Analysis

    Why 70% of Prediction Market Traders Lose Money — What the Data Actually Says

    A WSJ investigation and independent academic study of 2.4 million accounts reveal that 67–77% of prediction market profits go to a tiny group of professional traders. Here is what the data says about who wins, who loses, and why.

    By PredictionMarkets.usFriday, May 8, 20269 min read
    Why 70% of Prediction Market Traders Lose Money — What the Data Actually Says

    The question gets asked constantly on Reddit, in Discord servers, and in DMs to prediction market influencers: Can you actually make money on Kalshi and Polymarket?

    As of this month, we have the most comprehensive empirical answer ever assembled. And it's sobering — but not for the reason most critics assume.

    The WSJ Investigation: The Numbers Are Stark

    A Wall Street Journal analysis published May 4, 2026, examined 1.6 million Polymarket accounts that have traded since November 2022. The headline finding: 67% of all profits on Polymarket went to just 0.1% of accounts — fewer than 2,000 users — who collectively netted nearly half a billion dollars.

    More than 70% of Polymarket users have lost money overall. The typical user is down somewhere between $1 and $100. The bottom 10% of traders lost an average of $4,000 each.

    Kalshi's own numbers are similarly lopsided. Spokeswoman Elisabeth Diana told the Journal that there are 2.9 unprofitable users for every profitable one, based on data from the past month. The platform doesn't publish total user counts or comprehensive profit data.

    "Casual traders are bleeding cash while a small number of sophisticated pros — including trading firms with access to vast streams of data — eat their lunch," the Journal reported.

    The Academic Study: Same Picture, Bigger Dataset

    An independent academic paper published in March 2026 — before the WSJ investigation — reached nearly identical conclusions using a larger dataset.

    Researchers Pat Akey (ESSEC Business School), Vincent Grégoire (HEC Montréal), Nicolas Harvie, and Charles Martineau (University of Toronto) analyzed 2.4 million Polymarket users, $67 billion in trading volume, and 588 million trades from 2022 through March 2026. Their paper, Who Wins and Who Loses In Prediction Markets? Evidence from Polymarket, found:

    • 68.8% of users lost money
    • The top 1% of traders captured 76.5% of all gains
    • The top 0.1% alone took more than half of total platform profits

    "Polymarket investors are, on average, trading against efficient prices they cannot profit from," the researchers wrote.

    CNBC reported in May 2026 that the combined trading volume of Polymarket and Kalshi has reached approximately $60 billion so far in 2026, and Bernstein analysts project the sector could hit $1 trillion in annual volume by 2030.

    Who Is Actually Making Money?

    The WSJ identified several categories of winning accounts:

    Quantitative trading firms with data infrastructure. Susquehanna International Group — co-founded by billionaire Jeff Yass — is believed to trade hundreds of millions of dollars through Kalshi each week, according to professional traders monitoring the order book. Jump Trading is also active on both platforms. Citadel Securities President Jim Esposito said in mid-April that his firm is "absolutely keeping an eye" on prediction markets.

    Algorithmic retail traders with professional-grade setups. Michael Boss, a former professional poker player and statistician, places 60 trades per minute on Kalshi and modifies his bids and asks 30 times a second. He's made more than $668,000 on the platform, mostly on sports markets. "Casual traders have no chance. Systematically," Boss told the WSJ.

    Speed-focused startup trading firms. Jonathan Stoll-Ryan, a college student, co-founded a firm that is among the top-five traders by volume on Kalshi's crypto markets. The company pays more than $200,000 per year for live data feeds, AI coding assistants, and server infrastructure, executing tens of thousands of trades daily.

    Market makers. The academic study found that the single strongest predictor of profitability was using limit orders rather than taking posted prices. Moving from pure taker to pure maker activity reduced the probability of losing money by 9.3 percentage points. Professional market makers earn the bid-ask spread on completed round trips — Kalshi and Polymarket both offer fee rebates or waivers to market makers who provide liquidity.

    The Mention Market Problem

    One category stands out as particularly brutal for retail traders: Kalshi's mention markets, where users bet on whether a public figure will say a specific word during a broadcast.

    The WSJ analyzed more than 35,000 completed mention markets on Kalshi and found that "yes" trades priced at a 50% winning probability paid out only about 40% of the time. In other words, bettors are systematically overpaying for these contracts.

    The average retail trader who buys "yes" at the first listed price — the most common behavior — loses 11% of their stake on mention markets, according to the Journal's analysis. That's worse than most Las Vegas slot machines, according to research from the University of Nevada, Las Vegas.

    Why do these markets skew so badly? The academic literature points to long-shot bias — the well-documented human tendency to overvalue unlikely events. Professional traders largely avoid mention markets precisely because no amount of data gives a reliable edge. The outcome depends on a single human utterance in a live broadcast.

    Monthly mention market volume on Kalshi has exploded since mid-2025, driven by social media campaigns and influencers promoting their wins. Bank of America analysts noted in an April report that "mention-market streams on social media often go viral and improve Kalshi's brand awareness." In February 2026 alone, Kalshi users bet nearly $181 million on mention markets.

