{
"slug": "consensus-watch-monthly-2026-08",
"series": "monthly",
"series_label": "MONTHLY · CONSENSUS WATCH",
"issue": 1,
"title": "Monthly Consensus Watch #1",
"language": "en",
"window_start": "2026-08-01T00:00:00+00:00",
"window_end": "2026-09-01T00:00:00+00:00",
"window_text": "Aug 1-31, 2026 UTC",
"cutoff": "2026-09-01T16:00:00+00:00",
"generated_at": "2026-09-04T07:26:21+00:00",
"methodology": "v1.1 (2026-08-10)",
"methodology_hash": "e66c7e8c864a2233",
"source": "monthly_metrics_2026-08.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": 19383,
"hit_rate": 0.4633
},
"market_state": {
"panel": "market_state_6p2",
"month": {
"start": "2026-08-01",
"end": "2026-09-01",
"days": 31
},
"prev_month_partial": {
"start": "2026-07-15",
"end": "2026-08-01",
"days": 17,
"partial": true
},
"current": {
"btc_net_pct": 24.95,
"realized_vol_ann_pct": 41.53,
"volume_top5_usd_bn": 65.9696,
"volume_top5_mom_pct": 135.9,
"avg_pairwise_corr": 0.79
},
"previous_partial_month": {
"btc_net_pct": -3.31,
"realized_vol_ann_pct": 27.33,
"volume_top5_usd_bn": 27.9598,
"avg_pairwise_corr": 0.79
},
"snapshot_line": "BTC net +24.95% (prior -3.31%) - ann vol 41.5% (was 27.3%) - TOP5 volume $65.97B, +136% m/m - pairwise corr 0.79 (was 0.79)",
"exchange": "binance",
"audit_filter_applied": true,
"source": "single-exchange (binance) daily candle recompute, market_state_2026-08.json; the prior-month block is the pack's own prev_month_partial (Jul 15-31)",
"audit_note": "Market-state row is computed on a single-exchange (binance) daily candle series, as in the weekly series after the audit that found the raw table mixes two exchanges. The prior-month block is PARTIAL: audited daily candles start 2026-07-15, so it covers 17 days (Jul 15-31), and the TOP5 volume figure compares a 31-day sum with a 17-day sum -- a level difference, not a like-for-like change.",
"weeks": {
"W1": {
"btc_net_pct": 2.09,
"realized_vol_ann_pct": 11.42,
"regime": "calm"
},
"W2": {
"btc_net_pct": -3.08,
"realized_vol_ann_pct": 10.03,
"regime": "calm"
},
"W3": {
"btc_net_pct": 23.58,
"realized_vol_ann_pct": 65.02,
"regime": "volatile"
},
"W4": {
"btc_net_pct": -0.07,
"realized_vol_ann_pct": 30.85,
"regime": "volatile"
}
},
"volatility_rule": "volatile if realized_vol_ann_pct >= median(W1..W4) else calm",
"median_realized_vol_ann_pct": 21.14,
"delta_month_vs_prev_partial": {
"btc_net_pct": 28.26,
"realized_vol_ann_pct": 14.2,
"avg_daily_range_pct": 0.38,
"volume_top5_usd_pct": 135.9,
"avg_pairwise_corr": 0.0,
"note": "August (31 days) vs 15-31 July (17 days, PARTIAL -- daily candles start 2026-07-15); volume is a sum, so volume_top5_usd_pct compares a 31-day sum with a 17-day sum and is NOT a like-for-like change"
},
"controls": {
"w4_matches_published": {
"pins_source": "market_state_2026-08-24.json -> 'current' (the published wave-4 week)",
"block": "W4",
"published": {
"btc_net_pct": -0.07,
"realized_vol_ann_pct": 30.85,
"volume_top5_usd_bn": 19.7224,
"avg_pairwise_corr": 0.85
},
"got": {
"btc_net_pct": -0.07,
"realized_vol_ann_pct": 30.85,
"volume_top5_usd_bn": 19.7224,
"avg_pairwise_corr": 0.85
},
"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
},
"w3_matches_published_previous": {
"pins_source": "market_state_2026-08-24.json -> 'previous' (the wave-4 run's prev week)",
"block": "W3",
"published": {
"btc_net_pct": 23.58,
"realized_vol_ann_pct": 65.02,
"volume_top5_usd_bn": 25.9715,
"avg_pairwise_corr": 0.81
},
"got": {
"btc_net_pct": 23.58,
"realized_vol_ann_pct": 65.02,
"volume_top5_usd_bn": 25.9715,
"avg_pairwise_corr": 0.81
},
"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": "W4 and W3 of this run are exactly the two weeks the wave-4 market panel computed, with the same audit filter, so both must reproduce. Non-fatal: a MISMATCH means the exchange filter or the symbol set moved -- send the log, do not publish the row."
},
"regime_days": {
"2026-08-01": {
"move_pct": -0.13,
"regime": "flat"
},
"2026-08-02": {
"move_pct": 0.81,
"regime": "flat"
},
"2026-08-03": {
"move_pct": 0.09,
"regime": "flat"
},
"2026-08-04": {
"move_pct": 0.93,
"regime": "flat"
},
"2026-08-05": {
"move_pct": 0.98,
"regime": "flat"
},
"2026-08-06": {
"move_pct": -0.57,
"regime": "flat"
},
"2026-08-07": {
"move_pct": 0.88,
"regime": "flat"
},
"2026-08-08": {
"move_pct": 0.16,
"regime": "flat"
},
"2026-08-09": {
"move_pct": 0.09,
"regime": "flat"
},
"2026-08-10": {
"move_pct": -1.63,
"regime": "trend"
},
"2026-08-11": {
"move_pct": -0.56,
"regime": "flat"
},
"2026-08-12": {
"move_pct": -0.02,
"regime": "flat"
},
"2026-08-13": {
"move_pct": -0.25,
"regime": "flat"
},
"2026-08-14": {
"move_pct": -0.67,
"regime": "flat"
},
"2026-08-15": {
"move_pct": 0.07,
"regime": "flat"
},
"2026-08-16": {
"move_pct": -0.35,
"regime": "flat"
},
"2026-08-17": {
"move_pct": 2.55,
"regime": "trend"
},
"2026-08-18": {
"move_pct": 0.21,
"regime": "flat"
},
"2026-08-19": {
"move_pct": 7.82,
"regime": "trend"
},
"2026-08-20": {
"move_pct": 5.07,
"regime": "trend"
},
"2026-08-21": {
"move_pct": 7.26,
"regime": "trend"
},
"2026-08-22": {
"move_pct": -1.67,
"regime": "trend"
},
"2026-08-23": {
"move_pct": 0.69,
"regime": "flat"
},
"2026-08-24": {
"move_pct": 1.61,
"regime": "trend"
},
"2026-08-25": {
"move_pct": -0.58,
"regime": "flat"
},
"2026-08-26": {
"move_pct": 0.63,
"regime": "flat"
},
"2026-08-27": {
"move_pct": 1.53,
"regime": "trend"
},
"2026-08-28": {
"move_pct": -2.91,
"regime": "trend"
},
"2026-08-29": {
"move_pct": 0.48,
"regime": "flat"
},
"2026-08-30": {
"move_pct": -0.7,
"regime": "flat"
},
"2026-08-31": {
"move_pct": 1.19,
"regime": "flat"
}
},
"alive_slice_context": {
"abs_move_1d_pct": {
"A": 1.536,
"B": 1.979
},
"btc_realized_vol_ann_pct_hourly": {
"A": 33.46,
"B": 36.92
}
}
},
"tables": {
"pooled": {
"agree": {
"n": 15878,
"hit_rate": 0.4554,
"wilson_95": [
0.4477,
0.4632
]
},
"disagree": {
"n": 211,
"hit_rate": 0.436,
"wilson_95": [
0.3708,
0.5035
]
},
"lift_pp": 1.9,
"thin_or_tied": 3294,
"by_margin": {
"2": {
"n": 138,
"hit_rate": 0.6014
},
"3": {
"n": 1964,
"hit_rate": 0.4593
},
"4": {
"n": 2630,
"hit_rate": 0.4513
},
"5": {
"n": 4482,
"hit_rate": 0.4581
},
"6": {
"n": 6664,
"hit_rate": 0.4511
}
},
"total_scoreable": 16089,
"caveat": "The disagree bucket is n=211 over the whole month; the two Wilson intervals overlap. Prior window Jul 11-31 (partial): agree 44.6% / disagree 55.3%, lift -10.6pp, on n=16,557 agree and n=436 disagree. These are dependent observations pooled across four weeks and two volatility regimes."
