{
 "slug": "model-watch-2026-09-28",
 "series": "weekly",
 "series_label": "WEEKLY · MODEL WATCH",
 "issue": 9,
 "title": "Weekly Model Watch #9",
 "language": "en",
 "window_start": "2026-09-28T00:00:00+00:00",
 "window_end": "2026-10-05T00:00:00+00:00",
 "window_text": "Sep 28-Oct 4, 2026 UTC",
 "cutoff": "2026-10-05T16:00:00+00:00",
 "generated_at": "2026-10-07T07:28:57+00:00",
 "methodology": "v1.1 (2026-08-10)",
 "methodology_hash": "e66c7e8c864a2233",
 "source": "weekly_metrics_2026-09-28.json",
 "hit_rule": "direction: exit tp1/tp2 -> hit, sl -> miss, expiry -> sign of gross pnl",
 "bibtex_key": "mm_modelwatch_2026w40",
 "research_question": "who led the week, which tickers were readable, who saw the turn first, and does a model agreeing with itself across timeframes/horizons make its call stronger?",
 "base_field_hit": {
  "n": 4331,
  "hit_rate": 0.4964
 },
 "market_state": {
  "panel": "market_state_6p2",
  "week": {
   "start": "2026-09-28",
   "end": "2026-10-05"
  },
  "prev_week": {
   "start": "2026-09-21",
   "end": "2026-09-28"
  },
  "current": {
   "btc_net_pct": 2.44,
   "realized_vol_ann_pct": 21.13,
   "volume_top5_usd_bn": 17.9648,
   "volume_top5_wow_pct": -19.9,
   "avg_pairwise_corr": 0.57
  },
  "previous_as_published_issue8": {
   "btc_net_pct": 4.06,
   "realized_vol_ann_pct": 52.25,
   "volume_top5_usd_bn": 22.4265,
   "avg_pairwise_corr": 0.71
  },
  "snapshot_line": "BTC net +2.44% (prior +4.06%) - ann vol 21.1% (was 52.2%) - TOP5 volume $17.96B, -20% w/w - pairwise corr 0.57 (was 0.71)",
  "exchange": "binance",
  "audit_filter_applied": true,
  "source": "single-exchange (binance) daily candle recompute, market_state_2026-09-28.json; prior-week values as published in issue #8",
  "audit_note": "Market-state row is computed on a single-exchange (binance) daily candle series, as in issue #8 after the audit that found the raw table mixes two exchanges; issue #8 values are as published.",
  "prev_week_control": {
   "published_issue3": {
    "btc_net_pct": 4.06,
    "realized_vol_ann_pct": 52.25,
    "volume_top5_usd_bn": 22.4265,
    "avg_pairwise_corr": 0.71
   },
   "published_issue8": {
    "btc_net_pct": 4.06,
    "realized_vol_ann_pct": 52.25,
    "volume_top5_usd_bn": 22.4265,
    "avg_pairwise_corr": 0.71
   },
   "got": {
    "btc_net_pct": 4.06,
    "realized_vol_ann_pct": 52.25,
    "volume_top5_usd_bn": 22.4265,
    "avg_pairwise_corr": 0.71
   },
   "match": {
    "btc_net_pct": true,
    "realized_vol_ann_pct": true,
    "volume_top5_usd_bn": true,
    "avg_pairwise_corr": true
   },
   "tolerance": {
    "btc_net_pct": 0.05,
    "realized_vol_ann_pct": 0.5,
    "volume_top5_usd_bn": 0.05,
    "avg_pairwise_corr": 0.02
   },
   "all_match": true,
   "note": "previous week of this run == published issue-#8 week; the four Snapshot pins must reproduce (key published_issue3 kept for the builder schema, published_issue8 is the add-only alias)"
  },
  "regime_days": {
   "2026-09-28": {
    "move_pct": -1.19,
    "regime": "flat"
   },
   "2026-09-29": {
    "move_pct": 0.22,
    "regime": "flat"
   },
   "2026-09-30": {
    "move_pct": -0.22,
    "regime": "flat"
   },
   "2026-10-01": {
    "move_pct": 1.54,
    "regime": "trend"
   },
   "2026-10-02": {
    "move_pct": -0.39,
    "regime": "flat"
   },
   "2026-10-03": {
    "move_pct": 0.4,
    "regime": "flat"
   },
   "2026-10-04": {
    "move_pct": 1.81,
    "regime": "trend"
   }
  },
  "alive_slice_context": {
   "abs_move_1d_pct": {
    "A": 1.835,
    "B": 0.873
   },
   "btc_realized_vol_ann_pct_hourly": {
    "A": 35.8,
    "B": 32.24
   }
  }
 },
 "key_finding": {
  "label": "KEY FINDING · OBSERVATION (one weekly window)",
  "text": "Field accuracy went 46.8% -> 49.6% (+2.8pp) with 5 of 7 models improving, and no weekly title was awarded, for the ninth issue running.",
  "evidence_level": "observation"
 },
 "tables": {
  "leaderboard": {
   "rows": [
    {
     "model": "grok-4.6",
     "n": 431,
     "hit_rate": 0.5267,
     "wilson_95": [
      0.4795,
      0.5734
     ],
     "coverage": 0.326,
     "model_label": "grok-4.6 (incl. 4.5-era, Aug 24 00:00-09:22 UTC)",
     "insufficient": false
    },
    {
     "model": "deepseek-v4-pro",
     "n": 687,
     "hit_rate": 0.5095,
     "wilson_95": [
      0.4721,
      0.5467
     ],
     "coverage": 0.5165,
     "model_label": "deepseek-v4-pro",
     "insufficient": false
    },
    {
     "model": "claude-opus-5",
     "n": 534,
     "hit_rate": 0.5075,
     "wilson_95": [
      0.4652,
      0.5497
     ],
     "coverage": 0.4015,
     "model_label": "claude-opus-5",
     "insufficient": false
    },
    {
     "model": "claude-fable-5",
     "n": 623,
     "hit_rate": 0.4992,
     "wilson_95": [
      0.4601,
      0.5383
     ],
     "coverage": 0.4684,
     "model_label": "claude-fable-5",
     "insufficient": false
    },
    {
     "model": "gemini-3.1-pro",
     "n": 694,
     "hit_rate": 0.4899,
     "wilson_95": [
      0.4529,
      0.5271
     ],
     "coverage": 0.5218,
     "model_label": "gemini-3.1-pro",
     "insufficient": false
    },
    {
     "model": "qwen-3.8-max",
     "n": 711,
     "hit_rate": 0.4895,
     "wilson_95": [
      0.4529,
      0.5262
     ],
     "coverage": 0.535,
     "model_label": "qwen-3.8-max",
     "insufficient": false
    },
    {
     "model": "gpt-5.6-sol",
     "n": 651,
     "hit_rate": 0.4654,
     "wilson_95": [
      0.4274,
      0.5038
     ],
     "coverage": 0.4951,
     "model_label": "gpt-5.6-sol",
     "insufficient": false
    }
   ],
   "title_awarded": false,
   "title_rule": ">=5pp gap AND non-overlapping 95% CI, N>=10",
   "top2_gap_pp": 1.7,
   "movement_vs_prior_week": [
    {
     "model": "grok-4.6",
     "hit_rate_issue8_published": 0.4693,
     "hit_rate_issue9": 0.5267,
     "delta_pp": 5.7,
     "rank_issue9": 1
    },
    {
     "model": "deepseek-v4-pro",
     "hit_rate_issue8_published": 0.4389,
     "hit_rate_issue9": 0.5095,
     "delta_pp": 7.1,
     "rank_issue9": 2
    },
    {
     "model": "claude-opus-5",
     "hit_rate_issue8_published": 0.4758,
     "hit_rate_issue9": 0.5075,
     "delta_pp": 3.2,
     "rank_issue9": 3
    },
    {
     "model": "claude-fable-5",
     "hit_rate_issue8_published": 0.4674,
     "hit_rate_issue9": 0.4992,
     "delta_pp": 3.2,
     "rank_issue9": 4
    },
    {
     "model": "gemini-3.1-pro",
     "hit_rate_issue8_published": 0.5139,
     "hit_rate_issue9": 0.4899,
     "delta_pp": -2.4,
     "rank_issue9": 5
    },
    {
     "model": "qwen-3.8-max",
     "hit_rate_issue8_published": 0.4453,
     "hit_rate_issue9": 0.4895,
     "delta_pp": 4.4,
     "rank_issue9": 6
    },
    {
     "model": "gpt-5.6-sol",
     "hit_rate_issue8_published": 0.4677,
     "hit_rate_issue9": 0.4654,
     "delta_pp": -0.2,
     "rank_issue9": 7
    }
   ],
   "note": "No weekly title is awarded this issue -- for the ninth time. Title rule: **>=5pp gap AND non-overlapping 95% CI, N>=10** per side. grok-4.6's lead over deepseek-v4-pro is 1.7pp and their CIs overlap, so neither condition is met."
