BETA · descriptive · not a signal
The Engine Read track record
Engine Read is our internal multi-layer directional research engine. This page is
its measured record, the hit-rate history the accuracy loop computes
every round, published with the same rule as everything else we ship: every number
carries its sample size and its confidence bound, and the unflattering parts print
first. It is not the liquidation-map accuracy ledger (that lives at
/proof), it is not a signal feed,
and nothing here has cleared the promotion kill-gate.
window: 7 days · headline horizon 1.0h
measured 2026-11-16T12:19:01Z
anti-Goodhart: CLEAN
The served stream, and the caveat that comes first
Read this before the numbers. In this window the
engine labelled 100.0% of resolved hours “sideways”, so its
3-class hit rate 61.5% (sample size n=9,100, Wilson 95% lower
bound 60.5%) tracks the sideways base rate 61.5% with a
residual of 0.0%. In plain terms: calling “sideways” almost
always is what produced this number: it is honest, it is measured, and it is
not evidence of directional skill. The directional test is the shadow stream
below, and today it is BETA.
| Segment | 3-class hit rate | n |
Moved fraction | Directional calls | Directional wrong |
| last 24h | 60.8% | 1,092 | 39.2% | 0 | 0 |
| 24–48h ago | 38.3% | 1,344 | 61.7% | 0 | 0 |
| prior 5 days | 66.3% | 6,664 | 33.7% | 0 | 0 |
Segments of the same 7-day window. “Moved
fraction” is the share of hours price moved beyond the sideways tolerance,
the hours a sideways call gets wrong.
The directional shadow stream: pre-registered verdict
Every directional call the engine would have emitted is logged and graded against
a pre-registered graduation ladder (unvalidated → beta → validated)
with fixed evidence floors; the ladder cannot be moved to fit the results.
The verdict below is computed by the same function the guardian journal records;
it fails closed to the stricter verdict on any degraded input.
Verdict: BETA · outlook (descriptive
annotation, never a gate input): converging_negative. Shadow directional hit rate
11.7% [11.0%, 12.5%]
over n=7,545 resolved calls
(47 resolved days), against a sideways
base rate of 69.2% (Wilson low
68.2%). BETA: enough resolved evidence to publish the track record with its bound; the beat-the-base-rate gate has NOT cleared (no edge claim).
| Direction | Shadow hit rate | n |
| downward | 12.9% | 1,889 |
| upward | 11.3% | 5,656 |
Conditioned on hours that actually moved (n=2,323),
the shadow stream’s directional hit rate is 38.1%.
- beta: cleared: n resolved 7,545 / 60 ✓; resolved days 47 / 14 ✓
- validated: not cleared: n resolved 7,545 / 120 ✓; resolved days 47 / 28 ✓
Calibration record: every horizon, every confidence bucket
combined corpus: 201,590 samples
147,806 scorable
combined 3-class: 36.2%
combined directional: 26.8%
(random baseline 50.0%)
| Horizon | Hit rate | n scorable |
Directional hit | High-conf hit | n high-conf |
| 1h | 69.9% | 39,560 | — | — | — |
| 4h | 68.7% | 39,429 | — | — | — |
| 24h | 56.9% | 38,311 | 19.5% | 44.9% | 3,698 |
| 168h | 44.7% | 30,425 | — | — | — |
| 720h | 0.0% | 81 | — | — | — |
Per-horizon hit rates from the calibration artifact
(model v1.cal-20261116, calibrated
2026-11-16T09:48:34Z, scoring basis
per_horizon_gate_worst_case_fail_closed). “Hit” at short horizons is
dominated by sideways calls; see the disclosure above.
Is the engine’s own confidence calibrated? Measured, not assumed
| Confidence bucket | Expected hit rate |
Measured hit rate | n |
| 50-60 | — | 36.8% | — |
| 60-70 | — | 15.3% | — |
| 70-80 | — | 17.4% | — |
| 80-95 | — | 19.5% | — |
Expected = the bucket midpoint the confidence claim implies.
Today the higher-confidence buckets UNDERPERFORM their implied rates:
the engine’s confidence is not yet calibrated, and this page says so
rather than borrowing credibility from a number.
P&L directional edge check (counterfactual):
net as-emitted -3.90 USD over
99 actual directional
calls; edge positive: no.
The engine has not demonstrated a positive directional edge after costs.
Where it is worst: published, not hidden
- AVAXUSDT: 48.5% (n=650, 335 sideways-calls that moved)
- XRPUSDT: 53.2% (n=650, 304 sideways-calls that moved)
- SUIUSDT: 53.2% (n=650, 304 sideways-calls that moved)
- ADAUSDT: 54.6% (n=650, 295 sideways-calls that moved)
- DOTUSDT: 54.6% (n=650, 295 sideways-calls that moved)
The weakest symbols in the current window, straight from the
measurement. A track record that only shows the wins is marketing.
The honesty block
Validation path: RETIRE-RECOMMEND. Gate evidence today:
counterfactual net as-emitted P&L -3.90 USD (negative is a fail);
the blocked-stream (calls the engine withheld) hit 17.6%
over n=1,695. Outstanding:
dir_hit_rate 0.195 < 0.4 resume bar; consecutive_clean_windows 0 < 2; counterfactual as-emitted edge_positive != true (net -3.898463); blocked candidate stream antipredictive (0.176 < 0.4, n=1695) — gates correctly holding.
Kill-gate not cleared. Nothing on this page has cleared
the promotion gate (net_usd>0 ∧ ci95_low>0) that guards every
Hunter Killer strategy claim. This is a research engine’s measured record,
published because measured honesty is the product: not a signal, not a
forecast, not financial advice.
Basis: walk-forward, out-of-sample grading by the
accuracy loop; this page reads the loop’s measurement artifacts
read-only and never recomputes a figure. Historical, not predictive.
Related public records: the liquidation-map accuracy
ledger at /proof and /accuracy, the cascade anatomy at
/contagion, the hypothetical reference desk at /reference-desk.
Measurement stamp: loop state 2026-11-16T12:19:01Z ·
verdict BETA · outlook converging_negative