conviction = per-wallet top-20% stake (p80), not flat $200

Swap the flat $200 conviction cutoff for a per-wallet percentile (top
20% of each wallet's own stake sizes) everywhere it was used:

- cache.py: canonical CONV_PCTILE=0.80 + conv_cutoff() helper (matches
  the dashboard's pctl: filter >0, sort, linear interp)
- conviction_scan.py: per-wallet quantile_cont(size,0.8) in SQL, was
  `size >= 200`
- validate_timing.py, pnl_focused.py: use cache.conv_cutoff

Rationale + validation: p80 reproduces flat-$200's win-rate lift on the
sharps while adapting to scale (a whale's $200 isn't conviction, a
minnow's is). Re-running the pipeline under p80: scan finds 218 profile
wallets (was 69), forward 62/83 profitable (p~0), +16% pooled ROI — edge
persists out-of-sample. Regenerated conviction_wallets.json /
watch_sharps.json; docs updated. skill.py/strategy.py/insider.py
untouched (score over all bets / size as copyability heuristic only).

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

live/README.md

@@ -76,9 +76,11 @@ archetype (beats underdog prices) is where real alpha may live — test it with
 
 ## The repeatable find (`conviction_scan.py` + `validate_timing.py`)
 
-The best result. Score wallets on their **high-conviction (≥$200) bets only**: the
-edge is wallets that win 70–80% on genuinely-uncertain (~0.4–0.6) markets — real
-skill, not favorite-riding.
+The best result. Score wallets on their **high-conviction bets only — the top 20%
+by stake size (per-wallet p80, not a flat $200)**: the edge is wallets that win
+70–80% on genuinely-uncertain (~0.4–0.6) markets — real skill, not favorite-riding.
+The per-wallet cutoff reproduces flat-$200's win-rate lift while adapting to each
+wallet's scale (a whale's $200 bet isn't conviction; a minnow's is).
 
 - `conviction_scan.py` — train pre-June / validate June on conviction bets →
   69 matches, **25/37 profitable forward (p=0.024)**. → `conviction_wallets.json`.