},
"by_cell": [
{
"agree": {
"n": 9857,
"hit_rate": 0.4519,
"wilson_95": [
0.4421,
0.4617
]
},
"disagree": {
"n": 125,
"hit_rate": 0.48,
"wilson_95": [
0.3943,
0.5669
]
},
"lift_pp": -2.8,
"thin_or_tied": 2128,
"by_margin": {
"2": {
"n": 93,
"hit_rate": 0.5914
},
"3": {
"n": 1284,
"hit_rate": 0.4361
},
"4": {
"n": 1680,
"hit_rate": 0.428
},
"5": {
"n": 2670,
"hit_rate": 0.4603
},
"6": {
"n": 4130,
"hit_rate": 0.4579
}
},
"n_subjects": 12110,
"cell": "fh_1h_tf_1h",
"fh": "1h",
"tf": "1h"
},
{
"agree": {
"n": 2548,
"hit_rate": 0.449,
"wilson_95": [
0.4298,
0.4684
]
},
"disagree": {
"n": 25,
"hit_rate": 0.4,
"wilson_95": [
0.234,
0.5926
]
},
"lift_pp": 4.9,
"thin_or_tied": 466,
"by_margin": {
"2": {
"n": 9,
"hit_rate": 0.6667
},
"3": {
"n": 268,
"hit_rate": 0.4478
},
"4": {
"n": 425,
"hit_rate": 0.4565
},
"5": {
"n": 726,
"hit_rate": 0.438
},
"6": {
"n": 1120,
"hit_rate": 0.4518
}
},
"n_subjects": 3039,
"cell": "fh_4h_tf_1h",
"fh": "4h",
"tf": "1h"
},
{
"agree": {
"n": 2483,
"hit_rate": 0.4611,
"wilson_95": [
0.4416,
0.4808
]
},
"disagree": {
"n": 54,
"hit_rate": 0.3889,
"wilson_95": [
0.2704,
0.5221
]
},
"lift_pp": 7.2,
"thin_or_tied": 549,
"by_margin": {
"2": {
"n": 30,
"hit_rate": 0.5333
},
"3": {
"n": 320,
"hit_rate": 0.5219
},
"4": {
"n": 395,
"hit_rate": 0.5089
},
"5": {
"n": 744,
"hit_rate": 0.4946
},
"6": {
"n": 994,
"hit_rate": 0.3954
}
},
"n_subjects": 3086,
"cell": "fh_4h_tf_4h",
"fh": "4h",
"tf": "4h"
},
{
"agree": {
"n": 454,
"hit_rate": 0.5396,
"wilson_95": [
0.4937,
0.585
]
},
"disagree": {
"n": 5,
"hit_rate": 0.2,
"wilson_95": [
0.0362,
0.6245
]
},
"lift_pp": 34.0,
"thin_or_tied": 80,
"by_margin": {
"2": {
"n": 6,
"hit_rate": 1.0
},
"3": {
"n": 56,
"hit_rate": 0.5536
},
"4": {
"n": 75,
"hit_rate": 0.5733
},
"5": {
"n": 156,
"hit_rate": 0.4359
},
"6": {
"n": 161,
"hit_rate": 0.6025
}
},
"n_subjects": 539,
"cell": "fh_1d_tf_1d",
"fh": "1d",
"tf": "1d"
},
{
"agree": {
"n": 536,
"hit_rate": 0.4534,
"wilson_95": [
0.4117,
0.4957
]
},
"disagree": {
"n": 2,
"hit_rate": 0.0,
"wilson_95": [
0.0,
0.6576
]
},
"lift_pp": 45.3,
"thin_or_tied": 71,
"by_margin": {
"3": {
"n": 36,
"hit_rate": 0.6667
},
"4": {
"n": 55,
"hit_rate": 0.5455
},
"5": {
"n": 186,
"hit_rate": 0.3763
},
"6": {
"n": 259,
"hit_rate": 0.4595
}
},
"n_subjects": 609,
"cell": "fh_1d_tf_4h",
"fh": "1d",
"tf": "4h"
}
],
"by_margin": [
{
"margin": 2,
"n": 138,
"hit_rate": 0.6014
},
{
"margin": 3,
"n": 1964,
"hit_rate": 0.4593
},
{
"margin": 4,
"n": 2630,
"hit_rate": 0.4513
},
{
"margin": 5,
"n": 4482,
"hit_rate": 0.4581
},
{
"margin": 6,
"n": 6664,
"hit_rate": 0.4511
}
],
"by_margin_month_over_month": [
{
"margin": 3,
"prior_window_hit_rate": 0.427,
"prior_window_n": 1609,
"monthly1_hit_rate": 0.4593,
"monthly1_n": 1964,
"prior_source": "month.prior_window.consensus.by_margin",
"prior_note": null
},
{
"margin": 4,
"prior_window_hit_rate": 0.4353,
"prior_window_n": 2394,
"monthly1_hit_rate": 0.4513,
"monthly1_n": 2630,
"prior_source": "month.prior_window.consensus.by_margin",
"prior_note": null
},
{
"margin": 5,
"prior_window_hit_rate": 0.4308,
"prior_window_n": 3628,
"monthly1_hit_rate": 0.4581,
"monthly1_n": 4482,
"prior_source": "month.prior_window.consensus.by_margin",
"prior_note": null
},
{
"margin": 6,
"prior_window_hit_rate": 0.4562,
"prior_window_n": 11531,
"monthly1_hit_rate": 0.4511,
"monthly1_n": 6664,
"prior_source": "month.prior_window.consensus.by_margin",
"prior_note": null
}
],
"per_model": [
{
"agree": {
"n": 2391,
"hit_rate": 0.458,
"wilson_95": [
0.4381,
0.478
]
},
"disagree": {
"n": 13,
"hit_rate": 0.5385,
"wilson_95": [
0.2914,
0.7679
]
},
"lift_pp": -8.0,
"thin_or_tied": 408,
"by_margin": {
"2": {
"n": 25,
"hit_rate": 0.72
},
"3": {
"n": 292,
"hit_rate": 0.4384
},
"4": {
"n": 406,
"hit_rate": 0.4409
},
"5": {
"n": 716,
"hit_rate": 0.4665
},
"6": {
"n": 952,
"hit_rate": 0.458
}
},
"model": "claude-fable-5"
},
{
"agree": {
"n": 2105,
"hit_rate": 0.4727,
"wilson_95": [
0.4514,
0.494
]
},
"disagree": {
"n": 4,
"hit_rate": 0.5,
"wilson_95": [
0.15,
0.85
]
},
"lift_pp": -2.7,
"thin_or_tied": 172,
"by_margin": {
"2": {
"n": 21,
"hit_rate": 0.5714
},
"3": {
"n": 194,
"hit_rate": 0.5155
},
"4": {
"n": 317,
"hit_rate": 0.5079
},
"5": {
"n": 621,
"hit_rate": 0.4605
},
"6": {
"n": 952,
"hit_rate": 0.458
}
},
"model": "claude-opus-5"
},
{
"agree": {
"n": 2065,
"hit_rate": 0.4489,
"wilson_95": [
0.4276,
0.4704
]
},
"disagree": {
"n": 39,
"hit_rate": 0.4103,
"wilson_95": [
0.2708,
0.5658
]
},
"lift_pp": 3.9,
"thin_or_tied": 474,
"by_margin": {
"2": {
"n": 21,
"hit_rate": 0.4762
},
"3": {
"n": 247,
"hit_rate": 0.4615
},
"4": {
"n": 330,
"hit_rate": 0.4303
},
"5": {
"n": 515,
"hit_rate": 0.4583
},
"6": {
"n": 952,
"hit_rate": 0.4464
}
},
"model": "deepseek-v4-pro"
},
{
"agree": {
"n": 2249,
"hit_rate": 0.4544,
"wilson_95": [
0.4339,
0.4751
]
},
"disagree": {
"n": 89,
"hit_rate": 0.4719,
"wilson_95": [
0.3715,
0.5746
]
},
"lift_pp": -1.7,
"thin_or_tied": 695,
"by_margin": {
"2": {
"n": 20,
"hit_rate": 0.65
},
"3": {
"n": 308,
"hit_rate": 0.4643
},
"4": {
"n": 369,
"hit_rate": 0.4417
},
"5": {
"n": 600,
"hit_rate": 0.4617
},
"6": {
"n": 952,
"hit_rate": 0.4475
}
},
"model": "gemini-3.1-pro"
},
{
"agree": {
"n": 2459,
"hit_rate": 0.4453,
"wilson_95": [
0.4258,
0.465
]
},
"disagree": {
"n": 8,
"hit_rate": 0.375,
"wilson_95": [
0.1368,
0.6943
]
},
"lift_pp": 7.0,
"thin_or_tied": 503,
"by_margin": {
"2": {
"n": 9,
"hit_rate": 0.4444
},
"3": {
"n": 335,
"hit_rate": 0.4358
},
"4": {
"n": 445,
"hit_rate": 0.436
},
"5": {
"n": 718,
"hit_rate": 0.4387
},
"6": {
"n": 952,
"hit_rate": 0.458
}
},
"model": "gpt-5.6-sol"
},
{
"agree": {
"n": 2218,
"hit_rate": 0.4599,
"wilson_95": [
0.4392,
0.4807
]
},
"disagree": {
"n": 43,
"hit_rate": 0.4419,
"wilson_95": [
0.3043,
0.5889
]
},
"lift_pp": 1.8,
"thin_or_tied": 447,
"by_margin": {
"2": {
"n": 22,
"hit_rate": 0.6364
},
"3": {
"n": 258,
"hit_rate": 0.4496
},
"4": {
"n": 354,
"hit_rate": 0.4633
},
"5": {
"n": 632,
"hit_rate": 0.4731
},
"6": {
"n": 952,
"hit_rate": 0.4485
}
},
"model": "grok-4.6"
},
{
"agree": {
"n": 2391,
"hit_rate": 0.4504,
"wilson_95": [
0.4306,
0.4704
]
},
"disagree": {
"n": 15,
"hit_rate": 0.2,
"wilson_95": [
0.0705,
0.4519
]
},
"lift_pp": 25.0,
"thin_or_tied": 595,
"by_margin": {
"2": {
"n": 20,
"hit_rate": 0.6
},
"3": {
"n": 330,
"hit_rate": 0.4697
},
"4": {
"n": 409,
"hit_rate": 0.4499
},
"5": {
"n": 680,
"hit_rate": 0.45
},
"6": {
"n": 952,
"hit_rate": 0.4412
}
},
"model": "qwen-3.8-max"
}
],
"herd_worst_cards": [
{
"slot": "2026-08-04T19:01:00+00:00",
"symbol": "BTC",
"fh": "1h",
"tf": "1h",
"side": "long",
"n_models": 7,
"mean_conf": 71.7,
"hit_rate": 0.0
},
{
"slot": "2026-08-11T12:01:00+00:00",
"symbol": "BNB",
"fh": "4h",
"tf": "1h",
"side": "long",
"n_models": 7,
"mean_conf": 71,
"hit_rate": 0.0
},
{
"slot": "2026-08-02T06:01:00+00:00",
"symbol": "SOL",
"fh": "1h",
"tf": "1h",
"side": "long",
"n_models": 6,
"mean_conf": 70.8,
"hit_rate": 0.0
},
{
"slot": "2026-08-03T16:01:00+00:00",
"symbol": "XRP",
"fh": "4h",
"tf": "1h",
"side": "long",
"n_models": 7,
"mean_conf": 70,
"hit_rate": 0.0
},
{
"slot": "2026-08-05T20:01:00+00:00",
"symbol": "BTC",
"fh": "4h",
"tf": "1h",
"side": "long",
"n_models": 7,
"mean_conf": 69.9,
"hit_rate": 0.0
}
],
"counters": {
"forecasts_total": 42654,
"ok_in_gate": 40979,
"out_of_gate": 1604,
"out_of_gate_by_fh": {
"1m": 698,
"1w": 906
},
"invalid": 71,
"mature": 40979,
"mature_directional": 19383,
"mature_sideways": 21596,
"pending_next_issue": 0,
"pending_by_fh": {},
"late_closes": 0,
"uptime": [
{
"fh": "1h",
"tf": "1h",
"slots_seen": 741,
"slots_expected": 744
},
{
"fh": "4h",
"tf": "4h",
"slots_seen": 185,
"slots_expected": 186
},
{
"fh": "4h",
"tf": "1h",
"slots_seen": 185,
"slots_expected": 186
},
{
"fh": "1d",
"tf": "1d",
"slots_seen": 31,
"slots_expected": 31
},
{
"fh": "1d",
"tf": "4h",
"slots_seen": 31,
"slots_expected": 31
}
],
"regime_days": {
"2026-08-01": {
"move_pct": -0.13,
"regime": "flat"
},
"2026-08-02": {
"move_pct": 0.81,
"regime": "flat"
},
"2026-08-03": {
"move_pct": 0.09,
"regime": "flat"
},
"2026-08-04": {
"move_pct": 0.93,
"regime": "flat"
},
"2026-08-05": {
"move_pct": 0.98,
"regime": "flat"
},
"2026-08-06": {
