Grade Evidence Track Record
Backtest / Walk-Forward Simulation — Not Live Returns
The current artifact covers a walk-forward backtest over 2021-06-28 – 2026-06-27, using the same detectors that power live screening. They are not live trading results and do not represent what any user actually earned.
Survivorship bias: Backtested on currently-listed symbols only. Stocks that delisted, went bankrupt, or were acquired during the test period are excluded — these tend to be losers. Reported figures are higher than what a live trader would have observed — survivors inflate backtest results. Assume the true historical expectancy is lower.
Entry model: Entry is simulated at the next trading day's open after the signal fires, not the signal bar's close. The engine sees only data up to and including the current bar — no future data is used. No slippage, commissions, or market impact are modeled.
Grade as quality filter: The grade is a setup-quality filter, not a forecast of return. It blends entry efficiency, trend/MA structure, relative strength, and volume into one label.
Grade cohort evidence
Performance evidence unavailable
The authoritative autoresearch artifact is incomplete, so no grade-cohort performance claim is shown.
Missing evidence
costs.fees, costs.slippage, grades[0].trades, grades[0].netProfitFactor, grades[0].maxDrawdownPct, grades[0].holdTime.medianDays, grades[0].holdTime.averageDays, grades[1].trades, grades[1].netProfitFactor, grades[1].maxDrawdownPct, grades[1].holdTime.medianDays, grades[1].holdTime.averageDays, grades[2].trades, grades[2].netProfitFactor, grades[2].maxDrawdownPct, grades[2].holdTime.medianDays, grades[2].holdTime.averageDays, grades[3].trades, grades[3].netProfitFactor, grades[3].maxDrawdownPct, grades[3].holdTime.medianDays, grades[3].holdTime.averageDays, grades[4].trades, grades[4].netProfitFactor, grades[4].maxDrawdownPct, grades[4].holdTime.medianDays, grades[4].holdTime.averageDays, ungradedBaseline, regimesFormula version: v3-riskfloor-1 · Backtest window: 2021-06-28 – 2026-06-27· Numbers last computed: 2026-06-27
Optional methodology · strategy health
Retired strategy evidence
Of 23 candidate setups run through the same anti-overfit gates, 12 passed strictly, 1 are provisional (selection region validated, execution params held at defaults), 4 remain unverifiable on current data, and 6 were killed as empirically broken. We show the dead ones by name.
RSI Reversion
Killed — a raw profit factor of 1.67 over 533 out-of-sample trades did not survive our anti-overfit gate (parameter-robustness + shuffle-permutation + multiple-testing Sharpe haircut) — the edge is statistically indistinguishable from luck. Dropped from live picks.
Bear Flag
Killed — a raw profit factor of 0.67 over 29 out-of-sample trades did not survive our anti-overfit gate (parameter-robustness + shuffle-permutation + multiple-testing Sharpe haircut) — the edge is statistically indistinguishable from luck. Dropped from live picks.
RSI Overbought
Killed — a raw profit factor of 1.51 over 512 out-of-sample trades did not survive our anti-overfit gate (parameter-robustness + shuffle-permutation + multiple-testing Sharpe haircut) — the edge is statistically indistinguishable from luck. Dropped from live picks.
Swing Condor
Killed — a raw profit factor of 1.72 over 40 out-of-sample trades did not survive our anti-overfit gate (parameter-robustness + shuffle-permutation + multiple-testing Sharpe haircut) — the edge is statistically indistinguishable from luck. Dropped from live picks.
Frog-in-the-Pan Momentum
Killed — a raw profit factor of 1.84 over 88 out-of-sample trades did not survive our anti-overfit gate (parameter-robustness + shuffle-permutation + multiple-testing Sharpe haircut) — the edge is statistically indistinguishable from luck. Dropped from live picks.
ATR Stretch Reversion
Killed — a raw profit factor of 1.35 over 1,693 out-of-sample trades did not survive our anti-overfit gate (parameter-robustness + shuffle-permutation + multiple-testing Sharpe haircut) — the edge is statistically indistinguishable from luck. Dropped from live picks.
Full per-strategy verdicts — holdout profit factor, robustness, permutation p-value, and Sharpe haircut for every setup — are on the performance page →
Full methodology
- Backtesting methodology → walk-forward approach, regime approximation, entry model
- Strategy metrics methodology → win-rate computation, survivorship bias, "limited data" handling
Provenance
- Schema version: 2.0
- Formula version: v3-riskfloor-1
- Metric: avg_r
- Data last computed: 2026-06-27
Machine-readable feed at /track-record.json. Build-time companion at /track-record.md.