{
 "slug": "consensus-watch-2026-09-14",
 "series": "weekly",
 "series_label": "WEEKLY · CONSENSUS",
 "issue": 7,
 "title": "Consensus Watch #7",
 "language": "en",
 "window_start": "2026-09-14T00:00:00+00:00",
 "window_end": "2026-09-21T00:00:00+00:00",
 "window_text": "Sep 14-20, 2026 UTC",
 "cutoff": "2026-09-21T16:00:00+00:00",
 "generated_at": "2026-09-22T07:27:22+00:00",
 "methodology": "v1.1 (2026-08-10)",
 "methodology_hash": "e66c7e8c864a2233",
 "source": "weekly_metrics_2026-09-14.json",
 "research_question": "if a model's forecast matches the consensus of the other models, is the forecast more reliable?",
 "method": "leave-one-out strict majority, >=3 directional peers",
 "base_field_hit": {
  "n": 5598,
  "hit_rate": 0.5075
 },
 "market_state": {
  "panel": "market_state_6p2",
  "week": {
   "start": "2026-09-14",
   "end": "2026-09-21"
  },
  "prev_week": {
   "start": "2026-09-07",
   "end": "2026-09-14"
  },
  "current": {
   "btc_net_pct": 5.64,
   "realized_vol_ann_pct": 51.01,
   "volume_top5_usd_bn": 18.7386,
   "volume_top5_wow_pct": 16.8,
   "avg_pairwise_corr": 0.93
  },
  "previous_as_published_issue6": {
   "btc_net_pct": -4.36,
   "realized_vol_ann_pct": 19.68,
   "volume_top5_usd_bn": 16.0492,
   "avg_pairwise_corr": 0.74
  },
  "snapshot_line": "BTC net +5.64% (prior -4.36%) - ann vol 51.0% (was 19.7%) - TOP5 volume $18.74B, +17% w/w - pairwise corr 0.93 (was 0.74)",
  "exchange": "binance",
  "audit_filter_applied": true,
  "source": "single-exchange (binance) daily candle recompute, market_state_2026-09-14.json; prior-week values as published in issue #6",
  "audit_note": "Market-state row is computed on a single-exchange (binance) daily candle series, as in issue #6 after the audit that found the raw table mixes two exchanges; issue #6 values are as published.",
  "prev_week_control": {
   "published_issue3": {
    "btc_net_pct": -4.36,
    "realized_vol_ann_pct": 19.68,
    "volume_top5_usd_bn": 16.0492,
    "avg_pairwise_corr": 0.74
   },
   "published_issue6": {
    "btc_net_pct": -4.36,
    "realized_vol_ann_pct": 19.68,
    "volume_top5_usd_bn": 16.0492,
    "avg_pairwise_corr": 0.74
   },
   "got": {
    "btc_net_pct": -4.36,
    "realized_vol_ann_pct": 19.68,
    "volume_top5_usd_bn": 16.0492,
    "avg_pairwise_corr": 0.74
   },
   "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-#6 week; the four Snapshot pins must reproduce (key published_issue3 kept for the builder schema, published_issue6 is the add-only alias)"
  },
  "regime_days": {
   "2026-09-14": {
    "move_pct": 1.79,
    "regime": "trend"
   },
   "2026-09-15": {
    "move_pct": -3.35,
    "regime": "trend"
   },
   "2026-09-16": {
    "move_pct": 0.86,
    "regime": "flat"
   },
   "2026-09-17": {
    "move_pct": 0.21,
    "regime": "flat"
   },
   "2026-09-18": {
    "move_pct": 5.82,
    "regime": "trend"
   },
   "2026-09-19": {
    "move_pct": 0.46,
    "regime": "flat"
   },
   "2026-09-20": {
    "move_pct": 0.13,
    "regime": "flat"
   }
  },
  "alive_slice_context": {
   "abs_move_1d_pct": {
    "A": 1.347,
    "B": 2.693
   },
   "btc_realized_vol_ann_pct_hourly": {
    "A": 31.4,
    "B": 35.07
   }
  }
 },
 "tables": {
  "pooled": {
   "agree": {
    "n": 4825,
    "hit_rate": 0.5059,
    "wilson_95": [
     0.4918,
     0.52
    ]
   },
   "disagree": {
    "n": 52,
    "hit_rate": 0.3846,
    "wilson_95": [
     0.2647,
     0.5204
    ]
   },
   "lift_pp": 12.1,
   "thin_or_tied": 721,
   "by_margin": {
    "2": {
     "n": 45,
     "hit_rate": 0.4222
    },
    "3": {
     "n": 424,
     "hit_rate": 0.5212
    },
    "4": {
     "n": 685,
     "hit_rate": 0.5547
    },
    "5": {
     "n": 990,
     "hit_rate": 0.4424
    },
    "6": {
     "n": 2681,
     "hit_rate": 0.5159
    }
   },
   "total_scoreable": 4877,
   "caveat": "Read with context. The disagree bucket is n=52 this week against 26 in issue #6 and 26 in issue #5; the two Wilson intervals overlap. This is a single week of dependent observations: not evidence about herding in general, and not a claim that the sign will hold next week."
