We take 96 trades a day and lose -3.3% on each one. The losses are not spread evenly — they are concentrated in trades that die within an hour (-$22,401 of the -$23,077 total). Two mechanical defects cause most of it: a stop that cannot fill where it is set because exits run on a 5-minute cron, and no quality gate on entry — 6% of what we block is blocked for capacity, only 9% for quality. Fix those two and the book moves roughly $3,267 without finding a single new signal.
No — but not for the right reasons. Of 1,189,252 evaluations, 1,159,699 were blocked. The system therefore looks selective. But when you split the block reasons, 6% are capacity limits — we were already full, so the signal was dropped regardless of quality. Only 9% were rejected because the setup was judged bad.
That means selection is effectively arbitrary with respect to quality: which trade gets in depends on when it arrived and whether a slot was free, not on whether it was good. That is why the book tracks the random control so closely.
| evaluations | 1,189,252 |
| blocked | 1,159,699 |
| entered | 29,553 |
| scored here | 5,653 |
| reason | n | % of blocks | kind | what it means |
|---|---|---|---|---|
| strict_gate_reject | 416,435 | 35.9% | DATA | — |
| no_price | 212,810 | 18.4% | DATA | no price available — cannot enter |
| mc_below_floor | 128,774 | 11.1% | DATA | — |
| admission_reject | 91,543 | 7.9% | QUALITY | admission gate judged it bad |
| mc_resolved_dexscr | 84,927 | 7.3% | DATA | — |
| quality_tx_unavailable | 73,393 | 6.3% | DATA | — |
| src_cap | 56,048 | 4.8% | CAPACITY | per-source slot limit hit |
| decision_reject | 18,272 | 1.6% | QUALITY | decision engine refused |
| sm_distrib | 14,949 | 1.3% | DATA | smart money selling into it |
| quality_tx_reject | 14,073 | 1.2% | DATA | — |
| total_cap | 12,277 | 1.1% | CAPACITY | global open-position limit hit |
| fomo_large_tx_reject | 12,235 | 1.1% | DATA | — |
| nonpositive_or_invalid_size | 10,156 | 0.9% | DATA | — |
| source_cap | 3,866 | 0.3% | DATA | — |
| throttled | 3,402 | 0.3% | CAPACITY | rate limiter |
| shadow_mc_cap | 2,735 | 0.2% | DATA | — |
| young_defer | 2,019 | 0.2% | DATA | — |
| score_override | 811 | 0.1% | DATA | — |
| shadow_mc_deadline | 434 | 0.0% | DATA | — |
| ranker_subfloor_admit | 305 | 0.0% | DATA | — |
| oggy_net_cap | 144 | 0.0% | DATA | — |
| fomo_large_tx_unavailable | 74 | 0.0% | DATA | — |
| buys_band | 17 | 0.0% | DATA | — |
| verdict | n | % book | expectancy | win | P&L |
|---|---|---|---|---|---|
| NEVER_ARMED | 4802 | 84.9% | -8.0% | 11% | -$37,513 |
| CLEAN_WIN | 247 | 4.4% | +27.5% | 100% | $6,015 |
| BE_STOP_SAVE | 225 | 4.0% | -8.0% | 0% | -$1,768 |
| ROUND_TRIP | 143 | 2.5% | -21.5% | 0% | -$2,257 |
| TAIL_WIN | 124 | 2.2% | +190.4% | 100% | $17,795 |
| STOP_GAP | 61 | 1.1% | -74.1% | 0% | -$4,474 |
| DEAD_ZONE_ARM | 48 | 0.8% | -24.3% | 0% | -$847 |
| TIME_STOP | 2 | 0.0% | -3.4% | 0% | -$6 |
| ORDINARY_LOSS | 1 | 0.0% | -22.7% | 0% | -$23 |
| chain | n | exp | win | P&L |
|---|---|---|---|---|
| sol | 1377 | -1.6% | 19% | -$4,270 |
| robinhood | 1163 | -5.2% | 14% | -$6,775 |
| bsc | 1094 | -5.1% | 19% | -$6,262 |
| eth | 1073 | -1.0% | 15% | -$2,262 |
| base | 945 | -3.7% | 10% | -$3,489 |
| regime | n | exp | win | P&L |
|---|---|---|---|---|
| poor | 2608 | -2.9% | 15% | -$10,384 |
| ground | 2559 | -3.8% | 16% | -$11,219 |
| extreme_poor | 486 | -2.5% | 20% | -$1,474 |
A source only has edge if it beats random entry (n=4, expectancy -22.7%, tail 0.0%). Win rate is deliberately not the deciding column — tail rate is, because 2% of trades generate all the profit.