  },
  "ticker": {
   "rows": [
    {
     "symbol": "BTC",
     "n": 944,
     "hit_rate": 0.5328,
     "wilson_95": [
      0.5009,
      0.5645
     ],
     "issue8_hit_rate": 0.4886
    },
    {
     "symbol": "ETH",
     "n": 786,
     "hit_rate": 0.5076,
     "wilson_95": [
      0.4727,
      0.5425
     ],
     "issue8_hit_rate": 0.4579
    },
    {
     "symbol": "XRP",
     "n": 806,
     "hit_rate": 0.5025,
     "wilson_95": [
      0.468,
      0.5369
     ],
     "issue8_hit_rate": 0.4996
    },
    {
     "symbol": "BNB",
     "n": 899,
     "hit_rate": 0.4794,
     "wilson_95": [
      0.4469,
      0.5121
     ],
     "issue8_hit_rate": 0.3821
    },
    {
     "symbol": "SOL",
     "n": 896,
     "hit_rate": 0.4598,
     "wilson_95": [
      0.4274,
      0.4926
     ],
     "issue8_hit_rate": 0.4874
    }
   ],
   "best": "BTC",
   "hardest": "SOL",
   "all_above_51pct": false
  },
  "pairs_top": [
   {
    "model": "grok-4.6",
    "symbol": "BTC",
    "n": 93,
    "hit_rate": 0.5806
   },
   {
    "model": "qwen-3.8-max",
    "symbol": "BTC",
    "n": 145,
    "hit_rate": 0.5655
   },
   {
    "model": "claude-fable-5",
    "symbol": "BTC",
    "n": 131,
    "hit_rate": 0.5573
   },
   {
    "model": "grok-4.6",
    "symbol": "ETH",
    "n": 73,
    "hit_rate": 0.5479
   },
   {
    "model": "grok-4.6",
    "symbol": "XRP",
    "n": 81,
    "hit_rate": 0.5432
   },
   {
    "model": "deepseek-v4-pro",
    "symbol": "XRP",
    "n": 125,
    "hit_rate": 0.536
   }
  ],
  "pairs_worst": [
   {
    "model": "qwen-3.8-max",
    "symbol": "BNB",
    "n": 150,
    "hit_rate": 0.4667
   },
   {
    "model": "gpt-5.6-sol",
    "symbol": "BNB",
    "n": 136,
    "hit_rate": 0.4485
   },
   {
    "model": "qwen-3.8-max",
    "symbol": "SOL",
    "n": 144,
    "hit_rate": 0.4444
   },
   {
    "model": "claude-fable-5",
    "symbol": "SOL",
    "n": 135,
    "hit_rate": 0.4296
   },
   {
    "model": "gpt-5.6-sol",
    "symbol": "SOL",
    "n": 130,
    "hit_rate": 0.4154
   }
  ],
  "reversal_events": {
   "definition": "4h grid, trend=sign of prior 24h, counter-move >=2.0% (BTC/ETH) sustained 8h; caller: directional hit, FH 4h/1d, slot in [T-12h, T+FH]",
   "events": [
    {
     "symbol": "BTC",
     "t": "2026-10-02T12:00:00+00:00",
     "new_side": "short",
     "move_8h_pct": -2.28,
     "n_callers": 0,
     "first_caller": null,
     "caller_models": []
    },
    {
     "symbol": "ETH",
     "t": "2026-10-02T12:00:00+00:00",
     "new_side": "short",
     "move_8h_pct": -2.75,
     "n_callers": 0,
     "first_caller": null,
     "caller_models": []
    }
   ],
   "scope_v1": "BTC/ETH only",
   "status": "ok"
  },
  "cross_tf": {
   "fh_4h_tf_4h_vs_1h": {
    "agree": {
     "n": 442,
     "hit_rate": 0.491,
     "wilson_95": [
      0.4446,
      0.5374
     ],
     "by_regime": {
      "trend": {
       "n": 121,
       "hit_rate": 0.6694
      },
      "flat": {
       "n": 321,
       "hit_rate": 0.4237
      }
     }
    },
    "disagree": {
     "n": 5,
     "hit_rate": 0.0,
     "wilson_95": [
      -0.0,
      0.4345
     ],
     "by_regime": {
      "trend": {
       "n": 0,
       "hit_rate": null
      },
      "flat": {
       "n": 5,
       "hit_rate": 0.0
      }
     }
    },
    "mixed_pairs": 100,
    "base": {
     "n": 548,
     "hit_rate": 0.4872
    },
    "lift_pp": 0.4
   },
   "fh_1d_tf_1d_vs_4h": {
    "agree": {
     "n": 33,
     "hit_rate": 0.6364,
     "wilson_95": [
      0.4662,
      0.7781
     ],
     "by_regime": {
      "trend": {
       "n": 10,
       "hit_rate": 1.0
      },
      "flat": {
       "n": 23,
       "hit_rate": 0.4783
      }
     }
    },
    "disagree": {
     "n": 1,
     "hit_rate": 1.0,
     "wilson_95": [
      0.2065,
      1.0
     ],
     "by_regime": {
      "trend": {
       "n": 0,
       "hit_rate": null
      },
      "flat": {
       "n": 1,
       "hit_rate": 1.0
      }
     }
    },
    "mixed_pairs": 62,
    "base": {
     "n": 96,
     "hit_rate": 0.5729
    },
    "lift_pp": 6.3
   }
  },
  "cross_fh": {
   "fh_4h_confirmed_by_1h": {
    "agree": {
     "n": 404,
     "hit_rate": 0.5322,
     "wilson_95": [
      0.4834,
      0.5803
     ],
     "by_regime": {
      "trend": {
       "n": 115,
       "hit_rate": 0.7217
      },
      "flat": {
       "n": 289,
       "hit_rate": 0.4567
      }
     }
    },
    "disagree": {
     "n": 14,
     "hit_rate": 0.2143,
     "wilson_95": [
      0.0757,
      0.4759
     ],
     "by_regime": {
      "trend": {
       "n": 1,
       "hit_rate": 1.0
      },
      "flat": {
       "n": 13,
       "hit_rate": 0.1538
      }
     }
    },
    "mixed_pairs": 129,
    "base": {
     "n": 548,
     "hit_rate": 0.4872
    },
    "lift_pp": 4.5
   },
   "fh_1d_confirmed_by_4h": {
    "agree": {
     "n": 25,
     "hit_rate": 0.76,
     "wilson_95": [
      0.5657,
      0.885
     ],
     "by_regime": {
      "trend": {
       "n": 10,
       "hit_rate": 1.0
      },
      "flat": {
       "n": 15,
       "hit_rate": 0.6
