C
Trading Autopsy
RSI(2) Mean Reversion · QQQ, SPY · 1d bars
· 2012-01-01 → 2026-07-06 · report generated 2026-07-06
Verdict: Mixed evidence. Some real signal, some real problems — trade small or not at all.
- Decent sample size (270 trades).
- Out-of-sample performance holds up (Sharpe 0.62 vs 0.20 in-sample) — a good sign it isn't curve-fit.
- Profitable in only 2/4 time segments — returns are concentrated in one period, not repeatable edge.
- Max drawdown -16.2% — most people abandon a strategy long before -30%.
- Monte Carlo: with the same trades reshuffled, a -28.8% drawdown was plausible — the realized -16.2% was partly luck of the ordering.
- Doubling fees+slippage turns the strategy unprofitable — the edge is thinner than the costs assumption. Real-world fills will decide everything.
- Buy & hold returned 984.9% over the same period vs the strategy's 25.9% — all this effort underperformed doing nothing.
The strategy as tested
Direction: long. Entry when ALL of:
rsi(2) < 10close > sma(200)
Exit: stop at 2.0× ATR(14); time exit after 10 bars; signal: rsi(2) > 70.
Costs modeled: 0.05% fee per side + 5 bps adverse slippage.
Fills occur at the next bar's open after each signal — no same-bar fantasy fills.
When a stop and target are both touched in one bar, the stop is assumed to hit first (pessimistic).
Headline results (full period)
| Total return | +25.9% |
| CAGR | +1.6% |
| Buy & hold same period | +984.9% |
| Max drawdown | -16.2% |
| Sharpe ratio | 0.32 |
| Sortino ratio | 0.19 |
| Trades | 270 (19/yr) |
| Win rate | 66.7% |
| Profit factor | 1.24 |
| Avg trade | +0.22% |
| Avg holding time | 4.5 days |
| Total fees & slippage paid | $1,491 |
Monthly returns
| Jan | Feb | Mar | Apr | May | Jun | Jul | Aug | Sep | Oct | Nov | Dec |
|---|
| 2012 | | +0.0% | +0.0% | +0.0% | +0.0% | +0.0% | +0.0% | +0.0% | +0.0% | -0.8% | +2.0% | +0.9% |
|---|
| 2013 | +0.9% | +0.2% | +0.0% | +0.0% | +0.0% | +3.3% | -0.2% | +0.8% | -0.4% | +0.1% | +0.9% | +1.0% |
|---|
| 2014 | -0.9% | -1.1% | +0.7% | +1.9% | +0.0% | +0.1% | -0.7% | +0.4% | +0.9% | +1.0% | +0.0% | +1.0% |
|---|
| 2015 | +0.5% | +1.7% | +1.3% | +0.7% | +0.4% | -0.5% | +0.2% | -1.5% | +0.0% | +0.0% | -0.9% | +1.5% |
|---|
| 2016 | -1.6% | +0.0% | +0.0% | -1.4% | +0.1% | -1.3% | +0.0% | +0.7% | +2.8% | -0.4% | -1.8% | +1.1% |
|---|
| 2017 | +0.3% | +0.0% | +0.8% | +0.6% | +0.8% | +0.0% | +0.2% | +1.5% | -0.1% | +0.5% | +0.3% | -0.3% |
|---|
| 2018 | +0.9% | -4.9% | -2.3% | +1.1% | +0.0% | -0.6% | +0.9% | +1.3% | -0.0% | -3.5% | +0.0% | +0.0% |
|---|
| 2019 | +0.0% | +0.0% | +0.3% | +0.0% | -3.4% | +0.3% | +0.0% | -0.3% | -0.8% | -0.8% | +0.0% | +0.9% |
|---|
| 2020 | +0.2% | -3.2% | +1.8% | +0.0% | +0.0% | +0.6% | +0.0% | +0.8% | +1.8% | -2.4% | +4.7% | +0.2% |
|---|
| 2021 | -0.9% | +0.0% | +1.3% | +0.0% | +1.4% | +0.7% | +1.9% | +2.0% | -0.9% | -0.2% | -1.0% | -1.1% |
|---|
| 2022 | -4.6% | +0.0% | +0.0% | +0.0% | +0.0% | +0.0% | +0.0% | +0.0% | +0.0% | +0.0% | +0.0% | +0.0% |
|---|
| 2023 | +0.0% | -0.6% | +2.2% | +1.3% | +1.2% | +0.0% | +0.0% | -0.2% | -1.0% | -1.9% | +0.1% | +0.0% |
|---|
| 2024 | +0.7% | +0.5% | +0.0% | -0.7% | +0.3% | +0.8% | -0.9% | +1.5% | +2.0% | -0.1% | +4.0% | +1.9% |
|---|
| 2025 | +2.7% | -5.4% | +0.0% | +0.0% | +1.1% | +0.0% | +0.0% | +2.2% | +0.0% | +1.2% | +0.6% | +0.7% |
|---|
| 2026 | +1.9% | +2.0% | -2.8% | +0.0% | +0.0% | -1.2% | +0.0% | | | | | |
|---|
Out-of-sample test
The strategy was evaluated on data after 2022-02-22 that plays no role in the
in-sample window. Overfit strategies collapse here.
| In-sample | Out-of-sample |
|---|
| Total return | +9.8% | +14.6% |
| Sharpe | 0.20 | 0.62 |
| Max drawdown | -16.2% | -6.1% |
| Win rate | 63.4% | 74.7% |
| Trades | 191 | 79 |
Consistency across time
| Period | Return | Max DD | Trades |
|---|
| 2012-01-03 – 2015-08-18 | +17.3% | -3.8% | 65 |
| 2015-08-18 – 2019-04-02 | -5.0% | -11.3% | 64 |
| 2019-04-02 – 2022-11-10 | -1.4% | -8.2% | 62 |
| 2022-11-10 – 2026-07-02 | +14.6% | -6.1% | 79 |
Monte Carlo drawdown analysis
1,000 reshuffles of the realized trade sequence — how bad could the drawdown have been
with the exact same trades in a different order?
| Median max drawdown (reshuffled) | -19.5% |
| 95th-percentile max drawdown | -28.8% |
| Worst simulated drawdown | -45.3% |
| Simulations ending profitable | 100% |
Cost sensitivity
| Zero costs | As modeled | 2× costs |
|---|
| Total return | +62.4% |
+25.9% | -6.6% |
This report is independent historical research provided for educational purposes only. It is not financial advice, not a recommendation to buy or sell any security, and not a prediction of future results. Backtested performance has inherent limitations: it is hypothetical, does not reflect actual capital at risk, and past performance does not guarantee future results. Trading involves substantial risk of loss.