Prediction Markets Weekly: Feb 22-28, 2026
Trump Mentions and Tehran Bombs: The Week Prediction Markets Got Very Real
The week started boring and ended with geopolitical chaos, insider trading allegations, and the biggest single-day volume spike in prediction market history. If you thought betting on what Trump would say during his State of the Union was wild, wait until you hear about the traders who made $1.2M betting on when the U.S. would bomb Iran.
The Headline Numbers (Week Ending Feb 22)
Total volume held steady at $5.25B, down just 1.4% from the prior week. But these numbers only capture through Feb 22, before the Iran strikes sent Friday into absolute overdrive. Here's the breakdown:
Kalshi: $2.59B (up 6.7% WoW) Polymarket: $1.83B (down 3.2% WoW) Combined: $4.41B (83.9% of total market)
Kalshi is now at 49.3% market share, inching closer to that psychological 50% mark. Sports volume jumped 10.9% to $2.21B even without NFL games, thanks to NBA's return from All-Star break and college basketball heating up as Selection Sunday (March 15) approaches.
The story gets way more interesting when you add in the final few days of the month. Feb 28 alone saw Polymarket hit $478M in single-day volume, a record that obliterated the previous high. Politics accounted for $220M of that as Iran markets went nuclear (pun very much intended).
State of the Union: $17M on Mentions, Handshakes, and Tie Colors
Trump delivered his Feb 24 State of the Union address, and prediction markets turned it into a $17M+ spectacle. Mention markets were the star of the show:
Kalshi deployed 45 words and phrases you could bet on. The favorites hit exactly as expected: "America/American 25+ times" at 96% (free money), "250" for America's 250th birthday at 93%, and "fraud" at 88%. Trump said "America" or "American" 67 times, easily clearing the 25-mention threshold.
The surprise winner? "Discombobulator" drew $62K in volume at 9% odds. Trump didn't say it, but the fact that thousands traded on whether the president would use the word "discombobulator" on national television tells you everything about where we are as a society.
Polymarket focused on the meta-questions: speech length (100+ minutes hit at 66% odds, paid out easily as Trump rambled for 112 minutes), mentions of specific people (Biden at 95%, got mentioned 14 times), and the genuinely absurd stuff like tie color (red won at 100%) and whether VP Vance would clap more than 100 times during the speech (that market surged from 22% to 66% as traders realized a 100+ minute speech meant a lot of standing ovations).
Combined SOTU volume: $17M+ across both platforms, with Kalshi taking $8.6M and Polymarket grabbing $2M in mention markets alone. The event proved that mention markets drive outsized engagement relative to their actual importance, a lesson both platforms learned during the Super Bowl and are now banking on for every major speech.
February 28: When Prediction Markets Became War Markets
Then Friday happened.
The U.S. and Israel launched coordinated strikes on Iran, killing Supreme Leader Ayatollah Ali Khamenei and triggering the single biggest day in prediction market history. Polymarket processed $478M in notional volume on Feb 28 alone, with politics accounting for nearly half at $220M.
The Iran strike timing contract accumulated $529M in total volume, with Feb 28 specific bets drawing $89.6M. Every daily contract from Feb 28 through early March instantly resolved to "yes" after the strikes began, meaning anyone who bought the specific date before the attack collected on a binary bet about when the U.S. military would bomb another country.
Markets that exploded after the strikes:
Khamenei out by March 31: $58M volume on Polymarket, $54M on Kalshi
Iranian regime fall by June 30: jumped from low-20s to 54%
U.S.-Iran ceasefire by April 30: 78% probability
Next Supreme Leader markets: dozens of contracts on succession
The Insider Trading Problem
And then the controversy hit. Blockchain analytics firm Bubblemaps flagged six wallets that made approximately $1.2M by correctly betting on the Feb 28 strike, with most positions opened just hours before the military operation. The largest wallet turned $61,000 into $493,000. Another netted $120,000 from a $30,000 position.
