The short version: reported volume is self-published by venues that benefit from it being large, it has been demonstrably fabricated at enormous scale, and almost nothing you'd want to conclude from it survives contact with that fact. Depth and spread are the numbers that can't be faked as easily.

Feeding time at the colony is loud. Seven cats, a lot of movement, tails everywhere, the distinct impression of a great deal happening.

Then you look in the bowl afterwards and work out that two of them ate, one of them circled the bowl four times without touching anything, and the rest were mostly there because the others were. The activity was real. The activity was also almost entirely uninformative about how much food went anywhere.

Crypto volume figures are that bowl.

1. What volume is supposed to mean

In an ordinary market, volume answers: how much actually changed hands? It's useful because it tells you something about conviction and something about liquidity — whether you could take a position of a given size without moving the price against yourself.

That interpretation relies on an assumption so basic nobody states it: that the trades happened between parties who actually wanted the asset.

Remove that and volume stops measuring anything. You're left with a count of events, and events are cheap to manufacture.

2. Why the number is inflated

The incentives are blunt and they all point one way.

Rankings drive customers. Listing sites sort exchanges by volume. Traders go where the volume is, because they believe that means liquidity. Bigger reported number, more real users, more real fees.

Listings are sold. Projects pay to be listed, and they pay more at venues that appear busy. An exchange's reported volume is, in part, its rate card.

It costs almost nothing to fake. An exchange trading against itself pays itself the fees. Two bots passing the same coins back and forth generate unlimited volume at a cost of roughly zero. Nothing about the reported figure is audited.

Nobody is checking. There is no standard definition, no required methodology, no regulator receiving the filings for most venues. Each exchange publishes a number it computed itself, by whatever means it likes.

In March 2019, the asset manager Bitwise presented an analysis to the SEC arguing that roughly 95% of reported bitcoin spot volume was fake or non-economic, and that the genuine market was a small fraction of the headline figure — concentrated on about ten venues. The finding was widely covered and badly needed.

It also deserves its asterisk, which I'll give in the honest section below.

3. How to tell real activity from noise

You can't audit an exchange. You can look at things that are harder to fabricate.

Order book depth. How much can be bought or sold within 1% or 2% of the mid price before it moves? Fake volume doesn't produce real depth, because real depth means real capital sitting there exposed to being hit. This is the single best substitute.

Spread. A genuinely liquid market has a tight bid-ask. Wide spreads with enormous reported volume is an obvious contradiction.

Slippage on a real order. The most honest test available to you personally: place a modest order and see what you actually get filled at. It's a test an exchange can't study for.

Regulated and reporting venues. Volumes from venues with audited reporting and a regulator who'd prosecute fabrication are a different class of number from self-published figures.

Consistency over time. Real volume is spiky — it explodes on news and dries up in quiet weeks. Suspiciously smooth volume across a flat month is a machine, not a market.

4. What this changes for you

Don't use volume to confirm a move. "Breakout on high volume" is a staple of chart commentary, and on aggregated crypto data a large part of that volume may be synthetic. You are confirming a signal with a number somebody manufactured.

Do use depth to size yourself. If you're moving an amount that matters to you, depth tells you whether you'll get the price on the screen. Volume won't.

Treat "most traded" lists as advertising. They rank by the metric most easily gamed, which means they're closer to a leaderboard of who wanted the ranking most.

Watch where real flows are now visible. Spot ETFs publish creations and redemptions under reporting obligations with real consequences for lying. That's a genuinely different quality of data, and a lot of institutional activity now passes through it.

5. The uncomfortable half

The 95% study came from an ETF applicant. Bitwise was, at that time, seeking the SEC's approval for a bitcoin ETF, and the argument that the real market was small, orderly and concentrated on regulated venues happened to be exactly the argument that supported their application. That doesn't make the methodology wrong — it was well-received and much of it has held up — but anyone handing you a statistic that conveniently supports their business is handing you a statistic to check, including when the statistic is one you like.

It's a 2019 number and this is 2026. The market has changed enormously: ETFs, regulated venues, institutional desks, real market makers. Quoting "95% of volume is fake" as a present-tense fact is sloppy — the figure describes a market that no longer exists in that shape. The incentive to inflate persists wherever rankings and listing fees do; the magnitude almost certainly does not.

And the people selling you "real volume" have incentives too. Firms publishing adjusted or verified volume metrics are choosing which venues count, by criteria they set. Those criteria are usually reasonable and always arguable, and the resulting number is an estimate shaped by a methodology, not a measurement. Replacing a number you can't trust with a number you've decided to trust is progress only if you know why.

Low volume is also not automatically a problem. Quiet markets are normal, and most of Bitcoin's history has been quiet punctuated by brief noise. Thin trading matters if you need to move size; if you're buying a fixed amount monthly and holding, it's an irrelevance you can safely ignore.

The useful habit isn't distrusting every figure. It's asking, before you lean on one: who published this, what does it cost them if it's wrong, and could anyone check?

Seven cats around a bowl is not seven cats eating. Before you read anything into how busy a market looks, find out who counted — and what they got for the number being big. 🐾⚡