Published October 27, 2026 | Version v1.0.1

No Free Alpha: Transaction Costs, Luck, and the Limits of Algorithmic, LLM-Driven and Copy Trading for Retail Crypto Investors

Authors/Creators

Description

An end-to-end, cost-aware evaluation of the trading strategies offered to retail crypto investors in 2026: technical-indicator bots, Freqtrade community strategies, an LLM (Claude) trading agent, copy trading of Hyperliquid wallets selected by four finders (including a skill finder using block-bootstrap tests, Benjamini-Hochberg false-discovery-rate control and empirical-Bayes shrinkage) replayed walk-forward on the complete public fill ledger (457 days, 1.5 million wallets), and a daily trend rotation with a survivorship-free universe. After fees and slippage, no strategy beat holding Bitcoin over its full test window; past copy-trading profits did not predict future ones (rank correlation 0.04 over 33,927 wallet-weeks).

Notes

If you use this research or code, please cite it as below.

Files

AmmarHSufyan/no-free-alpha-v1.0.1.zip

Files (2.8 MB)

Name Size
md5:3d25be5ae176b0982072d4c68ff65b77
2.8 MB Preview Download

Additional details

Related works