# Crypto short-horizon reversal: two decisive negative trials

**Author:** Arhan Canli, Founder and Quantitative Researcher, Canli Capital  
**Family key:** `crypto_short_horizon_reversal` · **System:** ALPHAC / AlphaForge  
**Status:** public research record; not peer reviewed; not an investment solicitation

## Finding and boundary

Residual reversal asks whether short-horizon price dislocations mean-revert after removing common
movement. Two charged identities tested a 24-hour signal and a 24/72-hour blend. Their immutable
annualized Sharpe summaries are -2.1527 and -2.1140 over 1,574 observations. Both falsify the
implemented direction net of its historical construction.

Short-term reversal has conventional-market precedent, including Lehmann's evidence on weekly
return reversals ([Journal of Finance](https://doi.org/10.1111/j.1540-6261.1990.tb05110.x)), but
venue microstructure, funding, liquidation, and continuous crypto trading make transfer unsafe.
The blend and single horizon are one mechanism and cannot be counted as independent sleeves.

No full family curves, DSR, capacity, live execution, or Alpaca record are established, so no
missing metric is reconstructed. **Decision: FAIL / zero sleeves.** Exact keys, hashes,
configurations, and moments are public in
[`crypto_reversal_family.json`](/glassbox/crypto_reversal_family.json). The work was authored and
directed by **Arhan Canli**; the negative result remains permanently charged to the union.