"move_pct": -0.57,
"regime": "flat"
},
"2026-08-07": {
"move_pct": 0.88,
"regime": "flat"
},
"2026-08-08": {
"move_pct": 0.16,
"regime": "flat"
},
"2026-08-09": {
"move_pct": 0.09,
"regime": "flat"
},
"2026-08-10": {
"move_pct": -1.63,
"regime": "trend"
},
"2026-08-11": {
"move_pct": -0.56,
"regime": "flat"
},
"2026-08-12": {
"move_pct": -0.02,
"regime": "flat"
},
"2026-08-13": {
"move_pct": -0.25,
"regime": "flat"
},
"2026-08-14": {
"move_pct": -0.67,
"regime": "flat"
},
"2026-08-15": {
"move_pct": 0.07,
"regime": "flat"
},
"2026-08-16": {
"move_pct": -0.35,
"regime": "flat"
},
"2026-08-17": {
"move_pct": 2.55,
"regime": "trend"
},
"2026-08-18": {
"move_pct": 0.21,
"regime": "flat"
},
"2026-08-19": {
"move_pct": 7.82,
"regime": "trend"
},
"2026-08-20": {
"move_pct": 5.07,
"regime": "trend"
},
"2026-08-21": {
"move_pct": 7.26,
"regime": "trend"
},
"2026-08-22": {
"move_pct": -1.67,
"regime": "trend"
},
"2026-08-23": {
"move_pct": 0.69,
"regime": "flat"
},
"2026-08-24": {
"move_pct": 1.61,
"regime": "trend"
},
"2026-08-25": {
"move_pct": -0.58,
"regime": "flat"
},
"2026-08-26": {
"move_pct": 0.63,
"regime": "flat"
},
"2026-08-27": {
"move_pct": 1.53,
"regime": "trend"
},
"2026-08-28": {
"move_pct": -2.91,
"regime": "trend"
},
"2026-08-29": {
"move_pct": 0.48,
"regime": "flat"
},
"2026-08-30": {
"move_pct": -0.7,
"regime": "flat"
},
"2026-08-31": {
"move_pct": 1.19,
"regime": "flat"
}
},
"grok_era": {
"ids_in_window": [
"grok-4.5",
"grok-4.6"
],
"n_grok45": 4520,
"n_grok46": 1570,
"n_grok45_rows": 4520,
"n_grok46_rows": 1570,
"first_grok46_slot": "2026-08-24T10:01:00+00:00",
"last_grok45_slot": "2026-08-24T09:01:00+00:00",
"first_grok45_slot": "2026-08-01T00:01:00+00:00",
"flip_at": "2026-08-24T09:22:00+00:00",
"flip_inside_window": true,
"prev_window": {
"n_grok45_rows": 6608,
"n_grok46_rows": 0
},
"merged_model_id": "grok-4.6",
"merged_label": "grok-4.6 (incl. 4.5-era, Aug 1-24)",
"lineage_ids": [
"grok-4.5",
"grok-4.6"
],
"lineage_source": "benchmarks.config.lineage_ids(head) for each of the 7 heads + MODEL_SUCCESSION[claude-opus-5=('claude-opus-4.8',), grok-4.6=('grok-4.5',), qwen-3.8-max=('qwen-3.7-max',)]",
"lineage_warnings": [],
"rows_remapped_a_plus_b": 11128,
"note": "window MONTH [2026-08-01T00:00:00+00:00, 2026-09-01T00:00:00+00:00): grok-4.5 rows 4520, grok-4.6 rows 1570; flip 2026-08-24T09:22:00+00:00 is inside this 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 1-24)'). See lineage_era for the same facts across all 7 heads.",
"cells_with_two_grok_rows": 0
},
"lineage_era": {
"window_key": "MONTH",
"heads": [
"claude-fable-5",
"claude-opus-5",
"deepseek-v4-pro",
"gemini-3.1-pro",
"gpt-5.6-sol",
"grok-4.6",
"qwen-3.8-max"
],
"lineage_map": {
"claude-opus-4.8": "claude-opus-5",
"grok-4.5": "grok-4.6",
"qwen-3.7-max": "qwen-3.8-max"
},
"legacy_by_head": {
"claude-fable-5": [],
"claude-opus-5": [
"claude-opus-4.8"
],
"deepseek-v4-pro": [],
"gemini-3.1-pro": [],
"gpt-5.6-sol": [],
"grok-4.6": [
"grok-4.5"
],
"qwen-3.8-max": [
"qwen-3.7-max"
]
},
"lineage_source": "benchmarks.config.lineage_ids(head) for each of the 7 heads + MODEL_SUCCESSION[claude-opus-5=('claude-opus-4.8',), grok-4.6=('grok-4.5',), qwen-3.8-max=('qwen-3.7-max',)]",
"lineage_resolved_heads": [
"claude-fable-5",
"claude-opus-5",
"deepseek-v4-pro",
"gemini-3.1-pro",
"gpt-5.6-sol",
"grok-4.6",
"qwen-3.8-max"
],
"lineage_warnings": [],
"per_head": {
"claude-fable-5": {
"legacy_ids": [],
"legacy_ids_seen_in_window": [],
"n_rows_remapped": {},
"n_rows_remapped_prev_window": {},
"by_legacy_id": {},
"first_legacy_slot": null,
"last_legacy_slot": null,
"first_successor_slot": "2026-08-01T00:01:00+00:00",
"last_successor_slot": "2026-08-31T23:01:00+00:00",
"n_rows_successor": 6095,
"spliced_in_window": false,
"label": "claude-fable-5"
},
"claude-opus-5": {
"legacy_ids": [
"claude-opus-4.8"
],
"legacy_ids_seen_in_window": [],
"n_rows_remapped": {
"claude-opus-4.8": 0
},
"n_rows_remapped_prev_window": {
"claude-opus-4.8": 6315
},
"by_legacy_id": {
"claude-opus-4.8": {
"n_rows_remapped": {
"MONTH": 0,
"prev_window": 6315
},
"first_slot": null,
"last_slot": null,
"first_slot_prev_window": "2026-07-15T18:01:00+00:00",
"last_slot_prev_window": "2026-07-30T08:01:00+00:00"
}
},
"first_legacy_slot": null,
"last_legacy_slot": null,
"first_successor_slot": "2026-08-01T00:01:00+00:00",
"last_successor_slot": "2026-08-31T23:01:00+00:00",
"n_rows_successor": 6095,
"spliced_in_window": false,
"label": "claude-opus-5"
},
"deepseek-v4-pro": {
"legacy_ids": [],
"legacy_ids_seen_in_window": [],
"n_rows_remapped": {},
"n_rows_remapped_prev_window": {},
"by_legacy_id": {},
"first_legacy_slot": null,
"last_legacy_slot": null,
"first_successor_slot": "2026-08-01T00:01:00+00:00",
"last_successor_slot": "2026-08-31T23:01:00+00:00",
"n_rows_successor": 6095,
"spliced_in_window": false,
"label": "deepseek-v4-pro"
},
"gemini-3.1-pro": {
"legacy_ids": [],
"legacy_ids_seen_in_window": [],
"n_rows_remapped": {},
"n_rows_remapped_prev_window": {},
"by_legacy_id": {},
"first_legacy_slot": null,
"last_legacy_slot": null,
"first_successor_slot": "2026-08-01T00:01:00+00:00",
"last_successor_slot": "2026-08-31T23:01:00+00:00",
"n_rows_successor": 6094,
"spliced_in_window": false,
"label": "gemini-3.1-pro"
},
"gpt-5.6-sol": {
"legacy_ids": [],
"legacy_ids_seen_in_window": [],
"n_rows_remapped": {},
"n_rows_remapped_prev_window": {},
"by_legacy_id": {},
"first_legacy_slot": null,
"last_legacy_slot": null,
"first_successor_slot": "2026-08-01T00:01:00+00:00",
"last_successor_slot": "2026-08-31T23:01:00+00:00",
"n_rows_successor": 6095,
"spliced_in_window": false,
"label": "gpt-5.6-sol"
},
"grok-4.6": {
"legacy_ids": [
"grok-4.5"
],
"legacy_ids_seen_in_window": [
"grok-4.5"
],
"n_rows_remapped": {
"grok-4.5": 4520
},
"n_rows_remapped_prev_window": {
"grok-4.5": 6608
},
"by_legacy_id": {
"grok-4.5": {
"n_rows_remapped": {
"MONTH": 4520,
"prev_window": 6608
},
"first_slot": "2026-08-01T00:01:00+00:00",
"last_slot": "2026-08-24T09:01:00+00:00",
"first_slot_prev_window": "2026-07-15T18:01:00+00:00",
"last_slot_prev_window": "2026-07-31T23:01:00+00:00"
}
},
"first_legacy_slot": "2026-08-01T00:01:00+00:00",
"last_legacy_slot": "2026-08-24T09:01:00+00:00",
"first_successor_slot": "2026-08-24T10:01:00+00:00",
"last_successor_slot": "2026-08-31T23:01:00+00:00",
"n_rows_successor": 1570,
"spliced_in_window": true,
"label": "grok-4.6 (incl. 4.5-era, Aug 1-24)"
},
"qwen-3.8-max": {
"legacy_ids": [
"qwen-3.7-max"
],
"legacy_ids_seen_in_window": [
"qwen-3.7-max"
],
"n_rows_remapped": {
"qwen-3.7-max": 870
},
"n_rows_remapped_prev_window": {
"qwen-3.7-max": 6609
},
"by_legacy_id": {
"qwen-3.7-max": {
"n_rows_remapped": {
"MONTH": 870,
"prev_window": 6609
},
"first_slot": "2026-08-01T00:01:00+00:00",
"last_slot": "2026-08-05T09:01:00+00:00",
"first_slot_prev_window": "2026-07-15T18:01:00+00:00",
"last_slot_prev_window": "2026-07-31T23:01:00+00:00"
}
},
"first_legacy_slot": "2026-08-01T00:01:00+00:00",
"last_legacy_slot": "2026-08-05T09:01:00+00:00",
"first_successor_slot": "2026-08-05T10:01:00+00:00",
"last_successor_slot": "2026-08-31T23:01:00+00:00",
"n_rows_successor": 5220,
"spliced_in_window": true,
"label": "qwen-3.8-max (incl. 3.7-era, Aug 1-5)"
}
},
"labels": {
"claude-fable-5": "claude-fable-5",
"claude-opus-5": "claude-opus-5",
"deepseek-v4-pro": "deepseek-v4-pro",
"gemini-3.1-pro": "gemini-3.1-pro",
"gpt-5.6-sol": "gpt-5.6-sol",
"grok-4.6": "grok-4.6 (incl. 4.5-era, Aug 1-24)",
"qwen-3.8-max": "qwen-3.8-max (incl. 3.7-era, Aug 1-5)"
},
"rows_remapped_window": 5390,
"rows_remapped_prev_window": 19532,
"note": "Only the 7 current models are tracked; earlier versions are folded into their successors' lineage. Every per-model aggregate in this pack sees the merged head id; the raw id survives in model_raw. Labels and era spans are generated from the pulled rows, not hardcoded.",
"cells_with_two_rows_per_head": {
"claude-fable-5": 0,
"claude-opus-5": 0,
"deepseek-v4-pro": 0,
"gemini-3.1-pro": 0,
"gpt-5.6-sol": 0,
"grok-4.6": 0,
"qwen-3.8-max": 0
}
},
"series_density": {
"1d": {
"days_with_slots": 31,
"expected_days": 31,
"full_daily_coverage": true,
"in_fh_gate": true,
"n_slots": 31,
