Perp Open Interest Rank

OI only matters next to price

Open interest is a widely under-used "crowding" gauge. The absolute level by itself isn't that informative — BTC perp is permanently in the largest tier on OKX, and for the current reading you should go by the figure in the table above. What's informative is the direction of OI change relative to the direction of price change.

The four basic combinations: (1) price up + OI up → new longs entering, the trend is confirmed; (2) price up + OI down → shorts being squeezed out, a weak rally; (3) price down + OI up → new shorts entering, the trend is confirmed; (4) price down + OI down → longs unwinding, a weak pullback. Cases 2 and 4 are "false moves" that rarely have legs.

Another common read: an altcoin where OI climbs steeply over a short stretch while price barely shifts — that's a sudden crowding pulse, usually leveraged money positioning. A sharp OI build-up like that often comes ahead of a larger move, though OI itself doesn't tell you which way, and this is a desk observation rather than a statistical finding. Pair it with long/short ratio: see the LSR dashboard.

Don't trade off the snapshot

OI updates here on a 60-second cadence, but a real leg up or down can play out well inside that 60-second window. By the time you see an OI spike worth chasing, you may be looking at the tail of the move.

Also: this view is OKX-only. The "real" total OI for BTC perp is the sum across OKX, Binance, Bybit, and a dozen others (aggregators like Coinglass do the legwork). If your thesis depends on total market positioning, the OKX-only number isn't enough.

OI is a total, not a direction

Open interest counts how many contracts are still unclosed, and every long contract had a short contract on the other side of it the moment it was filled. Inside a single contract, therefore, longs and shorts are identical by construction, and splitting OI into "how many are long versus short" doesn't work as a method — the long/short OI ratios circulating online are mostly a mix-up of two different definitions. A rising OI says one thing only: more leveraged notional is tied up in this contract, meaning more capital and more triggerable liquidation orders are sitting in the market. Which side that capital is on takes a purpose-built metric — the account ratio and the position ratio.

Those four price-and-OI combinations are also where the conventional read should stop. One step further and you're over the line. In each of the items below, the first half is what the combination can support, and the second half is the part people tack on out of habit that it can't:

• Price up + OI up — Can say: new money is opening positions, so the move isn't only short covering. Can't say: therefore it keeps going up. The more fresh positions there are, the more there is to liquidate when it turns.
• Price down + OI up — Can say: someone is opening into the decline. Can't say: what's being opened must be shorts. It can just as easily be early bottom-fishing longs getting trapped.
• Price up + OI down — Can say: positions are being closed, so part of the rally is short covering. Can't say: the rally is meaningless. A move driven by closing can still travel a long way, it just has no new money relaying it.
• Price down + OI down — Can say: longs are leaving and leverage is shrinking. Can't say: the risk has been flushed out. There's no fixed mark for how much shrinkage counts as clean.

Back to the ranking itself: a high place isn't an event. Large-cap, high-volume coins hold the top rows year round because that's structural, not because something just happened. What's worth pulling out separately is relative size — OI against circulating market cap, or a contract's own 24h and 7d rate of OI change. A mid-cap coin whose OI jumps a level inside a day carries far more information than BTC sitting in first place again. Whether that crowding turns into a liquidation cascade depends on where those positions have their liquidation prices bunched; the arithmetic is in leverage and liquidation price maths.