      }
     }
    },
    "disagree": {
     "n": 1,
     "hit_rate": 1.0,
     "wilson_95": [
      0.2065,
      1.0
     ],
     "by_regime": {
      "trend": {
       "n": 0,
       "hit_rate": null
      },
      "flat": {
       "n": 1,
       "hit_rate": 1.0
      }
     }
    },
    "mixed_pairs": 70,
    "base": {
     "n": 96,
     "hit_rate": 0.5729
    },
    "lift_pp": 18.7
   }
  },
  "regime_days": {
   "2026-09-28": {
    "move_pct": -1.19,
    "regime": "flat"
   },
   "2026-09-29": {
    "move_pct": 0.22,
    "regime": "flat"
   },
   "2026-09-30": {
    "move_pct": -0.22,
    "regime": "flat"
   },
   "2026-10-01": {
    "move_pct": 1.54,
    "regime": "trend"
   },
   "2026-10-02": {
    "move_pct": -0.39,
    "regime": "flat"
   },
   "2026-10-03": {
    "move_pct": 0.4,
    "regime": "flat"
   },
   "2026-10-04": {
    "move_pct": 1.81,
    "regime": "trend"
   }
  },
  "side_mix": [
   {
    "model": "claude-fable-5",
    "n": 1330,
    "long": 0.3203,
    "short": 0.1481,
    "sideways": 0.5316,
    "ls_ratio": 2.16
   },
   {
    "model": "claude-opus-5",
    "n": 1330,
    "long": 0.2737,
    "short": 0.1278,
    "sideways": 0.5985,
    "ls_ratio": 2.14
   },
   {
    "model": "deepseek-v4-pro",
    "n": 1330,
    "long": 0.3113,
    "short": 0.2053,
    "sideways": 0.4835,
    "ls_ratio": 1.52
   },
   {
    "model": "gemini-3.1-pro",
    "n": 1330,
    "long": 0.3195,
    "short": 0.2023,
    "sideways": 0.4782,
    "ls_ratio": 1.58
   },
   {
    "model": "gpt-5.6-sol",
    "n": 1315,
    "long": 0.2837,
    "short": 0.2114,
    "sideways": 0.5049,
    "ls_ratio": 1.34
   },
   {
    "model": "grok-4.6",
    "n": 1322,
    "long": 0.1906,
    "short": 0.1354,
    "sideways": 0.674,
    "ls_ratio": 1.41
   },
   {
    "model": "qwen-3.8-max",
    "n": 1329,
    "long": 0.3439,
    "short": 0.1911,
    "sideways": 0.465,
    "ls_ratio": 1.8
   }
  ],
  "long_short_ratio_field": {
   "value": 1.71,
   "method": "mean of the per-model long-share / short-share ratios"
  },
  "consensus_skew": {
   "cells_ge5_same_side": 501,
   "n_mature": 3247,
   "hit_rate": 0.5112
  },
  "fail_of_week": {
   "model": "gemini-3.1-pro",
   "symbol": "XRP",
   "side": "long",
   "confidence": 85,
   "slot": "2026-10-04T08:01:00+00:00",
   "fh": "1h",
   "tf": "1h",
   "exit_reason": "expiry",
   "pnl_net_usd": -0.29
  },
  "counters": {
   "forecasts_total": 10290,
   "ok_in_gate": 9286,
   "out_of_gate": 976,
   "out_of_gate_by_fh": {
    "1w": 490,
    "1m": 486
   },
   "invalid": 28,
   "mature": 9286,
   "mature_directional": 4331,
   "mature_sideways": 4955,
   "pending_next_issue": 0,
   "pending_by_fh": {},
   "late_closes": 0,
   "uptime": [
    {
     "fh": "1h",
     "tf": "1h",
     "slots_seen": 168,
     "slots_expected": 168
    },
    {
     "fh": "4h",
     "tf": "4h",
     "slots_seen": 42,
     "slots_expected": 42
    },
    {
     "fh": "4h",
     "tf": "1h",
     "slots_seen": 42,
     "slots_expected": 42
    },
    {
     "fh": "1d",
     "tf": "1d",
     "slots_seen": 7,
     "slots_expected": 7
    },
    {
     "fh": "1d",
     "tf": "4h",
     "slots_seen": 7,
     "slots_expected": 7
    }
   ],
   "regime_days": {
    "2026-09-28": {
     "move_pct": -1.19,
     "regime": "flat"
    },
    "2026-09-29": {
     "move_pct": 0.22,
     "regime": "flat"
    },
    "2026-09-30": {
     "move_pct": -0.22,
     "regime": "flat"
    },
    "2026-10-01": {
     "move_pct": 1.54,
     "regime": "trend"
    },
    "2026-10-02": {
     "move_pct": -0.39,
     "regime": "flat"
    },
    "2026-10-03": {
     "move_pct": 0.4,
     "regime": "flat"
    },
    "2026-10-04": {
     "move_pct": 1.81,
     "regime": "trend"
    }
   },
   "grok_era": {
    "ids_in_window": [
     "grok-4.6"
    ],
    "n_grok45": 0,
    "n_grok46": 1470,
    "n_grok45_rows": 0,
    "n_grok46_rows": 1470,
    "first_grok46_slot": "2026-09-28T00:01:00+00:00",
    "last_grok45_slot": null,
    "first_grok45_slot": null,
    "flip_at": "2026-08-24T09:22:00+00:00",
    "flip_inside_window": false,
    "prev_window": {
     "n_grok45_rows": 0,
     "n_grok46_rows": 1470
    },
    "merged_model_id": "grok-4.6",
    "merged_label": "grok-4.6 (incl. 4.5-era, Aug 24 00:00-09:22 UTC)",
    "lineage_ids": [
     "grok-4.5",
     "grok-4.6"
    ],
    "lineage_source": "benchmarks.config.lineage_ids('grok-4.6') + MODEL_SUCCESSION[grok-4.6]=('grok-4.5',)",
    "lineage_warnings": [],
    "rows_remapped_a_plus_b": 0,
    "note": "grok-4.6 went live 2026-08-24T09:22:00+00:00 -- five windows back (inside issue #4's window); all per-model aggregates use the lineage splice grok-4.5 -> grok-4.6 (one row, labelled 'grok-4.6 (incl. 4.5-era, Aug 24 00:00-09:22 UTC)'). Both week A (prev window, issue #8) and week B (this window) are pure grok-4.6.",
    "cells_with_two_grok_rows": 0
   },
   "series_density": {
    "1d": {
     "days_with_slots": 7,
     "expected_days": 7,
     "full_daily_coverage": true,
     "in_fh_gate": true,
     "n_slots": 7,
