Hyperliquid Spot–Perpetual Aggressor Disagreement: HBAR, XLM and ETC

Hypothesis

Test whether sustained Binance spot buying against perpetual selling, or spot selling against perpetual buying, predicts multi-day adjustment toward the spot flow. Trade Hyperliquid perpetuals on completed 4-HOUR decision bars; Binance spot and USD-M instruments supply actual aggressor-signed trade-tick signals only. Baseline accumulation is 48 hours and holding time is 72 hours. Expected frequency is 12–30 entries per asset/year, an untested prediction. Market orders execute at signal-bar close, the registry equivalent of next-bar-open execution on continuous venues. Protective stops trigger intrabar through platform matching. Use all common available history; do not assume Hyperliquid's approximately 2.3-year 4H API coverage can be extended retrospectively. Input definitions follow [Binance public-data documentation](https://github.com/binance/binance-public-data); fees follow [Hyperliquid official documentation](https://hyperliquid.gitbook.io/hyperliquid-docs/trading/fees). These sources document inputs and costs, not evidence for this edge.

This addresses the largest direction imbalance: long-only represents 77.8% of experiments against a maximum target of 55%. It also adds Hyperliquid execution and cross-venue signal scope, both underrepresented. corpus_concentration was empty; snapshot checks found one primary run each for HBARUSD.HYPERLIQUID and ETCUSD.HYPERLIQUID and three for XLMUSD.HYPERLIQUID, without establishing secondary-use counts. Prior-art searches found single-venue taker/price divergence, spot-led breakouts and thin-volume reversals. This design instead requires opposite actual spot/perpetual aggressor imbalances over identical windows and trades against the perpetual price displacement. It uses neither breakout confirmation nor low-volume ranking. Binance provides paired historical flow observations; no Binance position is traded.

hold_expired: data_review:blocked_data held 524 h, past its 144 h limit, unresolved (owner: data). Not a verdict on the strategy; revivable once resolved. Recommended: reverify after a deployed sufficiency fix (dated futures and delisted symbols are measured against their own life since 2026-10-16); a leg listed too recently passes on a known date; otherwise the developer drops the leg

Implementation

Trades HBAR, XLM and ETC Hyperliquid perpetuals toward Binance spot aggressor flow when paired USD-M aggressor flow opposes it and the Hyperliquid price displaced against spot flow. Uses completed 4-hour bars, 48-hour baseline flow windows and 72-hour holds. Market execution at signal-bar close substitutes for next-bar-open execution on continuous venues; 4% reduce-only stops trigger intrabar.

Uses actual aggressor-signed quote notional from Binance trade ticks, with identical (start,end] calendar windows and no future buckets. Both feeds must have observed ticks in every window bucket; absent paired data never invokes a price proxy. Binance instruments supply signals only. Each Hyperliquid leg is capped at 15% of primary-venue equity, total gross at 45%, and nominal stop risk at 0.6% per leg. Position age comes from the engine's persisted fill timestamp, while stop levels rest as orders. Price displacement is measured on the traded Hyperliquid perpetual over the same window. Uses common available history without manufacturing earlier Hyperliquid data; the expected trade frequency and edge remain untested.

Outcome Summary

A multi-leg hypothesis that needs Binance trade-tick data plus Hyperliquid 4H bars should have its data sufficiency for every leg confirmed, including how recently each leg was listed, before coding, so the run does not stall in data review.

The run sat in a data_review:blocked_data hold for 524 hours, past the 144-hour limit, and the hold expired unresolved (owner: data). The record says this is not a verdict on the strategy and the run can be revived once the data issue is resolved.

It tested whether sustained disagreement between Binance spot and USD-M perpetual aggressor-signed trade flow (spot buying vs perp selling, or the reverse) over a 48-hour window on 4-hour bars predicts a multi-day price adjustment toward the spot flow, trading HBAR, XLM and ETC perpetuals on Hyperliquid with a 72-hour hold.

No performance was measured: the backtest report is empty and no optimization was run, so there are no return, Sharpe or trade-count figures.