📊 Earnings Pipeline Replay — the deployed system, replayed day-by-day
James's test (2026-07-12): rebuild each morning's earnings watchlist exactly as the LIVE scanner would have (earnings wire → liquid top-10), run the play, score the execution. This is the system as DEPLOYED — not the idealized backtest. ·
live ·
trust report
Bottom line (2026-06-23 → 2026-07-10): ALL fired: n=11 · win 45% · avgR -0.16 · total -1.8R · A-grade only (gap ≥5%): n=6 · win 50% · avgR +0.01 · total +0.1R. Every result is SIMULATED (the play's exit rules on real prices) — no real orders. Compare against the concept backtest (+0.12R avg, 53% win over 90 days) before trusting live.
2026-07-10 · wire 10 earnings names → watched 3 liquid (DAL, WDFC, NRIX) → 0 fired+0.0R
No gap-down reclaimed & held — no signal (that is the system being picky, not broken).
2026-07-09 · wire 10 earnings names → watched 7 liquid (PEP, ASX, LEVI, RXT, PSMT, AZZ, SMPL) → 3 fired+0.2R
| Time | Sym | Grade (gap) | Entry | Stop | Target | R (sim) | Exit |
|---|
| 09:46 | PEP | B (-3.79%) | 136.50 | 134.62 | 142.10 | +0.72R | eod |
| 09:34 | LEVI | B (-3.77%) | 23.60 | 22.44 | 24.15 | +0.48R | target |
| 10:46 | RXT | B (-4.64%) | 5.04 | 4.51 | 5.84 | -1.00R | stop |
2026-07-08 · wire 8 earnings names → watched 2 liquid (PENG, KRUS) → 1 fired+1.1R
| Time | Sym | Grade (gap) | Entry | Stop | Target | R (sim) | Exit |
|---|
| 09:44 | KRUS | A (-10.9%) | 50.00 | 46.96 | 53.28 | +1.08R | target |
2026-07-07 · wire 3 earnings names → watched 2 liquid (RIVN, LAES) → 2 fired-2.0R
| Time | Sym | Grade (gap) | Entry | Stop | Target | R (sim) | Exit |
|---|
| 09:42 | RIVN | A (-12.15%) | 17.88 | 17.19 | 19.16 | -1.00R | stop |
| 09:40 | LAES | C (-2.83%) | 2.92 | 2.90 | 3.06 | -1.00R | stop |
2026-07-06 · wire 1 earnings names → watched 0 liquid (none) → 0 fired+0.0R
No gap-down reclaimed & held — no signal (that is the system being picky, not broken).
2026-07-02 · wire 5 earnings names → watched 1 liquid (GBX) → 0 fired+0.0R
No gap-down reclaimed & held — no signal (that is the system being picky, not broken).
2026-07-01 · wire 9 earnings names → watched 5 liquid (STZ, GIS, FDS, MSM, PRGS) → 0 fired+0.0R
No gap-down reclaimed & held — no signal (that is the system being picky, not broken).
2026-06-30 · wire 2 earnings names → watched 2 liquid (AVAV, CNXC) → 1 fired+0.6R
| Time | Sym | Grade (gap) | Entry | Stop | Target | R (sim) | Exit |
|---|
| 10:06 | CNXC | A (-20.68%) | 21.13 | 19.07 | 23.13 | +0.62R | eod |
2026-06-29 · wire 4 earnings names → watched 1 liquid (REPL) → 0 fired+0.0R
No gap-down reclaimed & held — no signal (that is the system being picky, not broken).
2026-06-26 · wire 7 earnings names → watched 1 liquid (FDXF) → 0 fired+0.0R
No gap-down reclaimed & held — no signal (that is the system being picky, not broken).
2026-06-25 · wire 18 earnings names → watched 10 liquid (MU, MKC, DRI, SNX, BB, JEF, AYI, CMC, FUL, WGO) → 2 fired-2.0R
| Time | Sym | Grade (gap) | Entry | Stop | Target | R (sim) | Exit |
|---|
| 09:42 | JEF | B (-3.81%) | 53.67 | 52.19 | 57.00 | -1.00R | stop |
| 09:34 | FUL | A (-5.0%) | 60.28 | 58.30 | 64.03 | -1.00R | stop |
2026-06-24 · wire 9 earnings names → watched 4 liquid (CBRS, FDX, PAYX, KBH) → 0 fired+0.0R
No gap-down reclaimed & held — no signal (that is the system being picky, not broken).
2026-06-23 · wire 4 earnings names → watched 3 liquid (CCL, SUNB, KFY) → 2 fired+0.4R
| Time | Sym | Grade (gap) | Entry | Stop | Target | R (sim) | Exit |
|---|
| 09:42 | CCL | A (-9.26%) | 27.72 | 26.97 | 29.77 | +1.35R | eod |
| 10:24 | SUNB | A (-6.67%) | 77.04 | 75.23 | 81.97 | -1.00R | stop |
Funnel = same code path as the live scanner (kti earnings wire pre-mkt + prior-day after-hours · liquidity ≥$30M prior-day dollar-volume · ETFs excluded · top 10). Detector = gap ≤−2% vs close, reclaim & HOLD session VWAP, stop under LOD, target = 2-day AVWAP. Times ET. Generated 2026-07-12.