feat: add split-safe fundamentals research protocol

docs/research/fundamentals-weight-backtest.md

@@ -1,6 +1,7 @@
 # Point-in-time fundamentals weight backtest
 
-Status: pre-registered local research. Running it does not change production.
+Status: initial experiment completed; split-safe follow-up registered locally.
+Running either protocol does not change production.
 
 ## Question
 
@@ -10,7 +11,7 @@ position sizing, capacity, costs, ATR trail, and post-stop re-entry policy remai
 unchanged. This isolates the incremental value of fundamentals as a ranking
 overlay.
 
-## Registered experiment
+## Completed initial experiment
 
 The control is the production 80/20 residual-momentum / volatility rank. The
 runner tests three fundamental composites at weights 10%, 20%, 30%, and 40%:
@@ -28,11 +29,38 @@ tracked universe. Missing composite scores are neutral at 50. The formula is:
 There are 13 registered portfolio trials including the control. That complete
 count is used by the Deflated Sharpe calculation.
 
-Historical P/E and FCF yield are excluded. Stored Alpaca bars are split-adjusted,
-while filing-time EPS and shares are not guaranteed to use today's split basis;
-mixing them without point-in-time split factors can manufacture valuation moves.
-Earnings surprise is also excluded because the completed Dolt SUE study already
-failed its promotion bar for this strategy.
+No overlay passed the registered train and validation requirements. Review also
+found that filing-time diluted EPS and shares are not guaranteed to use the same
+split basis across periods. That makes EPS growth and share-count change unsafe
+for historical ranking without point-in-time split factors. The initial result
+remains an auditable rejection of its registered arms, but it is not evidence
+that split-safe fundamentals have no value.
+
+## Registered split-safe follow-up
+
+Run with `--protocol split-safe`. This is a smaller sensitivity experiment:
+
+- Quality: operating margin, FCF margin, and low net-debt/EBITDA. At least two
+  inputs must exist.
+- Growth: revenue growth only.
+- Balanced: equal weight to the quality and growth sub-scores. Both must exist.
+- Overlay weights: 5%, 10%, and 15%.
+
+Diluted-EPS growth and share-count change are excluded completely: they are not
+ranked, do not enter composites, and do not appear in factor-IC output. Historical
+P/E and FCF yield remain excluded for the same split-basis reason. Earnings
+surprise remains excluded because the completed Dolt SUE study failed its
+promotion bar for this strategy.
+
+There are 10 registered portfolio trials including the control. The split-safe
+report uses 10 in its Deflated Sharpe calculation. It has a separate score cache
+and `fundamentals-splitsafe-*` output prefix, so it cannot be confused with or
+silently reuse the initial experiment's scores.
+
+The test window has already been inspected during the initial experiment. Keep
+the original date boundaries and development selection discipline, but treat the
+follow-up as sensitivity evidence. Any promotion still requires forward paper
+evidence.
 
 ## Point-in-time rule
 
@@ -93,6 +121,8 @@ different path to the launcher.
 
 ## 3. Run it
 
+The launcher now defaults to the registered `split-safe` follow-up:
+
 ```bash
 ./scripts/run_fundamentals_macbook.sh \
   backtest_snapshots/fundamentals-backtest.sqlite
@@ -106,16 +136,24 @@ WORKERS=8 ./scripts/run_fundamentals_macbook.sh \
   backtest_snapshots/fundamentals-backtest.sqlite
 ```
 
+To reproduce the completed initial matrix instead, opt in explicitly:
+
+```bash
+PROTOCOL=original ./scripts/run_fundamentals_macbook.sh \
+  backtest_snapshots/fundamentals-backtest.sqlite
+```
+
 The first run builds two caches under `reports/.cache`: production candidate
-replay and point-in-time fundamental scores. If interrupted, rerun the same
-command; valid caches are reused. Cache keys include snapshot size and mtime, so
-a new snapshot triggers a rebuild.
+replay and protocol-specific point-in-time fundamental scores. If interrupted,
+rerun the same command; valid caches are reused. Cache keys include the protocol,
+snapshot size, and mtime, so neither a protocol switch nor a new snapshot can
+reuse incompatible scores.
 
 ## 4. Bring the result back
 
 The final line names one ZIP such as:
 
-`reports/fundamentals-overlay-20260723-180000.zip`
+`reports/fundamentals-splitsafe-20260723-180000.zip`
 
 That ZIP contains: