Alpha Decay

Alpha decay is the reduction in predictive power of a trading signal over time. A signal that worked in the past may weaken as markets evolve, competitors copy the idea, or transaction costs rise.

Common Causes

Alpha often decays due to crowding and competition. Structural changes in market microstructure, such as changes in tick size or liquidity, can also weaken a signal. Overfitting to historical data produces fragile signals that fail out of sample.

Measuring Decay

Decay can be measured by tracking the information coefficient, hit rate, or signal return over time. Another common measure is signal half-life, the time it takes for a signal to lose half its predictive power. Monitoring live performance drift relative to backtests provides early warnings.

Impact on Strategy Design

High turnover strategies are more sensitive to decay because small changes in signal quality can erase returns after costs. Portfolio diversification across multiple independent signals can reduce reliance on any single alpha source.

Mitigation Strategies

Mitigation includes refreshing models, expanding data sources, and improving execution to reduce costs. Some teams rotate signals or use adaptive weighting based on recent performance. A disciplined research process helps separate true signal from noise.

Conclusion

Alpha decay is inevitable in competitive markets. Strategies that plan for decay and monitor it continuously are more likely to survive long term.