Funding Rate, Open Interest and Long/Short Ratios Explained

Perpetual futures and the funding rate

A perpetual future is a futures contract with no expiry. A dated future converges on the spot price at expiry, but a perpetual has no such day, so it needs a separate mechanism to keep its price close to spot. That mechanism is the funding rate. When the future trades above spot (the index price), the funding rate moves toward positive; when it trades below, toward negative. When it is positive longs pay shorts, and when it is negative shorts pay longs. It is not a fee the exchange collects but money that passes between traders, and it changes hands only if you hold a position at the settlement time. The mechanism holds price in place through cost. When the future is expensive, holding a long costs money while a short gets paid, so trades that buy spot and sell the future to capture the gap increase, pulling the two prices back together. That is why the funding rate is also read as a number showing which side of the futures market was more eager.

How Binance sets the funding rate

On Binance USDⓈ-M perpetual futures, large coins such as Bitcoin and Ethereum settle every 8 hours (at 00:00, 08:00 and 16:00 UTC), many coins settle every 4 hours, and some settle every hour. On the 8-hour cycle, the funding rate is the premium (how far above the index the future traded) plus "interest − premium" clamped to within ±0.05%, and the interest component defaults to 0.01% per 8 hours. So whenever the premium is between −0.04% and +0.06%, the funding rate comes out at exactly 0.01%, which is why that value shows up so often on charts. The amount paid or received is the position value (mark price × quantity) × the funding rate. A stretch of 0.01% means less that there is no crowding than that the premium is sitting inside that clamped range. Also remember that with leverage, the same funding rate weighs on your margin several times more heavily.

• 10,000 USDT long, funding +0.01%: pays 1 USDT per settlement
• Same position, funding −0.05%: receives 5 USDT
• A 4-hour cycle means 6 settlements a day (3 on an 8-hour cycle)

How to read a funding rate chart

Because the funding rate reflects how far above spot the future traded, a strongly positive rate is read as longs buying more aggressively, bidding the price up as they go. A pattern that persists over several settlements says more than a single spike. Coins on different cycles have to be converted to an 8-hour basis before you compare them. This site's Funding Rate Dashboard labels an 8-hour-equivalent rate of +0.05% or more "Longs crowded" and −0.05% or less "Shorts dominant", but these are only names for ranges, not forecasts of price direction. Also keep in mind that the values in the table are estimates that keep changing until just before settlement. The order of checks is the size of the value, how long it has lasted, and what price did over the same time. If price is flat while funding rises, it is read as buying piling up only on the futures side; if funding stays low while price rises, many people take spot buying to be leading the advance. Newly listed coins often see funding spike sharply, which makes them hard to compare on the same scale.

Open interest: how many contracts are open

Open interest (OI) is the total number of contracts that have not yet been closed. In futures, every long contract has a short contract on the other side, so OI rises when both sides of a trade are opening new positions, falls when both sides are closing existing ones (liquidations included), and stays the same when one side is entering and the other is leaving. Viewed alongside price, it therefore lets you gauge whether a move came from new positions or from existing positions exiting. When price and OI both drop sharply, as in the plunge in the figure, it is commonly read as a wave of long closures and liquidations. OI measured in USDT rises just because price rises, so measuring it in number of coins is more accurate. Volume and OI are different things. Volume is how many contracts changed hands during the period, and OI is what remains open as a result, so heavy volume with flat OI means the same positions simply changed hands. Still, as with the figure, the fact that the two lines moved together cannot confirm who did the closing.

Four combinations of price and OI

Pairing the direction of price with the direction of OI gives four combinations. The usual interpretations are listed below, but all of them are guesses. Every contract has both a long and a short, so rising OI means new longs and new shorts were created in equal amounts. Saying "new longs came in" is an inference, drawn from the direction of price, about which side entered more urgently at market. Keep in mind too that this is data from one exchange for one coin. A widely used yardstick is the size of the change. OI moving 1 to 2% may just be everyday ripples, and many people consider the combinations meaningful only when OI shifts substantially within a few hours. A rise led by short covering is not new buying but existing shorts buying to close, so it is sometimes read as likely to lose steam once the covering is done. Conversely, price ↓ with OI ↑ only means new shorts piled in, not that those shorts are right.