  },
  "by_cell": [
   {
    "agree": {
     "n": 2952,
     "hit_rate": 0.5085,
     "wilson_95": [
      0.4904,
      0.5265
     ]
    },
    "disagree": {
     "n": 40,
     "hit_rate": 0.425,
     "wilson_95": [
      0.2851,
      0.5781
     ]
    },
    "lift_pp": 8.3,
    "thin_or_tied": 468,
    "by_margin": {
     "2": {
      "n": 33,
      "hit_rate": 0.303
     },
     "3": {
      "n": 312,
      "hit_rate": 0.5192
     },
     "4": {
      "n": 460,
      "hit_rate": 0.5696
     },
     "5": {
      "n": 600,
      "hit_rate": 0.44
     },
     "6": {
      "n": 1547,
      "hit_rate": 0.5191
     }
    },
    "n_subjects": 3460,
    "cell": "fh_1h_tf_1h",
    "fh": "1h",
    "tf": "1h"
   },
   {
    "agree": {
     "n": 785,
     "hit_rate": 0.5261,
     "wilson_95": [
      0.4911,
      0.5608
     ]
    },
    "disagree": {
     "n": 3,
     "hit_rate": 0.3333,
     "wilson_95": [
      0.0615,
      0.7923
     ]
    },
    "lift_pp": 19.3,
    "thin_or_tied": 110,
    "by_margin": {
     "2": {
      "n": 3,
      "hit_rate": 1.0
     },
     "3": {
      "n": 44,
      "hit_rate": 0.5455
     },
     "4": {
      "n": 80,
      "hit_rate": 0.575
     },
     "5": {
      "n": 168,
      "hit_rate": 0.4821
     },
     "6": {
      "n": 490,
      "hit_rate": 0.5286
     }
    },
    "n_subjects": 898,
    "cell": "fh_4h_tf_4h",
    "fh": "4h",
    "tf": "4h"
   },
   {
    "agree": {
     "n": 787,
     "hit_rate": 0.5413,
     "wilson_95": [
      0.5064,
      0.5758
     ]
    },
    "disagree": {
     "n": 8,
     "hit_rate": 0.25,
     "wilson_95": [
      0.0715,
      0.5907
     ]
    },
    "lift_pp": 29.1,
    "thin_or_tied": 113,
    "by_margin": {
     "2": {
      "n": 6,
      "hit_rate": 0.5
     },
     "3": {
      "n": 56,
      "hit_rate": 0.625
     },
     "4": {
      "n": 110,
      "hit_rate": 0.6545
     },
     "5": {
      "n": 174,
      "hit_rate": 0.5
     },
     "6": {
      "n": 441,
      "hit_rate": 0.5193
     }
    },
    "n_subjects": 908,
    "cell": "fh_4h_tf_1h",
    "fh": "4h",
    "tf": "1h"
   },
   {
    "agree": {
     "n": 146,
     "hit_rate": 0.3562,
     "wilson_95": [
      0.2831,
      0.4366
     ]
    },
    "disagree": {
     "n": 1,
     "hit_rate": 0.0,
     "wilson_95": [
      0.0,
      0.7935
     ]
    },
    "lift_pp": 35.6,
    "thin_or_tied": 19,
    "by_margin": {
     "2": {
      "n": 3,
      "hit_rate": 1.0
     },
     "3": {
      "n": 4,
      "hit_rate": 0.0
     },
     "4": {
      "n": 15,
      "hit_rate": 0.0
     },
     "5": {
      "n": 12,
      "hit_rate": 0.5
     },
     "6": {
      "n": 112,
      "hit_rate": 0.3839
     }
    },
    "n_subjects": 166,
    "cell": "fh_1d_tf_1d",
    "fh": "1d",
    "tf": "1d"
   },
   {
    "agree": {
     "n": 155,
     "hit_rate": 0.3161,
     "wilson_95": [
      0.2481,
      0.393
     ]
    },
    "disagree": {
     "n": 0,
     "hit_rate": null,
     "wilson_95": [
      null,
      null
     ]
    },
    "lift_pp": null,
    "thin_or_tied": 11,
    "by_margin": {
     "3": {
      "n": 8,
      "hit_rate": 0.0
     },
     "4": {
      "n": 20,
      "hit_rate": 0.0
     },
     "5": {
      "n": 36,
      "hit_rate": 0.0
     },
     "6": {
      "n": 91,
      "hit_rate": 0.5385
     }
    },
    "n_subjects": 166,
    "cell": "fh_1d_tf_4h",
    "fh": "1d",
    "tf": "4h"
   }
  ],
  "by_margin": [
   {
    "margin": 2,
    "n": 45,
    "hit_rate": 0.4222
   },
   {
    "margin": 3,
    "n": 424,
    "hit_rate": 0.5212
   },
   {
    "margin": 4,
    "n": 685,
    "hit_rate": 0.5547
   },
   {
    "margin": 5,
    "n": 990,
    "hit_rate": 0.4424
   },
   {
    "margin": 6,
    "n": 2681,
    "hit_rate": 0.5159
   }
  ],
  "by_margin_week_over_week": [
   {
    "margin": 3,
    "issue6_hit_rate_pct": 42.8,
    "issue7_hit_rate": 0.5212,
    "issue7_n": 424
   },
   {
    "margin": 4,
    "issue6_hit_rate_pct": 46.9,
