeniewski. (The materials have begun arriving, and the Python script has already been written!) This project is inspired by this example made by Andrew, Justin, Derik, and Joe2 of Cavendish Labs. Watch the video! It’s incredible.
Tuesday, November 5th, 2pm-1am, Lighthaven, Berkeley. View this market and this Partiful for more details. I’d love to see y’all there!
If you can’t make it, manifold.markets/tv will have some type of livestream as well. And keep your eyes on @ManifoldMarkets on Twitter!
Wasn’t I Promised Senate Races?
Okay, fine.
For each race, I’ll start with some stats — forecasts and past results. I don’t know much about The Hill’s Senate forecast, but it’s the only free one I could find from a major newspaper. Some of their results seem rather dubious to me, but I’ve included them anyway for context!
Then we’ve got expert forecasts — the Cook Political Report, Sabato’s Crystal Ball, and Inside Elections — which actually do have long track records of success. They use words like Safe/Likely/Lean/Tossup instead of probabilities, though. There’s the polling averages; they’re from post-Nate Silver 538 but they’re pretty reliable for this sort of thing. And finally some old election results.
So Let’s Start With the Classics
There’s Texas, a red state that has tantalized Democrats, and Florida, a former swing state that has been drifting away. All the way back in November 2022, I made a market asking where the Republican margin would be higher. It says 75% Florida. So all eyes are on Texas.
Texas: Colin Allred (D) vs. Sen. Ted Cruz (R)
Democrat win chances (Manifold, sweepstakes, 73 traders): 18%
Democrat win chances (Manifold, play-money, 130 traders): 17%
Expert forecasts: Lean R (Cook), Likely R (Sabato), Lean R (Inside Elections)
Polling average (538): R+4; Cruz 48.8%, Allred 45.0%
Election result in 2018, a Democratic wave year: R+2.5 (Cruz vs. O’Rourke)
Election result in 2012, amid Obama's reelection: R+16 (Cruz vs. Sadler)
Recent Texas presidential election results: Trump+6 (2020), Trump+9 (2016), Romney+16 (2012)
Manifolder commentary:
Richard Hanania: I think if there’s one place [Republicans might lose seats], it’s Texas, where Ted Cruz is running against Colin Allred, a UC Berkeley School of Law graduate, member of the House, and former linebacker for the Tennessee Titans … I say Allred is underpriced, but my view is mostly vibes based. Polls that show a tight race report that Allred’s favorability is positive, while Cruz is underwater … I’m going to give Allred a 30% chance of pulling off the upset.
Cremieux: My first bet will be safe and not worth all that much: I think Ted Cruz is going to win reelection, and I'll throw $20 at it.
Harold Reasons: If there is a god ted Cruz will lose so we can get one that doesn't leave his voters everytime there is a disaster. Poop on cruz3
Conflux commentary:
This is a cool election for me, because my friend’s mom worked on Colin Allred’s campaign when he was elected US representative in a 2018 upset victory. The same year, Beto O’Rourke waged a highly publicized campaign against the unpopular Ted Cruz for Senate (in an article from 2018, Clare Malone wrote “[O’Rourke has] gone on Ellen and The Late Show With Stephen Colbert, and has been compared to a Kennedy an embarrassing number of times in profiles … we’re in the midst of that self-mythologizing, magical moment for O’Rourke”) before running for president in 2020. Beto lost both races — the Senate race narrowly, the presidential race not so much.
Since 2018-Beto was such a good year for Democrats, can they really go beyond that in 2024 in order to do the unthinkable and win in Texas? They have a popular candidate, and the Republicans don’t. Democrats have proven, with Arizona, that they can sometimes win previously red states in the Southwest. But the fact is, Cruz is winning polls. Overall, I think a 20% chance is about right here, but there’s room for debate.
