id: 1ddbd98eea4b44529f0377e40ba9dd66
parent_id: 5e40d11e78fc4b0dac951b9c69fc866d
item_type: 1
item_id: d105cf38c62146daa0b7339bdc64aba5
item_updated_time: 1781722326237
title_diff: "[{\"diffs\":[[0,\"(fixes +\"],[1,\" value bet floor +\"],[0,\" sweeps)\"]],\"start1\":29,\"start2\":29,\"length1\":16,\"length2\":34}]"
body_diff: "[{\"diffs\":[[0,\" Commits\"],[-1,\"\\\n\\\n###\"],[0,\" \"],[1,\"(\"],[0,\"super_ma\"]],\"start1\":34,\"start2\":34,\"length1\":22,\"length2\":18},{\"diffs\":[[0,\"rvin\"],[-1,\" (4 commits\"],[1,\")\\\n\\\n### Session 1 (early\"],[0,\")\\\n1.\"]],\"start1\":52,\"start2\":52,\"length1\":19,\"length2\":31},{\"diffs\":[[0,\"### \"],[-1,\"super-marvin-userscripts (1 commit)\\\n1. **`887f8f6`** — `fix: syn\"],[1,\"Session 2 (evening — autonomous)\\\n4. **D4 fix** — `live_server/src/table_state.rs`: trivial `to_call` (< SB) guard prevents display-artifact `cur_bet=1001` corruption\\\n5. **Value bet floor** — `holdem_bots/src/cash/\"],[0,\"th\"],[1,\"r\"],[0,\"es\"],[-1,\"ize missing BB + guard reconstruction re-entry`\\\n\\\n## Issues Fixed\"],[1,\"holds.rs` + `rollout_postflop_base.rs`: new `value_bet_floor` parameter (default 1.0 = disabled). When `strength > value_bet_floor`, bets first-in regardless of multi-way power adjustment. Structural fix for S1 (passive value betting in multi-way pots).\\\n\\\n## Issues Resolved This Session\"],[0,\"\\\n\\\n| \"]],\"start1\":329,\"start2\":329,\"length1\":140,\"length2\":512},{\"diffs\":[[0,\" Issue |\"],[1,\" Status |\"],[0,\" Fix |\\\n|\"]],\"start1\":840,\"start2\":840,\"length1\":16,\"length2\":25},{\"diffs\":[[0,\"------|-----\"],[1,\"---|-----\"],[0,\"|\\\n| BB posti\"]],\"start1\":866,\"start2\":866,\"length1\":24,\"length2\":33},{\"diffs\":[[0,\"aptured \"],[1,\"(D1) | ✅ FIXED \"],[0,\"| Usersc\"]],\"start1\":907,\"start2\":907,\"length1\":16,\"length2\":31},{\"diffs\":[[0,\" in HU |\"],[1,\" ✅ FIXED |\"],[0,\" Gen1 HU\"]],\"start1\":1006,\"start2\":1006,\"length1\":16,\"length2\":26},{\"diffs\":[[0,\"s=0.083 \"],[1,\"(S2) | ✅ FIXED \"],[0,\"| `draw_\"]],\"start1\":1069,\"start2\":1069,\"length1\":16,\"length2\":31},{\"diffs\":[[0,\"on turn \"],[1,\"(S1) | ✅ ADDRESSED \"],[0,\"| `bet_a\"]],\"start1\":1150,\"start2\":1150,\"length1\":16,\"length2\":35},{\"diffs\":[[0,\"x = 3.0`\"],[1,\" + `value_bet_floor` (sweep pending)\"],[0,\" |\\\n| Dup\"]],\"start1\":1190,\"start2\":1190,\"length1\":16,\"length2\":52},{\"diffs\":[[0,\"attach |\"],[1,\" ✅ FIXED |\"],[0,\" `initia\"]],\"start1\":1266,\"start2\":1266,\"length1\":16,\"length2\":26},{\"diffs\":[[0,\"d |\\\n\"],[-1,\"\\\n## Sweep Results & Conclusions\\\n\\\n**Threshold tuning produces NO robust improvement.** Tested `bet_adj_max`, bet thresholds, call thresholds, draw restrictions across 30K × 3 seeds and 100K × 2 seeds. Results:\\\n\\\n| Test | Seed 42 | Seed 137 | Average |\\\n|------|---------|----------|---------|\\\n| bam=1.5+low_bets (30K) | +1.67 | +3.81 | +2.74 |\\\n| bam=1.5+low_bets (100K) | +5.12 | **-4.76** | **+0.18** |\\\n\\\nThe 30K results showed promise but 100K confirmation revealed it was seed-dependent variance. Average across 2 seeds at 100K hands is essentially zero.