• Price ↑ OI ↑: read as a rise led by new longs coming in
• Price ↑ OI ↓: a rise led by shorts closing out (short covering)
• Price ↓ OI ↑: a drop led by new shorts coming in
• Price ↓ OI ↓: a drop led by longs closing out and being liquidated

There are three long/short ratios

Across the whole futures market, long and short contracts are always equal in number, so a long/short ratio measures something other than contract count. Binance publishes three ratios, and each looks at something different. The all-accounts ratio is closer to a headcount of which way people lean, while the top-trader position ratio is closer to where the money sits. The two sometimes move in opposite directions, and there is no rule for which one is right. A ratio of 1.5 means 60% long and 40% short (long share = ratio ÷ (1 + ratio)). In an account-count ratio a small account and a large one each get one vote, so it tends to tilt toward whichever side has more small accounts. A common mistake is to treat a high long ratio as strong buying power; since longs and shorts are always equal in contracts, it may simply mean that fewer accounts on the short side hold larger positions. Keeping all three ratios on screen and watching where they disagree leads to fewer misreadings than watching just one.

• All accounts: accounts holding longs ÷ accounts holding shorts
• Top trader accounts: long accounts ÷ short accounts within the top 20% by margin
• Top trader positions: long position size ÷ short position size within the top 20%

The taker buy/sell ratio

Every trade has a side that posted an order and waited (the maker) and a side that took that order right away (the taker). The taker buy/sell ratio is the volume bought at market divided by the volume sold at market, and a value above 1 means buyers were the more urgent side during that period. The filled volume itself is always equal on the buy and sell sides, so this ratio shows not who bought more, but which side filled immediately instead of waiting in the order book. On short bars it can swing widely on just a few large orders. People in a hurry use market orders because getting filled matters more than the price, and liquidations are also executed at market, so in a plunge forced selling shows up as taker sells. A moment when the ratio leans hard one way can therefore be determined trading or forced closing. Looking at the pattern over several bars together with how price reacts is less noisy than looking at a single bar.

Crowding measures the size of the risk, not the direction

If the funding rate is strongly positive, OI is rising and the long ratio is high as well, you can read it as leverage building up on the long side. If price then moves sharply the other way, the chain reaction of liquidations triggering further liquidations can grow larger. Liquidated longs are sold at market, pushing price down further, and that drop reaches the next set of liquidation prices. But these numbers do not tell you when the crowding will unwind, or how much further price will go before it does. Crowding can stay high for weeks while price keeps rising, and those betting the other way can be the ones flushed out first. It is more accurate to treat crowding as information about how large a move could become if a sudden shift comes, rather than about the next direction. You can watch actual liquidations cluster on the Liquidation Radar. This article explains indicators and does not recommend any position.

Limits of the data

Binance publishes open interest, long/short ratio and taker ratio statistics for only the most recent 30 days. That makes it hard to compare with months or years ago, or to test anything over a long period. They are also figures from a single exchange, so moves on other exchanges and in the spot market are not captured. Exchanges differ in how they calculate funding, how often they settle and who trades on them, so the same coin can show different numbers, and it is hard to say that any one exchange represents the whole market. Even within 30 days, a single big event can shift the average for the period a lot, so when you judge something "higher than usual", it helps to note which period you are comparing against. The mechanics of leverage and liquidation are covered separately in the coin topic's article "What happens when you trade with borrowed money", and how to take in the various market indicators in "What market indicators actually tell you".

What looks different in crypto and stocks

The funding rate exists only in crypto perpetual futures. On the stock side, Korean index and single-stock futures and US index futures all have expiry dates, so the gap between futures and spot (the basis) shrinks on its own as expiry approaches. Stock futures do have open interest, but it repeatedly drops and rebuilds as positions roll to the next contract around expiry, so that change should not be read as a change in crowding. In Korea, margin loan balances are a common gauge of retail leverage, and many people watch them alongside short interest even for heavily traded names like Samsung Electronics and SK Hynix. These are published daily or less often, though, so you cannot follow them minute by minute as in crypto. US short interest is compiled and published about twice a month. Price limits differ too. Crypto futures have none, so a liquidation cascade can run its full course within minutes, whereas Korean stocks stop within ±30% a day, and forced selling from under-margined credit accounts usually comes around the next trading day's open. So the same crowding unwinds at a different speed and in a different shape.

Checklist in practice

When you keep these numbers beside a live chart, first know that they update at different speeds. Price bars change with every trade, but OI and long/short ratios are statistics compiled every few minutes, and the funding rate is an estimate that keeps changing until settlement. Pairing the price change of a bar still in progress with the last OI reading compares two different moments. Around settlement times, position adjustments to avoid or collect funding can also overlap and shake short bars.

• Check the coin's settlement cycle and convert funding to an 8-hour basis
• Look at the trend over several settlements rather than a single value
• Read OI in number of coins, paired with the price change of closed bars
• Check whether a long/short ratio measures account count or position size
• Read crowding as the size of a move in a sudden shift, not as direction
• Note the 30-day, single-exchange limits and hold off on conclusions