Florida: Debbie Mucarsel-Powell (D) vs. Sen. Rick Scott (R)
Democrat win chances (Manifold, 44 traders): 10% (it’s been 18% and 27% at other points today)
Expert forecasts: Likely R (Cook), Likely R (Sabato), Likely R (Inside Elections)
Polling average (538): R+5; Scott 48.2%, Mucarsel-Powell 43.5%
Election result in 2018, a Democratic wave year: R+0.1 (Scott vs. Nelson)
Election result in 2012, amid Obama's reelection: D+13 (Nelson vs. Mack)
Recent Florida presidential election results: Trump+3 (2020), Trump+1.2 (2016), Obama+0.9 (2012)
Manifold commentary (from Discord, I couldn’t find any on the website):
Gabrielle: Interesting Senate predictions from The Hill, they're seeing FL as more likely Dem than MT … And notably they give Dems better odds in FL than in TX
SemioticRivalry: Montana is and always has been likely R, bordering on safe. as is florida
there's no way they beat Rick Scott in this environment
if anyone disagrees i have a million mana and nothing to spend it on
Conflux commentary:
What is it about the Florida Senate race that makes it seem so hopeless and devoid of energy? It was a virtual tie in 2018, as you can see, but Florida has been drifting away from Democrats. In 2022, Marco Rubio won his Senate race by a 16-point margin. A prestigious New York Times poll for the presidency showed Trump winning by 13 in the state, and Manifold gives a 38% chance that Trump will win Florida by more than 8 points (a bet that Keith Rabois could have made with Nate Silver after their Twitter feud).
But sometimes, in elections, trends don’t continue. Often, there is reversion to the mean, where an outlier result — like Republicans in 2022 — does not shape the future trend. Rick Scott is leading Former Rep. Debbie Mucarsel-Powell in the polls by about 5%, but that is within a normal polling error. Even if people don’t talk about it, Democrats aren’t out of the woods in Florida. Still, the most likely outcome is a clear reelection of Scott.
As this article was in the process of being revised, there was some activity on the market. Gabrielle bet it up to 27%, toward the Hill figure, but Semiotic brought it down to 18%. I tend to agree with Semiotic on this.
Now it’s time for the Nebraska election, where there’s somehow a 19% chance of a Democrat.
Nebraska (regular): Dan Osborn (Independent) vs. Sen. Deb Fischer (R)
Osborn win chances (The Hill model, which treats Osborn as a Democrat and seems a bit dubious): 14%
Who will Osborn caucus with? (Manifold, 15 traders): 77% Democrats, 6% Republicans, 17% neither
Expert forecasts: Likely R (Cook), Likely R (Sabato), Likely R (Inside Elections)
Polling average (538): R+1; Fischer 44.2%, Osborn 43.1%
Election result in 2018, a Democratic wave year: R+19 (Fischer vs. Raybould)
Election result in 2012, amid Obama's reelection: R+16 (Fischer vs. Kerrey)
Recent Nebraska presidential election results: Trump +19 (2020), Trump+25 (2016), Romney+22 (2012)
Manifolder commentary from April:
Magnus: There’s still a decent chance the Democrats will endorse Osborn. Either way, this isn’t going to be a close race.
Timothy Johnson: He might be better off if they don't officially endorse him…
There was a single poll in November that showed Biden down by 18 (53-35), and Dan Osborn winning 40-38. I don't quite believe that, but I could believe that he'll do much better than Biden.
Manifolder commentary from more recently:
Timothy Johnson: The latest poll showed Fischer winning by just one point, 39-38.
SemioticRivalry (on Discord): not gonna totally rule it out … osborn is doing the right stuff
Conflux commentary:
Nebraska is a red state. No Democrat has ever come close to winning there.
But Dan Osborn isn’t a Democrat. He’s a veteran, labor union leader (who led a strike at Kellogg’s Omaha plant), and independent. For now, anyway — if he wins, Manifold gives him a 77% chance of joining Bernie Sanders, Angus King, and the other members of the Democratic caucus in the Senate. (Not caucusing with any party in the Senate would deny him committee assignments and other perks.) Multiply his 24% chance of victory with his 77% chance of caucusing with Democrats and you get a crazy 19% overall chance of a Democratic senator from Nebraska.
