Open Interest, Volume and Liquidations: Three Different Clocks

A derivatives dashboard may place open interest, trading volume and liquidation totals next to each other. Their proximity suggests a single story about leverage. Their definitions describe different things.

Open interest measures positions that remain outstanding at a point in time. Volume measures trading during a period. Liquidation data records a particular kind of forced closure, subject to the reporting system’s coverage.

A useful interpretation starts by separating those clocks.

Outstanding does not mean newly traded

CME Group’s explanation distinguishes contracts that remain open from all contracts traded during a period.

Consider a simplified market with ten outstanding contracts, counted once per contract. If a new buyer and new seller create one additional contract, open interest increases to eleven. In a separate scenario starting again from ten, two existing counterparties close a contract and open interest falls to nine. If an existing exposure changes hands while the opposite side remains open, trading occurs without increasing the outstanding count.

Volume can therefore rise while open interest remains flat. The same exposure can change hands repeatedly.

Nor does every new contract indicate a one-sided bullish conviction. A contract has opposing sides. Participants may be expressing a view, hedging another asset or maintaining a multi-leg position. The aggregate count does not reveal those motives.

Counting conventions can double the apparent market

Before comparing two open-interest figures, check whether each counts a contract once or adds both its long and short sides.

The CoinMarketCap methodology explicitly distinguishes single-sided and double-sided reporting. Bybit’s open-interest API documentation, consulted on 22 September 2026, separately describes a both-sides field and a single-side field.

Units also vary by contract. That Bybit documentation identifies BTC as the unit for a BTCUSDT linear contract and USD for a BTCUSD inverse contract. A field name by itself is not sufficient evidence that two values can be added.

An aggregate must reconcile these differences before summing them. Otherwise, a change in feed coverage or counting convention can masquerade as a surge in positions.

Dollar open interest mixes quantity and price

Suppose a hypothetical market has 100 units of open exposure, each valued at 10 USD. Its notional value is 1,000 USD. If the price doubles while the quantity remains unchanged, the notional becomes 2,000 USD.

The dollar measure has doubled without a doubling in position quantity.

The reverse is also possible: a falling price can reduce reported notional even while more contracts are opened. A responsible chart therefore pairs notional with the underlying quantity, or explicitly states that the price component has not been separated.

Notional is also different from collateral. A billion dollars of position value is not evidence that a billion dollars of fresh cash entered the venue.

A data check that stopped a headline

On 22 September 2026 at 19:31:05.536 UTC, CryptX retrieved a CoinMarketCap derivatives summary. The response included outstanding-position figures, rolling-volume fields and Bitcoin liquidation totals.

However, it did not include a dataset observation timestamp or an explicit exchange universe. An unusually large volume field also required reconciliation. We retained the raw response for review and withheld its headline market totals from this article.

This is a finding about the evidence available to us, not a claim that every underlying value is wrong. A missing timestamp cannot be repaired by substituting the moment our request completed. Nor can an extreme-looking value be silently corrected by changing its unit.

Liquidation feeds have their own boundaries

Liquidation totals are not a substitute for the change in open interest. Positions can close voluntarily; new positions can open during the same interval; and the valuation basis can differ.

Feed design also matters. Bybit’s all-liquidation stream documents a 500-millisecond push frequency and identifies its price field as a bankruptcy price. That is a specific definition, not an invitation to call every price field the market execution price.

Its API changelog records replacement of an older liquidation stream that pushed only one liquidation per second. Combining historical feeds without accounting for such changes can create an artificial jump in observed activity.

A zero from a partial or delayed feed means that the returned dataset reports zero. Establishing zero across the whole market requires evidence of complete, timely coverage.

Compare like periods before suggesting a cause

An end-of-hour open-interest observation and a rolling 24-hour liquidation total do not describe the same interval. A comparison should identify both endpoints for the stock measure and the exact window for the flow measure.

It should also distinguish last price, mark price and index price where relevant. Bybit’s ticker specification exposes these as separate fields because they serve different purposes.

Even after harmonization, simultaneous price changes and liquidations do not alone establish the full causal story. The sequence of events and the venue coverage remain important.

A careful derivatives dashboard therefore needs three clear labels: what remains open, what traded, and what was forcibly closed. When the clocks or definitions cannot be aligned, the strongest analysis explains the gap instead of inventing a market narrative.

Disclaimer: For informational and research purposes only. This is not financial advice. Market data may be delayed or inaccurate; verify material information with primary sources.