Trading desks are moving in: prediction markets are shedding their “niche betting” tag and becoming a bona fide trading venue for quant firms.
Chicago powerhouse DRW — a multi-decade force in derivatives, fixed income and crypto trading since 1992 — has quietly opened a dedicated prediction-market desk focused on platforms like Polymarket and Kalshi. Job listings from the firm ask for traders who can watch prices on multiple platforms in real time, spot cross-platform mispricings, and execute microstructure and cross-platform arbitrage along with sub-second, news-driven momentum trades. Those are the same playbooks quants perfected in crypto derivatives, now being redeployed on sports and political event markets.
DRW isn’t alone. Algorithmic market maker Wintermute is recruiting traders with prediction-market experience, proprietary trading firm IMC is seeking quantitative traders for binary-event contracts, and even large crypto exchanges such as OKX and Crypto.com have posted similar roles. The hiring surge signals that institutional trading shops increasingly view prediction markets as a mature asset class with exploitable inefficiencies — and enough volume to matter.
Volume is the catalyst. Polymarket reportedly processed between $22 billion and $40 billion in trades across political, economic and sports markets in 2025 — up from essentially nothing three years earlier. Sports have become a major focus: Polymarket’s market for the UEFA Champions League Winner has seen $256 million in volume, the 2026 NBA Champion market $399 million, and the 2026 NHL Stanley Cup market $79 million. Together those three sports markets account for roughly $730 million — approaching the annual volume of some mid-sized European sports betting exchanges.
But institutional traders aren’t necessarily trying to out-predict veteran sports bettors. Harry Crane, a Rutgers statistician who studies prediction-market calibration, says the sharpest sports pricing still comes from specialized sports betting groups. Instead, quants are exploiting short-term market dynamics — information lag, latency, and fragmented liquidity across venues — rather than superior domain knowledge about outcomes.
A recent example makes that clear. On May 14, Andy Burnham’s chances in Polymarket’s “Next UK Prime Minister” market jumped from $0.24 to $0.43 amid leadership rumblings. Betfair, the large London exchange, had already marked Burnham at roughly $0.50. Polymarket took hours to catch up. For an experienced quant, that price gap is classic arbitrage: buy the underpriced contracts and hedge or sell on the faster venue. In one hypothetical, a trader buying $10,000 of Burnham contracts at $0.24 and exiting around Betfair’s level could have locked in roughly $7,900 of profit — with no need to wait for the political outcome.
Prediction markets add operational complexity, though. Currency and settlement differences (Betfair settles in sterling while Polymarket settles in crypto) demand infrastructure to move capital across currencies, exchanges, and settlement systems — precisely the advantage large trading shops can leverage.
Two structural features make prediction markets particularly attractive right now:
- Information lag: centralized exchanges and traditional sportsbooks often price events faster than decentralized platforms, creating exploitable windows.
- Liquidity fragmentation: the same event can trade across Polymarket, Kalshi, and legacy sportsbooks — no single venue always captures the entire market consensus.
On the modeling side, quants are applying familiar quantitative tools. Soccer traders use Dixon-Coles Poisson models to estimate goal distributions; basketball traders employ Bayesian hierarchical models that update team-strength estimates as information arrives. The aim is the same as in other markets: identify model-implied probabilities that diverge from market prices and capture “closing line value” (CLV) — the degree to which a trader’s price forecast beats the final market price.
Crane cautions that institutional capital alone won’t necessarily improve market accuracy: the sharpest sports players have long dominated pricing. Still, the talent migration is underway. Crypto market makers are learning sports analytics and expected-goals models, while traditional sports bet specialists are being recruited by crypto firms.
The infrastructure shift is visible too. Onchain derivatives venue HyperLiquid — which once processed over $10 billion in daily volume at peak — is preparing prediction markets for the 2026 World Cup, which will create thousands of correlated binary outcomes across 64 games.
Bottom line: trading desks are staffing up and building the tech to exploit latency, market structure, and cross-platform inefficiencies. Whether institutions can outpace decades-honed sharp bettors on pure outcome forecasting remains an open question. What’s no longer in doubt is that prediction markets are now on institutional trading maps — and the competition has begun.
Read more AI-generated news on: undefined/news
