Research Note 02

Momentum: Evidence from a Binance Replication

Tests whether a cross-sectional momentum spread survives on Binance USDⓈ-M perpetuals net of realistic costs, not just in an academic backtest. Liu, Tsyvinski & Wu document momentum as statistically significant across a broad crypto sample; Starkiller Capital's practitioner backtest beat the market in relative terms through the 2021–2022 bear market without an absolute profit, and named a 125-basis-point cost threshold that kills the effect. The project's own replication (832 Binance perpetual symbols, daily panel, 2020–2026) finds a gross weekly spread of +0.573%, but a Binance taker fee plus slippage consumes about 70% of it, leaving the net spread indistinguishable from zero on all three significance tests (bootstrap CI, sign-flip, permutation). A literature-mandated second specification (J=2/K=2) also fails, its sign flipping with an arbitrary rebalancing-phase choice. Momentum is real in the literature; it does not clear this project's cost bar on Binance.

This page preserves the approved research abstract in full. The working repository contains the complete note, methodology, code, data, and references.