    "issue7_hit_rate": 0.5547,
    "issue7_n": 685
   },
   {
    "margin": 5,
    "issue6_hit_rate_pct": 40.4,
    "issue7_hit_rate": 0.4424,
    "issue7_n": 990
   },
   {
    "margin": 6,
    "issue6_hit_rate_pct": 34.2,
    "issue7_hit_rate": 0.5159,
    "issue7_n": 2681
   }
  ],
  "per_model": [
   {
    "agree": {
     "n": 713,
     "hit_rate": 0.5133,
     "wilson_95": [
      0.4767,
      0.5498
     ]
    },
    "disagree": {
     "n": 0,
     "hit_rate": null,
     "wilson_95": [
      null,
      null
     ]
    },
    "lift_pp": null,
    "thin_or_tied": 86,
    "by_margin": {
     "2": {
      "n": 8,
      "hit_rate": 0.25
     },
     "3": {
      "n": 61,
      "hit_rate": 0.541
     },
     "4": {
      "n": 101,
      "hit_rate": 0.5941
     },
     "5": {
      "n": 160,
      "hit_rate": 0.4562
     },
     "6": {
      "n": 383,
      "hit_rate": 0.517
     }
    },
    "model": "claude-fable-5"
   },
   {
    "agree": {
     "n": 659,
     "hit_rate": 0.5144,
     "wilson_95": [
      0.4763,
      0.5524
     ]
    },
    "disagree": {
     "n": 1,
     "hit_rate": 0.0,
     "wilson_95": [
      0.0,
      0.7935
     ]
    },
    "lift_pp": 51.4,
    "thin_or_tied": 51,
    "by_margin": {
     "2": {
      "n": 6,
      "hit_rate": 0.1667
     },
     "3": {
      "n": 40,
      "hit_rate": 0.55
     },
     "4": {
      "n": 87,
      "hit_rate": 0.5747
     },
     "5": {
      "n": 143,
      "hit_rate": 0.4336
     },
     "6": {
      "n": 383,
      "hit_rate": 0.5326
     }
    },
    "model": "claude-opus-5"
   },
   {
    "agree": {
     "n": 736,
     "hit_rate": 0.4905,
     "wilson_95": [
      0.4545,
      0.5266
     ]
    },
    "disagree": {
     "n": 12,
     "hit_rate": 0.4167,
     "wilson_95": [
      0.1933,
      0.6805
     ]
    },
    "lift_pp": 7.4,
    "thin_or_tied": 188,
    "by_margin": {
     "2": {
      "n": 9,
      "hit_rate": 0.4444
     },
     "3": {
      "n": 72,
      "hit_rate": 0.5833
     },
     "4": {
      "n": 119,
      "hit_rate": 0.5294
     },
     "5": {
      "n": 153,
      "hit_rate": 0.451
     },
     "6": {
      "n": 383,
      "hit_rate": 0.4778
     }
    },
    "model": "deepseek-v4-pro"
   },
   {
    "agree": {
     "n": 695,
     "hit_rate": 0.5252,
     "wilson_95": [
      0.488,
      0.5621
     ]
    },
    "disagree": {
     "n": 30,
     "hit_rate": 0.4333,
     "wilson_95": [
      0.2738,
      0.608
     ]
    },
    "lift_pp": 9.2,
    "thin_or_tied": 158,
    "by_margin": {
     "2": {
      "n": 7,
      "hit_rate": 0.5714
     },
     "3": {
      "n": 74,
      "hit_rate": 0.5405
     },
     "4": {
      "n": 99,
      "hit_rate": 0.5556
     },
     "5": {
      "n": 132,
      "hit_rate": 0.447
     },
     "6": {
      "n": 383,
      "hit_rate": 0.5405
     }
    },
    "model": "gemini-3.1-pro"
   },
   {
    "agree": {
     "n": 720,
     "hit_rate": 0.5083,
     "wilson_95": [
      0.4719,
      0.5447
     ]
    },
    "disagree": {
     "n": 2,
     "hit_rate": 0.5,
     "wilson_95": [
      0.0945,
      0.9055
     ]
    },
    "lift_pp": 0.8,
    "thin_or_tied": 99,
    "by_margin": {
     "2": {
      "n": 3,
      "hit_rate": 0.3333
     },
     "3": {
      "n": 70,
      "hit_rate": 0.4714
     },
     "4": {
      "n": 105,
      "hit_rate": 0.5429
     },
     "5": {
      "n": 159,
      "hit_rate": 0.4465
     },
     "6": {
      "n": 383,
      "hit_rate": 0.5326
     }
    },
    "model": "gpt-5.6-sol"
   },
   {
    "agree": {
     "n": 577,
     "hit_rate": 0.4922,
     "wilson_95": [
      0.4516,
      0.5329
     ]
    },
    "disagree": {
     "n": 6,
     "hit_rate": 0.1667,
     "wilson_95": [
      0.0301,
      0.5635
     ]
    },
    "lift_pp": 32.6,
    "thin_or_tied": 46,
    "by_margin": {
     "2": {
      "n": 5,
      "hit_rate": 0.6
     },
     "3": {
      "n": 32,
      "hit_rate": 0.375
     },
     "4": {