\\\n\\\n**Root cause of high variance**: With `reset_bankroll_each_hand = true`, BB/100 is dominated by a small number of large-pot outcomes. Parameter changes flip decisions in marginal spots, which cascades into different big-pot results, swinging the delta by ±5 BB/100 per seed.\\\n\\\n**Recommendation**: Stop threshold micro-tuning. Focus on structural improvements:\\\n- Gen3 opponent modeling (range estimation)\\\n- Gen4 RL\\\n- Preflop chart improvements (hand ranges vs heuristics)\\\n\\\n## Production Config (cash_nl.toml) — Current State\\\n```toml\\\nmc_samples = 50000\\\ncall_turn_base = 0.82\\\npower_callers_weight = 0.3\\\ncall_pot_odds_norm = 60.0\\\nraise_cbet_into_raiser = 0.82\\\nnpot_raise_ceiling = 0.35\\\ndraw_raise_min_strength = 0.15   # NEW: prevents pure-draw raises\\\nbet_adj_max = 3.0                # NEW: caps multi-way bet adjustment\\\n```\\\n\\\n## Audit Notes\\\n- Deleted audits #4, #5 (all issues fixed)\\\n- Audit #2 updated: QJo fixed, AQo/QTs partially addressed by new thresholds\"],[1,\"| Duplicate hand blocks / $0 pots (L1) | ✅ FIXED | `flushed_to_history` flag |\\\n| Blind reconciliation leaks (L2) | ✅ FIXED | sb_seat/bb_seat reset in start_hand |\\\n| Corrupt blinds after reload (D2) | ✅ RESOLVED | localStorage fallback + server reconcile + auto-reload |\\\n| Executor first raise didn't fire (D3) | ✅ RESOLVED | parseAmount k/M suffix fix (confirmed working audit #6) |\\\n| `cur_bet=1001 to_call=1` off-by-one (D4) | ✅ FIXED | Trivial to_call (< SB) guard |\\\n\\\n## All audit notes deleted — issues fully resolved.\\\n\\\n## Production Config (cash_nl.toml) — Current State\\\n```toml\\\nmc_samples = 50000\\\ncall_turn_base = 0.82\\\npower_callers_weight = 0.3\\\ncall_pot_odds_norm = 60.0\\\nraise_cbet_into_raiser = 0.82\\\nnpot_raise_ceiling = 0.35\\\ndraw_raise_min_strength = 0.15\\\nbet_adj_max = 3.0\\\n```\\\n\\\n## Value Bet Floor — Sweep Results (pending)\\\n\\\n2K hand preliminary (seed 42): directional signal — lower floor = more value betting = better vs chump/flock.\\\n\\\n| value_bet_floor | Var BB/100 | Base BB/100 | Delta |\\\n|-----------------|------------|-------------|-------|\\\n| 1.0 (disabled)  | 148.50     | 183.81      | -35.31 |\\\n| 0.75            | 167.00     | 188.58      | -21.58 |\\\n| 0.70            | 177.46     | 186.93      | -9.47 |\\\n| 0.65            | 188.56     | 187.28      | +1.28 |\\\n| 0.60            | 200.07     | 185.34      | +14.73 |\\\n\\\n30K × 2 seeds sweep in progress. If promising, 100K confirmation to follow.\\\n\\\n## Previous Sweep Conclusions\\\nThreshold tuning of EXISTING parameters produced no robust improvement (seed-dependent variance dominates). The `value_bet_floor` is a NEW structural parameter — it adds a parallel decision path (absolute strength check), not a tuning of an existing threshold. This is qualitatively different from the previous sweeps that failed.\\\n\"]],\"start1\":1303,\"start2\":1303,\"length1\":1515,\"length2\":1775}]"
metadata_diff: {"new":{},"deleted":[]}
encryption_cipher_text: 
encryption_applied: 0
updated_time: 2026-06-17T18:57:13.006Z
created_time: 2026-06-17T18:57:13.006Z
type_: 13