     "n_rows": 490,
     "days": [
      "2026-09-28",
      "2026-09-29",
      "2026-09-30",
      "2026-10-01",
      "2026-10-02",
      "2026-10-03",
      "2026-10-04"
     ],
     "missing_days": []
    },
    "1h": {
     "days_with_slots": 7,
     "expected_days": 7,
     "full_daily_coverage": true,
     "in_fh_gate": true,
     "n_slots": 168,
     "n_rows": 5880,
     "days": [
      "2026-09-28",
      "2026-09-29",
      "2026-09-30",
      "2026-10-01",
      "2026-10-02",
      "2026-10-03",
      "2026-10-04"
     ],
     "missing_days": []
    },
    "1m": {
     "days_with_slots": 7,
     "expected_days": 7,
     "full_daily_coverage": true,
     "in_fh_gate": false,
     "n_slots": 7,
     "n_rows": 490,
     "days": [
      "2026-09-28",
      "2026-09-29",
      "2026-09-30",
      "2026-10-01",
      "2026-10-02",
      "2026-10-03",
      "2026-10-04"
     ],
     "missing_days": []
    },
    "1w": {
     "days_with_slots": 7,
     "expected_days": 7,
     "full_daily_coverage": true,
     "in_fh_gate": false,
     "n_slots": 7,
     "n_rows": 490,
     "days": [
      "2026-09-28",
      "2026-09-29",
      "2026-09-30",
      "2026-10-01",
      "2026-10-02",
      "2026-10-03",
      "2026-10-04"
     ],
     "missing_days": []
    },
    "4h": {
     "days_with_slots": 7,
     "expected_days": 7,
     "full_daily_coverage": true,
     "in_fh_gate": true,
     "n_slots": 42,
     "n_rows": 2940,
     "days": [
      "2026-09-28",
      "2026-09-29",
      "2026-09-30",
      "2026-10-01",
      "2026-10-02",
      "2026-10-03",
      "2026-10-04"
     ],
     "missing_days": []
    }
   },
   "model_roster": {
    "current": [
     "claude-fable-5",
     "claude-opus-5",
     "deepseek-v4-pro",
     "gemini-3.1-pro",
     "gpt-5.6-sol",
     "grok-4.6",
     "qwen-3.8-max"
    ],
    "current_count": 7,
    "current_raw_ids_in_window": [
     "claude-fable-5",
     "claude-opus-5",
     "deepseek-v4-pro",
     "gemini-3.1-pro",
     "gpt-5.6-sol",
     "grok-4.6",
     "qwen-3.8-max"
    ],
    "archived_legacy": [
     "grok-4.5",
     "qwen-3.7-max",
     "claude-opus-4.8"
    ],
    "archived_legacy_count": 3,
    "archived_legacy_detail": [
     {
      "model": "grok-4.5",
      "n_rows_since": 11128,
      "first_slot": "2026-07-15T18:01:00+00:00",
      "last_slot": "2026-08-24T09:01:00+00:00"
     },
     {
      "model": "qwen-3.7-max",
      "n_rows_since": 7479,
      "first_slot": "2026-07-15T18:01:00+00:00",
      "last_slot": "2026-08-05T09:01:00+00:00"
     },
     {
      "model": "claude-opus-4.8",
      "n_rows_since": 6315,
      "first_slot": "2026-07-15T18:01:00+00:00",
      "last_slot": "2026-07-30T08:01:00+00:00"
     }
    ],
    "archived_excluding_lineage_merged": [
     "qwen-3.7-max",
     "claude-opus-4.8"
    ],
    "tracked_total": 10,
    "canon_expected": {
     "current": 7,
     "archived_legacy": 4,
     "tracked_total": 11
    },
    "matches_canon_11_7_4": false,
    "roster_since": "2026-07-11T00:00:00+00:00",
    "note": "current = merged-lineage model ids emitting inside the window (grok-4.5 rows are spliced into grok-4.6); archived legacy = raw ids seen since 2026-07-11 that no longer emit in the window -- grok-4.5 is one of them by raw id even though its rows are spliced. Models retired before 2026-07-11 are not visible to this query."
   }
  },
  "uptime": [
   {
    "fh": "1h",
    "tf": "1h",
    "slots_seen": 168,
    "slots_expected": 168
   },
   {
    "fh": "4h",
    "tf": "4h",
    "slots_seen": 42,
    "slots_expected": 42
   },
   {
    "fh": "4h",
    "tf": "1h",
    "slots_seen": 42,
    "slots_expected": 42
   },
   {
    "fh": "1d",
    "tf": "1d",
    "slots_seen": 7,
    "slots_expected": 7
   },
   {
    "fh": "1d",
    "tf": "4h",
    "slots_seen": 7,
    "slots_expected": 7
   }
  ],
  "ok_rate_by_model": {
   "grok-4.6": 0.9946,
   "deepseek-v4-pro": 1.0,
   "qwen-3.8-max": 0.9966,
   "claude-opus-5": 1.0,
   "claude-fable-5": 1.0,
   "gemini-3.1-pro": 1.0,
   "gpt-5.6-sol": 0.9898
  },
  "calibration_bridge": {
   "best_calibrated_model": "grok-4.6",
   "brier": 0.2566,
   "hit_rate_leader": "grok-4.6"
  }
 },
 "key_findings": [
  "**OBSERVATION -- No weekly title again.** grok-4.6 leads descriptively at 52.7% [47.9%, 57.3%]; the gap to #2 (deepseek-v4-pro, 50.9%) is 1.7pp. 5 of 7 models improved week over week; issue #8's leader gemini-3.1-pro is now 5 of 7.",
  "**1 of 5 tickers were readable.** BTC is the week's most readable coin (53.3%, n=944) and SOL the hardest at 46.0%; 1 of the 5 assets cleared 51%. Pair of the week: grok-4.6 x BTC 58.1% (n=93); worst: gpt-5.6-sol x SOL 41.5% (n=130).",
  "**0 of 2 qualifying turns found a caller.** BTC short -2.28%, 0 caller(s), first NOBODY; ETH short -2.75%, 0 caller(s), first NOBODY.",
  "**Self-agreement read as a lift in every cell.** Cross-TF/FH lifts ran +0.4pp to +18.7pp (issue #8: -2.5pp to +3.8pp). On trend days, 4h TF-agreement hit 66.9% (n=121) against 42.4% (n=321) on flat days."