      "n": 64,
      "hit_rate": 0.5312
     },
     "5": {
      "n": 93,
      "hit_rate": 0.4086
     },
     "6": {
      "n": 383,
      "hit_rate": 0.5144
     }
    },
    "model": "grok-4.6"
   },
   {
    "agree": {
     "n": 725,
     "hit_rate": 0.4966,
     "wilson_95": [
      0.4603,
      0.5329
     ]
    },
    "disagree": {
     "n": 1,
     "hit_rate": 0.0,
     "wilson_95": [
      0.0,
      0.7935
     ]
    },
    "lift_pp": 49.7,
    "thin_or_tied": 93,
    "by_margin": {
     "2": {
      "n": 7,
      "hit_rate": 0.5714
     },
     "3": {
      "n": 75,
      "hit_rate": 0.52
     },
     "4": {
      "n": 110,
      "hit_rate": 0.5545
     },
     "5": {
      "n": 150,
      "hit_rate": 0.44
     },
     "6": {
      "n": 383,
      "hit_rate": 0.4961
     }
    },
    "model": "qwen-3.8-max"
   }
  ],
  "herd_worst_cards": [
   {
    "slot": "2026-09-18T20:01:00+00:00",
    "symbol": "XRP",
    "fh": "4h",
    "tf": "4h",
    "side": "long",
    "n_models": 7,
    "mean_conf": 70,
    "hit_rate": 0.0
   },
   {
    "slot": "2026-09-16T09:01:00+00:00",
    "symbol": "BNB",
    "fh": "1h",
    "tf": "1h",
    "side": "short",
    "n_models": 7,
    "mean_conf": 69.6,
    "hit_rate": 0.0
   },
   {
    "slot": "2026-09-14T19:01:00+00:00",
    "symbol": "BTC",
    "fh": "1h",
    "tf": "1h",
    "side": "long",
    "n_models": 7,
    "mean_conf": 68.6,
    "hit_rate": 0.0
   },
   {
    "slot": "2026-09-19T04:01:00+00:00",
    "symbol": "XRP",
    "fh": "4h",
    "tf": "4h",
    "side": "long",
    "n_models": 7,
    "mean_conf": 68.4,
    "hit_rate": 0.0
   },
   {
    "slot": "2026-09-20T00:01:00+00:00",
    "symbol": "BTC",
    "fh": "1d",
    "tf": "1d",
    "side": "long",
    "n_models": 7,
    "mean_conf": 68.4,
    "hit_rate": 0.0
   }
  ],
  "counters": {
   "forecasts_total": 10290,
   "ok_in_gate": 9308,
   "out_of_gate": 973,
   "out_of_gate_by_fh": {
    "1w": 488,
    "1m": 485
   },
   "invalid": 9,
   "mature": 9308,
   "mature_directional": 5598,
   "mature_sideways": 3710,
   "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": {
    "2026-09-14": {
     "move_pct": 1.79,
     "regime": "trend"
    },
    "2026-09-15": {
     "move_pct": -3.35,
     "regime": "trend"
    },
    "2026-09-16": {
     "move_pct": 0.86,
     "regime": "flat"
    },
    "2026-09-17": {
     "move_pct": 0.21,
     "regime": "flat"
    },
    "2026-09-18": {
     "move_pct": 5.82,
     "regime": "trend"
    },
    "2026-09-19": {
     "move_pct": 0.46,
     "regime": "flat"
    },
    "2026-09-20": {
     "move_pct": 0.13,
     "regime": "flat"
    }
   },
   "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": "2026-09-14T00:01:00+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 -- three 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 #6) 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": [
      "2026-09-14",
      "2026-09-15",
      "2026-09-16",
      "2026-09-17",
      "2026-09-18",
      "2026-09-19",
      "2026-09-20"
     ],
     "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": [
      "2026-09-14",
      "2026-09-15",
      "2026-09-16",
      "2026-09-17",
      "2026-09-18",
      "2026-09-19",
      "2026-09-20"
     ],
     "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": [
      "2026-09-14",
      "2026-09-15",
      "2026-09-16",
      "2026-09-17",
      "2026-09-18",
      "2026-09-19",
      "2026-09-20"
     ],
     "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": [
      "2026-09-14",
      "2026-09-15",
      "2026-09-16",
      "2026-09-17",
      "2026-09-18",
      "2026-09-19",
      "2026-09-20"
     ],
     "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": [
      "2026-09-14",
      "2026-09-15",
      "2026-09-16",
      "2026-09-17",
      "2026-09-18",
      "2026-09-19",
      "2026-09-20"
     ],
     "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."
