What Is a Normal Funding Rate?.

A normal funding rate is slightly positive. The baseline on most venues is 0.01 percent per eight hours, which annualizes to roughly 11 percent paid by longs to shorts, and a healthy major like BTC or ETH spends most of its life oscillating in a band around that level. The baseline is not an accident of market mood: it is built into the formula. Venues including Hyperliquid and Variational set a fixed interest component of 0.00125 percent per hour inside the funding calculation, representing the cost of borrowing dollars versus holding crypto, so when the perp tracks its index perfectly, funding does not read zero, it reads that small positive number.

That gives you the first calibration: zero is not neutral. A funding rate of exactly zero on a major is already mildly informative, telling you short pressure is strong enough to cancel the structural positive bias. The resting state of a perp market is a slow trickle from longs to shorts, and deviations should be measured from that trickle, not from zero.

The second calibration is by asset class, because “normal” is not one number. Majors are the tight case: deep books and heavy arbitrage keep their funding pinned near baseline, and a major drifting to a few multiples of baseline is already a statement about positioning. Mid-caps run looser, with wider and longer-lived swings. The long tail, new listings and memecoins and thin markets, barely has a normal at all: funding there can sit at levels that would be historic on BTC and revert within hours, because a thin book lets modest flows push the perp far from its index. A reading that means nothing on a memecoin would be a five-alarm event on Bitcoin, and any tool comparing them on one scale is misleading you. New listings are their own regime, with unstable funding for days or weeks while the market discovers positioning, which is part of why fresh tickers dominate screener “opportunities” and why those opportunities are usually mirages, as covered in the 1,500% APR that doesn’t exist.

The third calibration is regime. The baseline holds across eras but the distribution around it moves with the cycle: in euphoric stretches the whole market runs hot-positive for weeks, with majors sustaining several times baseline, while in deleveraging stretches funding across the board goes flat to negative, a market-wide reading that is one of the more reliable temperature gauges in crypto. The same number can be unremarkable in one regime and extreme in another, which is why a snapshot without history is nearly useless.

Which leads to the practical method: judge any funding rate by its distance from that asset’s own recent norm, in that regime, at that venue’s interval, normalized per the interval differences. Not by its raw size, and not against another asset’s scale. Extremes against an asset’s own baseline are where the information is, and the deeply negative tail of that distribution, where shorts are paying heavily and the squeeze math starts, is specifically what my screener hunts. What those extremes actually set up is the subject of the two trades hidden in every funding dislocation.