Midway through election night, Ashley Hinson's odds in the Iowa Senate race doubled, from 13 cents to 27, shortly after one of the panelists called her a live shot. Earlier in the day, a Democratic vice presidential nomination contract sat at 1 cent. By 5:55 p.m. it was at 6.
Those are the moves PredictQ's Midterms Live Sweat Show was built to catch. Justin Herzig was joined by Jordan Cooper, Squirrel Patrol and Grayson Nichols to trade and talk through the night as results came in. Brian Hartman pulled the key moments into an edited episode, and this article walks through the lessons: how to build a thesis before the first vote is counted, how to think about polling error, why the 2028 markets started moving on election night, and why the best trade often isn't in the headline market.
Watch the Episode Here - Read the Breakdown Below
Start With a Simple Thesis
Jordan Cooper came into the night with a framework he describes as "keep it simple, stupid." He's bet on the past four elections and done well in all of them, and he doesn't build elaborate models.
"When it comes to the polls, you care more about what people do than what they say," Jordan said. "I care about data that shows who actually showed up and did something. What they say is a descriptive element you can adjust for after the fact."
His approach borrows from his background in digital marketing, specifically Facebook lookalike audiences, where you take a small sample of known customers and look for more people who resemble them. Applied to elections, that means finding past races that look like this one.
"Here are the deciding factors. A lame-duck president who's hugely unpopular. There's a war going on. The economy is the top issue. The incumbent party is on the defensive. The closest resembling election to that is 2006, and you know what happened in 2006? A blue wave."
Then the polls came in with a generic ballot between D+8 and D+11. "You'd literally have to think that this time it's different," Jordan said. "Trump has been in office or in our general vicinity for 10 years. I don't think this time is different."
The takeaway for traders: before election night, write down the historical comparison you're working from and the few factors that drive it. When prices move on early results, you have something to check them against.
Polling Error and Correlation
The panel pushed on whether the polls themselves could be trusted. One question was whether pollsters had overcorrected. The New York Times/Siena poll made a large effort to reach more Trump voters after 2024, but the 2022 midterm polls were fine without that adjustment, and Trump voters don't reliably turn out in midterms.
Jordan's answer came back to correlation, a concept anyone from fantasy sports already knows well.
"People from daily fantasy just understand correlation more than the general public, and probabilistic distributions of outcomes more than the general public," he said. "In a presidential year, the top of the ticket is the same in all 50 states, so if the polling is off by three, it's going to be off by three pretty much everywhere. In a midterm, if you're voting for Collins or Jackson in Maine, you're not voting for Talarico or Paxton in Texas. You can have polling errors that are localized to certain states. Everyone being afraid of a humongous, nationwide polling miss was a non-starter for me."
That distinction matters for how you size positions. In a presidential year, a set of state-level bets on the same party is effectively one big bet. In a midterm, those bets are less tied together, so a miss in one state doesn't automatically mean a miss everywhere.
Watching for the Reversal
Even with a strong view, the group kept questioning what the early returns were telling them. Later in the night, Squirrel Patrol flagged the market starting to turn.
"I think things are starting to shift a little bit in real time," Squirrel said. "Grayson, you just mentioned you thought Hinson had a really good shot at winning, and her odds have just doubled, from 13 to 27. Mississippi is looking pretty solidly red. Nebraska is turning pretty solidly red. There's obviously a lot of Democratic strength tonight, but maybe we're getting a little carried away with some of the East Coast results."
East Coast states report first, and the early picture can pull prices further than the full night will support. If you're trading live, the question to keep asking is whether a move reflects new information about the race you're in, or just the order in which states report.
The 2028 Markets Start Moving on Election Night
Grayson Nichols was already looking two years out. Nobody knows what will happen by 2028, he said, but election night attention can move a candidate's price in the short term, and that creates a trade with a defined exit.