   }
  }
 },
 "herd_note": "539 cells had 6+ directional models on one side; unanimous_cells_ge6 per weekly metrics",
 "smart_note": "Smart-vs-Outlook deferred: as-of weights pipeline not yet recording; will appear once weights are logged per-slot (methodology 3.2.3 fallback)",
 "key_finding": {
  "label": "KEY FINDING · OBSERVATION (one weekly window)",
  "text": "Herding ran at 98.9% of leave-one-out scoreable calls and the pooled lift came in at +12.1pp: agree-hit 50.6% against 38.5% on disagreement, on a disagree bucket of n=52.",
  "evidence_level": "observation"
 },
 "key_findings": [
  "**OBSERVATION -- 98.9% of the field's leave-one-out calls ran with the herd.** Of 4,877 LOO-scoreable directional calls, 4,825 (98.9%) sided with peers' majority (issue #6: 99.3%); agree-hit was 50.6% [49.2%, 52.0%] vs. 38.5% [26.5%, 52.0%] for the 52 disagreers, a pooled lift of +12.1pp. Descriptive, one week, not a claim of a durable edge.",
  "**The margin curve took a seventh reading.** 3 peers 52.1% (was 42.8%), 4 peers 55.5% (was 46.9%), 5 peers 44.2% (was 40.4%), full unanimity 51.6% (was 34.2%). 3 of the 4 tiers sit above the 50.7% field base, and 2 of the 4 kept the side of the base they took in issue #6.",
  "**Daily-horizon unanimity split around the field base.** 1d/1d margin-6 hit 38.4% (n=112) and 1d/4h margin-6 hit 53.8% (n=91); issue #6 printed 0.0% and 14.3% on the same two cells.",
  "**Herd failures did not disappear.** 539 cells had 6+ directional models on one side (issue #6: 417); the 5 worst sat at 68.4-70.0 mean confidence and resolved 0.0%-0.0%."
 ],
 "market_check": {
  "label": "MARKET CHECK · OBSERVATION (two adjacent calendar weeks)",
  "text": "OBSERVATION — mean |1d move| 1.35% -> 2.69% (+100% rel), BTC realized vol 31.4% -> 35.1% (ann., hourly); field directional accuracy 39.6% -> 50.7% (+11.2pp), 7 of 7 models improved, sim win-rate up for 7 of 7, field sim PnL -$1018.93 -> -$337.72 (adjacent calendar weeks Sep 7-13 vs Sep 14-20; descriptive, one pair of weeks, not a claim).",
  "robustness": "Robustness: raw price-sign accuracy 38.7% -> 54.2% — the same direction as the trade-based hit rule.",
  "weeks": {
   "A": {
    "days": "2026-09-07..2026-09-13",
    "slots": [
     "2026-09-07T00:00:00+00:00",
     "2026-09-14T00:00:00+00:00"
    ],
    "cutoff": "2026-09-14T16:00:00+00:00"
   },
   "B": {
    "days": "2026-09-14..2026-09-20",
    "slots": [
     "2026-09-14T00:00:00+00:00",
     "2026-09-21T00:00:00+00:00"
    ],
    "cutoff": "2026-09-21T16:00:00+00:00"
   }
  },
  "price_source": "crypto_spot/1h@:00",
  "accuracy_field": {
   "A": {
    "n": 4468,
    "hit_rate": 0.3957,
    "wilson_95": [
     0.3815,
     0.4101
    ]
   },
   "B": {
    "n": 5598,
    "hit_rate": 0.5075,
    "wilson_95": [
     0.4944,
     0.5206
    ]
   },
   "delta_pp": 11.2
  },
  "accuracy_per_model": [
   {
    "model": "claude-fable-5",
    "model_label": "claude-fable-5",
    "A": {
     "n": 624,
     "hit_rate": 0.3654,
     "wilson_95": [
      0.3285,
      0.4039
     ]
    },
    "B": {
     "n": 799,
     "hit_rate": 0.5232,
     "wilson_95": [
      0.4885,
      0.5576
     ]
    },
    "delta_pp": 15.8
   },
   {
    "model": "claude-opus-5",
    "model_label": "claude-opus-5",
    "A": {
     "n": 510,
     "hit_rate": 0.3647,
     "wilson_95": [
      0.3241,
      0.4073
     ]
    },
    "B": {
     "n": 711,
     "hit_rate": 0.5162,
     "wilson_95": [
      0.4795,
      0.5527
     ]
    },
    "delta_pp": 15.1
   },
   {
    "model": "gemini-3.1-pro",
    "model_label": "gemini-3.1-pro",
    "A": {
     "n": 673,
     "hit_rate": 0.4175,
     "wilson_95": [
      0.3808,
      0.4552
     ]
    },
    "B": {
     "n": 883,
     "hit_rate": 0.5289,
     "wilson_95": [
      0.4959,
      0.5616
     ]
    },
    "delta_pp": 11.1
   },
   {
    "model": "qwen-3.8-max",
    "model_label": "qwen-3.8-max",
    "A": {
     "n": 707,
     "hit_rate": 0.4003,
     "wilson_95": [
      0.3648,
      0.4368
     ]
    },
    "B": {
     "n": 819,
     "hit_rate": 0.5079,
     "wilson_95": [
      0.4737,
      0.5421
     ]
    },
    "delta_pp": 10.8
   },
   {
    "model": "grok-4.6",
    "model_label": "grok-4.6 (incl. 4.5-era, Aug 24 00:00-09:22 UTC)",
    "A": {
     "n": 518,
     "hit_rate": 0.3958,
     "wilson_95": [
      0.3546,
      0.4385
     ]
    },