 ],
 "market_check": {
  "label": "MARKET CHECK · OBSERVATION (two adjacent calendar weeks)",
  "text": "OBSERVATION — mean |1d move| 1.83% -> 0.87% (-52% rel), BTC realized vol 35.8% -> 32.2% (ann., hourly); field directional accuracy 46.8% -> 49.6% (+2.8pp), 5 of 7 models improved, sim win-rate up for 5 of 7, field sim PnL -$315.68 -> -$408.24 (adjacent calendar weeks Sep 21-27 vs Sep 28-Oct 4; descriptive, one pair of weeks, not a claim).",
  "robustness": "Robustness: raw price-sign accuracy 49.2% -> 49.0% — the opposite direction to the trade-based hit rule.",
  "weeks": {
   "A": {
    "days": "2026-09-21..2026-09-27",
    "slots": [
     "2026-09-21T00:00:00+00:00",
     "2026-09-28T00:00:00+00:00"
    ],
    "cutoff": "2026-09-28T16:00:00+00:00"
   },
   "B": {
    "days": "2026-09-28..2026-10-04",
    "slots": [
     "2026-09-28T00:00:00+00:00",
     "2026-10-05T00:00:00+00:00"
    ],
    "cutoff": "2026-10-05T16:00:00+00:00"
   }
  },
  "price_source": "crypto_spot/1h@:00",
  "accuracy_field": {
   "A": {
    "n": 4958,
    "hit_rate": 0.4683,
    "wilson_95": [
     0.4545,
     0.4822
    ]
   },
   "B": {
    "n": 4331,
    "hit_rate": 0.4964,
    "wilson_95": [
     0.4815,
     0.5113
    ]
   },
   "delta_pp": 2.8
  },
  "accuracy_per_model": [
   {
    "model": "deepseek-v4-pro",
    "model_label": "deepseek-v4-pro",
    "A": {
     "n": 777,
     "hit_rate": 0.4389,
     "wilson_95": [
      0.4044,
      0.474
     ]
    },
    "B": {
     "n": 687,
     "hit_rate": 0.5095,
     "wilson_95": [
      0.4721,
      0.5467
     ]
    },
    "delta_pp": 7.1
   },
   {
    "model": "grok-4.6",
    "model_label": "grok-4.6 (incl. 4.5-era, Aug 24 00:00-09:22 UTC)",
    "A": {
     "n": 537,
     "hit_rate": 0.4693,
     "wilson_95": [
      0.4274,
      0.5116
     ]
    },
    "B": {
     "n": 431,
     "hit_rate": 0.5267,
     "wilson_95": [
      0.4795,
      0.5734
     ]
    },
    "delta_pp": 5.7
   },
   {
    "model": "qwen-3.8-max",
    "model_label": "qwen-3.8-max",
    "A": {
     "n": 750,
     "hit_rate": 0.4453,
     "wilson_95": [
      0.4101,
      0.4811
     ]
    },
    "B": {
     "n": 711,
     "hit_rate": 0.4895,
     "wilson_95": [
      0.4529,
      0.5262
     ]
    },
    "delta_pp": 4.4
   },
   {
    "model": "claude-fable-5",
    "model_label": "claude-fable-5",
    "A": {
     "n": 736,
     "hit_rate": 0.4674,
     "wilson_95": [
      0.4316,
      0.5035
     ]
    },
    "B": {
     "n": 623,
     "hit_rate": 0.4992,
     "wilson_95": [
      0.4601,
      0.5383
     ]
    },
    "delta_pp": 3.2
   },
   {
    "model": "claude-opus-5",
    "model_label": "claude-opus-5",
    "A": {
     "n": 620,
     "hit_rate": 0.4758,
     "wilson_95": [
      0.4368,
      0.5151
     ]
    },
    "B": {
     "n": 534,
     "hit_rate": 0.5075,
     "wilson_95": [
      0.4652,
      0.5497
     ]
    },
    "delta_pp": 3.2
   },
   {
    "model": "gpt-5.6-sol",
    "model_label": "gpt-5.6-sol",
    "A": {
     "n": 744,
     "hit_rate": 0.4677,
     "wilson_95": [
      0.4321,
      0.5037
     ]
    },
    "B": {
     "n": 651,
     "hit_rate": 0.4654,
     "wilson_95": [
      0.4274,
      0.5038
     ]
    },
    "delta_pp": -0.2
   },
   {
    "model": "gemini-3.1-pro",
    "model_label": "gemini-3.1-pro",
    "A": {
     "n": 794,
     "hit_rate": 0.5139,
     "wilson_95": [
      0.4791,
      0.5485
     ]
    },
    "B": {
     "n": 694,
     "hit_rate": 0.4899,
     "wilson_95": [
      0.4529,
      0.5271
     ]
    },
    "delta_pp": -2.4
   }
  ],
  "raw_sign_field": {
   "A": {
    "n_scored": 4954,
    "hit_rate": 0.4923
   },
   "B": {
    "n_scored": 4311,
    "hit_rate": 0.4901
   }
  },
  "abs_move_1d": {
   "A": 1.835,
   "B": 0.873,
   "delta_pct_rel": -52.4
  },
  "btc_realized_vol_ann_pct_hourly": {
   "A": 35.8,
   "B": 32.24
  },
  "trading_field": {
   "A": {
    "n_trades": 4958,
    "wr": 0.3998,
    "pnl_net_usd": -315.68,
    "pnl_gross_usd": 180.12
   },
   "B": {
    "n_trades": 4331,
    "wr": 0.4103,
    "pnl_net_usd": -408.24,
    "pnl_gross_usd": 24.86
   }
  },
  "verdict": {
   "models_with_hit_improved": "5 of 7",
   "models_with_rawsign_improved": "3 of 7",
   "models_with_wr_improved": "5 of 7",
   "field_hit_delta_pp": 2.8,
   "field_pnl_net": {
    "A": -315.68,
    "B": -408.24
   },
   "abs_move_delta": {
    "1h": {
     "A": 0.358,
     "B": 0.314,
     "delta_pct_rel": -12.3
    },
    "4h": {
     "A": 0.766,
     "B": 0.706,
     "delta_pct_rel": -7.8
    },
    "1d": {
     "A": 1.835,
     "B": 0.873,
     "delta_pct_rel": -52.4
    }
   },
   "btc_vol_ann_pct": {
    "A": 35.8,
    "B": 32.24
   },
   "note": "descriptive, two adjacent calendar weeks; hit rule = methodology v1.1 trade-based; raw_sign = price-sign robustness check (closes at :00 vs slots at :01)"
  },
  "lineage_note": "lineage splice ACTIVE: grok-4.5 rows are aggregated into grok-4.6 (flip 2026-08-24T09:22:00+00:00, five windows back, inside issue #4's window). Week A (2026-09-21..2026-09-27) is pure grok-4.6 (it reproduces the published issue #8); week B is pure grok-4.6."