"n_rows": 2170,
"days": [
"2026-08-01",
"2026-08-02",
"2026-08-03",
"2026-08-04",
"2026-08-05",
"2026-08-06",
"2026-08-07",
"2026-08-08",
"2026-08-09",
"2026-08-10",
"2026-08-11",
"2026-08-12",
"2026-08-13",
"2026-08-14",
"2026-08-15",
"2026-08-16",
"2026-08-17",
"2026-08-18",
"2026-08-19",
"2026-08-20",
"2026-08-21",
"2026-08-22",
"2026-08-23",
"2026-08-24",
"2026-08-25",
"2026-08-26",
"2026-08-27",
"2026-08-28",
"2026-08-29",
"2026-08-30",
"2026-08-31"
],
"missing_days": []
},
"1h": {
"days_with_slots": 31,
"expected_days": 31,
"full_daily_coverage": true,
"in_fh_gate": true,
"n_slots": 741,
"n_rows": 25935,
"days": [
"2026-08-01",
"2026-08-02",
"2026-08-03",
"2026-08-04",
"2026-08-05",
"2026-08-06",
"2026-08-07",
"2026-08-08",
"2026-08-09",
"2026-08-10",
"2026-08-11",
"2026-08-12",
"2026-08-13",
"2026-08-14",
"2026-08-15",
"2026-08-16",
"2026-08-17",
"2026-08-18",
"2026-08-19",
"2026-08-20",
"2026-08-21",
"2026-08-22",
"2026-08-23",
"2026-08-24",
"2026-08-25",
"2026-08-26",
"2026-08-27",
"2026-08-28",
"2026-08-29",
"2026-08-30",
"2026-08-31"
],
"missing_days": []
},
"1m": {
"days_with_slots": 10,
"expected_days": 31,
"full_daily_coverage": false,
"in_fh_gate": false,
"n_slots": 10,
"n_rows": 700,
"days": [
"2026-08-01",
"2026-08-23",
"2026-08-24",
"2026-08-25",
"2026-08-26",
"2026-08-27",
"2026-08-28",
"2026-08-29",
"2026-08-30",
"2026-08-31"
],
"missing_days": [
"2026-08-02",
"2026-08-03",
"2026-08-04",
"2026-08-05",
"2026-08-06",
"2026-08-07",
"2026-08-08",
"2026-08-09",
"2026-08-10",
"2026-08-11",
"2026-08-12",
"2026-08-13",
"2026-08-14",
"2026-08-15",
"2026-08-16",
"2026-08-17",
"2026-08-18",
"2026-08-19",
"2026-08-20",
"2026-08-21",
"2026-08-22"
]
},
"1w": {
"days_with_slots": 13,
"expected_days": 31,
"full_daily_coverage": false,
"in_fh_gate": false,
"n_slots": 13,
"n_rows": 910,
"days": [
"2026-08-03",
"2026-08-10",
"2026-08-17",
"2026-08-22",
"2026-08-23",
"2026-08-24",
"2026-08-25",
"2026-08-26",
"2026-08-27",
"2026-08-28",
"2026-08-29",
"2026-08-30",
"2026-08-31"
],
"missing_days": [
"2026-08-01",
"2026-08-02",
"2026-08-04",
"2026-08-05",
"2026-08-06",
"2026-08-07",
"2026-08-08",
"2026-08-09",
"2026-08-11",
"2026-08-12",
"2026-08-13",
"2026-08-14",
"2026-08-15",
"2026-08-16",
"2026-08-18",
"2026-08-19",
"2026-08-20",
"2026-08-21"
]
},
"4h": {
"days_with_slots": 31,
"expected_days": 31,
"full_daily_coverage": true,
"in_fh_gate": true,
"n_slots": 185,
"n_rows": 12939,
"days": [
"2026-08-01",
"2026-08-02",
"2026-08-03",
"2026-08-04",
"2026-08-05",
"2026-08-06",
"2026-08-07",
"2026-08-08",
"2026-08-09",
"2026-08-10",
"2026-08-11",
"2026-08-12",
"2026-08-13",
"2026-08-14",
"2026-08-15",
"2026-08-16",
"2026-08-17",
"2026-08-18",
"2026-08-19",
"2026-08-20",
"2026-08-21",
"2026-08-22",
"2026-08-23",
"2026-08-24",
"2026-08-25",
"2026-08-26",
"2026-08-27",
"2026-08-28",
"2026-08-29",
"2026-08-30",
"2026-08-31"
],
"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-4.5",
"grok-4.6",
"qwen-3.7-max",
"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": [],
"tracked_total": 10,
"canon_expected": {
"current": 7,
"archived_legacy": 4,
"tracked_total": 11
},
"matches_canon_11_7_4": false,
"roster_rule_w213": "7 models tracked (current line-up; earlier versions folded into their successors' lineage)",
"canon_expected_w213": {
"current": 7
},
"heads_w213": [
"claude-fable-5",
"claude-opus-5",
"deepseek-v4-pro",
"gemini-3.1-pro",
"gpt-5.6-sol",
"grok-4.6",
"qwen-3.8-max"
],
"matches_canon_7_current": true,
"heads_missing_in_window": [],
"heads_unexpected_in_window": [],
"roster_since": "2026-07-11T00:00:00+00:00",
"note": "current = merged-lineage model ids emitting inside the window (legacy ids are spliced into their successor); archived legacy = raw ids seen since 2026-07-11 that no longer emit in the window. Models retired before 2026-07-11 are not visible to this query. The canon checks are DESCRIPTIVE: if the data says otherwise, the data wins and is printed."
}
},
"by_week": {
"rows": [
{
"week": "W1",
"label": "Aug 3-9",
"n_scoreable": 3327,
"herd_share_pct": 99.4,
"agree_hit": 0.4039,
"disagree_hit": 0.6316,
"lift_pp": -22.8,
"unanimity_cells": 367,
"unanimity_n": 2380,
"unanimity_hit": 0.4147,
"published_lift_pp": -22.8,
"published_n_scoreable": 3327,
"published_unanimous_cells_ge6": 367
},
{
"week": "W2",
"label": "Aug 10-16",
"n_scoreable": 2929,
"herd_share_pct": 99.8,
"agree_hit": 0.4307,
"disagree_hit": 0.5,
"lift_pp": -6.9,
"unanimity_cells": 299,
"unanimity_n": 1961,
"unanimity_hit": 0.4238,
"published_lift_pp": -6.9,
"published_n_scoreable": 2929,
"published_unanimous_cells_ge6": 299
},
{
"week": "W3",
"label": "Aug 17-23",
"n_scoreable": 4357,
"herd_share_pct": 97.1,
"agree_hit": 0.5475,
"disagree_hit": 0.376,
"lift_pp": 17.1,
"unanimity_cells": 437,
"unanimity_n": 2887,
"unanimity_hit": 0.5539,
"published_lift_pp": 17.1,
"published_n_scoreable": 4357,
"published_unanimous_cells_ge6": 437
},
{
"week": "W4",
"label": "Aug 24-30",
"n_scoreable": 3884,
"herd_share_pct": 99.1,
"agree_hit": 0.4119,
"disagree_hit": 0.4722,
"lift_pp": -6.0,
"unanimity_cells": 408,
"unanimity_n": 2709,
"unanimity_hit": 0.3872,
"published_lift_pp": -6.0,
"published_n_scoreable": 3884,
"published_unanimous_cells_ge6": 408
},
{
"week": "month",
"label": "Aug 1-31",
"n_scoreable": 16089,
"herd_share_pct": 98.7,
"agree_hit": 0.4554,
"disagree_hit": 0.436,
"lift_pp": 1.9,
"unanimity_cells": 1672,
"unanimity_n": 10984,
"unanimity_hit": 0.4539
}
],
"source": "month_pull_split_by_slot_week (cutoff 2026-09-01T16:00:00+00:00)",
"unanimity_definition": "unanimity = a cell (slot x symbol x fh x tf) in which every directional forecast is on the same side and at least 6 models are present -- the same N the w201 herd block uses (counters/consensus.herd.unanimous_cells_ge6); the LOO peer floor is 3 peers.",
"published_cross_check": {
"W1": {
"n_scoreable": 3327,
"agree_n": 3308,
"agree_hit": 0.4039,
"agree_wilson_95": [
0.3873,
0.4207
],
"disagree_n": 19,
"disagree_hit": 0.6316,
"disagree_wilson_95": [
0.4104,
0.8085
],
"lift_pp": -22.8,
"herd_share_pct": 99.4,
"thin_or_tied": 715,
"by_margin_all": {
"2": {
"n": 3,
"hit_rate": 1.0
},
"3": {
"n": 404,
"hit_rate": 0.3045
},
"4": {
"n": 515,
"hit_rate": 0.433
},
"5": {
"n": 1140,
"hit_rate": 0.4991
},
"6": {
"n": 1246,
"hit_rate": 0.3355
}
},
"unanimous_cells_ge6": 367,
"source": "weeks.W1.consensus.loo_pooled (frozen cutoff)"
},
"W2": {
"n_scoreable": 2929,
"agree_n": 2923,
"agree_hit": 0.4307,
"agree_wilson_95": [
0.4129,
0.4488
],
"disagree_n": 6,
"disagree_hit": 0.5,
"disagree_wilson_95": [
0.1876,
0.8124
],
"lift_pp": -6.9,
"herd_share_pct": 99.8,
"thin_or_tied": 741,
"by_margin_all": {
"2": {
"n": 9,
"hit_rate": 0.6667
},
"3": {
"n": 428,
"hit_rate": 0.5023
},
"4": {
"n": 525,
"hit_rate": 0.3943
},
"5": {
"n": 792,
"hit_rate": 0.4104
},
"6": {
"n": 1169,
"hit_rate": 0.4328
}
},
"unanimous_cells_ge6": 299,
"source": "weeks.W2.consensus.loo_pooled (frozen cutoff)"
},
"W3": {
"n_scoreable": 4357,
"agree_n": 4232,
"agree_hit": 0.5475,
"agree_wilson_95": [
0.5325,
0.5624
],
"disagree_n": 125,
"disagree_hit": 0.376,
"disagree_wilson_95": [
0.296,
0.4634
],
"lift_pp": 17.1,
"herd_share_pct": 97.1,
"thin_or_tied": 773,
"by_margin_all": {
"2": {
"n": 81,
"hit_rate": 0.5926
},
"3": {
"n": 468,
"hit_rate": 0.5641
},
"4": {
"n": 700,
"hit_rate": 0.4886
},
"5": {
"n": 1128,
"hit_rate": 0.5443
},
"6": {
"n": 1855,
"hit_rate": 0.5655
}
},
"unanimous_cells_ge6": 437,
"source": "weeks.W3.consensus.loo_pooled (frozen cutoff)"
},
"W4": {
"n_scoreable": 3884,
"agree_n": 3848,
"agree_hit": 0.4119,
"agree_wilson_95": [
0.3964,
0.4275
],
"disagree_n": 36,
"disagree_hit": 0.4722,
"disagree_wilson_95": [
0.3199,
0.6299
],
"lift_pp": -6.0,
"herd_share_pct": 99.1,
"thin_or_tied": 692,
"by_margin_all": {
"2": {
"n": 21,
"hit_rate": 0.619
},
"3": {
"n": 456,
"hit_rate": 0.4934
},
"4": {
"n": 620,
"hit_rate": 0.4758
},
"5": {
"n": 924,
"hit_rate": 0.3593
},
"6": {
"n": 1827,
"hit_rate": 0.3941
}
},
"unanimous_cells_ge6": 408,
"source": "weeks.W4.consensus.loo_pooled (frozen cutoff)"
}
},
"weeks_matching_published_lift": [
"W1",
"W2",
"W3",
"W4"
]
},
"margin_by_week": {
"tiers": [
"3",
"4",
