Strategy Overview
The mean-reversion agent applies one of the oldest price-action theses in quantitative finance: what goes too far in one direction tends to snap back. Operationally, this means buying any name in its 24-stock universe when the RSI drops below 30 (oversold territory) and exiting when RSI climbs above 70 (overbought). The universe spans large-cap names across technology, financials, healthcare, consumer staples, energy, and industrials — liquid, well-covered stocks where price dislocations tend to be temporary rather than structural.
The logic is sound in calm, range-bound markets. Whether it holds up across regimes is the central question the validation data now forces us to confront.
Backtest Performance
Over 451 days and 38 paper trades, the strategy compounded to a 14.73% total return (7.98% CAGR), with a Sharpe of 0.58 and a maximum drawdown of 15.64%. The win rate of 70.59% is genuinely strong — nearly three in every four closed positions were profitable.
Two caveats temper that picture. First, turnover sits at 879%, meaning the notional value traded was nearly nine times the starting equity. With $1 in fees per trade (totalling $38), real-world friction would be modest on this scale, but slippage on sub-$200 fills in illiquid conditions could meaningfully erode edge. Second, a 15.64% max drawdown against a 7.98% CAGR is a wide ratio — the strategy risks giving back nearly two full years of gains in a single adverse stretch.
Walk-Forward Validation: Where It Gets Complicated
Headmars runs a four-fold walk-forward validation before promoting any strategy. The mean-reversion agent did not pass.
| Fold | Period | Return | Sharpe | Max DD | Trades |
|---|---|---|---|---|---|
| 1 | Aug 2024 – Jan 2025 | +2.06% | 0.57 | 4.24% | 8 |
| 2 | Jan 2025 – Jul 2025 | +11.10% | 1.32 | 10.53% | 10 |
| 3 | Jul 2025 – Dec 2025 | +2.21% | 0.46 | 8.80% | 6 |
| 4 | Dec 2025 – May 2026 | −2.84% | −0.33 | 14.96% | 10 |
Three of four folds were positive, and Fold 2 was exceptional (Sharpe 1.32). But Fold 4 — the most recent and therefore most predictive period — reversed sharply. The out-of-sample return of −2.84% with a Sharpe of −0.33 is the gate the strategy could not clear.
The probabilistic Sharpe ratio (PSR) of 0.785 suggests a reasonable probability the strategy's true Sharpe is above zero, but the Deflated Sharpe Ratio of 0.304 tells a harder story: after adjusting for the six trials run during research, the edge may not be statistically distinguishable from chance at conventional thresholds.
Recent Activity
The strategy has been quiet. Six consecutive daily scans between June 19 and June 26 executed zero trades, with cash held at $7,569 out of a $10,024 total portfolio — roughly 75% idle. The most recent execution was a 21-share buy of WMT at $115.75 on May 31. The current market environment does not appear to be generating RSI extremes in the monitored universe at a rate that triggers signals.
This cash-heaviness is a double-edged signal: it means the strategy is disciplined and not forcing trades, but it also means the opportunity set for RSI-driven entries is thin right now.
Risks and Outlook
Mean-reversion strategies are well-documented to underperform during trending or momentum-driven markets. The Fold 4 drawdown of nearly 15% coincides with a period when trend and momentum factors broadly outperformed, which is consistent with this known failure mode.
The strategy deserves continued observation rather than dismissal. Three positive folds show the thesis is not wrong, and the win rate is a real edge. But until a future walk-forward window confirms the out-of-sample result recovers, the validation gate exists for good reason — and it is doing its job.