    Kalshi's spokeswoman acknowledged the long-shot bias in mention markets, but said the platform's overall pricing is more accurate and that mention markets show better calibration in the four hours immediately preceding an event.

    The Paradox: The Market Works — Just Not For You

    Here's the uncomfortable truth that critics tend to miss: the academic study also confirms that prediction market prices are remarkably accurate. When a Polymarket contract trades at 30 cents, the underlying event happens approximately 30% of the time. At 80 cents, it happens about 80% of the time. The crowd wisdom thesis holds up.

    The problem for retail traders isn't that the market is wrong. It's that the prices are already right, which means you need genuine insight that others don't have to beat them. And when your competition is algorithmic trading firms with $200,000/year data subscriptions executing 60 trades per minute, "genuine insight" has a very high bar.

    The former Kalshi employee who coined the "fish" term for casual traders — Adhi Rajaprabhakaran — told the WSJ that while he still considers casual traders fish in general, their presence in the market actually serves an important function: it attracts sophisticated traders, which in turn makes prices more accurate. Retail money is the substrate that makes the information machine run.

    What This Means If You Want to Trade

    The data doesn't say "never trade prediction markets." It says: know where you're actually competitive.

    Markets where retail has historically done better:

    • Local/niche knowledge markets. The WSJ profiled a trader who made consistent gains on Detroit snowfall totals — a market where his local experience was an edge no algorithmic firm could easily replicate with data.
    • Sports with specialized expertise. Boss himself noted sports as his primary profit driver, not despite the competition, but because he could identify systematic pricing errors that generalist algorithms miss.
    • Political markets you follow closely. Retail traders with strong domain expertise in specific political races or regulatory proceedings can sometimes be ahead of market makers who rely on aggregated polling data.

    Markets where retail historically struggles most:

    • Mention markets (long-shot bias + no data edge)
    • Extremely liquid, widely followed markets (professionals dominate)
    • Markets with complex settlement rules that casual traders misread

    Structural tips from the academic research:

    • Use limit orders instead of market orders. The paper found this single change reduces the probability of losing by more than 9 percentage points.
    • Avoid contracts priced below 10 cents or above 90 cents. These "longshot" positions are where retail money goes to die — 63% of retail trades, according to the study, are placed at these extreme prices.
    • Trade smaller, more obscure markets. Professional capital concentrates on liquid, high-volume events. Smaller markets with lower volume sometimes offer genuine informational edges.

    The Regulatory Context

    The profit concentration data is now part of the political debate over prediction market regulation. Several bills introduced in the current Congressional session would impose new consumer disclosures, restrict advertising to casual users, or tighten definitions around what kinds of markets are permissible. The 41-state attorneys general coalition that filed a comment with the CFTC in late April 2026 cited the "gambling-like" behavior of retail users as evidence that state gaming oversight should apply.

    The platforms argue — correctly, based on the academic data — that their prices are accurate and provide genuine public value as information aggregation tools. That argument can be true simultaneously with the data showing that most retail users are net losers. Both things are real.

    The CFTC is still finalizing rules under its Advanced Notice of Proposed Rulemaking, with public comments closed as of April 30. Whatever framework emerges will define how aggressively platforms can market to retail users and what disclosures they must provide about risk.

    FAQ

    Do most people lose money on prediction markets? Yes. Based on the Wall Street Journal's analysis of 1.6 million Polymarket accounts and Kalshi's own spokeswoman, roughly 70% or more of active users are net losers.

    Who makes money on prediction markets? The winners are predominantly algorithmic trading firms, quantitative traders with professional data infrastructure, and market makers — participants who post limit orders rather than taking existing prices. The top 0.1% of Polymarket accounts captured 67% of profits, according to the WSJ.

    Are prediction markets rigged? The academic evidence says prices are well-calibrated and accurately reflect probabilities. The problem isn't manipulation — it's that accurate prices mean you need information others don't have to profit. Most retail traders don't.

    Is Kalshi or Polymarket better for casual traders? Neither platform publishes comprehensive outcome data by user type, but the dynamics are similar. Both attract professional trading firms. Kalshi's mention markets are documented as particularly bad for retail outcomes.

    Can you beat prediction markets as a retail trader? It's possible in niche markets where you have genuine local knowledge or specialized expertise. The academic study shows skill is real and persistent among the top performers. The question is whether your edge is large enough to overcome the execution advantage held by professional market makers.

    Bottom Line

    The new data from the Wall Street Journal and independent academic researchers tells a consistent story: prediction markets produce accurate prices and transfer money from a large pool of retail participants to a small group of sophisticated traders. Both things are true.

    If you're a retail trader, the market isn't cheating you — it's pricing you fairly. The problem is you're playing against opponents with significantly more information, faster execution, and greater discipline. That's not a flaw in the market design. That's what efficient markets do.

    You can trade prediction markets and do well. You just need to be honest about where your actual edge is — and stay out of the markets where you don't have one.

    Sources & Verification

    Back to Articles