"5",
"6"
],
"rows": [
{
"margin": 3,
"weeks": {
"W1": {
"n": 404,
"hit_rate": 0.3045
},
"W2": {
"n": 428,
"hit_rate": 0.5023
},
"W3": {
"n": 468,
"hit_rate": 0.5641
},
"W4": {
"n": 456,
"hit_rate": 0.4934
}
},
"month": {
"n": 1964,
"hit_rate": 0.4593
}
},
{
"margin": 4,
"weeks": {
"W1": {
"n": 515,
"hit_rate": 0.433
},
"W2": {
"n": 525,
"hit_rate": 0.3943
},
"W3": {
"n": 700,
"hit_rate": 0.4886
},
"W4": {
"n": 620,
"hit_rate": 0.4758
}
},
"month": {
"n": 2630,
"hit_rate": 0.4513
}
},
{
"margin": 5,
"weeks": {
"W1": {
"n": 1140,
"hit_rate": 0.4991
},
"W2": {
"n": 792,
"hit_rate": 0.4104
},
"W3": {
"n": 1128,
"hit_rate": 0.5443
},
"W4": {
"n": 924,
"hit_rate": 0.3593
}
},
"month": {
"n": 4482,
"hit_rate": 0.4581
}
},
{
"margin": 6,
"weeks": {
"W1": {
"n": 1246,
"hit_rate": 0.3355
},
"W2": {
"n": 1169,
"hit_rate": 0.4328
},
"W3": {
"n": 1855,
"hit_rate": 0.5655
},
"W4": {
"n": 1827,
"hit_rate": 0.3941
}
},
"month": {
"n": 6664,
"hit_rate": 0.4511
}
}
],
"source": "month_pull_split_by_slot_week (cutoff 2026-09-01T16:00:00+00:00)"
},
"prior_window_consensus": {
"loo_pooled": {
"n_scoreable": 16993,
"agree_n": 16557,
"agree_hit": 0.4464,
"agree_wilson_95": [
0.4388,
0.454
],
"disagree_n": 436,
"disagree_hit": 0.5528,
"disagree_wilson_95": [
0.5058,
0.5988
],
"lift_pp": -10.6,
"herd_share_pct": 97.4,
"excluded_thin_tied": 2548,
"thin_or_tied": 2548,
"min_peers": 3
},
"by_margin": [
{
"peers": 3,
"n": 1609,
"hit_rate": 0.427,
"wilson_95": [
0.403,
0.4513
]
},
{
"peers": 4,
"n": 2394,
"hit_rate": 0.4353,
"wilson_95": [
0.4155,
0.4552
]
},
{
"peers": 5,
"n": 3628,
"hit_rate": 0.4308,
"wilson_95": [
0.4148,
0.447
]
},
{
"peers": "unanimity",
"n": 11531,
"hit_rate": 0.4562,
"wilson_95": [
0.4471,
0.4653
],
"cells": 1456,
"min_models": 6
}
],
"loo_by_cell": {
"fh_1h_tf_15m": {
"n_scoreable": 2746,
"agree_n": 2706,
"agree_hit": 0.4579,
"disagree_n": 40,
"disagree_hit": 0.425,
"lift_pp": 3.3,
"n_subjects": 3338
},
"fh_1h_tf_1h": {
"n_scoreable": 7963,
"agree_n": 7684,
"agree_hit": 0.4641,
"disagree_n": 279,
"disagree_hit": 0.5663,
"lift_pp": -10.2,
"n_subjects": 9024
},
"fh_4h_tf_1h": {
"n_scoreable": 3198,
"agree_n": 3129,
"agree_hit": 0.4394,
"disagree_n": 69,
"disagree_hit": 0.5507,
"lift_pp": -11.1,
"n_subjects": 3567
},
"fh_4h_tf_4h": {
"n_scoreable": 2170,
"agree_n": 2138,
"agree_hit": 0.4415,
"disagree_n": 32,
"disagree_hit": 0.5625,
"lift_pp": -12.1,
"n_subjects": 2573
},
"fh_1d_tf_1d": {
"n_scoreable": 408,
"agree_n": 399,
"agree_hit": 0.2782,
"disagree_n": 9,
"disagree_hit": 0.6667,
"lift_pp": -38.8,
"n_subjects": 473
},
"fh_1d_tf_4h": {
"n_scoreable": 508,
"agree_n": 501,
"agree_hit": 0.3114,
"disagree_n": 7,
"disagree_hit": 0.5714,
"lift_pp": -26.0,
"n_subjects": 566
}
},
"loo_by_model": [
{
"n_scoreable": 2476,
"agree_n": 2466,
"agree_hit": 0.4428,
"disagree_n": 10,
"disagree_hit": 0.6,
"lift_pp": -15.7,
"model": "claude-fable-5",
"insufficient": false
},
{
"n_scoreable": 2916,
"agree_n": 2852,
"agree_hit": 0.4428,
"disagree_n": 64,
"disagree_hit": 0.5781,
"lift_pp": -13.5,
"model": "claude-opus-5",
"insufficient": false
},
{
"n_scoreable": 1701,
"agree_n": 1651,
"agree_hit": 0.4525,
"disagree_n": 50,
"disagree_hit": 0.36,
"lift_pp": 9.2,
"model": "deepseek-v4-pro",
"insufficient": false
},
{
"n_scoreable": 2507,
"agree_n": 2437,
"agree_hit": 0.4506,
"disagree_n": 70,
"disagree_hit": 0.5571,
"lift_pp": -10.7,
"model": "gemini-3.1-pro",
"insufficient": false
},
{
"n_scoreable": 2712,
"agree_n": 2696,
"agree_hit": 0.4414,
"disagree_n": 16,
"disagree_hit": 0.5625,
"lift_pp": -12.1,
"model": "gpt-5.6-sol",
"insufficient": false
},
{
"n_scoreable": 1977,
"agree_n": 1907,
"agree_hit": 0.4473,
"disagree_n": 70,
"disagree_hit": 0.5714,
"lift_pp": -12.4,
"model": "grok-4.6",
"insufficient": false
},
{
"n_scoreable": 2704,
"agree_n": 2548,
"agree_hit": 0.4505,
"disagree_n": 156,
"disagree_hit": 0.5897,
"lift_pp": -13.9,
"model": "qwen-3.8-max",
"insufficient": false
}
],
"unanimity": {
"cells": 1456,
"n_calls": 11531,
"hit_rate": 0.4562,
"wilson_95": [
0.4471,
0.4653
],
"insufficient": false,
"min_models": 6
},
"unanimity_definition": "unanimity = a cell (slot x symbol x fh x tf) in which every directional forecast is on the same side and at least 6 models are present -- the same N the w201 herd block uses (counters/consensus.herd.unanimous_cells_ge6); the LOO peer floor is 3 peers.",
"n_cells": 5077,
"definitions": {
"n_scoreable": "agree_n + disagree_n (subjects with >= min_peers directional peers, no tie)",
"excluded_thin_tied": "subjects dropped because the cell had < min_peers peers or a tied majority",
"lift_pp": "agree_hit - disagree_hit, percentage points",
"herd_share_pct": "agree_n / n_scoreable * 100",
"by_margin": "peers = peers on the majority side (capped at 6 as in w201); 'unanimity' = the w201 herd rule"
}
},
"verdict_block": {
"rule": "Verdict rule: each row states one reading of the month; a week agrees when it shows the same reading (same sign, or the same name). Confirmed = 3 or 4 of the 4 weeks agree with the month; Inconclusive = 2; Rejected = 0 or 1. Weekly cells come from the frozen weekly packs (7-line roster, lineage spliced); the month is the pooled Aug 1-31 pull.",
"header": [
"Metric",
"W1 Aug 3-9",
"W2 Aug 10-16",
"W3 Aug 17-23",
"W4 Aug 24-30",
"Month",
"Verdict"
],
"widths_mm": [
36,
23,
23,
23,
23,
18,
28
],
"weeks_source": "frozen weekly packs (weeks.W1..W4 of the monthly pack, 7-line roster, lineage spliced)",
"month_source": "the pooled Aug 1-31 pull (month block)",
"rows": [
{
"metric": "LOO lift, pp",
"weeks": {
"W1": {
"value": "-22.8",
"reading": "negative"
},
"W2": {
"value": "-6.9",
"reading": "negative"
},
"W3": {
"value": "+17.1",
"reading": "positive"
},
"W4": {
"value": "-6.0",
"reading": "negative"
}
},
"month": {
"value": "+1.9",
"reading": "positive"
},
"agree_count": 1,
"verdict": "Rejected (1/4)"
},
{
"metric": "Herd share >= 95%",
"weeks": {
"W1": {
"value": "99.4%",
"reading": ">= 95%"
},
"W2": {
"value": "99.8%",
"reading": ">= 95%"
},
"W3": {
"value": "97.1%",
"reading": ">= 95%"
},
"W4": {
"value": "99.1%",
"reading": ">= 95%"
}
},
"month": {
"value": "98.7%",
"reading": ">= 95%"
},
"agree_count": 4,
"verdict": "Confirmed (4/4)"
},
{
"metric": "Unanimity hit vs field hit",
"weeks": {
"W1": {
"value": "-8.9",
"reading": "negative"
},
"W2": {
"value": "-0.5",
"reading": "negative"
},
"W3": {
"value": "+2.5",
"reading": "positive"
},
"W4": {
"value": "-3.5",
"reading": "negative"
}
},
"month": {
"value": "-1.2",
"reading": "negative"
},
"agree_count": 3,
"verdict": "Confirmed (3/4)"
}
]
}
},
"herd_note": "1,672 cells had 6+ directional models on one side; unanimous_cells_ge6 per monthly 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 monthly window)",
"text": "Herding ran at 98.7% of leave-one-out scoreable calls across the month and the pooled lift came in at +1.9pp: agree-hit 45.5% against 43.6% on disagreement, on a disagree bucket of n=211.",
"evidence_level": "observation"
},
"key_findings": [
"**OBSERVATION -- 98.7% of the field's leave-one-out calls ran with the herd over the month.** Of 16,089 LOO-scoreable directional calls, 15,878 (98.7%) sided with peers' majority; agree-hit was 45.5% [44.8%, 46.3%] vs. 43.6% [37.1%, 50.3%] for the 211 disagreers, a pooled lift of +1.9pp. Descriptive, one month, not a claim of a durable edge.",
"**The margin curve read 45.9% / 45.1% / 45.8% / 45.1% from 3 peers to unanimity.** 0 of the 4 tiers sit above the 46.3% field base. The same four tiers read 42.7% / 43.5% / 43.1% / 45.6% in the prior window Jul 11-31 (partial), on n=1,609 / 2,394 / 3,628 / 11,531 there.",
"**Week by week the lift changed sign 2 times.** The four published weeks ran Aug 3-9 -22.8pp, Aug 10-16 -6.9pp, Aug 17-23 +17.1pp, Aug 24-30 -6.0pp; the month pools them at +1.9pp and the prior window read -10.6pp. The month is a weighted pool of the four weeks, not their mean: the weekly range is -22.8pp to +17.1pp.",
"**1,672 cells had 6+ directional models on one side across the month.** The 5 worst sat at 69.9-71.7 mean confidence and resolved 0.0%-0.0%; the prior window Jul 11-31 (partial) had 1,456 such cells, hitting 45.6% on n=11,531."
