research: interpret prod book universe x horizon matrix; ignore candidate cache

Four-arm results: 505 stays positive (softer on deep history); liquid breadth destroys book under current knobs. Stop tracking 1.3GB pkl cache under reports/.cache.
tests done
2026-07-19 14:25:06 +02:00 · 2026-07-19 14:21:05 +02:00
4 changed files with 241 additions and 9 deletions
@@ -46,3 +46,4 @@ backtest_snapshots/
 # Rebuildable pickle caches are local accelerators, not decision evidence.
 reports/*.pkl
 reports/*.pk1
+reports/.cache/
@@ -71,17 +71,57 @@ or depth changes the risk story.

 ## Results

-*(filled after run)*
+Generated: `2026-07-19T14:07:37` · artifact
+`reports/prod-book-universe-horizon-20260719-140737.json`  
+Snapshot: MacBook deep `research.sqlite` (506 prod + breadth prices; 2.39M raw
+GTL candidates). Strategy knobs = live production (residual 80, 80/20 high-vol
+rank, ATR trail 3×, hold 30, gate-reset, `fill_mode=close`).

-| arm | universe | entry start | Sharpe | SE | CAGR % | max DD % | trades | notes |
-|---|---|---|---:|---:|---:|---:|---:|---|
-| A | 505 | 2022-07-01 | | | | | | |
-| B | 505+liquid | 2022-07-01 | | | | | | |
-| C | 505 | 2016-07-01 | | | | | | |
-| D | 505+liquid | 2016-07-01 | | | | | | |
+> Survivorship: today's constituents backfilled. **Compare arms relatively.**
+> Absolute deep CAGR/Sharpe are not deployable forecasts.

---
+| arm | universe | entries from | Sharpe | SE | CAGR % | max DD % | total ret % | trades | win % | vs SPY |
+|---|---|---|---:|---:|---:|---:|---:|---:|---:|---:|
+| **A** | 505 only | 2022-07-01 | **1.32** | 0.49 | **31.8** | **18.9** | +205 | 374 | 35.6 | +96.5 |
+| **B** | 505 + liquid 1500 | 2022-07-01 | 0.14 | 0.50 | −1.8 | 55.3 | −7 | 706 | 29.3 | +95.0 |
+| **C** | 505 only | 2016-07-01 | **0.88** | 0.31 | **16.7** | **24.4** | +369 | 763 | 36.7 | +257 |
+| **D** | 505 + liquid 1500 | 2016-07-01 | −0.06 | 0.32 | −7.0 | 73.9 | −52 | 1567 | 28.0 | +254 |
+
+Qualified longs: A 1 448 · B 6 587 · C 2 450 · D 11 551.
+
+### Read (relative only)
+
+1. **Same strategy, broader liquid universe kills the book** (A→B and C→D).
+   Sharpe collapses; DD roughly triples; win rate drops ~6–8pp; trade count
+   ~doubles. This matches earlier breadth IC work: the production residual +
+   high-vol package is a **large-cap / prod-universe** edge, not a
+   “more names = better” edge.
+
+2. **Longer history on 505 stays positive but softer** (A→C). Sharpe 1.32 → 0.88,
+   CAGR 32% → 17%, DD 19% → 24%. Still well above the liquid-breadth arms.
+   Levels are optimistic (survivorship); the useful message is “edge does not
+   vanish when 2018/2020 are included,” not “expect 17% CAGR forever.”
+
+3. **Arm A vs older Phase‑A / short-window controls** (~Sharpe 1.7–2.1): this
+   matrix re-ranked on deep research.sqlite with a fixed entry start; numbers
+   need not match prior reports row-for-row. Use **this table for A–D
+   comparisons**, not for rewriting the production baseline number.
+
+4. **No production change implied.** Keep the live ~505 universe. Do not broaden
+   the tradable set to liquid Nasdaq under current knobs without a new
+   pre-registered design (and almost certainly a different rank/tilt package).

