BANGER
2026-06-09 · 6 min read · facts as of 2026-06-12

Can You Make Money on Polymarket?

Yes, you can make money on Polymarket. But the honest framing matters: most accounts fund price discovery rather than capture it. A working paper published in April 2026 by researchers from London Business School and Yale University put numbers on this for the first time at scale.

What the Research Actually Says

The study, titled "Prediction Market Accuracy: Crowd Wisdom or Informed Minority?", analyzed the complete transaction history of Polymarket: 98,906 events, 210,322 markets, $13.76 billion in trading volume, and 1.72 million accounts from 2023 to 2025. The authors used a sign-randomization test that replicated each trader's bets 10,000 times to separate genuine skill from luck.

The results are blunt. The study classifies accounts into four buckets: skilled winners (3.14% of accounts), lucky winners (29%), unlucky losers (61.4%), and unskilled losers (the remainder). The unlucky and unskilled loser groups absorb the entirety of aggregate losses. Market makers and skilled takers together represent roughly 3.5% of accounts but capture more than 30% of total gains. The paper's conclusion: the majority of participants fund accuracy rather than produce it.

Crucially, skill showed persistence. Skilled Polymarket traders retained their classification 44% of the time out-of-sample, compared to just 10% for skilled mutual funds. That is a meaningful signal that a small group of participants genuinely has edge, not just variance. Lucky winners, despite their positive profits, had no significant predictive power when tested on separate events.

Volume Is Real, but So Is Noise

Polymarket's growth is not in question. The platform recorded $21.5 billion in trading volume in 2025, according to Dune Analytics data. Monthly volume on Polymarket alone crossed $10 billion for the first time in March 2026, reaching $10.57 billion. Combined monthly volume on Polymarket and Kalshi reached roughly $24 billion in April 2026, according to a Pew Research Center analysis of data from The Block.

But volume headlines deserve a discount. A November 2025 Columbia University study estimated that roughly 25% of Polymarket's historical trading volume came from wash trading, where users bought and sold contracts back and forth without changing their net exposure. Suspected fake trades peaked at nearly 60% of weekly volume in December 2024. Sports and election markets were hit hardest. The researchers did not allege Polymarket was complicit; they pointed to the platform's no-fee structure and pseudonymous on-chain design as structural enablers.

The implication for traders is practical: volume figures are not a reliable proxy for real liquidity in a given market. Check the order book, not the headline number.

Where Edge Actually Comes From

If only 3.14% of accounts qualify as skilled winners, what makes that group different? The research and practitioner experience point to a few specific sources:

The US Regulatory Picture (as of June 2026)

The legal status of Polymarket for US persons has changed materially. In July 2025, Polymarket acquired QCX LLC (and its clearinghouse affiliate QC Clearing LLC, together known as QCEX), a CFTC-licensed derivatives exchange, for $112 million. In September 2025, the CFTC issued a no-action letter for QCX covering certain recordkeeping and reporting requirements. On November 25, 2025, the CFTC issued an Amended Order of Designation, permitting Polymarket to operate as a fully regulated US Designated Contract Market through its QCX subsidiary. Polymarket began onboarding US users through this new structure in December 2025 and removed its US waitlist in May 2026, now serving users in more than 40 states.

The federal picture is clearer than it was, but the state picture is not settled. Nevada has court-ordered restrictions on certain contracts, and multiple other states have active litigation against prediction market platforms broadly. The regulatory situation continues to evolve; check CFTC guidance and your state's specific rules before trading. Nothing in this post is legal advice.

Why Automation and Paper Testing Matter

The 3.14% of skilled accounts almost certainly are not clicking through a browser interface. Edge in prediction markets is behavioral and informational; it erodes quickly when it becomes crowded, and it requires consistency across dozens or hundreds of markets to compound. Manual trading makes that nearly impossible to maintain.

The first discipline is testing before risking capital. A strategy that looks sharp in your head will often look different when you paper-trade it against the live order book for 30 days. Fills, timing, and resolution lags behave differently than a spreadsheet suggests.

This is the use case Banger is built for. You write a Python strategy (subclassing banger.Strategy), run it in paper mode against real Polymarket and Kalshi order books, and only graduate to live when the metrics justify it. A declarative risk envelope (per-trade cap, daily loss stop, max open positions, kill switch) keeps a live strategy from running off-script. Banger never custodies funds; you bring your own venue API keys.

pip install bangertrades
banger run strategy.py --paper

The Honest Summary

Polymarket is a real market with real money and real edge for a small minority of participants. The April 2026 research confirms that 3.14% of accounts qualify as skilled winners and that their skill persists out-of-sample. It also confirms that the other 96.86% of accounts are not classified as skilled winners, with the loss-absorbing majority funding returns for the informed minority. The path into that skilled group runs through specific structural advantages: speed, thin-market mispricing, cross-venue arbitrage, and information. None of those advantages are exploitable consistently without systematic, automated execution and rigorous paper testing first.

Sources

Run your first strategy free

Paper-trade on Polymarket and Kalshi in minutes, with your own keys.

Start free