2025-08-01 · 1 min read
Why I Built a Pipeline That Found No Alpha
Project CHF v2 reports alpha_verified=false. Why a pipeline built to reject weak alpha claims is worth more than one built to find them.
The negative result
Project CHF (v1.0-research-release) found statistically significant candidate signals: best Rank IC 0.0275 at t-stat 7.10. After portfolio construction, transaction costs, and benchmarking against BTC, ETH, BTC/ETH 50-50, and equal-weight universes, no strategy cleared verified alpha.
alpha_verified = false is the successful outcome here.
Why this matters
A system built to find alpha is easy to fool. A system built to reject unsupported alpha claims is harder and more valuable. CHF's 9-agent pipeline enforces:
- Leakage-safe features and exact forward labels
- Purged + embargoed walk-forward screening
- Deterministic allocation (no research-stage alpha claims)
- Cost-aware backtesting against four benchmarks
Strongest backtest
linear_ridge / market_only / 30d hit 147.36% total return and 0.7521 Sharpe, but still lost to BTC (305.50%) and BTC/ETH 50-50 (178.04%). Alpha verification failed.
Takeaway
The pipeline did its job: surface candidates, test them against costs and benchmarks, and reject the ones that don't clear the bar.
Related project
Project CHF (Crypto Hedge Fund v2)
9-agent crypto quant pipeline. Best signal hit Rank IC 0.0275, but after walk-forward validation and costs it reports alpha_verified=false.