 ## Verdict

-**PENDING_HUMAN** after numbers land.
+**Descriptive matrix complete.**
+
+| question | answer from this matrix |
+|---|---|
+| Prod book @ ~4y / 505 | Positive (arm A) |
+| Same + liquid Nasdaq | **No** — large degradation (arm B) |
+| Prod book since 2016 / 505 | Still positive, milder (arm C) |
+| Same + liquid Nasdaq deep | **No** — worst arm (arm D) |
+
+**PENDING_HUMAN** only for whether to log “universe broaden under current knobs”
+as rejected in the main research index. Strategy knobs unchanged either way.
+
@@ -0,0 +1,137 @@
+{
+  "generated_at": "2026-07-19T14:07:37.458454",
+  "snapshot": "/Users/taathde3/git/lab/signal_platform/backtest_snapshots/research.sqlite",
+  "snapshot_meta": {
+    "prod_universe_n": 506,
+    "price_symbols_n": 4654,
+    "raw_candidates": 2389258,
+    "liquid_top_n": 1500,
+    "liquid_min_price": 5.0,
+    "short_start": "2022-07-01",
+    "long_start": "2016-07-01"
+  },
+  "strategy": {
+    "note": "Live production knobs \u2014 no modifications",
+    "momentum": "residual_12_1 gate 80",
+    "rank": "residual_high_vol_blend_80_20",
+    "fill_mode": "close",
+    "cost_per_side": 0.001,
+    "exit": {
+      "mode": "atr_trailing",
+      "trailing_pct": 12.0,
+      "atr_multiplier": 3.0,
+      "hold_days": 30
+    },
+    "max_positions": 10,
+    "risk_per_trade": 0.01,
+    "reentry": "gate_reset"
+  },
+  "arms": [
+    {
+      "id": "A_prod_4y_505",
+      "label": "Prod book \u00b7 ~4y \u00b7 505 only",
+      "start": "2022-07-01",
+      "universe": "prod_505",
+      "n_candidates": 81626,
+      "n_qualified_longs": 1448,
+      "fill_mode": "close",
+      "ranking_key": "residual_high_vol_blend_80_20_score",
+      "exit_policy": "atr_trail3",
+      "hold_days": 30,
+      "sharpe": 1.32,
+      "sharpe_se": 0.49,
+      "cagr_pct": 31.8,
+      "max_drawdown_pct": 18.9,
+      "total_return_pct": 204.7,
+      "calmar": 1.68,
+      "trades": 374,
+      "win_rate": 35.6,
+      "n_returns": 1009,
+      "psr": 0.9965,
+      "start_date": "2022-07-01",
+      "end_date": "2026-07-13",
+      "spy_return_pct": 96.5,
+      "final_equity": 30467.86
+    },
+    {
+      "id": "B_prod_4y_505_liquid",
+      "label": "Prod book \u00b7 ~4y \u00b7 505 + liquid top-1500",
+      "start": "2022-07-01",
+      "universe": "prod_plus_liquid",
+      "n_candidates": 267579,
+      "n_qualified_longs": 6587,
+      "fill_mode": "close",
+      "ranking_key": "residual_high_vol_blend_80_20_score",
+      "exit_policy": "atr_trail3",
+      "hold_days": 30,
+      "sharpe": 0.14,
+      "sharpe_se": 0.499,
+      "cagr_pct": -1.8,
+      "max_drawdown_pct": 55.3,
+      "total_return_pct": -7.1,
+      "calmar": -0.03,
+      "trades": 706,
+      "win_rate": 29.3,
+      "n_returns": 1013,
+      "psr": 0.6107,
+      "start_date": "2022-07-01",
+      "end_date": "2026-07-17",
+      "spy_return_pct": 95.0,
+      "final_equity": 9287.52
+    },
+    {
+      "id": "C_prod_2016_505",
+      "label": "Prod book \u00b7 since 2016-07 \u00b7 505 only",
+      "start": "2016-07-01",
+      "universe": "prod_505",
+      "n_candidates": 190179,
+      "n_qualified_longs": 2450,
+      "fill_mode": "close",
+      "ranking_key": "residual_high_vol_blend_80_20_score",
+      "exit_policy": "atr_trail3",
+      "hold_days": 30,
+      "sharpe": 0.88,
+      "sharpe_se": 0.314,
+      "cagr_pct": 16.7,
