TOTAL VOLUME:
$126.8b
24H VOL:
$122,314,324
24H TRANSACTIONS:
2,159,476,470
OPEN INTEREST:
$1,331,186,481
338,503
Markets across
31,526
events
MATCHED EVENTS:
2,864
PLATFORM COVERAGE:
5
Polymarket:
40%
VS.
Kalshi:
60%
Wash trading is a trader taking both sides of a contract to fake volume with zero real risk. Here's how it works, why on-chain platforms are more exposed, and the signals that give it away.
Jared Polites
Sep 12, 2026

TL;DR
Pulled up a market with heavy 24-hour volume and noticed the price hasn't moved at all? That gap, loud activity next to a flat chart, is usually the first sign you're looking at wash-traded volume rather than real disagreement.
Wash trading is when a trader buys and sells the same contract against themselves, or a linked wallet, to manufacture volume that never reflects real risk-taking. On decentralized prediction markets, this is structurally easier to pull off than on centralized exchanges, and it distorts the volume rankings researchers use to judge which markets actually matter.
A wash trade requires no opinion about the outcome. One party, or two wallets controlled by the same party, takes both sides of a contract, pays the spread and any fees, and walks away with a matched position and net exposure near zero. The trade still counts toward the platform's reported volume.
Repeat it enough times and a market with almost no genuine interest can look like one of the most actively traded events on the board.
The motive varies. Some wash trading is designed to farm token incentives or liquidity mining rewards tied to trading volume. Some is meant to make a market look liquid enough to attract real traders, who then provide the genuine volume the wash trader was hoping to bait. Either way, the volume number stops meaning what a researcher assumes it means.
Centralized exchanges like Kalshi require identity verification for every account and settle trades through a single custodial ledger. That doesn't make wash trading impossible, but it raises the cost: a wash trader needs multiple verified identities, and the exchange's compliance team has both the incentive and the regulatory obligation to look for exactly this pattern.
Decentralized platforms remove that friction by design. A single person can generate as many wallets as they want, fund each one, and trade between them with no identity check at any step.
If a platform also runs a token incentive program that rewards volume or liquidity provision, self-trading between two wallets can be directly profitable, independent of whether either wallet has a real view on the outcome. That combination, permissionless wallet creation plus volume-linked incentives, is what makes wash trading structurally cheaper on-chain than off-chain.
It is not an accusation against any single platform. It is a property of the architecture, and it means on-chain volume figures deserve more scrutiny than centralized ones, not less trust.
The CFTC's own enforcement guidance backs up why that scrutiny matters on the regulated side of the industry. Pre-arranged, noncompetitive trading and wash sales are named directly as prohibited practices under Section 4c(a) of the Commodity Exchange Act, enforceable on any CFTC-regulated exchange. An unregulated, on-chain venue doesn't carry that same enforcement backstop, which is exactly why the wallet-level checks below matter more there.
Aggregate volume alone won't tell you whether a market is wash traded. You need to look at the shape of the trading, not just the total.
Genuine price discovery leaves a mark on the chart. Wash trading doesn't. A large position with no wallet history behind it and no follow-through afterward reads a lot like the pattern our guide to spotting a manipulated market walks through in more depth. A wallet that keeps showing up on both sides of the same trade is worth checking against the wallet-concentration signals in our prediction market whale guide too.
PredictionHero aggregates data across Polymarket, Kalshi, Limitless, Predict.Fun, and Opinion so a researcher can compare the same event across platforms instantly. As of this writing, PredictionHero matches 3,225 events across all five platforms, out of more than 397,000 markets tracked in total.
That comparison is only useful if the volume behind each number is real. A market showing $2 million in daily volume should carry more weight in your read of consensus probability than one showing $200,000, but only if that $2 million reflects independent traders taking positions, not a handful of wallets rotating the same contract.
Say two platforms price the same event at 40% and 55%. If the 55% price sits on a market with a handful of wallets doing most of the round-tripping and flat price action despite heavy turnover, that number deserves less trust, not more, even though it looks like the more "active" market. The lower-volume, cleaner market with fewer wallets and visible price movement in response to news is the one worth weighting.
This is also why raw volume rankings, by themselves, are a weak filter. A platform's top-10-by-volume list treats a market with organic participation from a thousand traders the same as one dominated by a coordinated group cycling funds through a handful of wallets. Cross-referencing volume against wallet concentration and price stability is what separates a real consensus from a manufactured one, and it's the same discipline behind reading a volume figure correctly in the first place.
No, but it changes the economics. Kalshi's identity verification requirement makes systematic wash trading materially harder to sustain there than on a permissionless on-chain venue, which is one reason cross-platform comparison is useful in the first place.
If Polymarket, Limitless, Predict.Fun, or Opinion all show heavy volume with flat pricing on a market where Kalshi's version is thinly traded but moving, the divergence itself is informative. It doesn't prove wash trading on the higher-volume venues, but it's a legitimate reason to weight the platform with visible price discovery more heavily, the same logic behind checking why odds disagree across platforms in general.
Wash trading is when a trader takes both sides of the same contract, directly or through linked wallets, to create the appearance of trading activity without taking on real directional risk. The trades count toward reported volume even though no genuine disagreement about the outcome occurred.
Decentralized platforms let anyone create unlimited wallets without identity verification, and some pair trading with token incentives tied to volume. That combination makes self-trading both easy to execute and, in some cases, directly profitable. Centralized, identity-verified exchanges raise the cost of doing the same thing.
Look for round-trip trades between the same or linked wallets, volume that isn't moving the price, turnover concentrated in a small number of wallets, and volume that outpaces growth in open interest. Any one of these alone isn't proof, but several together are a strong signal.
Not automatically. High volume from many independent traders is a strong signal. High volume from a handful of wallets cycling the same contract is not. Check wallet concentration and price movement before treating volume as a proxy for accuracy.
On a CFTC-regulated exchange like Kalshi, wash sales and pre-arranged noncompetitive trading are explicitly prohibited under Section 4c(a) of the Commodity Exchange Act, and the CFTC's enforcement division has said it has full authority to police them on any designated contract market. On unregulated, on-chain platforms, enforcement depends on the platform's own terms of service rather than a federal regulator.
Cross-referencing volume against wallet concentration and price stability, on the actual platform a market trades on, catches more manufactured volume than reading a leaderboard alone. Compare odds, volume, and consensus probability across Polymarket, Kalshi, Limitless, Predict.Fun, and Opinion on PredictionHero.
PredictionHero aggregates publicly available prediction market data for informational purposes only. This is not financial advice. Prediction markets may not be available in all jurisdictions.
Related

Wash trading, thin-book spikes, resolution clustering, and self-dealing are the four patterns that give away a manipulated price. Here's how to check for each one.
Jared Polites · Aug 26

Volume is the total traded over time. Open interest is what's still at stake right now. Liquidity is what you can trade this instant, and most headlines quietly blur all three into one number.
Jared Polites · Aug 5

Position size alone proves nothing. Here's how to separate real information from ordinary risk-taking and wash trading, and why some platforms let you check a wallet's history while others don't.
Jared Polites · Sep 4