 },
 "why_it_matters": "MarketMania scores 7 model lines against the same market, hour after hour. Weekly Model Watch asks four practical questions about that week of calls: who led the week, which tickers were readable, who saw the turn first, and does a model agreeing with itself across timeframes/horizons make its call stronger? Each question gets its own block below, with its own table, its own N, and its own honesty caveats. The grok row is a pure grok-4.6 week in this issue and was one in issue #8 as well -- the 4.5 -> 4.6 flip sits five windows back -- so read the grok week-over-week row as a line, not a single model build.",
 "how_to_read_this": "Every block below is scored on the same pool: matured directional calls inside the 1h/4h/1d FH gate, hit by the methodology v1.1 direction rule. The weekly title is awarded only on a >=5pp gap AND non-overlapping 95% Wilson CIs with N>=10 per side -- a descriptive lead is not a title. Reversals use the 4h grid with a >=2.0% counter-move sustained 8h, BTC/ETH only in v1. Cross-TF/FH blocks ask whether a model agreeing with itself does any better, and are split trend vs flat wherever the week supplies both sides. The grok 4.5 -> 4.6 flip (Aug 24, 2026 09:22 UTC) sits five windows back (issue #4): both weeks of this issue are pure grok-4.6, with no 4.5-era rows in either window, so the grok week-over-week row is the fourth pure-4.6 against pure-4.6 comparison of the series.",
 "practical_implications": [
  "Do not read this week's leaderboard as a skill ranking: no title was awarded and the top CIs overlap.",
  "Treat per-ticker readability as weekly weather. Nine issues, and the best ticker changed hands in every one except issue #6 (SOL -> BNB -> BTC -> SOL -> BNB -> BNB -> ETH -> XRP -> BTC): BTC leads this week at 53.3%, after one issue of XRP.",
  "Self-agreement lifts run +0.4pp to +18.7pp with small disagree cells; carry the regime split into the monthly test, do not act on it."
 ],
 "limitations": [
  "Single week (Sep 28-Oct 4, 2026 UTC). Descriptive for this window only; no claim about next week or any model's underlying skill. This window ran 2 trend days of 7 (issue #8: 2 of 7); every comparison with issue #8 is a comparison across regimes as well as across weeks.",
  "Observations inside one window are not independent, and the 95% Wilson CIs shown throughout are descriptive, not inferential.",
  "The reversal detector remains v1 and BTC/ETH-only, on price series reconstructed from trade entry prices. 17 events across nine issues cannot characterise detector performance either way.",
  "Cross-TF and cross-FH disagreement cells are small (n=5, 1, 14, 1); cells under N=10 are insufficient and never used to rank models. Week-over-week deltas are computed on unrounded rates and may differ by 0.1pp from the difference of the rounded columns.",
  "Model lines in this window: claude-fable-5, claude-opus-5, deepseek-v4-pro, gemini-3.1-pro, gpt-5.6-sol, grok-4.6, qwen-3.8-max. DeepSeek announced routing of V4 Pro to V4.1 Flash from Sep 14, 04:00 UTC; an API check on Oct 7 still returns deepseek-v4-pro as the served model for our requests, with the same system fingerprint as on Sep 29, so this issue keeps the label; it will change the day the served id changes. On Sep 21, from 04:01 UTC, the provider returned HTTP 402 (insufficient balance) to 20 deepseek-v4-pro calls inside week A (the issue-#8 window), counted as invalid there (invalid 44 in issue #8); in this window the line's ok-rate is 100.0% (invalid 28 this window). Series density and lineage notes (wave 9): (1) rolling-1w forecasts emit daily since Aug 22, 2026 and rolling-1M daily since Aug 23, 2026, so inside this window (Sep 28-Oct 4) daily coverage is FULL, as it was in issues #4, #5, #6, #7 and #8: the 1w series has slots on 7 of 7 days and the 1M series on 7 of 7 days; both sit outside this report's FH gate (1h/4h/1d) and appear only in the exclusions counter (1w 490, 1M 486). (2) The grok line flipped 4.5 -> 4.6 at Aug 24, 2026 09:22 UTC, five windows back (inside the issue-#4 window): this window carries 1,470 grok rows and none from the 4.5 era, and neither does week A (the issue-#8 window), so every grok number in this issue and in issue #8 is a pure grok-4.6 line -- the grok week-over-week row is the fourth pure-4.6 against pure-4.6 comparison of the series. (3) qwen-3.8-max succeeded qwen-3.7-max on Aug 5, 2026 (fully before this window).",
  "Market-state row is computed on a single-exchange (binance) daily candle series, as in issue #8 after the audit that found the raw table mixes two exchanges; issue #8 values are as published.",
  "Research-to-date counter: directional forecasts resolved inside the FH gate (1h/4h/1d) since Jul 11, counted at the issue cutoff (Oct 5 16:00 UTC) -- the definition pinned in issue #3 and carried by issues #4, #5, #6, #7 and #8, which printed 57,929 at the Sep 28 16:00 cutoff. The pack reproduces that pin at the previous cutoff (control OK), so this issue's 62,260 is an additive step under one definition; counter deltas against issue #7 and earlier remain definitional."
 ],
 "notes": {
  "series_density_and_lineage_wave9": "Series density and lineage notes (wave 9): (1) rolling-1w forecasts emit daily since Aug 22, 2026 and rolling-1M daily since Aug 23, 2026, so inside this window (Sep 28-Oct 4) daily coverage is FULL, as it was in issues #4, #5, #6, #7 and #8: the 1w series has slots on 7 of 7 days and the 1M series on 7 of 7 days; both sit outside this report's FH gate (1h/4h/1d) and appear only in the exclusions counter (1w 490, 1M 486). (2) The grok line flipped 4.5 -> 4.6 at Aug 24, 2026 09:22 UTC, five windows back (inside the issue-#4 window): this window carries 1,470 grok rows and none from the 4.5 era, and neither does week A (the issue-#8 window), so every grok number in this issue and in issue #8 is a pure grok-4.6 line -- the grok week-over-week row is the fourth pure-4.6 against pure-4.6 comparison of the series. (3) qwen-3.8-max succeeded qwen-3.7-max on Aug 5, 2026 (fully before this window).",
  "grok_era": {
   "ids_in_window": [
    "grok-4.6"
   ],
   "n_grok45": 0,
   "n_grok46": 1470,
   "n_grok45_rows": 0,
   "n_grok46_rows": 1470,
   "first_grok46_slot": "2026-09-28T00:01:00+00:00",
   "last_grok45_slot": null,
   "first_grok45_slot": null,
   "flip_at": "2026-08-24T09:22:00+00:00",
   "flip_inside_window": false,
   "prev_window": {
    "n_grok45_rows": 0,
    "n_grok46_rows": 1470
   },
   "merged_model_id": "grok-4.6",
   "merged_label": "grok-4.6 (incl. 4.5-era, Aug 24 00:00-09:22 UTC)",
   "lineage_ids": [
    "grok-4.5",
    "grok-4.6"
   ],
