Made a copy of this market, but for after the 2026 election:

Bearish on Dems controlling the Senate. None of the Republican seats are competitive right now, and GOP is virtually guaranteed to take Joe Machin's seat. Meanwhile, Democrats have 7 competitive races they need to win if they want to maintain 50 seats. Plus the vice-presidency.

Most likely outcome is that GOP keeps the 49 seats they already have, plus Manchin and Tester's seats, as well as one of the "Toss UP" elections. That would put them at 52.

What if Senate is 50 R, 49 "D" (King and Sanders included), and Murkowski splits off post election but before the swearing in?

then we're screwed and probably Manifold Politics will have to do something silly like resolving 50% between a Senate D and Senate R option

@Najawin I suspect AP won't officially call Senate control until the new Senate is in session.

Upgrading this market to premium!

Currently, Nate Silver has the national odds at 53.7% Harris to 45.9% Trump.

The Economist has it at 51% Harris to 48% Trump.

Meanwhile our friends at Metaculus currently have 55% Harris to 45% Trump, and their Balance of Power market shows this:

Our friends at Metaculus love to be overconfident in all of their predictions

I've been (cleaning up) in the Metaculus quarterly competition this last month, and one thing I've noticed is as soon as a market falls below like ~10% (or above ~90%) probability, there's some sort of weird downward spiral effect where people notice the market probability dropping and adjust their own probabilities downwards, causing it to go even further down. You end up with some weird situations where a market gets to 0.1% in cases where that level of confidence is completely and utterly unwarranted, but it rarely teaches the traders a lesson, because even when the actual probability is 10x more, that will only go awry 1% of the time.

Quite amazing you can sell a trifecta at 25% while Tester is at 32%, and Moreno is at 54%.

Adding this together, Republicans have a 48.7% chance of winning presidency to Dems 51%. Strong arbitrage with this market.

What GOP wins elections, and Dems refuse to certify it?

@Joshua just bet no on R trifecta

Nah

or copy poly that's fine too

These odds are so bizarre

ok I took more of mars martian's mana

More concerned with these:

I think there’s way more than a 5% likelihood that one of these unusual scenarios occurs. Probably more like 20%. But I’d still lose mana 4 out of 5 times even if I’m right 😢

The senate is quite high volatility, as it can be swung easily by 1-2 elections. So not super correlated with general election margin. And house and Pres. election are not so correlated that they only go opposite ways 17% of the time!

(Especially in an election with as much uncertainty in the presidential outcome as this one)

D pres, R senate, R house: roughly correct, maybe higher

R pres, D senate, R house: <0.1%

R pres, D senate, D house <0.01%

D pres, D senate, R house <0.1%

There would simply have to be a red wave concentrated entirely in the coasts/blue states for that to happen

Maybe 2x as large as 2022

The senate is quite high volatility, as it can be swung easily by 1-2 elections. So not super correlated with general election margin. And house and Pres. election are not so correlated that they only go opposite ways 17% of the time

No, presidential years have much less ticket splitting. Also, winning the Presidency basically gives you an extra senate seat, so there's more correlation than you expect

0.01% is actually wild for R Pres, D senate, D house. Imagine a scenario where Kamala has a scandal (or Trump just experiences a massive popularity boost from assassination attempt) and loses while generally dems are popular and win house and senate. That happens in way more than 1 in 10,000 universes.

R Pres, D senate, R house seems quite plausible to me: republicans have a good election but dems have particularly likable senate candidates (and reps particularly unlikable ones) and pick up one seat (this kind of happened last midterms more or less!)

I don’t think that’s true about Pres years having less ticket splitting. It’s quite normal for popular senate candidates to run 5-10% ahead of a Pres candidate! If polling is any indication, before Biden pulled out, he was polling like 15% behind swing state dem senate candidates!

And yes, you’re right about the VP seat giving control of senate, but the senate is currently 50-50!

poly:

DRR 2.2

RDR 1.9

DDR 0.8

RDD 0.6

That’s really interesting that it replicates on Polymarket. I’ll be interested to see how that shifts once 538 and Nate come out with their usual models on how correlated these are

I don’t think that’s true about Pres years having less ticket splitting. It’s quite normal for popular senate candidates to run 5-10% ahead of a Pres candidate!

Ticket splitting has gone way down since 2020 in red states. Nobody runs 10 points ahead anymore. See: Steve Bullock, Phil Bredesen, Tim Ryan, Doug Jones

If polling is any indication, before Biden pulled out, he was polling like 15% behind swing state dem senate candidates!

The dem senate candidate's opponents are basically winning by name ID, happens every cycle with incumbents. The Dems are all incumbents (except AZ).

Also, only one split ticket in 2020, and zero in 2016