March Madness 2026: What Prediction Markets Are Saying Heading Into the Sweet 16

Two rounds are done. 48 teams are home. And the prediction markets have already been proven right about some things and dead wrong about others.

Kalshi has blown past $100 million in championship futures volume on this tournament alone — making March Madness 2026 the biggest sports trading event in prediction market history outside the Super Bowl. The $1 billion perfect bracket contest? Over. No perfect brackets survived past the first weekend, which surprises exactly nobody. The top entry — a user called "KingMe03" — missed only UCLA over UCF heading into Sunday and is in pole position for the $1 million consolation prize.

So with the Sweet 16 set, what are the markets actually telling us?

The Championship Odds Have Shifted Hard

Here's how the market has repriced since tip-off. Compare this to where things stood a week ago:

The biggest story: Michigan has overtaken Duke as the market favourite. Before the tournament, Duke was priced at 19% and Michigan at 18%. That gap has flipped — Michigan now sits around 23% implied probability (roughly +340 in betting odds), with Duke dropping to third behind Arizona.

Arizona's surge is the other headline. They weren't in the top 6 pre-tournament. Now they're the second favourite at around +400. The market likes their depth and how they've handled the first two rounds.

The Upset That Repriced Everything

The 9-seed Iowa Hawkeyes knocked out 1-seed Florida. That single result sent shockwaves through every championship futures market on the platform.

Remember the seed market from last week? It had 1-seeds at 65% to win the whole thing — historically high. With Florida gone, only three 1-seeds remain (Duke, Michigan, Arizona), and that 65% number has come down. The market was pricing this year's top seeds as stronger than usual. It was partly right — three of four survived — but that "partly" matters when you're trading real money.

Iowa at ~4% to win it all is the classic Cinderella pricing. Probably too high on pure basketball merit, probably too low if you believe in momentum and a favourable bracket path. Either way, the fact that a 9-seed has a tradeable championship contract is exactly what makes prediction markets more interesting than brackets.

The Sweet 16 Field — What the Market Sees

This year's Sweet 16 is all major-conference teams for the second straight year — the only two instances in tournament history. But the seedings are wild:

The Cinderellas still standing: 9-Iowa, 11-Texas (came through the First Four with three wins in four days), 6-Tennessee, and 5-St. John's. None of those four programs has ever won a national championship. The market isn't giving any of them better than mid-single-digit odds individually, but collectively they represent real chaos potential.

The chalk: Michigan, Duke, Arizona, Houston, Purdue are all still alive and holding the lion's share of championship probability. Michigan State, Illinois, Alabama, and Arkansas round out the field.

Thursday's marquee game: 2-Purdue vs 11-Texas. The Longhorns are the only double-digit seed left, and they've been on a tear. If Texas pulls this off, the championship market reprices again.

Why This Matters for Prediction Markets

This is the first NCAA Tournament where prediction markets are a genuine factor in how people follow the games. Kalshi's March Madness handle has already surpassed $100 million — with the Sweet 16, Elite Eight, Final Four, and Championship still to play. Projections for total tournament volume now sit well above $150 million.

The Billion Dollar Bracket contest pulled massive attention. Kalshi accepted up to 10 million entries, each requiring a verified account. Even with no winner, the marketing value was enormous — Fortune compared it to Warren Buffett's famous bracket challenge from 12 years ago.

And Polymarket is running March Madness markets too. Their liquidity has historically skewed toward politics and crypto, but the sports push is real. Which means you can now compare how two different crowds are pricing the same outcomes.

If Kalshi has Michigan at 23% and Polymarket has them at 20%, that 3-point gap is either noise or information. You can track those spreads in real time on predictions.io, which pulls the odds from both platforms into one view. During a tournament with games playing simultaneously across multiple venues, having a single dashboard instead of flipping between apps makes a real difference.

Markets to Watch This Week

Michigan's price trajectory — are they peaking or still climbing? If they tick above 25% at any point before the Elite Eight, that's a strong signal the sharp money has picked its horse.

The Arizona path — they've got a potentially clear run to the Final Four. Watch whether their price converges with Michigan or stays in the 15-18% range.

Iowa's Cinderella run — every win doubles their price. A Sweet 16 win could push them to 8-10%, which starts to feel like a real contender.

The Schiff-Curtis bill — Senators introduced bipartisan legislation this week to ban sports contracts on CFTC-regulated platforms. If it gains traction, it would fundamentally reshape Kalshi's sports business. Worth watching the political prediction markets for the meta-irony of prediction markets predicting their own regulation.

I'll be tracking all of these on predictions.io through the tournament. Follow the live markets there for the cross-platform view.

Frequently Asked Questions

Can you bet on March Madness on Kalshi?

Yes. Kalshi offers championship futures, game-level markets, and seed markets for the 2026 NCAA Tournament. Kalshi is CFTC-regulated and available to US users in all states at the federal level. Over $100 million has been traded on March Madness markets so far. Note: a bipartisan Senate bill introduced in March 2026 could restrict sports contracts on prediction market platforms if passed.

Does Polymarket have March Madness markets?

Yes. Polymarket has been expanding into sports markets in 2026, including March Madness. US access to Polymarket is currently via waitlist following their CFTC approval in late 2025, and varies by state.

How accurate are prediction markets for March Madness?

Prediction markets aggregate the opinions of thousands of traders putting real money behind their forecasts. They correctly priced Michigan and Duke as top contenders and were sceptical of some higher seeds that have since been eliminated. They're not perfect — Iowa's upset of Florida proves that — but they're one of the best publicly available probability sources, and they reprice in real time as games play out.

Who is favoured to win March Madness 2026?

As of March 24, Michigan leads prediction markets at approximately 23% implied probability, followed by Arizona (~18%) and Duke (~15%). Houston and Purdue are in the 6-7% range. The Sweet 16 begins March 26.

Track all March Madness prediction markets — Kalshi, Polymarket, and more — in one place at predictions.io.

Last updated: March 24, 2026. Markets move fast during the tournament — check live prices before trading.