   "lineage_source": "benchmarks.config.lineage_ids('grok-4.6') + MODEL_SUCCESSION[grok-4.6]=('grok-4.5',)",
   "lineage_warnings": [],
   "rows_remapped_a_plus_b": 0,
   "note": "grok-4.6 went live 2026-08-24T09:22:00+00:00 -- five windows back (inside issue #4's window); all per-model aggregates use the lineage splice grok-4.5 -> grok-4.6 (one row, labelled 'grok-4.6 (incl. 4.5-era, Aug 24 00:00-09:22 UTC)'). Both week A (prev window, issue #8) and week B (this window) are pure grok-4.6.",
   "cells_with_two_grok_rows": 0
  },
  "model_ids_line": [
   "claude-fable-5",
   "claude-opus-5",
   "deepseek-v4-pro",
   "gemini-3.1-pro",
   "gpt-5.6-sol",
   "grok-4.6",
   "qwen-3.8-max"
  ],
  "model_roster": {
   "current": [
    "claude-fable-5",
    "claude-opus-5",
    "deepseek-v4-pro",
    "gemini-3.1-pro",
    "gpt-5.6-sol",
    "grok-4.6",
    "qwen-3.8-max"
   ],
   "current_count": 7,
   "current_raw_ids_in_window": [
    "claude-fable-5",
    "claude-opus-5",
    "deepseek-v4-pro",
    "gemini-3.1-pro",
    "gpt-5.6-sol",
    "grok-4.6",
    "qwen-3.8-max"
   ],
   "archived_legacy": [
    "grok-4.5",
    "qwen-3.7-max",
    "claude-opus-4.8"
   ],
   "archived_legacy_count": 3,
   "archived_legacy_detail": [
    {
     "model": "grok-4.5",
     "n_rows_since": 11128,
     "first_slot": "2026-07-15T18:01:00+00:00",
     "last_slot": "2026-08-24T09:01:00+00:00"
    },
    {
     "model": "qwen-3.7-max",
     "n_rows_since": 7479,
     "first_slot": "2026-07-15T18:01:00+00:00",
     "last_slot": "2026-08-05T09:01:00+00:00"
    },
    {
     "model": "claude-opus-4.8",
     "n_rows_since": 6315,
     "first_slot": "2026-07-15T18:01:00+00:00",
     "last_slot": "2026-07-30T08:01:00+00:00"
    }
   ],
   "archived_excluding_lineage_merged": [
    "qwen-3.7-max",
    "claude-opus-4.8"
   ],
   "tracked_total": 10,
   "canon_expected": {
    "current": 7,
    "archived_legacy": 4,
    "tracked_total": 11
   },
   "matches_canon_11_7_4": false,
   "roster_since": "2026-07-11T00:00:00+00:00",
   "note": "current = merged-lineage model ids emitting inside the window (grok-4.5 rows are spliced into grok-4.6); archived legacy = raw ids seen since 2026-07-11 that no longer emit in the window -- grok-4.5 is one of them by raw id even though its rows are spliced. Models retired before 2026-07-11 are not visible to this query."
  },
  "series_density": {
   "1d": {
    "days_with_slots": 7,
    "expected_days": 7,
    "full_daily_coverage": true,
    "in_fh_gate": true,
    "n_slots": 7,
    "n_rows": 490,
    "days": [
     "2026-09-28",
     "2026-09-29",
     "2026-09-30",
     "2026-10-01",
     "2026-10-02",
     "2026-10-03",
     "2026-10-04"
    ],
    "missing_days": []
   },
   "1h": {
    "days_with_slots": 7,
    "expected_days": 7,
    "full_daily_coverage": true,
    "in_fh_gate": true,
    "n_slots": 168,
    "n_rows": 5880,
    "days": [
     "2026-09-28",
     "2026-09-29",
     "2026-09-30",
     "2026-10-01",
     "2026-10-02",
     "2026-10-03",
     "2026-10-04"
    ],
    "missing_days": []
   },
   "1m": {
    "days_with_slots": 7,
    "expected_days": 7,
    "full_daily_coverage": true,
    "in_fh_gate": false,
    "n_slots": 7,
    "n_rows": 490,
    "days": [
     "2026-09-28",
     "2026-09-29",
     "2026-09-30",
     "2026-10-01",
     "2026-10-02",
     "2026-10-03",
     "2026-10-04"
    ],
    "missing_days": []
   },
   "1w": {
    "days_with_slots": 7,
    "expected_days": 7,
    "full_daily_coverage": true,
    "in_fh_gate": false,
    "n_slots": 7,
    "n_rows": 490,
    "days": [
     "2026-09-28",
     "2026-09-29",
     "2026-09-30",
     "2026-10-01",
     "2026-10-02",
     "2026-10-03",
     "2026-10-04"
    ],
    "missing_days": []
   },
   "4h": {
    "days_with_slots": 7,
    "expected_days": 7,
    "full_daily_coverage": true,
    "in_fh_gate": true,
    "n_slots": 42,
    "n_rows": 2940,
    "days": [
     "2026-09-28",
     "2026-09-29",
     "2026-09-30",
     "2026-10-01",
     "2026-10-02",
     "2026-10-03",
     "2026-10-04"
    ],
    "missing_days": []
   }
  },
  "week_over_week_note": "Week-over-week columns compare back-to-back windows: Sep 21-27 (issue #8, and week A of this issue's alive slice) vs Sep 28-Oct 4 (this issue). \"Was\" values are the numbers published in issue #8; deltas are computed on unrounded rates.",
  "research_to_date_pin": "Research-to-date counter: directional forecasts resolved inside the FH gate (1h/4h/1d) since Jul 11, counted at the issue cutoff (Oct 5 16:00 UTC) -- the definition pinned in issue #3 and carried by issues #4, #5, #6, #7 and #8, which printed 57,929 at the Sep 28 16:00 cutoff. The pack reproduces that pin at the previous cutoff (control OK), so this issue's 62,260 is an additive step under one definition; counter deltas against issue #7 and earlier remain definitional.",
  "market_state_audit": "Market-state row is computed on a single-exchange (binance) daily candle series, as in issue #8 after the audit that found the raw table mixes two exchanges; issue #8 values are as published.",
  "engine": "Engine 1.1 has powered the sandbox since Aug 18, i.e. before this window; no cross-engine PnL comparisons are claimed.",
  "report_template": "PDF built with the canonical mm_report.py template module; preview card with report_preview.render_preview"
 },
 "living_series_note": "**Issue #9.** Weekly Model Watch is a living series. Issue #8's open questions -- \"does gemini-3.1-pro hold the top rank for a fifth issue, and does the title rule ever fire?\" and \"Does a qualifying reversal fire outside ETH for a second issue running, and does a second turn in the same pair ever draw a caller?\" -- read this issue as: grok-4.6 at the top (52.7%), not the line that led issues #5 to #8 (gemini-3.1-pro), so the rank did not hold for a fifth issue; no weekly title was awarded, for the ninth issue running, and 0 of 2 qualifying turns found a caller, on turns in BTC / ETH; the BTC turn makes a second issue running with a qualifying turn outside ETH, and with one turn per pair the second-turn question stays open. The grok row is pure grok-4.6 this issue and was pure grok-4.6 in issue #8; the 4.5 -> 4.6 flip (Aug 24, 2026 09:22 UTC) sits five windows back. Engine 1.1 has powered the sandbox since Aug 18, i.e. before this window; no cross-engine PnL comparisons are claimed.",
 "testing_next": [
  "Next issue: does grok-4.6 hold the top rank for a second issue, and does the title rule ever fire?",
  "Does a qualifying reversal fire outside ETH for a third issue running, and does a second turn in the same pair ever draw a caller?",
  "Monthly series: cross-TF/FH lifts with regime control across mixed weeks (trend vs. flat), block bootstrap."