],
"market_check": {
"label": "MARKET CHECK · OBSERVATION (August against the partial July)",
"text": "OBSERVATION — mean |1d move| 1.54% -> 1.98% (+29% rel), BTC realized vol 33.5% -> 36.9% (ann., hourly); field directional accuracy 45.0% -> 46.3% (+1.4pp), 5 of 7 models improved, sim win-rate up for 7 of 7, field sim PnL -$3095.02 -> -$1054.79 (Jul 11-31 (partial) vs Aug 1-31; descriptive, one pair of windows, not a claim).",
"robustness": "Robustness: raw price-sign accuracy 46.8% -> 46.0% — the same direction as the trade-based hit rule.",
"weeks": {
"A": {
"days": "2026-07-01..2026-07-31",
"slots": [
"2026-07-01T00:00:00+00:00",
"2026-08-01T00:00:00+00:00"
],
"cutoff": "2026-08-01T16:00:00+00:00",
"partial": true
},
"B": {
"days": "2026-08-01..2026-08-31",
"slots": [
"2026-08-01T00:00:00+00:00",
"2026-09-01T00:00:00+00:00"
],
"cutoff": "2026-09-01T16:00:00+00:00"
}
},
"price_source": "crypto_spot/1h@:00",
"accuracy_field": {
"A": {
"n": 19541,
"hit_rate": 0.4498,
"wilson_95": [
0.4429,
0.4568
]
},
"B": {
"n": 19383,
"hit_rate": 0.4633,
"wilson_95": [
0.4563,
0.4703
]
},
"delta_pp": 1.4
},
"accuracy_per_model": [
{
"model": "claude-opus-5",
"model_label": "claude-opus-5",
"A": {
"n": 3448,
"hit_rate": 0.4492,
"wilson_95": [
0.4327,
0.4659
]
},
"B": {
"n": 2281,
"hit_rate": 0.4779,
"wilson_95": [
0.4574,
0.4984
]
},
"delta_pp": 2.9
},
{
"model": "gemini-3.1-pro",
"model_label": "gemini-3.1-pro",
"A": {
"n": 2813,
"hit_rate": 0.4511,
"wilson_95": [
0.4328,
0.4696
]
},
"B": {
"n": 3033,
"hit_rate": 0.4761,
"wilson_95": [
0.4584,
0.4939
]
},
"delta_pp": 2.5
},
{
"model": "claude-fable-5",
"model_label": "claude-fable-5",
"A": {
"n": 2690,
"hit_rate": 0.445,
"wilson_95": [
0.4263,
0.4638
]
},
"B": {
"n": 2812,
"hit_rate": 0.4673,
"wilson_95": [
0.4489,
0.4858
]
},
"delta_pp": 2.2
},
{
"model": "gpt-5.6-sol",
"model_label": "gpt-5.6-sol",
"A": {
"n": 3085,
"hit_rate": 0.4431,
"wilson_95": [
0.4257,
0.4607
]
},
"B": {
"n": 2970,
"hit_rate": 0.4586,
"wilson_95": [
0.4407,
0.4765
]
},
"delta_pp": 1.6
},
{
"model": "deepseek-v4-pro",
"model_label": "deepseek-v4-pro",
"A": {
"n": 1869,
"hit_rate": 0.4505,
"wilson_95": [
0.4281,
0.4731
]
},
"B": {
"n": 2578,
"hit_rate": 0.4554,
"wilson_95": [
0.4362,
0.4747
]
},
"delta_pp": 0.5
},
{
"model": "grok-4.6",
"model_label": "grok-4.6 (incl. 4.5-era, Aug 1-24)",
"A": {
"n": 2265,
"hit_rate": 0.453,
"wilson_95": [
0.4326,
0.4735
]
},
"B": {
"n": 2708,
"hit_rate": 0.4527,
"wilson_95": [
0.4341,
0.4715
]
},
"delta_pp": -0.0
},
{
"model": "qwen-3.8-max",
"model_label": "qwen-3.8-max (incl. 3.7-era, Aug 1-5)",
"A": {
"n": 3371,
"hit_rate": 0.4568,
"wilson_95": [
0.4401,
0.4737
]
},
"B": {
"n": 3001,
"hit_rate": 0.4565,
"wilson_95": [
0.4388,
0.4744
]
},
"delta_pp": -0.0
}
],
"raw_sign_field": {
"A": {
"n_scored": 17840,
"hit_rate": 0.4684
},
"B": {
"n_scored": 19304,
"hit_rate": 0.4599
}
},
"abs_move_1d": {
"A": 1.536,
"B": 1.979,
"delta_pct_rel": 28.8
},
"btc_realized_vol_ann_pct_hourly": {
"A": 33.46,
"B": 36.92
},
"trading_field": {
"A": {
"n_trades": 19541,
"wr": 0.3188,
"pnl_net_usd": -3095.02,
"pnl_gross_usd": -1140.92
},
"B": {
"n_trades": 19383,
"wr": 0.3576,
"pnl_net_usd": -1054.79,
"pnl_gross_usd": 883.51
}
},
"verdict": {
"models_with_hit_improved": "5 of 7",
"models_with_rawsign_improved": "1 of 7",
"models_with_wr_improved": "7 of 7",
"field_hit_delta_pp": 1.4,
"field_pnl_net": {
"A": -3095.02,
"B": -1054.79
},
"abs_move_delta": {
"1h": {
"A": 0.288,
"B": 0.325,
"delta_pct_rel": 12.8
},
"4h": {
"A": 0.557,
"B": 0.724,
"delta_pct_rel": 30.0
},
"1d": {
"A": 1.536,
"B": 1.979,
"delta_pct_rel": 28.8
}
},
"btc_vol_ann_pct": {
"A": 33.46,
"B": 36.92
},
"note": "descriptive, two adjacent windows; 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 for all 7 heads: [('claude-opus-4.8', 'claude-opus-5'), ('grok-4.5', 'grok-4.6'), ('qwen-3.7-max', 'qwen-3.8-max')]. Window MONTH remapped 5390 row(s), the previous window 19532. Labels are generated from the pulled rows (see lineage_era).",
"prior_window_partial": true
},
"why_it_matters": "MarketMania publishes forecasts from 7 frontier model lines into the same slots every hour; 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. The four weekly issues inside this month answered it with lifts of Aug 3-9 -22.8pp, Aug 10-16 -6.9pp, Aug 17-23 +17.1pp, Aug 24-30 -6.0pp, and the pooled month reads +1.9pp. This monthly issue asks the same question at monthly n, across four weeks the pack labels as 2 calm and 2 volatile, using the same leave-one-out design built to avoid 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. 2 model lines are lineage splices this month (grok-4.6 (incl. 4.5-era, Aug 1-24); qwen-3.8-max (incl. 3.7-era, Aug 1-5)), and a spliced line never counts as two peers in a cell. The month is one pull at one cutoff (Mon Sep 1, 2026, 16:00 UTC); the four weekly blocks quoted alongside it are the published issues at their own frozen cutoffs.",
"practical_implications": [
"0 of the 4 margin tiers sit above the 46.3% field base this month; the four tier rates span 45.1% to 45.9%, a spread of +0.8pp, and the four weeks inside the month ran lifts of -22.8pp to +17.1pp against a pooled monthly lift of +1.9pp.",
"The pooled monthly lift is +1.9pp while the four weeks ran -22.8pp to +17.1pp and the prior window read -10.6pp.",
"The field base moved +1.4pp from the partial July to August (45.0% -> 46.3%), and every consensus tier is measured against that base."