Instead, he’s running as an independent. There was some drama in the state Democratic party about whether to field a candidate; initially, they planned to endorse Osborn on May 18, but only three days before, Osborn said he wouldn’t be “charmed” or “schmoozed” and didn’t want the Democrats’ help. The state party originally looked for a write-in candidate, but ultimately decided not to field anyone. So Osborn’s shrewd strategy may have paid off: it’s a two-way race between him and incumbent Deb Fischer, and he seems like a true independent. (There is a chance he’ll caucus with no one, which would make him a true independent.)
I wouldn’t have guessed this race would be competitive at all. There have only been a handful of Independent vs. Republican races in red states like this: Kansas in 2014, Alaska in 2020. It’ll be an uphill battle for Osborn here, but the polls are very close. This will be an interesting race.
It’s time to consider the others.
Do the Democrats have a shot in any of the remaining races? Let’s go through in alphabetical order!
For these, there are not enough polls for 538 to give a full average, so I’ll just list those conducted in September or October. And there is generally little Manifold activity for these, so the commentary will all be my own. It’ll be a lot of fun facts and historical context. These aren’t competitive races.
Indiana: Valerie McCray (D) vs. Jim Banks (R)
Democrat win chances (Manifold, 7 traders): 2% (after my bet)
Expert forecasts: Solid R (Cook), Safe R (Sabato), Solid R (Inside Elections)
Poll results (538): Banks+12, Banks+14, Banks+13, Banks+10
Election result in 2018, a Democratic wave year: R+6 (Braun vs. Donnelly)
Election result in 2012, amid Obama’s reelection: D+6 (Donnelly vs. Mourdock)
Recent Indiana presidential election results: Trump+16 (2020), Trump+19 (2016), Romney+10 (2012)
Commentary:
In 2012, longtime Republican senator Richard Lugar was primaried by Tea Party candidate Richard Mourdock. Then Mourdock lost to Joe Donnelly, a Democrat who published a basketball-themed crossword in the New York Times once. Despite 2018 being a good year for Democrats, Indiana proved too red for Donnelly to hold on, and he lost to Mike Braun. Braun is now running for governor, making this an open seat. But Indiana’s pretty red these days. If Sen. Donnelly couldn’t win in 2018, it seems far-fetched for Valerie McCray, a psychologist, to beat Rep. Jim Banks in 2024.
On October 12, a new user bought M100 in McCray shares, moving the market probability from 2% to 80%. I presume they didn’t realize they were doing this. Then, moneylab moved them back down to 5%. I’ve taken the liberty to drop the probability back down to 2%, where I think it belongs.
Mississippi: Ty Pinkins (D) vs. Sen. Roger Wicker (R)
Expert forecasts: Solid R (Cook), Safe R (Sabato), Solid R (Inside Elections)
Poll results (538): Wicker+13 (one partisan poll)
Election result in 2018, a Democratic wave year: R+19 (Wicker vs. Baria)
Election result in 2012, amid Obama's reelection: R+17 (Wicker vs. Gore)
Recent Mississippi presidential election results: Trump+17 (2020), Trump+18 (2016), Romney+12 (2012)
Commentary:
Today I learned: The 2012 election for this seat had Roger Wicker running against Albert N. Gore, who was apparently 92 and claimed to maybe be distantly related to the famous Al Gore.
Albert N. Gore, alas, did not come close to winning. Neither will Ty Pinkins.
