Audited builds
Each strategy below has been run on real market data across seven pairs at the timeframe it declares, after fees and slippage, then audited for correctness and for uniform presentation in the Gunbot GUI. Most declare 60m and are measured over three years; the sub-hourly ones are measured on their own bars over a shorter window, because measuring a 1-minute strategy on hourly bars measures a different strategy. Each report states its own window and bar size. Every defect found was fixed, and every fix is documented with the measurement that proves it.
| Strategy | Median return | Buy & hold | Edge | Max DD | In market | Profitable | Invariants | Uniform | Fixes |
|---|---|---|---|---|---|---|---|---|---|
| adxTrendMMhand-written (no shared engine) | -56.8% | 8.1% | -64.9% | 56.8% | 3.2% | 0 of 7 | 4 / 5 pass | 4 / 5 | 4DEFECTDEFECTDEFECTDEFECT |
| apexEnginehand-written (no shared engine) | -39.6% | 49.6% | -94.8% | 64.1% | 96.9% | 0 of 7 | 4 / 5 pass | 5 / 5 | 3DEFECTDEFECTNOTE |
| ASkewMM_VWAPneutral-mm | -0.9% | 49.6% | -68.8% | 57.1% | 99.2% | 3 of 7 | 5 / 5 pass | 4 / 5 | 3DEFECTDEFECTDEFECT |
| ATRChannel_MMmm-dca | -41.7% | 39.2% | -112.0% | 61.4% | 14.8% | 1 of 7 | 5 / 5 pass | 5 / 5 | 2NOTEDEFECT |
| btcRegimeScalerhand-written (no shared engine) | -78.2% | 49.6% | -127.8% | 80.5% | 33.5% | 0 of 7 | 4 / 5 pass | 5 / 5 | 5DEFECTDEFECTDEFECTDEFECTWARNING |
| ConnorsRSI_Compositedirectional | -1.4% | 39.2% | -40.1% | 3.2% | 0.5% | 1 of 7 | 5 / 5 pass | 5 / 5 | 4WARNINGDEFECTWARNINGNOTE |
| dipHunterhand-written (no shared engine) | -64.5% | 49.6% | -114.1% | 66.9% | 80.1% | 2 of 7 | 5 / 5 pass | 5 / 5 | 3DEFECTDEFECTWARNING |
| Donchian_Reversion_MMmm-dca | -85.3% | 49.6% | -134.9% | 85.6% | 10.5% | 0 of 7 | 4 / 5 pass | 5 / 5 | 2WARNINGDEFECT |
| donchianBreakoutMMhand-written (no shared engine) | -16.3% | 8.1% | -24.1% | 16.6% | 0.7% | 0 of 7 | 4 / 5 pass | 4 / 5 | 5DEFECTDEFECTDEFECTDEFECTWARNING |
| FIL_Grid_Enginefil-grid | -2.8% | 49.6% | -53.4% | 2.8% | 0.1% | 0 of 7 | 5 / 5 pass | 5 / 5 | 4DEFECTDEFECTDEFECTWARNING |
| fxLadderMMhand-written (no shared engine) | -95.9% | 39.2% | -134.7% | 96.6% | 92.3% | 0 of 7 | 4 / 5 pass | 4 / 5 | 6DEFECTDEFECTWARNINGDEFECTWARNINGNOTE |
| kalmanGridhand-written (no shared engine) | -1.4% | -0.8% | 0.1% | 1.4% | 0.0% | 0 of 7 | 5 / 5 pass | 5 / 5 | 4DEFECTWARNINGDEFECTWARNING |
| meanRevOscillatorhand-written (no shared engine) | 0.9% | -0.8% | 1.7% | 3.5% | 99.8% | 4 of 7 | 4 / 5 pass | 5 / 5 | 3DEFECTDEFECTWARNING |
| QGrid_DONCHIAN_PCT_ADX_Antiquantgrid | 3.1% | 49.6% | -46.5% | 3.4% | 95.8% | 6 of 7 | 5 / 5 pass | 3 / 5 | 1DEFECT |
| QGrid_KAMA_PCT_ADX_Antiquantgrid | 3.2% | 49.6% | -45.6% | 4.4% | 95.8% | 7 of 7 | 5 / 5 pass | 3 / 5 | 1DEFECT |
| rangeRiderhand-written (no shared engine) | 6.8% | 49.6% | -42.9% | 23.8% | 8.4% | 7 of 7 | 4 / 5 pass | 4 / 5 | 4DEFECTWARNINGDEFECTDEFECT |
| SuperTrend_MMmm-dca | -80.9% | 49.6% | -130.5% | 80.9% | 4.4% | 0 of 7 | 4 / 5 pass | 5 / 5 | 6DEFECTDEFECTDEFECTWARNINGDEFECTWARNING |
| volHarvesterhand-written (no shared engine) | -10.5% | 49.6% | -66.4% | 15.5% | 81.4% | 1 of 7 | 5 / 5 pass | 4 / 5 | 5DEFECTDEFECTDEFECTWARNINGDEFECT |
| volRegimeMMhand-written (no shared engine) | -10.5% | 49.6% | -66.4% | 15.5% | 81.4% | 1 of 7 | 5 / 5 pass | 4 / 5 | 5DEFECTDEFECTDEFECTWARNINGDEFECT |
| WaveTrend_LBdirectional | -24.9% | 49.6% | -74.1% | 26.9% | 2.5% | 0 of 7 | 5 / 5 pass | 5 / 5 | 4DEFECTDEFECTDEFECTWARNING |
Every figure is the median across the seven markets, not the best one. Ranking a strategy on its best pair measures which pair was picked, not what the strategy does.
How to read the return column
The edge column matters more than the return column, and both need the buy-and-hold column next to them. Over this window buy-and-hold was strongly positive on most pairs, so a strategy can show a respectable return while having badly underperformed simply holding the asset — and a strategy that sits in cash can show a small loss that would have looked like a large win in a bear window.
The buy-and-hold figure is measured over the same window as the strategy beside it, so the two are comparable row by row. They are not comparable between rows where the windows differ — a sub-hourly strategy measured over six months sits next to an hourly one measured over three years, and the edge column is the only column that survives that comparison.
How every audit is measured — window, cost model, fill model and the five correctness invariants →
The reports
What these reports do not claim
An audited strategy is correct, not profitable. Nothing here was measured on a live exchange, the engine models spot and long-only with a single price per bar, and no strategy in this pack should be run unattended on capital that matters.