Abstract:
Predictive signals operating at very short horizons often exhibit strong gross performance in backtests but fail to survive realistic transaction costs due to prohibitive turnover. This research note argues that the inability to monetize such signals directly does not imply the absence of economic value. We distinguish between monetizable alpha, which survives trading frictions as a standalone strategy, and informational alpha, which contributes value by conditioning the execution of other strategies. Using 5-minute intraday data on SPY from 2007 to 2026, we construct a simple streak-based mean-reversion signal that delivers a gross CAGR of approximately 31.9% and a Sharpe ratio exceeding 2, yet becomes unprofitable once standard commissions are applied. We then deploy this signal as a tactical execution overlay on a baseline intraday trend-following strategy, conditioning entry and exit timing on short-term counter-moves rather than trading the fast signal independently. The enhanced strategy improves net-of-fee CAGR by approximately 200 basis points and raises the Sharpe ratio from 0.87 to 0.99. These findings suggest that evaluating predictive signals solely on standalone net performance may understate their true economic relevance, and that a multi-horizon framework integrating fast and slow alphas can unlock value inaccessible to either signal in isolation
Read the QuanTip from here: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=6247138
QuanTip: Improving Performance with Fast Alphas — A Tactical Overlay for Intraday Trend Trading. Read on SSRN