signal_log: weight-research data points, scan_outcomes forward labels, watch provenance

New additive columns on every logged scan: bar_ts (primary closed-bar label, indexed with
timeframe), engine_fingerprint (hash of the scoring config incl. the regime weight table in
force - SignalAggregator.engine_config_fingerprint, twin of the gate fingerprint), code_version
(short git SHA), penalty_tags_json / penalty_total / momentum_fading / exhaustion_count, and
post-hoc watch_level / watch_kinds / watch_line written by violent_watch.attach_watch via
signal_log.update_watch. Each indicators_json entry now also carries eff_weight (base weight x
regime multiplier, the weight the composite actually used) and wscore.

engine: IndicatorSignal.effective_weight / weighted_score are stamped inside _compute_composite
(record-keeping only); regime_weight_table(config) is the single source of truth for the
multiplier table (override honoured). Composite output byte-identical (120-report hash proof).

New table scan_outcomes: model-free forward moves per scan (fwd_1h/4h/24h in $ and ATRs,
mfe/mae over 24h, burst_1h, bars_used, complete) measured on the stored candle stream with the
advisory model's continuity rule; labelled opportunistically at the end of log() (time-boxed,
throttled, idempotent) - no daemon; scripts/label_scan_outcomes.py drives the backlog
(--dry-run is read-only). analyze_signal_log --scan-outcomes: forward move + hit rate by
TF x signal/grade with a NEUTRAL baseline and burst rate by watch level. Readers dedup by the
first row per (timeframe, bar_ts).

Module 31 (10 tests). Suite: 463 passed. Backtester untouched (test-locked).

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01RWV2yLt3YXHHGVYHSQU2EE

analyze_signal_log.py

@@ -1152,6 +1152,101 @@ def export_csv(events: Sequence[SignalEvent], path: str) -> None:
     print(f"\nWrote {len(recs)} events to {path}")
 
 
+# ── Scan outcomes (model-free forward returns; no bar data needed) ────
+
+
+def scan_outcome_rows(db_path: str) -> List[dict]:
+    """One labelled scan per (timeframe, bar_ts) — the LOWEST signal id.
+
+    Several surfaces log the same bar, so the raw table is duplicated;
+    SQLite takes the bare columns from the row min() selected."""
+    con = open_readonly(db_path)
+    try:
+        if not con.execute(
+                "SELECT 1 FROM sqlite_master WHERE type='table'"
+                " AND name='scan_outcomes'").fetchone():
+            return []
+        return [
+            {"id": r[0], "timeframe": r[1], "signal": r[3] or "",
+             "quality": r[4] or "-", "watch_level": r[5],
+             "fwd_4h_atr": r[6], "burst_1h": r[7], "complete": r[8]}
+            for r in con.execute(
+                "SELECT MIN(o.signal_id), o.timeframe, o.bar_ts, s.signal,"
+                " s.quality, s.watch_level, o.fwd_4h_atr, o.burst_1h,"
+                " o.complete FROM scan_outcomes o"
+                " JOIN signals s ON s.id = o.signal_id"
+                " GROUP BY o.timeframe, o.bar_ts")
+        ]
+    finally:
+        con.close()
+
+
+def _scan_stats(rows: Sequence[dict], directional: bool) -> Optional[dict]:
+    """n / mean / median forward move and the hit rate for one bucket.
+
+    'Hit' = the 4 h move agreed with the signal; for the NEUTRAL baseline
+    it degrades to the share of UP moves."""
+    vals = sorted(r["fwd_4h_atr"] for r in rows if r["fwd_4h_atr"] is not None)
+    if not vals:
+        return None
+    hits = 0
+    for row in rows:
+        move = row["fwd_4h_atr"]
+        if move is None:
+            continue
+        sign = -1 if (directional and "SELL" in row["signal"]) else 1
+        hits += int(sign * move > 0)
+    mid = len(vals) // 2
+    return {"n": len(vals), "mean": sum(vals) / len(vals),
+            "median": vals[mid] if len(vals) % 2 else
+            (vals[mid - 1] + vals[mid]) / 2,
+            "hit_rate": hits / len(vals)}
+
+
+def print_scan_outcomes(rows: Sequence[dict]) -> None:
+    """Forward-move report: no replay, no model — just what happened."""
+    if not rows:
+        print("No scan_outcomes rows yet "
+              "(python scripts/label_scan_outcomes.py)")
+        return
+    complete = [r for r in rows if r["complete"]]
+    print(f"\n{'═' * 66}\n  SCAN OUTCOMES — 4 h forward move in ATRs "
+          f"({len(complete)} complete of {len(rows)} labelled bars)")
+    print(f"{'═' * 66}")
+    directional = [r for r in complete if r["signal"] not in ("", "NEUTRAL")]
+    neutral = [r for r in complete if r["signal"] == "NEUTRAL"]
+    for title, groups, keyfn, is_dir in (
+            ("by timeframe x signal", directional,
+             lambda r: (r["timeframe"], r["signal"]), True),
+            ("by timeframe x grade", directional,
+             lambda r: (r["timeframe"], r["quality"]), True),
+            ("NEUTRAL baseline", neutral,
+             lambda r: (r["timeframe"], "NEUTRAL"), False)):
+        print(f"\n── {title} " + "─" * (52 - len(title)))
+        buckets: Dict[Tuple[str, str], List[dict]] = {}
+        for row in groups:
+            buckets.setdefault(keyfn(row), []).append(row)
+        for key in sorted(buckets):
+            stat = _scan_stats(buckets[key], is_dir)
+            if stat is None:
+                continue
+            print(f"  {key[0]:6} {key[1]:12} n={stat['n']:5d}  "
+                  f"mean {stat['mean']:+.3f}  med {stat['median']:+.3f}  "
+                  f"{'hit' if is_dir else 'up '} "
+                  f"{100 * stat['hit_rate']:.0f}%")
+    print("\n── burst (>=3x ATR bar) within 1 h, by watch level " + "─" * 15)
+    levels: Dict[str, List[int]] = {}
+    for row in complete:
+        if row["burst_1h"] is None:
+            continue
+        levels.setdefault(row["watch_level"] or "NULL", []).append(
+            int(row["burst_1h"]))
+    for level in sorted(levels):
+        hits = levels[level]
+        print(f"  {level:6} n={len(hits):5d}  burst "
+              f"{100 * sum(hits) / len(hits):.1f}%")
+
+
 # ── main ──────────────────────────────────────────────────────────────
 
 
@@ -1172,8 +1267,15 @@ def main(argv: Optional[Sequence[str]] = None) -> int:
     ap.add_argument("--news", action="store_true",
                     help="news-sentiment lead/lag + outcome buckets "
                          "(requires --stored-candles)")
+    ap.add_argument("--scan-outcomes", action="store_true",
+                    help="labelled forward moves per scan (no bar data "
+                         "needed); see scripts/label_scan_outcomes.py")
     args = ap.parse_args(argv)
 
+    if args.scan_outcomes:
+        print_scan_outcomes(scan_outcome_rows(args.db))
+        return 0
+
     if args.news and not args.stored_candles:
         print("--news requires --stored-candles")
         return 2