 ],
 "related_research": [
  {
   "title": "Consensus Watch #9",
   "url": "https://marketmania.ai/research/reports/consensus-watch-2026-09-28.pdf"
  },
  {
   "title": "Weekly Calibration #9",
   "url": "https://marketmania.ai/research/reports/weekly-calibration-2026-09-28.pdf"
  },
  {
   "title": "Weekly Model Watch #8",
   "url": "https://marketmania.ai/research/reports/model-watch-2026-09-21.pdf"
  }
 ],
 "related_research_note": "The three weekly reports publish together as one issue each week; each links straight to the others' PDF and to its own previous issue. Direct links are the posting rule from wave 2 on.",
 "research_to_date": {
  "this_report": {
   "scored_observations": 4331
  },
  "platform": {
   "as_of_cutoff": "2026-10-05",
   "resolved_forecasts": 62260,
   "resolved_forecasts_definition": "directional forecasts resolved inside the FH gate (1h/4h/1d) since Jul 11, at the issue cutoff",
   "resolved_forecasts_since": "2026-07-11",
   "models_tracked": 10,
   "models_tracked_detail": "7 models tracked (current line-up; earlier versions folded into their successors' lineage)",
   "assets": 5,
   "forecast_horizons": 5,
   "cadence": "hourly",
   "published_research_reports": 45
  },
  "source": "weekly_metrics_2026-09-28.json",
  "control_reproduces_wave3_pin": true,
  "pin_note": "Research-to-date counter: directional forecasts resolved inside the FH gate (1h/4h/1d) since Jul 11, counted at the issue cutoff (Oct 5 16:00 UTC) -- the definition pinned in issue #3 and carried by issues #4, #5, #6, #7 and #8, which printed 57,929 at the Sep 28 16:00 cutoff. The pack reproduces that pin at the previous cutoff (control OK), so this issue's 62,260 is an additive step under one definition; counter deltas against issue #7 and earlier remain definitional."
 },
 "citation": {
  "bibtex_key": "mm_modelwatch_2026w40",
  "title": "Weekly Model Watch #9: weekly leaderboard, ticker/pair reads and cross-confirmation, Sep 28-Oct 4 2026",
  "author": "MarketMania Research",
  "year": 2026,
  "month": "October",
  "day": 7,
  "url": "https://marketmania.ai/research/reports/model-watch-2026-09-28.pdf",
  "note": "Methodology v1.1; window Sep 28-Oct 4, 2026 UTC; source weekly_metrics_2026-09-28.json"
 },
 "site_entry": {
  "slug": "model-watch-2026-09-28",
  "series": "weekly",
  "seriesLabel": "Weekly · Model Watch",
  "date": "October 7, 2026",
  "title": "Weekly Model Watch #9",
  "summary": "4,331 directional calls, four questions: who led the week, which tickers were readable, who saw the turn first, does self-agreement help. No weekly title again — grok-4.6 led at 52.7% (+5.7pp w/w) against deepseek-v4-pro at 50.9%, a gap of 1.7pp. BTC was the most readable ticker at 53.3% and SOL the hardest at 46.0%, with 1 of 5 assets above 51 percent. Reversals: 0 of 2 qualifying turns found a caller. Field accuracy moved 46.8% -> 49.6% (+2.8pp) with 5 of 7 models improving.",
  "stats": [
   "4,331 directional calls",
   "No weekly title again",
   "field +2.8pp w/w"
  ],
  "manifest_key_finding": "Field accuracy moved 46.8% -> 49.6% (+2.8pp) with 5 of 7 models improving; grok-4.6 led the week at 52.7% (+5.7pp w/w) and no weekly title was awarded, for the ninth issue running.",
  "files": {
   "pdf": "/research/reports/model-watch-2026-09-28.pdf",
   "md": "/research/reports/model-watch-2026-09-28.md",
   "json": "/research/reports/model-watch-2026-09-28.json"
  }
 },
 "social": {
  "telegram": {
   "photo_caption_html": "📊 <b>Weekly Model Watch #9</b> — weekly series (window Sep 28–Oct 4)\n\nWho led, what was readable, who saw the turn, does self-agreement help? 4,331 directional calls, 7 model lines:\n• <b>No weekly title again</b> — grok-4.6 leads at <b>52.7%</b> (+5.7pp w/w) against deepseek-v4-pro at 50.9%, a 1.7pp gap\n• Tickers: <b>BTC</b> most readable (53.3%); SOL hardest at 46.0% — 1 of 5 assets above 51%\n• Reversals: 0 of 2 qualifying turns found a caller\n• Market check: field accuracy 46.8% -> 49.6% (+2.8pp), 5 of 7 improved; |1d| move 1.83% -> 0.87%\n\nNote: grok is a pure 4.6 line in both weeks of this issue; the 4.5 -> 4.6 flip (Aug 24 09:22 UTC) sits five windows back\n\n📄 <a href=\"https://marketmania.ai/research/reports/model-watch-2026-09-28.pdf\">Full report (PDF)</a> · <a href=\"https://marketmania.ai/research\">all research</a>\n\n#cryptotrading #LLM #AI #crypto #trading #models #research",
   "document_caption_html": "📄 Weekly Model Watch #9 — full report (PDF).\nWeb copy: <a href=\"https://marketmania.ai/research/reports/model-watch-2026-09-28.pdf\">model-watch-2026-09-28.pdf</a> · machine-readable: <a href=\"https://marketmania.ai/research/reports/model-watch-2026-09-28.json\">JSON</a> · <a href=\"https://marketmania.ai/research\">all research</a>\n\n#cryptotrading #LLM #AI #crypto #trading #models #research",
   "caption_len": 891,
   "document_caption_len": 390,
   "limit": 1024
  },
  "x": {
   "main": "Weekly Model Watch #9 is out.\n\n4,331 directional LLM calls: field accuracy 46.8% -> 49.6% w/w, 5 of 7 models improved. grok-4.6 led at 52.7%; BTC was the most readable ticker.\n\nFull report: https://marketmania.ai/research/reports/model-watch-2026-09-28.pdf\n\n#AI #Crypto #Research #LLM",
   "main_tco_len": 241,
   "reply": "Hit-rate, w/w:\ngrok-4.6: 46.9 -> 52.7\ndeepseek: 43.9 -> 50.9\nopus-5: 47.6 -> 50.7\nfable-5: 46.7 -> 49.9\ngemini-3.1: 51.4 -> 49.0\nqwen-3.8: 44.5 -> 48.9\ngpt-5.6: 46.8 -> 46.5\nField: 46.8 -> 49.6. Wilson CIs; methodology v1.1.\nData: https://marketmania.ai/research/reports/model-watch-2026-09-28.json",
   "reply_tco_len": 254,
   "limit_tco": 280,
   "replies_max": 1
  }
 }
}