# Signal stats — do entry conditions predict win/loss? (2026-08-11)

Entry-time `SignalSnapshot` fields correlated against each trade's eventual P&L. MAE/MFE are deliberately excluded — they describe what happened AFTER entry, not something known before the buy (docs/METRICS.md's own separate diagnostic pair). Read-only, no journal write, no network calls — this only re-reads `data/trades.jsonl`.

## qsr (69 trades)

| Feature | n | corr with P&L (r) | mean · winners | mean · losers | winner−loser |
|---|---|---|---|---|---|
| RSI(2) | 24 | 0.558 | 50.48 | – | – |
| % from MA50 | 24 | 0.512 | -1.35 | – | – |
| RSI(14) | 24 | 0.482 | 46.71 | – | – |
| % change today | 24 | 0.419 | 3.75 | – | – |
| ATR % of price | 24 | -0.197 | 4.42 | – | – |
| % since open at entry | 62 | 0.132 | 0.46 | -0.33 | 0.79 |
| Gap from signal close | 62 | -0.126 | -0.42 | 0.33 | -0.75 |
| Relative volume | 24 | 0.122 | 0.44 | – | – |
| Minutes since open at entry | 62 | -0.119 | 113.63 | 174.75 | -61.12 |
| MA50 slope % | 24 | 0.110 | 0.31 | – | – |
| % from MA200 | 24 | 0.063 | 5.12 | – | – |

`r` ranges −1..+1 — closer to either end means the entry value and the trade's eventual P&L move together (or apart) fairly consistently across this sample; near 0 means this feature alone doesn't distinguish winners from losers here. `winner−loser` is the same idea in the feature's own units.

- **Market regime (SPY vs MA200):** bull 24 trades, 100% win, 4.40% avg · bear 0 trades, –% win, –% avg