The trading patterns were suspicious as hell:
Wallets funded within 24 hours of the attack
Bet specifically on Feb 28 rather than broader timeframes
Purchased "yes" shares hours before the strike
Many had no prior trading history
One account, "Magamyman," made $553K on the Khamenei market alone, with the first trade placed 71 minutes before news broke publicly. Sen. Chris Murphy (D-CT) immediately vowed to introduce legislation: "It's insane this is legal. People around Trump are profiting off war and death."
The Kalshi Death Carveout Fiasco
While Polymarket traders cashed out, Kalshi users got rugged. The platform had a "death carveout" in its Khamenei contract stating that if he left power due to death, the market would resolve at the last traded price before his death, not at 100%. This meant traders who correctly predicted he'd be out couldn't profit from his death.
The backlash was immediate and brutal. Traders who watched Kalshi promote the market for days on social media felt deceived by fine print. CEO Tarek Mansour defended the policy: "We don't list markets directly tied to death. When there are markets where potential outcomes involve death, we design the rules to prevent people from profiting from death."
Kalshi refunded all fees and net losses out of pocket, incurring a substantial loss to make users whole. But the damage to trust was done. Discord and Twitter lit up with accusations that Kalshi is "a scam" and that "centralized oracles will always bend to compliance over reality."
User Metrics: The Great Polymarket Consolidation
Buried in the week's data is an interesting trend: Polymarket's weekly active users dropped 18% from 309,991 to 254,730, but the platform processed 22.7M transactions (up from prior weeks). This means fewer users trading more frequently in larger sizes.
Whether this is healthy consolidation around engaged power users or early retail churn depends on your perspective. The fact that Polymarket added $580M in absolute volume the following week (week ending March 1) suggests the remaining user base is highly engaged and willing to deploy serious capital.
College Basketball Filling the NFL Void
Sports volume grew 10.9% to $2.21B even without football, driven by NBA and college basketball. With Selection Sunday on March 15, college hoops markets are deepening daily. Big Ten and ACC conference tournament games are drawing serious volume, and bracketology markets are starting to appear.
Kalshi's sports dominance continues: $2.0B+ in sports volume vs Polymarket's ~$721M. But Kalshi's 85.4% sports concentration creates regulatory vulnerability, while Polymarket's category diversity (sports 39.6%, crypto 26%, politics 25%) provides structural buffers when any single category cools.
February Monthly Wrap
February closed at $22.3B in total volume, down 14.5% from January but up significantly on a per-day basis ($795M/day vs $840M in January, just -5.4%). Kalshi posted $9.93B (+3.9% MoM), Polymarket posted $7.94B (+3.7% MoM). Both platforms grew month-over-month, the decline came from smaller platforms.
The Super Bowl spike is officially unwound, but the Iran strikes proved prediction markets can spike even harder on geopolitical events than sports. The question heading into March: will regulatory scrutiny from the insider trading controversy slow growth, or will the publicity just attract more degens?
What's Next
The ceasefire markets give just 4% chance of U.S.-Iran peace by March 2, jumping to 61% by March 31. Regime change markets are at 54% by June 30. Bitcoin bounced to $68K on the thesis that this resolves quickly, and prediction markets are pricing the same outcome.
March Madness bracket markets are coming online, Fed decision volume will spike on March 17-18, and Massachusetts geofencing deadline hits March 11. The Ninth Circuit oral arguments on federal preemption are scheduled for April.
The platform war meta-market on Manifold still has Polymarket at 47% vs Kalshi at 34% for 2026 volume crown (sports excluded). This week didn't change those odds much, both platforms proved they can spike on different catalysts.
But the real story is that prediction markets just demonstrated they can price geopolitical events faster than traditional news media, handle half a billion in volume in a single day, and create enough controversy to get senators introducing legislation. The industry is growing up fast, whether regulators and the public are ready or not.
Markets don't sleep. Neither do the degens betting on when bombs will drop.