    "B": {
     "n": 629,
     "hit_rate": 0.496,
     "wilson_95": [
      0.4571,
      0.535
     ]
    },
    "delta_pp": 10.0
   },
   {
    "model": "gpt-5.6-sol",
    "model_label": "gpt-5.6-sol",
    "A": {
     "n": 709,
     "hit_rate": 0.4006,
     "wilson_95": [
      0.3651,
      0.4371
     ]
    },
    "B": {
     "n": 821,
     "hit_rate": 0.497,
     "wilson_95": [
      0.4628,
      0.5311
     ]
    },
    "delta_pp": 9.6
   },
   {
    "model": "deepseek-v4-pro",
    "model_label": "deepseek-v4-pro",
    "A": {
     "n": 727,
     "hit_rate": 0.414,
     "wilson_95": [
      0.3788,
      0.4502
     ]
    },
    "B": {
     "n": 936,
     "hit_rate": 0.484,
     "wilson_95": [
      0.4521,
      0.516
     ]
    },
    "delta_pp": 7.0
   }
  ],
  "raw_sign_field": {
   "A": {
    "n_scored": 4466,
    "hit_rate": 0.3871
   },
   "B": {
    "n_scored": 5583,
    "hit_rate": 0.5416
   }
  },
  "abs_move_1d": {
   "A": 1.347,
   "B": 2.693,
   "delta_pct_rel": 99.9
  },
  "btc_realized_vol_ann_pct_hourly": {
   "A": 31.4,
   "B": 35.07
  },
  "trading_field": {
   "A": {
    "n_trades": 4468,
    "wr": 0.3212,
    "pnl_net_usd": -1018.93,
    "pnl_gross_usd": -572.13
   },
   "B": {
    "n_trades": 5598,
    "wr": 0.4337,
    "pnl_net_usd": -337.72,
    "pnl_gross_usd": 222.08
   }
  },
  "verdict": {
   "models_with_hit_improved": "7 of 7",
   "models_with_rawsign_improved": "7 of 7",
   "models_with_wr_improved": "7 of 7",
   "field_hit_delta_pp": 11.2,
   "field_pnl_net": {
    "A": -1018.93,
    "B": -337.72
   },
   "abs_move_delta": {
    "1h": {
     "A": 0.333,
     "B": 0.369,
     "delta_pct_rel": 10.8
    },
    "4h": {
     "A": 0.605,
     "B": 0.835,
     "delta_pct_rel": 38.0
    },
    "1d": {
     "A": 1.347,
     "B": 2.693,
     "delta_pct_rel": 99.9
    }
   },
   "btc_vol_ann_pct": {
    "A": 31.4,
    "B": 35.07
   },
   "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, three windows back, inside issue #4's window). Week A (2026-09-07..2026-09-13) is pure grok-4.6 (it reproduces the published issue #6); week B is pure grok-4.6."
 },
 "why_it_matters": "MarketMania publishes forecasts from 7 frontier model lines into the same slots every week; a natural question for anyone reading the feed is whether a model's call is worth more when it agrees with what everyone else is saying, or whether that agreement is just everyone making the same mistake together. Consensus Watch answers that question empirically, one week at a time, using a leave-one-out design built specifically to avoid the obvious trap of comparing a model to an index that includes its own vote.",
 "how_to_read_this": "For every model M and every forecast it made, we rebuild that slot's consensus using only the OTHER active models in the same symbol / forecast-horizon (FH) / timeframe (TF) cell -- M's own call never counts toward its own consensus. M is scored agree if its side (long or short) matches the strict majority of those peers, and disagree if it sits alone against that majority; a cell needs at least 3 directional peers to count at all, and tied peer splits are excluded rather than forced either way. LOO peers are genuinely external to the model being scored, so the comparison below is not comparing a model to itself. The grok rows are one line, so grok never counts as two peers in a cell. The grok 4.5 -> 4.6 flip (Aug 24, 2026 09:22 UTC) sits three 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 second pure-4.6 against pure-4.6 comparison of the series.",
 "practical_implications": [
  "Seven issues, seven margin-curve readings. The tier ordering moved again -- 4 peers still on top, but unanimity came off the bottom and the 5-peer tier took it -- and all four tiers printed a higher rate than in issue #6 (4 peers 46.9% -> 55.5%, full unanimity 34.2% -> 51.6%), so a consensus filter still should not carry weight in a config.",
  "The pooled lift is +12.1pp this week against -30.4pp in issue #6 and -27.6pp in issue #5. The sign flipped after three issues below zero: treat the number as this week's weather, not an edge.",
  "Read the consensus number together with the market check on the same page. The field base moved +11.2pp week over week; 4 of the 4 consensus tiers moved up and 0 down against issue #6 (3 peers +9.4pp, 4 peers +8.6pp, 5 peers +3.9pp, unanimity +17.3pp)."
