The thesis
Mean-reversion is one of the oldest ideas in systematic trading, and this strategy states it plainly: buy the oversold (RSI below 30), sell the overbought (RSI above 70). It runs on a universe of 24 US large-caps spanning tech, financials, healthcare, staples, and energy — AAPL, MSFT, NVDA, JPM, V, JNJ, PG, XOM, CAT and others. The bet is that short-term price extremes in liquid blue-chips snap back toward their mean. It is now live.
Recent activity
The cadence has been quiet. Across the last six scheduled runs (20–27 October), only two fired any orders: two executions on 23 October and one on 27 October, with every other day recording zero trades and zero rejections. Portfolio value has hovered tightly around its $10k base — $9,924.74 on the 20th, $10,051.40 by the 27th.
The trades themselves tell the mean-reversion story cleanly. On 23 October the strategy sold 8 shares of HD at $318.76, having bought the same block at $299.12 on 21 September — an overbought exit booking a gain of roughly $157 before fees. Likewise it sold 21 shares of WMT at $118.83 on 16 October against a May entry at $115.75. Fresh buys into GOOGL (7 @ $322.21) and CAT (3 @ $688.79) represent new oversold entries.
Backtest vs. validation
Here is where discipline matters. The full backtest looks appealing: 14.73% total return over 451 days, a 7.98% CAGR, a 70.59% win rate across 38 trades, and a 15.64% max drawdown. Fees were negligible at $38.
But the walk-forward validation did not pass. Of four folds, three were positive — fold 2 was excellent (+11.1%, Sharpe 1.32) — yet the most recent out-of-sample fold (Dec 2025 → May 2026) lost 2.84% with a negative Sharpe of −0.33 and a 14.96% drawdown. The deflated Sharpe ratio, which penalises for the 6 trials run, sits at just 0.304, well short of confidence despite a healthier PSR of 0.785.
The balance
Strengths: a high win rate, low fee drag, and a sound, well-understood thesis that produced real gains in three of four periods. The HD and WMT round-trips show the mechanism working as designed.
Risks: the headline return leans heavily on one strong fold, the newest unseen period was a loss, and turnover of 879% means performance is sensitive to execution and regime. A 70% win rate alongside a negative out-of-sample fold hints that occasional losers are outsized, or that the edge is regime-dependent.
The honest read: mean-reversion is live and behaving sensibly, but its validation is a caution, not a green light. We would want the next out-of-sample stretch to confirm the edge before trusting the backtest's 14.73% at face value.