FundingKilled

Cross-venue funding arbitrage on BTC / ETH / SOL / HYPE

0 of 4,216

hours that were profitable after costs

In plain English

The “free” price gap between exchanges was never once wide enough to cover trading costs, not in a single hour out of 4,216.

What was claimed

Two exchanges pay different funding rates on the same coin at the same time, so hold opposite positions on each and collect the difference risk-free.

The bar it had to clear

spread > ~11.5 bps round-trip (HL + Pacifica)

What the data said

max ever ~1 bp/hour; 0 of 4,216 hours

Why it's dead

The "free" funding spread between exchanges on majors (the big coins: BTC, ETH, SOL, HYPE) never gets wide enough to clear round-trip costs (the fees paid to get in and back out), and collapses before you can hold it. A permanent structural kill: the market's structure itself, not bad tuning or bad timing, makes the trade impossible.

The detail

Across a 63-day HL–Pacifica screen, |spread| never exceeded ~1 bp/hour, max single-hour observation HYPE 0.98 bps/h. Breaking even on an 11.5 bps round-trip needs the spread to run >11.7× its historical max; spreads also mean-revert (shrink back toward zero) with a 2–3h half-life, collapsing before any hold pays. 0 of 4,216 hourly windows were net-positive at 1/2/4/8h holds. No >156 bps cross-venue dislocation appears on BTC/ETH/SOL in 2023–2026 public data.

Kill date: 2026-05-09
Sample: n≈1,500 hrs/symbol
Method: Documented kill
Verdict: permanent structural kill