| source | n | expectancy | win | tail | vs control | P&L | call |
|---|---|---|---|---|---|---|---|
| volume_path | 1832 | -3.1% | 17% | 0.7% | +19.6pp | -$6,304 | KEEP |
| whalemover_rh | 737 | -2.8% | 13% | 0.5% | +19.9pp | -$2,470 | KEEP |
| source_strategy | 519 | -3.6% | 13% | 1.0% | +19.1pp | -$2,119 | KEEP |
| known_conv | 471 | -3.5% | 13% | 0.4% | +19.2pp | -$1,515 | KEEP |
| wallet_intel | 315 | -1.5% | 18% | 3.5% | +21.3pp | -$1,081 | KEEP |
| whalemover | 297 | -2.6% | 13% | 1.0% | +20.1pp | -$880 | KEEP |
| sa_accum | 290 | -4.0% | 16% | 1.4% | +18.8pp | -$1,564 | KEEP |
| clu:fomo | 235 | -9.1% | 17% | 0.9% | +13.6pp | -$2,290 | KEEP |
| ranker | 216 | -11.8% | 24% | 2.8% | +10.9pp | -$2,901 | KEEP |
| accum_engine | 176 | -0.9% | 15% | 1.7% | +21.8pp | -$391 | KEEP |
| alpha_brief | 82 | -7.6% | 20% | 1.2% | +15.1pp | -$571 | KEEP |
| wallet_goldilocks | 55 | -8.5% | 16% | 5.5% | +14.2pp | -$560 | KEEP |
| oggy_kol | 46 | +65.4% | 15% | 6.5% | +88.1pp | $1,947 | KEEP |
| whale_driver | 46 | -6.8% | 11% | 0.0% | +15.9pp | -$304 | WATCH |
| fresh_large_buy | 44 | -19.8% | 9% | 0.0% | +2.9pp | -$851 | WATCH |
| leading_conv | 40 | -3.3% | 5% | 0.0% | +19.4pp | -$133 | WATCH |
| hold | n | expectancy | win | tail | P&L | P&L |
|---|---|---|---|---|---|---|
| <1h | 4394 | -5.2% | 14% | 0.3% | -$22,401 | |
| 1-2h | 186 | -14.4% | 13% | 2.7% | -$2,800 | |
| 2-4h | 153 | -3.3% | 20% | 3.9% | -$1,111 | |
| 4-6h | 70 | +6.0% | 27% | 7.1% | $83 | |
| 6-24h | 263 | +15.2% | 30% | 7.6% | $1,796 | |
| 1-7d | 248 | +12.6% | 25% | 6.9% | $1,672 | |
| >7d | 327 | +1.1% | 24% | 1.5% | -$2 |
Bar height = |expectancy|. Teal above zero, red below. Best: 14:00 (+6%), 22:00 (+5%), 04:00 (+1%). Worst: 20:00 (-8%), 23:00 (-8%), 08:00 (-8%). Treat hour-of-day as a tiebreaker only — the sample per hour is 170–366 trades and these gaps are not yet significant.
Full journal: paper_journal.jsonl (29,553 entries) · rendered at /journal.html. Below: the two costliest trades in each failure class, each with the machine-written why and the fix it implies.
Realized universal-paper exits only; selection lift is graded separately in alpha_feature_scoreboard.json. Decision effect: NONE.
| arm | resolved / total | coverage | median return | promotion gate |
|---|---|---|---|---|
| M22_BURST_PROPENSITY_GE_5PCT | 506/530 | 95.5% | -10% | WAIT_MATCHED_CONTROL_OOS_FDR |
| M22_NR7 | 396/418 | 94.7% | -3% | WAIT_MATCHED_CONTROL_OOS_FDR |
| M22_PPV10 | 246/258 | 95.3% | -3% | WAIT_MATCHED_CONTROL_OOS_FDR |
| M22_PPV10_NR7 | 11/11 | 100.0% | -3% | WAIT_SAMPLE |
| M22_PPV10_WR20_BREAKOUT | 115/121 | 95.0% | -4% | WAIT_MATCHED_CONTROL_OOS_FDR |
| M22_SUPPLY_RECLAIM | 586/612 | 95.8% | -3% | WAIT_MATCHED_CONTROL_OOS_FDR |
| M22_UNRESOLVED_SUPPLY | 907/983 | 92.3% | -3% | WAIT_MATCHED_CONTROL_OOS_FDR |
| M22_WR20_BREAKOUT | 175/184 | 95.1% | -3% | WAIT_MATCHED_CONTROL_OOS_FDR |
Yes — and the absence of one is the root cause, not a missing feature. Today entry is decided by arrival order and free slots. A real admission brain would score every candidate before it competes for a slot, and spend the day's risk budget on the best N rather than the first N.