1 / 6 · 8 min
Funding in depth: rank, the spread between exchanges, and what it does not say
Seeing the rate is not enough — it has to be compared with its own history and with the neighbouring exchange. Both comparisons change the conclusion more than the number itself.
An ordinary futures contract has a settlement date, and by that date its price converges with the asset's on its own. A perpetual futures contract has no date and nothing to converge to. To keep it from drifting away from spot, a payment was invented: when the future trades above spot, longs pay shorts; when below, the other way round. That payment is funding.
One coin, different cost of holding
Every exchange computes funding from ITS OWN order book, and the rates diverge. The same coin in the same hour can cost 0.05% on Binance and 0.01% on Bybit — five times cheaper to hold a long, with nothing changed in the trade itself. This is neither rare nor a glitch: the venues have different participants. We collect the rate from all nine venues, and the spread between them is shown as a separate number.
What the spread is worth in practice
The rate is 0.05% on one exchange and 0.01% on another, a spread of 0.04% per eight hours. On a $10000 position that is $4 per period and $12 a day — purely for standing in the wrong place. Over a week it adds up to $84 at the same risk and the same price move. The second use is harder and not for everyone: a long on the cheap venue and a short on the expensive one earn the spread with no directional exposure. But that means two commissions, two margins and the risk that one exchange changes its rules — and on a small account the spread is eaten by costs entirely.
The rate is quoted per payment period — usually eight hours, more often on some venues. At a rate of 0.01% and a position of $10000 you pay $1 per eight hours, $3 a day. It looks trivial. But at 0.05% that is $5 per eight hours, $15 a day and about $450 a month on the same position — nearly five percent of its size, purely for being open.
What sitting through it costs
You hold a $10000 long for a week at an average funding rate of 0.05%. You will pay roughly $105. To break even the price must rise 1.05% — and that is BEFORE commissions. Now compare that with your instrument's usual weekly range: if it moves 3–4%, a third of your expected move has already been given away. This is exactly why just waiting on expensive funding is not a free strategy.
What funding shows and what it does not
It shows: the skew of positions. A high positive rate means longs outnumber shorts and are willing to pay for the privilege of standing there. That is a fact about how the sides are arranged. It does not show: where the price will go. A skew can persist for weeks and resolve either way — the crowd is sometimes right. The claim that high funding means a fall has no mechanism behind it; all it has is the observation that expensive positions are easier to close at the first shove.
Compare with its own history, not with zero
0.05% is a rare extreme on one coin and an ordinary Tuesday on another. The absolute number says nothing until you know what this funding USUALLY looks like on this instrument. That is why the screener shows a monthly rank next to the rate: 97 means it has only been higher three percent of the time. That is a statement you can work with.
Treating funding as an entry signal
The most common beginner's mistake: seeing an extreme rate and opening against the crowd. The problem is that funding gives no invalidation price. It does not say where you are wrong, and without that number a position turns into hope. If you decide to trade the skew, take the invalidation price from the chart — from structure, a level or volatility — and leave funding as a filter: it answers whether this is worth looking at, not where to enter.
Forgetting funding while holding
Entry and stop are computed, a target is set, and the cost of holding is not counted at all. At an extreme rate it eats more than the entry and exit commissions combined, and quietly: the money is debited three times a day and never appears as a separate line in the trade report. The rule is simple: if you hold longer than a day, compute funding BEFORE entering and subtract it from the expected profit.
Look at the extremes right now: the screener has a ready preset called Funding, which leaves the coins whose rate is far from their own normal. Notice two things: how many such coins there are at all (usually not many) and whether an extreme rate coincides with a strong price move. Often it does not — and that is the most useful observation of all.
Funding: how to read itCompute the cost of your last week
Take your most recent position that you held for more than a day. Find the funding rate for every period of that holding and add them up. Divide by the position size and you get the percentage you paid purely for time. Now compare it with the trade's result. For most people that number turns out comparable to the profit or larger; better to learn this on paper than on your tenth trade.
The funding rate is positive and very high. What does that claim about the market — and what does it NOT claim?
It claims: there are more longs and they are paying shorts for the right to stand there; holding a long is expensive right now. That is a fact about the arrangement of the sides and about your costs. It does NOT claim that price will fall. The skew can persist for weeks, and it speaks not about direction but about the fact that at the first sharp move down a lot of people will be closing — that is, about potential SHARPNESS, not about its side.
Why is the rate compared with its own history rather than with zero?
Because every instrument has its own normal funding level: on one, 0.01% is a ceiling; on another, it is background noise. Comparing with zero answers only who pays, not whether it is a lot or a little. A monthly rank turns the rate into a testable statement: it has only been higher three percent of the time — that is a comparison with the instrument's own history rather than with an abstract middle.