"I was looking at Ossoff," Grayson said. "He's going to have a big win, I imagine he's going to give a fiery victory speech, and his campaign is very good at going viral. I could see someone like him gaining traction in the markets, and you could get on him early and then exit in a week or two when the buzz starts to die down. To be fair, he's said he doesn't plan on running for president, but we've heard that from plenty of candidates who ended up running."
The same thing showed up on the Democratic vice presidential nomination board, where one contract went from 1 cent earlier in the day to 6 cents by 5:55 p.m. And Grayson pointed to James Talarico's price after he won the Texas Senate primary: "It ballooned really fast in a week. You can spot some of these things, plan a good exit strategy, since who knows what happens afterward."
There's also a flow-of-money angle. "You've got to assume a lot of people are collecting their Democrat bags tonight, and they're going to redeploy that money somewhere," he said.
The point of this trade is to own something attention is about to find, then sell into that attention. You're betting on the next two weeks of headlines, and the exit plan is part of the trade from the start.
Look Beyond the Headline Markets
Jordan's last point came straight from daily fantasy bankroll management.
"If you have a small bankroll, go for the markets with the biggest edge and the smallest liquidity," he said. "In daily fantasy, if you only have $100 on an NFL Sunday, maybe you don't play the Millionaire Maker or the $25 double-up. You find the small contests where you can predict what everyone else will play and take a different angle. There are edges there, but you can't get much money down. You can't get $50,000 down and keep that edge."
Prediction markets work the same way. "People get focused on the top line. Who's going to win the House, who's going to win the Senate, who's going to win Texas. If you think Talarico is going to win Texas, figure out the correlation. If that's the environment, he's probably going to win by at least seven. If you only have $1,000, it's much better to spend $100 on a 12-cent contract that correlates to what would be a 56-cent contract in the main market."
His summary: use the fact that big money can't fit in small markets to your advantage. Pick off the smaller, correlated contracts first, and leave the main markets to the people with six figures to deploy.
There's a cost to that, and Jordan ran into it live. "I'm looking at the order book on one of the closest races, and I can't even sell my position. That's how low the liquidity is." Small markets give you better prices on the way in and a harder time on the way out. Size them so you're comfortable holding to resolution.
Entry Price Matters Most on Longshots
Justin closed with two pieces of advice on longshots.
"If you can take something at 1 cent and it gets to 2 cents, that's the same as getting something at 25 cents and watching it go to 50. You're doubling your money either way," he said. "A lot of people looking at a longshot candidate don't consider whether they're getting it at 2 cents or 3 cents. They just want the longshot. But the difference between 50-to-1 at 2 cents and 100-to-1 at 1 cent is huge."
And the candidate doesn't have to win for the trade to work. "Get those 1-cent and 2-cent names, and all you need is that small movement. If you can project that this person is just going to be talked about, they don't need to win. You can sell out."
Where Do Things Stand Now?
None of these approaches is free. A strong historical thesis can still get carried by early returns that don't hold, which is exactly what Squirrel was flagging mid-night. Correlated side markets offer better prices, but as Jordan found, you may not be able to sell when you want to. And the 2028 attention trade only works if you actually take the exit. A candidate who says he isn't running may mean it, and a contract bought at 6 cents on election night can drift back toward 1 when the news cycle moves on.
As Brian put it at the end of the episode: having a view is only the beginning. The contract, the price and your ability to exit all matter too.
Which election-night moves do you think will hold, and which 2028 names are getting ahead of themselves? Bring your takes to the PredictQ Discord. New to these markets? Start with our Prediction Markets 101, and check the current sign-up offers before you open an account.
This article is adapted from an edited recording of PredictQ's Midterms Live Sweat Show, recorded live on election night (November 3, 2026). It is not an exact transcript, a directional prediction, a trade recommendation, or investment advice. Prices and commentary reflect the discussion at the time of recording, are point-in-time, and may have changed. PredictQ is a marketing partner of Kalshi and Polymarket and may receive compensation for referrals.