 ],
 "limitations": [
  "Single week (Sep 14-20, 2026 UTC). Every finding is descriptive for this window only.",
  "Observations are not independent: the same models watch overlapping symbol / FH / TF cells hour after hour. Wilson intervals here are descriptive, not inferential.",
  "Correlational, not causal: all models see the same market data, so 'agreement' and 'hit' can rise together simply because a slot was easy to read. LOO removes self-match bias, not this confound.",
  "The pooled lift rests on 52 disagreeing calls; per-model counts run 0-30. 721 calls (12.9% of the 5,598-call mature directional universe) were thin or tied and are excluded from every consensus table.",
  "This window ran 3 trend days of 7 (issue #6: 2 of 7); every comparison with issue #6 is a comparison across regimes as well as across weeks. Series density and lineage notes (wave 7): (1) rolling-1w forecasts emit daily since Aug 22, 2026 and rolling-1M daily since Aug 23, 2026, so inside this window (Sep 14-20) daily coverage is FULL, as it was in issues #4, #5 and #6: 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 488, 1M 485). (2) The grok line flipped 4.5 -> 4.6 at Aug 24, 2026 09:22 UTC, three 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-#6 window), so every grok number in this issue and in issue #6 is a pure grok-4.6 line -- the grok week-over-week row is the second 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 #6 after the audit that found the raw table mixes two exchanges; issue #6 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 (Sep 21 16:00 UTC) -- the definition pinned in issue #3 and carried by issues #4, #5 and #6, which printed 47,373 at the Sep 14 16:00 cutoff. The pack reproduces that pin at the previous cutoff (control OK), so this issue's 52,971 is an additive step under one definition; counter deltas against issue #5 and earlier remain definitional."
 ],
 "notes": {
  "series_density_and_lineage_wave7": "Series density and lineage notes (wave 7): (1) rolling-1w forecasts emit daily since Aug 22, 2026 and rolling-1M daily since Aug 23, 2026, so inside this window (Sep 14-20) daily coverage is FULL, as it was in issues #4, #5 and #6: 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 488, 1M 485). (2) The grok line flipped 4.5 -> 4.6 at Aug 24, 2026 09:22 UTC, three 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-#6 window), so every grok number in this issue and in issue #6 is a pure grok-4.6 line -- the grok week-over-week row is the second 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": "2026-09-14T00:01:00+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 -- three 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 #6) 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": [
     "2026-09-14",
     "2026-09-15",
     "2026-09-16",
     "2026-09-17",
     "2026-09-18",
     "2026-09-19",
     "2026-09-20"
    ],
    "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": [
     "2026-09-14",
     "2026-09-15",
     "2026-09-16",
     "2026-09-17",
     "2026-09-18",
     "2026-09-19",
     "2026-09-20"
    ],
    "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": [
     "2026-09-14",
     "2026-09-15",
     "2026-09-16",
     "2026-09-17",
     "2026-09-18",
     "2026-09-19",
     "2026-09-20"
    ],
    "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": [
     "2026-09-14",
     "2026-09-15",
     "2026-09-16",
     "2026-09-17",
     "2026-09-18",
     "2026-09-19",
     "2026-09-20"
    ],
    "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": [
     "2026-09-14",
     "2026-09-15",
     "2026-09-16",
     "2026-09-17",
     "2026-09-18",
     "2026-09-19",
     "2026-09-20"
    ],
    "missing_days": []
   }
  },
  "week_over_week_note": "Week-over-week columns compare back-to-back windows: Sep 7-13 (issue #6, and week A of this issue's alive slice) vs Sep 14-20 (this issue). \"Was\" values are the numbers published in issue #6; 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 (Sep 21 16:00 UTC) -- the definition pinned in issue #3 and carried by issues #4, #5 and #6, which printed 47,373 at the Sep 14 16:00 cutoff. The pack reproduces that pin at the previous cutoff (control OK), so this issue's 52,971 is an additive step under one definition; counter deltas against issue #5 and earlier remain definitional.",
  "market_state_audit": "Market-state row is computed on a single-exchange (binance) daily candle series, as in issue #6 after the audit that found the raw table mixes two exchanges; issue #6 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"
 },
 "bibtex_key": "mm_consensus_2026w38",
 "citation": {
  "bibtex_key": "mm_consensus_2026w38",
  "title": "Consensus Watch #7: does agreeing with the crowd make an LLM's market call safer?",
  "author": "MarketMania Research",
  "year": 2026,
  "month": "September",
  "day": 22,
  "url": "https://marketmania.ai/research/reports/consensus-watch-2026-09-14.pdf",
  "note": "Methodology v1.1; window Sep 14-20, 2026 UTC; source weekly_metrics_2026-09-14.json"
 },
 "living_series_note": "**Issue #7.** Consensus Watch is a living weekly comparison; each issue appends one more week of leave-one-out agreement-vs-hit data. Issue #6 asked whether the pooled agree-vs-disagree sign keeps the same direction for a fourth issue running: it reads +12.1pp this week after -30.4pp in issue #6 (agree 50.6% vs disagree 38.5% on n=52), and whether the same margin tiers hold their side of the field base for a fourth issue: 2 of the 4 tiers did, and the 4-peer tier came in above the field base. The grok row is pure grok-4.6 this issue and was pure grok-4.6 in issue #6; the 4.5 -> 4.6 flip (Aug 24, 2026 09:22 UTC) sits three 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: the pooled agree-vs-disagree lift flipped to +12.1pp after three issues below zero -- does the positive sign hold for a second issue?",
  "The 4-peer tier has sat above the field base two issues running -- does it hold for a third?",
  "Monthly series: the agreement curve across regimes at monthly n."
