Open interest counts how many derivative positions are currently open and have not yet been closed or liquidated. Volume counts how many contracts changed hands over a period of time. The two numbers can move in completely different directions on the same day, and reading them together, rather than either alone, is what tells you whether a price move is being driven by fresh money or by traders being pushed out.
Stock versus flow
CoinGlass’s educational material describes open interest as a measure of “the depth of market participation, while volume measures its intensity.” Crypto.news makes the same distinction in different language: volume is a flow, something that resets each period, like monthly spending; open interest is a stock, something that carries forward, like an account balance. A market can see enormous volume in a session and end with open interest completely unchanged, because every contract that traded was also closed by someone else on the other side. Nothing about total exposure in the market moved, even though the tape looked busy.
How a single trade moves the number, or doesn’t
Crypto.news works through this with a simple illustration, starting from a market where 100 contracts are outstanding. If a new buyer and a new seller enter the market to trade with each other — neither had a prior position — a new contract is created and open interest rises to 101. If an existing long and an existing short trade with each other and both exit, the contract is extinguished and open interest falls to 99. But if an existing long sells their position to a new buyer, no new contract is created and none is destroyed; a position is simply transferred from one holder to another, and open interest stays at 100. Volume records one contract traded in every one of these cases. Only the underlying combination of who is opening and who is closing determines what happens to open interest.
CoinGlass describes the same logic in its own terms: new positions on both sides increase open interest, positions closing on both sides decrease it, and a transfer of an existing position — a new buyer taking over from a seller who wants out — leaves the total unchanged.
Contracts versus notional dollars
Both sources note that exchanges publish open interest in two different units. One is a raw count of contracts or coins. The other is notional value — that count multiplied by the current price of the underlying asset, usually expressed in dollars. CoinGlass calls this a “dual-denomination approach,” distinguishing it further by margin type, such as USDT-margined contracts versus coin-margined ones. Crypto.news flags the practical consequence: a notional open interest figure can rise or fall purely because the price of the underlying asset moved, even if not a single contract opened or closed. A dashboard showing open interest climbing in dollar terms during a rally may simply be reporting the same number of contracts worth more, not new positions being added.
Does open interest tell you which way price will move?
This is where the two primary sources in this evidence set diverge, and it is the point most worth reading carefully. Amberdata’s blog, published 1 September 2023, lays out a set of interpretive combinations. When price rises alongside rising open interest, Amberdata says this can indicate fresh money entering the market, and calls that bullish if the increase is driven by long positions. Falling open interest during a downtrend, in Amberdata’s framework, can suggest holders being forced to liquidate, which it describes as bearish and a possible sign that a selling climax is approaching. These are specific, named combinations — five of them in Amberdata’s write-up — each assigned a directional read.
CoinGlass’s own explanation of open interest takes a more guarded position. It states plainly that open interest “is directionally neutral,” revealing only the total scale of the standoff between long and short positions, not which side has the advantage. CoinGlass argues that open interest has to be read inside a wider framework, alongside price action, volume, funding rates and liquidation data, before it says anything about market sentiment at all.
Both statements can be true at once, but they are not the same claim. Amberdata is offering a set of trading heuristics for interpreting price and open interest together. CoinGlass is making a more basic point about what the open interest number itself can and cannot show in isolation. Neither outlet in this evidence set publishes data testing whether Amberdata’s combinations actually predict what happens to price afterward. The quadrant framework is a widely used way of talking about open interest — it is not a demonstrated forecasting tool.
The common misreading
The mistake is treating a rising open interest figure as itself a bullish or bearish signal, or as proof of which side of the market is about to win. CoinGlass’s own framing argues against this directly: the number measures the size of the standoff, not the likely outcome of it. What it reliably tells you, per CoinGlass, is that a large amount of leveraged exposure is sitting in the market and has not yet been resolved — which is also why CoinGlass links large open interest to the risk of cascading liquidations and auto-deleveraging when a market moves sharply against one side.
What this page does not tell you
This page cannot tell you whether Amberdata’s price-and-open-interest combinations reliably predict subsequent price behavior. Amberdata’s blog post presents them as a trading framework; none of the three sources examined here supply backtested results, win rates, or sample data showing how often each combination played out as described.
It also cannot tell you whether an aggregated “total open interest” figure on any given market dashboard nets out consistently across venues. Open interest is venue-specific, and exchanges differ in contract size, margin currency — USDT-margined versus coin-margined, per CoinGlass — and reporting cutoffs. None of the sources here describe how, or whether, aggregators reconcile those differences when they sum figures across exchanges.
Finally, this page cannot correct for the notional-versus-contract-count distortion in real time. A notional open interest figure moves whenever the underlying price moves, independent of whether any contracts opened or closed. Reading that distortion out of a live dashboard number requires the contract-count figure alongside price at each point in time, which is not something either primary source in this evidence set supplies as a running dataset.
Sources
Every fact above is attributed to one of these reports. Where they disagree, the article says so.