+      "max_drawdown_pct": 24.4,
+      "total_return_pct": 369.4,
+      "calmar": 0.68,
+      "trades": 763,
+      "win_rate": 36.7,
+      "n_returns": 2519,
+      "psr": 0.9975,
+      "start_date": "2016-07-01",
+      "end_date": "2026-07-13",
+      "spy_return_pct": 256.9,
+      "final_equity": 46938.66
+    },
+    {
+      "id": "D_prod_2016_505_liquid",
+      "label": "Prod book \u00b7 since 2016-07 \u00b7 505 + liquid top-1500",
+      "start": "2016-07-01",
+      "universe": "prod_plus_liquid",
+      "n_candidates": 649305,
+      "n_qualified_longs": 11551,
+      "fill_mode": "close",
+      "ranking_key": "residual_high_vol_blend_80_20_score",
+      "exit_policy": "atr_trail3",
+      "hold_days": 30,
+      "sharpe": -0.06,
+      "sharpe_se": 0.316,
+      "cagr_pct": -7.0,
+      "max_drawdown_pct": 73.9,
+      "total_return_pct": -52.0,
+      "calmar": -0.1,
+      "trades": 1567,
+      "win_rate": 28.0,
+      "n_returns": 2523,
+      "psr": 0.4287,
+      "start_date": "2016-07-01",
+      "end_date": "2026-07-17",
+      "spy_return_pct": 254.1,
+      "final_equity": 4800.5
+    }
+  ],
+  "survivorship_banner": "Today's constituents backfilled. Relative arm comparison only.",
+  "pending_human": true
+}
@@ -0,0 +1,54 @@
+# Production book × universe × horizon — results
+
+Generated: `2026-07-19T14:07:37.458454`
+
+> Survivorship: today's constituents backfilled. Compare arms relatively; do not treat deep CAGR/Sharpe levels as deployable forecasts.
+
+## Arms
+
+| arm | universe | entry start | Sharpe | SE | CAGR % | max DD % | ret % | trades | qual longs | span |
+|---|---|---|---:|---:|---:|---:|---:|---:|---:|---|
+| A_prod_4y_505 | prod_505 | 2022-07-01 | 1.32 | 0.49 | 31.8 | 18.9 | 204.7 | 374 | 1448 | 2022-07-01→2026-07-13 |
+| B_prod_4y_505_liquid | prod_plus_liquid | 2022-07-01 | 0.14 | 0.499 | -1.8 | 55.3 | -7.1 | 706 | 6587 | 2022-07-01→2026-07-17 |
+| C_prod_2016_505 | prod_505 | 2016-07-01 | 0.88 | 0.314 | 16.7 | 24.4 | 369.4 | 763 | 2450 | 2016-07-01→2026-07-13 |
+| D_prod_2016_505_liquid | prod_plus_liquid | 2016-07-01 | -0.06 | 0.316 | -7.0 | 73.9 | -52.0 | 1567 | 11551 | 2016-07-01→2026-07-17 |
+
+## Config (production, unchanged)
+
+```json
+{
+  "note": "Live production knobs \u2014 no modifications",
+  "momentum": "residual_12_1 gate 80",
+  "rank": "residual_high_vol_blend_80_20",
+  "fill_mode": "close",
+  "cost_per_side": 0.001,
+  "exit": {
+    "mode": "atr_trailing",
+    "trailing_pct": 12.0,
+    "atr_multiplier": 3.0,
+    "hold_days": 30
+  },
+  "max_positions": 10,
+  "risk_per_trade": 0.01,
+  "reentry": "gate_reset"
+}
+```
+
+## Snapshot
+
+```json
+{
+  "prod_universe_n": 506,
+  "price_symbols_n": 4654,
+  "raw_candidates": 2389258,
+  "liquid_top_n": 1500,
+  "liquid_min_price": 5.0,
+  "short_start": "2022-07-01",
+  "long_start": "2016-07-01"
+}
+```
+
+PENDING_HUMAN — descriptive matrix only; no auto promotion.
+
+JSON: `reports/prod-book-universe-horizon-20260719-140737.json`
+
Author	SHA1	Message	Date
dennisthiessen	1c38a94dd0	research: interpret prod book universe x horizon matrix; ignore candidate cache Four-arm results: 505 stays positive (softer on deep history); liquid breadth destroys book under current knobs. Stop tracking 1.3GB pkl cache under reports/.cache.	2026-07-19 14:25:06 +02:00
Dennis Thiessen	a4d5ed7a93	tests done	2026-07-19 14:21:05 +02:00