 ],
 "related_research": [
  {
   "title": "Weekly Calibration #7",
   "url": "https://marketmania.ai/research/reports/weekly-calibration-2026-09-14.pdf"
  },
  {
   "title": "Weekly Model Watch #7",
   "url": "https://marketmania.ai/research/reports/model-watch-2026-09-14.pdf"
  },
  {
   "title": "Consensus Watch #6",
   "url": "https://marketmania.ai/research/reports/consensus-watch-2026-09-07.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": {
   "loo_scored_observations": 4877
  },
  "platform": {
   "as_of_cutoff": "2026-09-21",
   "resolved_forecasts": 52971,
   "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": 31
  },
  "source": "weekly_metrics_2026-09-14.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 (Sep 21 16:00 UTC) -- the definition pinned in issue #3 and carried by issues #4, #5 and #6, which printed 47,373 at the Sep 14 16:00 cutoff. The pack reproduces that pin at the previous cutoff (control OK), so this issue's 52,971 is an additive step under one definition; counter deltas against issue #5 and earlier remain definitional."
 },
 "site_entry": {
  "slug": "consensus-watch-2026-09-14",
  "series": "weekly",
  "seriesLabel": "Weekly · Consensus",
  "date": "September 22, 2026",
  "title": "Consensus Watch #7",
  "summary": "9,308 mature forecasts asked whether agreeing with the crowd makes an LLM market call safer. Herding ran at 98.9% of leave-one-out scoreable calls (issue #6: 99.3%), and the pooled lift came in at +12.1pp: agree-hit 50.6% against 38.5% on disagreement, on a disagree bucket of n=52 and a 50.7% field base. The margin curve read 52.1% / 55.5% / 44.2% / 51.6% from 3 peers to full unanimity. Field accuracy moved 39.6% -> 50.7% (+11.2pp) week over week, with 7 of 7 models improving.",
  "stats": [
   "4,877 peer-judged calls",
   "98.9% ran with the herd",
   "lift +12.1pp vs disagree"
  ],
  "manifest_key_finding": "Herding ran at 98.9% of leave-one-out scoreable calls and the pooled lift came in at +12.1pp: agree-hit 50.6% against 38.5% on disagreement (disagree bucket n=52), against a 50.7% field base.",
  "files": {
   "pdf": "/research/reports/consensus-watch-2026-09-14.pdf",
   "md": "/research/reports/consensus-watch-2026-09-14.md",
   "json": "/research/reports/consensus-watch-2026-09-14.json"
  }
 },
 "social": {
  "telegram": {
   "photo_caption_html": "🧭 <b>Consensus Watch #7</b> — weekly series (window Sep 14–20)\n\nResearch question: if a model's forecast matches the consensus of the other models, is it more reliable?\n\n4,877 leave-one-out scored calls, 7 model lines:\n• Herding ran at <b>98.9%</b> (issue #6: 99.3%) — agree-hit 50.6% vs <b>38.5%</b> when a model left the crowd (field base 50.7%, disagree n=52)\n• Margin curve: 3-peer 52.1%, 4-peer 55.5%, 5-peer 44.2%, unanimity 51.6%\n• Daily horizon: 1d/1d unanimity 38.4%, 1d/4h unanimity 53.8%\n• Market check: field accuracy 39.6% -> 50.7% (+11.2pp), 7 of 7 models improved; avg |1d| move 1.35% -> 2.69%\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 three windows back\n\n📄 <a href=\"https://marketmania.ai/research/reports/consensus-watch-2026-09-14.pdf\">Full report (PDF)</a> · <a href=\"https://marketmania.ai/research\">all research</a>\n\n#cryptotrading #LLM #AI #crypto #trading #consensus #research",
   "document_caption_html": "📄 Consensus Watch #7 — full report (PDF).\nWeb copy: <a href=\"https://marketmania.ai/research/reports/consensus-watch-2026-09-14.pdf\">consensus-watch-2026-09-14.pdf</a> · machine-readable: <a href=\"https://marketmania.ai/research/reports/consensus-watch-2026-09-14.json\">JSON</a> · <a href=\"https://marketmania.ai/research\">all research</a>\n\n#cryptotrading #LLM #AI #crypto #trading #consensus #research",
   "caption_len": 961,
   "document_caption_len": 402,
   "limit": 1024
  },
  "x": {
   "main": "Consensus Watch #7 is out.\n\n4,877 peer-judged LLM market calls: 98.9% ran with the herd, and agreeing scored 50.6% against 38.5% for the 52 calls that left it (+12.1pp).\n\nFull report: https://marketmania.ai/research/reports/consensus-watch-2026-09-14.pdf\n\n#AI #Crypto #Research #Consensus",
   "main_tco_len": 241,
   "reply": "Margin curve, w/w (hit):\n3 peers: 42.8 -> 52.1\n4 peers: 46.9 -> 55.5\n5 peers: 40.4 -> 44.2\nunanimity: 34.2 -> 51.6\nField base: 39.6 -> 50.7. LOO 3+ peers, methodology v1.1.\nData: https://marketmania.ai/research/reports/consensus-watch-2026-09-14.json",
   "reply_tco_len": 202,
   "limit_tco": 280,
   "replies_max": 1
  }
 }
}