A rolling 30-day event study using official hourly funding and completed HYPE candle closes. Inspect the sample before turning a popular market narrative into a trading rule.
Not by itself. The all-hours 24-hour median HYPE return was +0.86% in this rolling 30-day window. Across 23 elevated funding episode starts, the 24-hour median was +2.22%, 1.36 percentage points above the baseline. Across 35 negative episode starts, it was -0.10%, 0.96 percentage points below the baseline. This descriptive sample does not support treating either funding extreme as an automatic reversal or continuation signal.
Source window
Episode-level comparison
Primary episode rows use the first hour after a threshold crossing. Each result is compared with all eligible hours in the same window.
Every aligned hourly funding observation in the window.
Same-window hourly baseline
First hour after funding moves above the 0.125 bps hourly base interest rate.
Descriptive sample, no significance test
First hour after funding moves below 0 bps, when shorts pay longs.
Descriptive sample, no significance test
Text and image include the baseline, both episode samples, source time, and limits. No affiliate link or trading signal.
The rolling values are frozen at the displayed source time.
Inspect the observations
The latest six starts from each regime are shown directly. Newer rows can have pending forward windows.
Interpretation
It describes what HYPE did after comparable funding conditions in one recent rolling window and how those returns differed from the same-window hourly baseline. It can challenge a simplistic story, expose a small sample, and give funding context before a trade. It cannot establish that funding caused the move or that the next episode will behave the same way.
The CSV includes every funding timestamp, regime label, episode flag, reference close, and available 1h, 4h, and 24h return. Empty cells mean the forward candle had not completed or could not be aligned.
Download CSVReproducible method
Request two bounded 15-day funding windows and one 30-day set of 1-hour candles from Hyperliquid.
Match each funding timestamp to the hourly candle close completed at that same UTC boundary.
Label funding above 0.125 bps as elevated, below 0 bps as negative, and collapse consecutive hours into episodes.
Calculate 1h, 4h, and 24h returns, then compare episode medians and positive shares with all eligible hours.
If you have already decided to trade and are eligible, review the referral discount and its limits before opening Hyperliquid. This study is independent of that decision.