Missouri: Lucas Kunce (D) vs. Sen. Josh Hawley (R)
Expert forecasts: Solid R (Cook), Safe R (Sabato), Solid R (Inside Elections)
Poll results (538): Hawley+8, Hawley+12, Hawley+5, Hawley+4, Hawley+4, Hawley+15
Election result in 2018, a Democratic wave year: R+6 (Hawley vs. McCaskill)
Election result in 2012, amid Obama's reelection: D+16 (McCaskill vs. Akin)
Recent Missouri presidential election results: Trump+16 (2020), Trump+19 (2016), Romney+10 (2012)
Commentary:
Claire McCaskill’s victory in this seat in 2012 is a remarkable story, which she wrote up here. It begins “It was August 7, 2012, and I was standing in my hotel room in Kansas City about to shotgun a beer for the first time in my life. I had just made the biggest gamble of my political career—a $1.7 million gamble—and it had paid off. Running for reelection to the U.S. Senate as a Democrat from Missouri, I had successfully manipulated the Republican primary so that in the general election I would face the candidate I was most likely to beat. And this is how I had promised my daughters we would celebrate.”
But despite her popularity, she lost in 2018 to Josh Hawley, who became the first senator to say he’d object to certifying Joe Biden’s 2020 victory. He’s generally considered somewhat far-right. His 2024 opponent, Lucas Kunce, has run for office before, but never won. He’s an attorney, veteran, and national security director for the American Economic Liberties Project. While Missouri has the potential to be more competitive than the other races in this category — The Hill has Kunce’s chances at 8% — I’d trust Manifold’s 4% more. My prediction: Hawley will win by greater than his 2018 margin against McCaskill.
Nebraska (special): Preston Love Jr. (D) vs. Sen. Pete Ricketts (R)
Expert forecasts: Solid R (Cook), Safe R (Sabato), Solid R (Inside Elections)
Poll results (538): Ricketts+16, Ricketts+18
Election result in 2018, a Democratic wave year: R+19 (Fischer vs. Raybould)
Election result in 2012, amid Obama's reelection: R+16 (Fischer vs. Kerrey)
Recent Nebraska presidential election results: Trump+19 (2020), Trump+25 (2016), Romney+22 (2012)
Commentary:
In 2020, the Democratic Senate nominee in Nebraska was Chris Janicek. He was urged to drop out after scandals including sexual harassment. The Democratic party switched to supporting Preston Love, Jr. as a write-in candidate. Janicek got 24% of the vote anyway, and Love got only 6%. They lost to Ben Sasse, who voted to convict Trump in his impeachment trial, but later left the Senate to serve as president of the University of Florida — a position he has somewhat unceremoniously left.
All this to say, in addition to the exciting election in Nebraska, there’s a special election to fill Sasse’s vacancy. The Republican nominee is Pete Ricketts, governor from 2015–2023, heir to TD Ameritrade,4 and the man appointed to fill the seat. The Democratic nominee: Preston Love, Jr., who got 6% of the vote in 2020.
North Dakota: Katrina Christiansen (D) vs. Sen. Kevin Cramer (R)
Expert forecasts: Solid R (Cook), Safe R (Sabato), Solid R (Inside Elections)
Poll results (538): Cramer+22, Cramer+9, Cramer+29 (the last two are partisan polls)
Election result in 2018, a Democratic wave year: R+11 (Cramer vs. Heitkamp)
Election result in 2012, amid Obama's reelection: D+1 (Heitkamp vs. Berg)
Recent North Dakota presidential election results: Trump+33 (2020), Trump+36 (2016), Romney+20 (2012)
Commentary:
Like Indiana and Missouri, this seat was won by a Democrat in 2012 who then lost in 2018. Heidi Heitkamp’s loss was even more decisive. I don’t think anyone’s holding their breath for 2024.