],
"limitations": [
"The prior window is a PARTIAL month: forecasts run only from Jul 11 and the audited daily candles only from Jul 15, so every \"prior\" column, the gray series of the page-1 chart and the market-check A column describe Jul 11-31 (partial), not a full July. It is window A of the same pull, slots 2026-07-01 to 2026-08-01, forecasts counted from Jul 11 (first slot actually seen 2026-07-15 18:01 UTC) and audited candles from Jul 15, matured at the prior cutoff Sat Aug 1, 2026, 16:00 UTC under the same engine, the same FH gate and the same hit rule as August -- 19,541 directional calls against 19,383 in the month. Month-over-month deltas are a step between two windows of different length.",
"Two cutoffs live in this issue. The month is frozen at Mon Sep 1, 2026, 16:00 UTC; the four weekly blocks are frozen at their own published cutoffs (W1 Aug 3-9, W2 Aug 10-16, W3 Aug 17-23, W4 Aug 24-30). A forecast still pending at its week's cutoff is immature in the weekly pack and mature in the month pull, so the month does not equal the sum of the weeks (19,383 vs 17,418 directional, plus 1,965 on the 3 remainder days) -- that gap is arithmetic, not a data problem.",
"Observations are not independent: the same models watch overlapping symbol / FH / TF cells hour after hour for 31 days. Wilson intervals here are descriptive, not inferential, and pooling a month does not make them inferential.",
"The pooled lift rests on 211 disagreeing calls; per-model counts run 4-89. 3,294 calls (17.0% of the 19,383-call mature directional universe) were thin or tied and are excluded from every consensus table. Correlational, not causal: LOO removes self-match bias, not the confound that an easy slot lifts agreement and hit together.",
"This window ran 9 trend days of 31, and the pack's own volatility rule splits the four Mon-Sun weeks as W1 calm, W2 calm, W3 volatile, W4 volatile; every month-over-month comparison is a comparison across regimes as well as across windows. Series density and lineage notes (monthly #1): (1) rolling-1w forecasts emit daily since Aug 22, 2026 and rolling-1M daily since Aug 23, 2026, so across this 31-day window daily coverage is PARTIAL as measured -- the 1w series has slots on 13 of 31 days and the 1M series on 10 of 31 days; both sit outside this report's FH gate (1h/4h/1d) and appear only in the exclusions counter (1w 906, 1M 698). (2) 2 model lines are lineage splices inside this window: grok-4.6 (incl. 4.5-era, Aug 1-24); qwen-3.8-max (incl. 3.7-era, Aug 1-5). (3) Every per-model aggregate in this issue sees the merged head id; the raw id is kept in the pack, and 5,390 rows were remapped inside the month against 19,532 in the prior window.",
"Market-state row is computed on a single-exchange (binance) daily candle series, as in the weekly series after the audit that found the raw table mixes two exchanges. The prior-month block is PARTIAL: audited daily candles start 2026-07-15, so it covers 17 days (Jul 15-31), and the TOP5 volume figure compares a 31-day sum with a 17-day sum -- a level difference, not a like-for-like change.",
"Research-to-date counter: directional forecasts resolved inside the FH gate (1h/4h/1d) since Jul 11, counted at the issue cutoff (Mon Sep 1, 2026, 16:00 UTC) -- the D2 definition pinned in the weekly series; this issue prints 39,073. The pack carries no prior-cutoff pin for a monthly window (the prior-cutoff pin is null by construction), so the control this issue is the weekly reproduction: the W4 block of this pack reproduces the published weekly #4 on 7 of 7 checks. Counter deltas against pre-wave-3 issues remain definitional."
],
"notes": {
"series_density_and_lineage_monthly1": "Series density and lineage notes (monthly #1): (1) rolling-1w forecasts emit daily since Aug 22, 2026 and rolling-1M daily since Aug 23, 2026, so across this 31-day window daily coverage is PARTIAL as measured -- the 1w series has slots on 13 of 31 days and the 1M series on 10 of 31 days; both sit outside this report's FH gate (1h/4h/1d) and appear only in the exclusions counter (1w 906, 1M 698). (2) 2 model lines are lineage splices inside this window: grok-4.6 (incl. 4.5-era, Aug 1-24); qwen-3.8-max (incl. 3.7-era, Aug 1-5). (3) Every per-model aggregate in this issue sees the merged head id; the raw id is kept in the pack, and 5,390 rows were remapped inside the month against 19,532 in the prior window.",
"grok_era": {
"ids_in_window": [
"grok-4.5",
"grok-4.6"
],
"n_grok45": 4520,
"n_grok46": 1570,
"n_grok45_rows": 4520,
"n_grok46_rows": 1570,
"first_grok46_slot": "2026-08-24T10:01:00+00:00",
"last_grok45_slot": "2026-08-24T09:01:00+00:00",
"first_grok45_slot": "2026-08-01T00:01:00+00:00",
"flip_at": "2026-08-24T09:22:00+00:00",
"flip_inside_window": true,
"prev_window": {
"n_grok45_rows": 6608,
"n_grok46_rows": 0
},
"merged_model_id": "grok-4.6",
"merged_label": "grok-4.6 (incl. 4.5-era, Aug 1-24)",
"lineage_ids": [
"grok-4.5",
"grok-4.6"
],
"lineage_source": "benchmarks.config.lineage_ids(head) for each of the 7 heads + MODEL_SUCCESSION[claude-opus-5=('claude-opus-4.8',), grok-4.6=('grok-4.5',), qwen-3.8-max=('qwen-3.7-max',)]",
"lineage_warnings": [],
"rows_remapped_a_plus_b": 11128,
"note": "window MONTH [2026-08-01T00:00:00+00:00, 2026-09-01T00:00:00+00:00): grok-4.5 rows 4520, grok-4.6 rows 1570; flip 2026-08-24T09:22:00+00:00 is inside this 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 1-24)'). See lineage_era for the same facts across all 7 heads.",
"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-4.5",
"grok-4.6",
"qwen-3.7-max",
"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": [],
"tracked_total": 10,
"canon_expected": {
"current": 7,
"archived_legacy": 4,
"tracked_total": 11
},
"matches_canon_11_7_4": false,
"roster_rule_w213": "7 models tracked (current line-up; earlier versions folded into their successors' lineage)",
"canon_expected_w213": {
"current": 7
},
"heads_w213": [
"claude-fable-5",
"claude-opus-5",
"deepseek-v4-pro",
"gemini-3.1-pro",
"gpt-5.6-sol",
"grok-4.6",
"qwen-3.8-max"
],
"matches_canon_7_current": true,
"heads_missing_in_window": [],
"heads_unexpected_in_window": [],
"roster_since": "2026-07-11T00:00:00+00:00",
"note": "current = merged-lineage model ids emitting inside the window (legacy ids are spliced into their successor); archived legacy = raw ids seen since 2026-07-11 that no longer emit in the window. Models retired before 2026-07-11 are not visible to this query. The canon checks are DESCRIPTIVE: if the data says otherwise, the data wins and is printed."
},
"series_density": {
"1d": {
"days_with_slots": 31,
"expected_days": 31,
"full_daily_coverage": true,
"in_fh_gate": true,
"n_slots": 31,
"n_rows": 2170,
"days": [
"2026-08-01",
"2026-08-02",
"2026-08-03",
"2026-08-04",
"2026-08-05",
"2026-08-06",
"2026-08-07",
"2026-08-08",
"2026-08-09",
"2026-08-10",
"2026-08-11",
"2026-08-12",
"2026-08-13",
"2026-08-14",
"2026-08-15",
"2026-08-16",
"2026-08-17",
"2026-08-18",
"2026-08-19",
"2026-08-20",
"2026-08-21",
"2026-08-22",
"2026-08-23",
"2026-08-24",
"2026-08-25",
"2026-08-26",
"2026-08-27",
"2026-08-28",
"2026-08-29",
"2026-08-30",
"2026-08-31"
],
"missing_days": []
},
"1h": {
"days_with_slots": 31,
"expected_days": 31,
"full_daily_coverage": true,
"in_fh_gate": true,
"n_slots": 741,
"n_rows": 25935,
"days": [
"2026-08-01",
"2026-08-02",
"2026-08-03",
"2026-08-04",
"2026-08-05",
"2026-08-06",
"2026-08-07",
"2026-08-08",
"2026-08-09",
"2026-08-10",
"2026-08-11",
"2026-08-12",
"2026-08-13",
"2026-08-14",
"2026-08-15",
"2026-08-16",
"2026-08-17",
"2026-08-18",
"2026-08-19",
"2026-08-20",
"2026-08-21",
"2026-08-22",
"2026-08-23",
"2026-08-24",
"2026-08-25",
"2026-08-26",
"2026-08-27",
"2026-08-28",
"2026-08-29",
"2026-08-30",
"2026-08-31"
],
"missing_days": []
},
"1m": {
"days_with_slots": 10,
"expected_days": 31,
"full_daily_coverage": false,
"in_fh_gate": false,
"n_slots": 10,
"n_rows": 700,
"days": [
"2026-08-01",
"2026-08-23",
"2026-08-24",
"2026-08-25",
"2026-08-26",
"2026-08-27",
"2026-08-28",
"2026-08-29",
"2026-08-30",
"2026-08-31"
],
"missing_days": [
"2026-08-02",
"2026-08-03",
"2026-08-04",
"2026-08-05",
"2026-08-06",
"2026-08-07",
"2026-08-08",
"2026-08-09",
"2026-08-10",
"2026-08-11",
"2026-08-12",
"2026-08-13",
"2026-08-14",
"2026-08-15",
"2026-08-16",
"2026-08-17",
"2026-08-18",
"2026-08-19",
"2026-08-20",
"2026-08-21",
"2026-08-22"
]
},
"1w": {
"days_with_slots": 13,
"expected_days": 31,
"full_daily_coverage": false,
"in_fh_gate": false,
"n_slots": 13,
"n_rows": 910,
"days": [
"2026-08-03",
"2026-08-10",
"2026-08-17",
"2026-08-22",
"2026-08-23",
"2026-08-24",
"2026-08-25",
"2026-08-26",
"2026-08-27",
"2026-08-28",
"2026-08-29",
"2026-08-30",
"2026-08-31"
],
"missing_days": [
"2026-08-01",
"2026-08-02",
"2026-08-04",
"2026-08-05",
"2026-08-06",
"2026-08-07",
"2026-08-08",
"2026-08-09",
"2026-08-11",
"2026-08-12",
"2026-08-13",
"2026-08-14",
"2026-08-15",
"2026-08-16",
"2026-08-18",
"2026-08-19",
"2026-08-20",
"2026-08-21"
]
},
"4h": {
"days_with_slots": 31,
"expected_days": 31,
"full_daily_coverage": true,
"in_fh_gate": true,
"n_slots": 185,
"n_rows": 12939,
"days": [
"2026-08-01",
"2026-08-02",
"2026-08-03",
"2026-08-04",
"2026-08-05",
"2026-08-06",
"2026-08-07",
"2026-08-08",
"2026-08-09",
"2026-08-10",
"2026-08-11",
"2026-08-12",
"2026-08-13",
"2026-08-14",
"2026-08-15",
"2026-08-16",
"2026-08-17",
"2026-08-18",
"2026-08-19",
