The number that gets thrown around is that somewhere between half and two thirds of reported global crypto volume is fake, and that estimate keeps showing up across separate studies from Forbes, Bitwise, and the National Bureau of Economic Research. Some exchanges report 10 to 100 times more volume than they actually have. It gets worse on unregulated venues and thin tokens, but there is some of it almost everywhere, so it is worth knowing how to spot.
Wash trading is just buying and selling the same asset at the same time so it looks like something is happening. Your net position never changes. You are only manufacturing volume. In regulated markets that is illegal and people get prosecuted for it. In crypto the enforcement is spotty, and plenty of exchanges quietly benefit from it because a bigger reported number pulls in more users and more listing fees.
Why anyone bothers
Exchanges like inflated volume because the rankings on CoinMarketCap and CoinGecko lean partly on it. Higher up the list means more visibility, more signups, and more listing revenue. Some venues run rebate structures where market makers get paid for posting limit orders, which means you can wash trade at zero cost or even get paid for it. The exchange is literally subsidizing its own fake volume.
Token projects do it because listing agreements often carry minimum volume thresholds. A token that cannot show enough activity risks getting delisted, so the project or its contracted market maker keeps the tape moving to look alive. That is a different thing from real market making, where the maker is posting genuine liquidity and eating actual inventory risk.
And individual traders wash trade when there is a reward for it. If an exchange hands out fee rebates for high volume or airdrops tokens weighted by how much you trade, people will churn against themselves to farm the reward, as long as the cost of trading stays below what they collect.
The tells
The good news is that fake volume leaves fingerprints. The simplest one is the gap between volume and price impact. Real volume moves price. If a token prints $50 million a day and the chart barely twitches, that volume is not doing any price discovery. Genuine markets show volume and volatility moving together, so big volume tends to come with big moves. Washed markets show the opposite, huge volume sitting on a suspiciously flat price.
Order book depth against reported volume is another one. An exchange claiming $100 million a day on a pair while showing only $200,000 of depth within 2% of the mid is telling you two things that cannot both be true. Real volume needs real liquidity behind it, and the two should scale together.
Trade size distribution is worth looking at too. Organic markets roughly follow a power law, a lot of small trades, fewer medium ones, and the occasional whale. Bots don't do that. They cluster around fixed sizes or spit out an unnaturally uniform spread of trades because whatever is running the wash loop uses predictable sizing.
- Volume with no price movement. High turnover, flat chart, no volatility to match.
- Depth that doesn't back the volume. Big daily number, thin order book near the mid.
- Odd trade sizing. Clustering around fixed amounts or a distribution that is too clean to be human.
- Volume that never sleeps. Same rate at 3am on a dead Sunday as during a news spike.
That last one is the timing tell, and it is one of my favorites. Real volume tracks market hours and news. It spikes when things get wild and dies off when nothing is happening. A wash bot just runs, so the volume comes out weirdly steady 24/7 no matter what the rest of the market is doing.
Where to check
A few platforms publish adjusted volume that tries to strip the fake stuff out. CoinGecko's Trust Score grades exchanges on volume patterns, spreads, and depth together. CoinMarketCap rolled out its own liquidity and volume metrics. Kaiko and Amberdata sell institutional-grade data with quality assessments baked in. The Blockchain Transparency Institute used to publish exchange-by-exchange breakdowns of real versus fake volume, and while people argued about their exact methodology, the broad finding that a lot of exchanges heavily inflate their numbers has held up under independent work.
The cheapest check you can run yourself is comparing the same token across venues. If it trades $30 million on Coinbase and $300 million on some tier-three exchange, that tier-three number is almost certainly padded. Real volume tends to track exchange size and user base, so a small venue posting enormous volume is a flag on its own.
Why I actually care about this
It comes down to not making decisions off numbers that were never real. If you size a position off reported volume, you can badly overestimate how easily you get in and out. A token showing $20 million a day sounds fine until you realize $18 million of it is wash, real liquidity is $2 million, and your $50,000 order is going to move the price on the way in and again on the way out.
It also poisons anything volume-based. VWAP, volume profile, OBV, all of it assumes the volume means something. Feed those inflated numbers and they hand you clean-looking signals built on nothing. When we score venues and route execution at Blockcircle, this is a big part of why we lean toward exchanges with demonstrated real volume even when they report smaller totals. The order book you can actually trade against beats the impressive number every time, and it usually means tighter real spreads and less counterparty risk to go with it.
So before you trust a volume figure, spend two minutes on it. Compare the token across a couple of exchanges, glance at the depth near the mid, and ask whether the volume and the price action agree. That is usually enough to tell you whether you are looking at a market or a treadmill.