Tennessee: Gloria Johnson (D) vs. Sen. Marsha Blackburn (R)
Expert forecasts: Solid R (Cook), Safe R (Sabato), Solid R (Inside Elections)
Poll results (538): Blackburn+23, Blackburn+23, Blackburn+20
Election result in 2018, a Democratic wave year: R+11 (Blackburn vs. Bredesen)
Election result in 2012, amid Obama's reelection: R+34 (Corker vs. Clayton)
Recent Tennessee presidential election results: Trump+23 (2020), Trump+26 (2016), Romney+21 (2012)
Commentary:
In 2018, Taylor Swift gave a surprise endorsement to Phil Bredesen, the popular former governor of Tennessee who was making a longshot bid for Senate. While Bredesen outperformed the partisan lean of Tennessee, he still lost by 11 to Congressman Marsha Blackburn. This time, the Democratic nominee is State Rep. Gloria Johnson, who seems interesting but is not as well-known. It’s hard to imagine her even doing better than Bredesen.
Utah: Caroline Gleich (D) vs. John Curtis (R)
Expert forecasts: Solid R (Cook), Safe R (Sabato), Solid R (Inside Elections)
Poll results (538): Curtis+29, Curtis+30, Curtis+19
Election result in 2018, a Democratic wave year: R+32 (Romney vs. Wilson)
Election result in 2012, amid Obama's reelection: R+35 (Hatch vs. Howell)
Recent Utah presidential election results: Trump+20 (2020), Trump+18 (2016), Romney+48 (2012)
Commentary:
In 2018, Mitt Romney became the oldest freshman senator. He’s not running for reelection, and the Republican nominee this time is Rep. John Curtis. I couldn’t find any good fun facts about this race.
Wyoming: Scott Morrow (D) vs. Sen. John Barrasso (R)
Expert forecasts: Solid R (Cook), Safe R (Sabato), Solid R (Inside Elections)
Poll results (538): None
Election result in 2018, a Democratic wave year: R+37 (Barrasso vs. Trauner)
Election result in 2012, amid Obama's reelection: R+54 (Barrasso vs. Chesnut)
Recent Wyoming presidential election results: Trump+43 (2020), Trump+46 (2016), Romney+41 (2012)
Commentary:
Zero polls. In all of the other races, someone — maybe a campaign — bothered to commission at least one poll. But in Wyoming, they know there’s no point.
I hope you appreciated the roundup of the Republican-held Senate races up for election this year! (Sorry to any East-Coasters who felt lied to when I promised a post on Tuesday — Manifold is headquartered in San Francisco so only Pacific Time is real.) Next Tuesday, I’ll cover the Democrat-held seats, from Jon Tester’s uphill battle in Montana to the nail-biter in Ohio to the Arizona seat held by Kyrsten Sinema. These should be more exciting and competitive. And then next Tuesday will be Election Day. Let me know if there’s anything else you’d like to see before the election!
One Last Market
Before you go, my tradition is to highlight a "fun" market that’s little known. This one is classic strutheo: betting on it requires combining forecasts of two unrelated things. In this case, the longevity of Dick Cheney, and the Great Internet Mersenne Prime Search (which just found a new prime!)
As usual, I’ve given the market a M1,000 subsidy to encourage accuracy.
I’m Jacob Cohen (aka Conflux), a student at Stanford University. I blog at tinyurl.com/confluxblog, host puzzles at puzzlesforprogress.net, and don’t release any new episodes of the Market Manipulation Podcast on your favorite podcast platform. Check out Manifold Politics at manifold.markets/elections, or via its Politics or 2024 Election categories. And if you enjoyed this newsletter, don’t forget to forward, like, share, comment, and/or subscribe to Above the Fold on Substack!
Thanks to my proofreaders, who contributed a lot to this post — particularly Gabrielle, who told me about The Hill’s forecast, and my dorm-mate Emmanuel, who suggested good line edits but keeps stealing my water bottle.
As of now it’s 75% in the sweepstakes market and 82% in the play-money market. Arbitrage opportunity…
Joe M. Cavanagh was erroneously left off the original tweet. He is considered by many to be a romantic catch.
To clarify, I don’t think this quote has any forecasting value, I just think it’s funny.
At least, his dad was the founder. Does that make him automatically the heir?
Thanks for throwing the late party, I think I will need it.