# QSR stop-loss cap sweep — account `Paper03QSR`

Generated 2026-09-11T16:03:45.696Z · 15 buy event(s) (7 anchor/runner duplicate row(s) collapsed), 15 symbol(s)

Current live setting: `QSR_V1.maxStopPct` = **8%** cap on `2×ATR14`. Take-profit held fixed at `hardSellPct` = 5%.

**Research only** — re-simulated from real daily bars, stop-before-target pessimism, no trailing-runner leg, no portfolio constraints, and does NOT re-model position sizing (a wider stop would size fewer shares for the same $ risk in the real system — see this script's header). Never merge into the journal.

## All symbols combined

Read **%/day**, not %/trade — a wider stop cap only ever loses more when it's hit (never less), so per-trade expectancy alone doesn't tell the whole story either way. %/day accounts for how long capital sat tied up before resolving.

| Stop cap % | Buy events | Win % | Exp %/trade | Avg days held | **Exp %/day** | Hit TP | Stopped | Timed out |
|---|---|---|---|---|---|---|---|---|
| 4% | 15 | 73.3% | +2.60% | 2.3 | **+1.147%** | 11 | 4 | 0 |
| 5% | 15 | 80.0% | +3.07% | 2.5 | **+1.243%** | 12 | 3 | 0 |
| 6% | 15 | 80.0% | +2.98% | 2.5 | **+1.208%** | 12 | 3 | 0 |
| 7% | 15 | 80.0% | +2.98% | 2.5 | **+1.208%** | 12 | 3 | 0 |
| 8% ← current | 15 | 80.0% | +2.98% | 2.5 | **+1.208%** | 12 | 3 | 0 |
| 9% | 15 | 80.0% | +2.98% | 2.5 | **+1.208%** | 12 | 3 | 0 |
| 10% | 15 | 80.0% | +2.98% | 2.5 | **+1.208%** | 12 | 3 | 0 |
| 12% | 15 | 80.0% | +2.98% | 2.5 | **+1.208%** | 12 | 3 | 0 |

- Best by **%/trade**: 5% cap (3.07%/trade, avg 2.5 days held)
- Best by **%/day** (capital efficiency): **5%** cap (1.243%/day, 3.07%/trade over 2.5 days)
- Current 8%: 1.208%/day (2.98%/trade over 2.5 days)

## Symbol-day clustered (the honest significance view)

15 buy event(s) collapse to **15 independent symbol-day(s)** (1.0 per cluster). Means below are per symbol-day; the CI is computed on those, not on the raw rows.

| Stop cap % | Symbol-days | Mean %/trade | 95% CI | Differs from zero? |
|---|---|---|---|---|
| 4% | 15 | +2.60% | +0.52% … +4.68% | yes |
| 5% | 15 | +3.07% | +1.04% … +5.10% | yes |
| 6% | 15 | +2.98% | +0.85% … +5.10% | yes |
| 7% | 15 | +2.98% | +0.85% … +5.10% | yes |
| 8% ← current | 15 | +2.98% | +0.85% … +5.10% | yes |
| 9% | 15 | +2.98% | +0.85% … +5.10% | yes |
| 10% | 15 | +2.98% | +0.85% … +5.10% | yes |
| 12% | 15 | +2.98% | +0.85% … +5.10% | yes |

Best mean at symbol-day level: **5%** cap (+3.07%/trade) vs current 8% (+2.98%/trade). **Read the CIs before acting** — if they overlap heavily, the levels are not distinguishable on this sample regardless of which mean is highest.

## Per symbol

Sample sizes here are small (docs/METRICS.md wants ~20/bucket) — directional only.

| Symbol | Buy events | %/day @8% (current) | Best cap (%/day) | %/day @best |
|---|---|---|---|---|
| ADI | 1 | +0.833% | 4% | +0.833% |
| AON | 1 | -1.890% | 5% | -1.667% |
| ASX | 1 | +1.667% | 4% | +1.667% |
| CIEN | 1 | +2.500% | 4% | +2.500% |
| DDOG | 1 | +5.000% | 4% | +5.000% |
| ETN | 1 | +1.667% | 4% | +1.667% |
| FLEX | 1 | +5.000% | 4% | +5.000% |
| GLW | 1 | +5.000% | 4% | +5.000% |
| LRCX | 1 | +2.500% | 5% | +2.500% |
| NOK | 1 | +5.000% | 4% | +5.000% |
| NVO | 1 | -1.878% | 5% | -1.667% |
| PANW | 1 | +1.000% | 4% | +1.000% |
| PCAR | 1 | -2.009% | 4% | -2.000% |
| PWR | 1 | +2.500% | 4% | +2.500% |
| TER | 1 | +2.500% | 4% | +2.500% |