Fix post-cluster neutral strength ablation

docs/research/sr-levels-and-exits.md

@@ -498,6 +498,29 @@ Run on macOS:
 Do not validate any traffic arm yet. First establish whether geometry, pivots, or
 the range-occupancy grid reproduces production on pre-2024 training data.
 
+Initial result:
+
+| arm | Sharpe | CAGR | MaxDD | net avg R | ex-top-5% |
+|---|---:|---:|---:|---:|---:|
+| production control | 1.28 | 28.8% | 21.4% | 0.230 | 0.066 |
+| pivots only | 1.38 | 32.9% | 25.2% | 0.236 | 0.076 |
+| traffic grid only | **1.72** | **41.8%** | **16.8%** | **0.271** | **0.136** |
+
+The traffic grid is the first research arm to beat control simultaneously on
+Sharpe, CAGR, drawdown, and robust expectancy. It retains 264 production setups,
+adds 240 positive-expectancy setups, and removes 412 weaker setups. This supports
+the hypothesis that the useful legacy feature is the volume-weighted range-
+occupancy grid, not pivots or visually meaningful S/R.
+
+The first geometry-neutral result is not interpretable: neutral strength was set
+before the shared zone cluster, which summed multiple 50-strength constituents
+back to 100. Neutrality is now enforced after clustering. Rerun only that arm:
+
+```bash
+.venv/bin/python scripts/run_sr_v2_matrix.py traffic \
+  --only-arm legacy_geometry_neutral --workers 14
+```
+
 **Next runs, if picked back up:**
 
 - A **per-name target model** for clear-air setups instead of a constant k×ATR. This