"2026-08-20",
"2026-08-21",
"2026-08-22",
"2026-08-23",
"2026-08-24",
"2026-08-25",
"2026-08-26",
"2026-08-27",
"2026-08-28",
"2026-08-29",
"2026-08-30",
"2026-08-31"
],
"missing_days": []
}
},
"month_over_month_note": "Month-over-month columns compare back-to-back windows: Jul 11-31 (partial) vs Aug 1-31 (this issue). \"Was\" values are the pack's own prior-window block -- window A of the same pull, same engine, same FH gate and same hit rule, matured at the prior cutoff Sat Aug 1, 2026, 16:00 UTC. The prior month is PARTIAL: forecasts run from Jul 11 and audited candles from Jul 15, so it holds 19,541 directional calls against 19,383 in August. 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 (Mon Sep 1, 2026, 16:00 UTC) -- the D2 definition pinned in the weekly series; this issue prints 39,073. The pack carries no prior-cutoff pin for a monthly window (the prior-cutoff pin is null by construction), so the control this issue is the weekly reproduction: the W4 block of this pack reproduces the published weekly #4 on 7 of 7 checks. Counter deltas against pre-wave-3 issues remain definitional.",
"market_state_audit": "Market-state row is computed on a single-exchange (binance) daily candle series, as in the weekly series after the audit that found the raw table mixes two exchanges. The prior-month block is PARTIAL: audited daily candles start 2026-07-15, so it covers 17 days (Jul 15-31), and the TOP5 volume figure compares a 31-day sum with a 17-day sum -- a level difference, not a like-for-like change.",
"engine": "Engine 1.1 has powered the sandbox since Aug 18, i.e. inside 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",
"lineage_era": {
"window_key": "MONTH",
"heads": [
"claude-fable-5",
"claude-opus-5",
"deepseek-v4-pro",
"gemini-3.1-pro",
"gpt-5.6-sol",
"grok-4.6",
"qwen-3.8-max"
],
"lineage_map": {
"claude-opus-4.8": "claude-opus-5",
"grok-4.5": "grok-4.6",
"qwen-3.7-max": "qwen-3.8-max"
},
"legacy_by_head": {
"claude-fable-5": [],
"claude-opus-5": [
"claude-opus-4.8"
],
"deepseek-v4-pro": [],
"gemini-3.1-pro": [],
"gpt-5.6-sol": [],
"grok-4.6": [
"grok-4.5"
],
"qwen-3.8-max": [
"qwen-3.7-max"
]
},
"lineage_source": "benchmarks.config.lineage_ids(head) for each of the 7 heads + MODEL_SUCCESSION[claude-opus-5=('claude-opus-4.8',), grok-4.6=('grok-4.5',), qwen-3.8-max=('qwen-3.7-max',)]",
"lineage_resolved_heads": [
"claude-fable-5",
"claude-opus-5",
"deepseek-v4-pro",
"gemini-3.1-pro",
"gpt-5.6-sol",
"grok-4.6",
"qwen-3.8-max"
],
"lineage_warnings": [],
"per_head": {
"claude-fable-5": {
"legacy_ids": [],
"legacy_ids_seen_in_window": [],
"n_rows_remapped": {},
"n_rows_remapped_prev_window": {},
"by_legacy_id": {},
"first_legacy_slot": null,
"last_legacy_slot": null,
"first_successor_slot": "2026-08-01T00:01:00+00:00",
"last_successor_slot": "2026-08-31T23:01:00+00:00",
"n_rows_successor": 6095,
"spliced_in_window": false,
"label": "claude-fable-5"
},
"claude-opus-5": {
"legacy_ids": [
"claude-opus-4.8"
],
"legacy_ids_seen_in_window": [],
"n_rows_remapped": {
"claude-opus-4.8": 0
},
"n_rows_remapped_prev_window": {
"claude-opus-4.8": 6315
},
"by_legacy_id": {
"claude-opus-4.8": {
"n_rows_remapped": {
"MONTH": 0,
"prev_window": 6315
},
"first_slot": null,
"last_slot": null,
"first_slot_prev_window": "2026-07-15T18:01:00+00:00",
"last_slot_prev_window": "2026-07-30T08:01:00+00:00"
}
},
"first_legacy_slot": null,
"last_legacy_slot": null,
"first_successor_slot": "2026-08-01T00:01:00+00:00",
"last_successor_slot": "2026-08-31T23:01:00+00:00",
"n_rows_successor": 6095,
"spliced_in_window": false,
"label": "claude-opus-5"
},
"deepseek-v4-pro": {
"legacy_ids": [],
"legacy_ids_seen_in_window": [],
"n_rows_remapped": {},
"n_rows_remapped_prev_window": {},
"by_legacy_id": {},
"first_legacy_slot": null,
"last_legacy_slot": null,
"first_successor_slot": "2026-08-01T00:01:00+00:00",
"last_successor_slot": "2026-08-31T23:01:00+00:00",
"n_rows_successor": 6095,
"spliced_in_window": false,
"label": "deepseek-v4-pro"
},
"gemini-3.1-pro": {
"legacy_ids": [],
"legacy_ids_seen_in_window": [],
"n_rows_remapped": {},
"n_rows_remapped_prev_window": {},
"by_legacy_id": {},
"first_legacy_slot": null,
"last_legacy_slot": null,
"first_successor_slot": "2026-08-01T00:01:00+00:00",
"last_successor_slot": "2026-08-31T23:01:00+00:00",
"n_rows_successor": 6094,
"spliced_in_window": false,
"label": "gemini-3.1-pro"
},
"gpt-5.6-sol": {
"legacy_ids": [],
"legacy_ids_seen_in_window": [],
"n_rows_remapped": {},
"n_rows_remapped_prev_window": {},
"by_legacy_id": {},
"first_legacy_slot": null,
"last_legacy_slot": null,
"first_successor_slot": "2026-08-01T00:01:00+00:00",
"last_successor_slot": "2026-08-31T23:01:00+00:00",
"n_rows_successor": 6095,
"spliced_in_window": false,
"label": "gpt-5.6-sol"
},
"grok-4.6": {
"legacy_ids": [
"grok-4.5"
],
"legacy_ids_seen_in_window": [
"grok-4.5"
],
"n_rows_remapped": {
"grok-4.5": 4520
},
"n_rows_remapped_prev_window": {
"grok-4.5": 6608
},
"by_legacy_id": {
"grok-4.5": {
"n_rows_remapped": {
"MONTH": 4520,
"prev_window": 6608
},
"first_slot": "2026-08-01T00:01:00+00:00",
"last_slot": "2026-08-24T09:01:00+00:00",
"first_slot_prev_window": "2026-07-15T18:01:00+00:00",
"last_slot_prev_window": "2026-07-31T23:01:00+00:00"
}
},
"first_legacy_slot": "2026-08-01T00:01:00+00:00",
"last_legacy_slot": "2026-08-24T09:01:00+00:00",
"first_successor_slot": "2026-08-24T10:01:00+00:00",
"last_successor_slot": "2026-08-31T23:01:00+00:00",
"n_rows_successor": 1570,
"spliced_in_window": true,
"label": "grok-4.6 (incl. 4.5-era, Aug 1-24)"
},
"qwen-3.8-max": {
"legacy_ids": [
"qwen-3.7-max"
],
"legacy_ids_seen_in_window": [
"qwen-3.7-max"
],
"n_rows_remapped": {
"qwen-3.7-max": 870
},
"n_rows_remapped_prev_window": {
"qwen-3.7-max": 6609
},
"by_legacy_id": {
"qwen-3.7-max": {
"n_rows_remapped": {
"MONTH": 870,
"prev_window": 6609
},
"first_slot": "2026-08-01T00:01:00+00:00",
"last_slot": "2026-08-05T09:01:00+00:00",
"first_slot_prev_window": "2026-07-15T18:01:00+00:00",
"last_slot_prev_window": "2026-07-31T23:01:00+00:00"
}
},
"first_legacy_slot": "2026-08-01T00:01:00+00:00",
"last_legacy_slot": "2026-08-05T09:01:00+00:00",
"first_successor_slot": "2026-08-05T10:01:00+00:00",
"last_successor_slot": "2026-08-31T23:01:00+00:00",
"n_rows_successor": 5220,
"spliced_in_window": true,
"label": "qwen-3.8-max (incl. 3.7-era, Aug 1-5)"
}
},
"labels": {
"claude-fable-5": "claude-fable-5",
"claude-opus-5": "claude-opus-5",
"deepseek-v4-pro": "deepseek-v4-pro",
"gemini-3.1-pro": "gemini-3.1-pro",
"gpt-5.6-sol": "gpt-5.6-sol",
"grok-4.6": "grok-4.6 (incl. 4.5-era, Aug 1-24)",
"qwen-3.8-max": "qwen-3.8-max (incl. 3.7-era, Aug 1-5)"
},
"rows_remapped_window": 5390,
"rows_remapped_prev_window": 19532,
"note": "Only the 7 current models are tracked; earlier versions are folded into their successors' lineage. Every per-model aggregate in this pack sees the merged head id; the raw id survives in model_raw. Labels and era spans are generated from the pulled rows, not hardcoded.",
"cells_with_two_rows_per_head": {
"claude-fable-5": 0,
"claude-opus-5": 0,
"deepseek-v4-pro": 0,
"gemini-3.1-pro": 0,
"gpt-5.6-sol": 0,
"grok-4.6": 0,
"qwen-3.8-max": 0
}
},
"frozen_cutoffs": "Two cutoffs live in this issue. The month is frozen at Mon Sep 1, 2026, 16:00 UTC; the four weekly blocks are frozen at their own published cutoffs (W1 Aug 3-9, W2 Aug 10-16, W3 Aug 17-23, W4 Aug 24-30). A forecast still pending at its week's cutoff is immature in the weekly pack and mature in the month pull, so the month does not equal the sum of the weeks (19,383 vs 17,418 directional, plus 1,965 on the 3 remainder days) -- that gap is arithmetic, not a data problem.",
"partial_prior_window": "The prior window is a PARTIAL month: forecasts run only from Jul 11 and the audited daily candles only from Jul 15, so every \"prior\" column, the gray series of the page-1 chart and the market-check A column describe Jul 11-31 (partial), not a full July. It is window A of the same pull, slots 2026-07-01 to 2026-08-01, forecasts counted from Jul 11 (first slot actually seen 2026-07-15 18:01 UTC) and audited candles from Jul 15, matured at the prior cutoff Sat Aug 1, 2026, 16:00 UTC under the same engine, the same FH gate and the same hit rule as August -- 19,541 directional calls against 19,383 in the month. Month-over-month deltas are a step between two windows of different length."
},
"bibtex_key": "mm_consensus_monthly_2026m08",
"citation": {
"bibtex_key": "mm_consensus_monthly_2026m08",
"title": "Monthly Consensus Watch #1: does agreeing with the crowd make an LLM's market call safer?",
"author": "MarketMania Research",
"year": 2026,
"month": "September",
"day": 4,
"url": "https://marketmania.ai/research/reports/consensus-watch-monthly-2026-08.pdf",
"note": "Methodology v1.1; window Aug 1-31, 2026 UTC; source monthly_metrics_2026-08.json"
},
"living_series_note": "**Issue #1.** Monthly Consensus Watch is a living monthly comparison; each issue appends one more calendar month of leave-one-out agreement-vs-hit data. This first issue sets the baseline: herd share 98.7%, pooled lift +1.9pp (agree 45.5% vs disagree 43.6% on n=211), four weekly lifts inside the month running -22.8pp to +17.1pp, and a prior window Jul 11-31 (partial) at herd share 97.4% and lift -10.6pp. 2 model lines are lineage splices this month (grok-4.6 (incl. 4.5-era, Aug 1-24); qwen-3.8-max (incl. 3.7-era, Aug 1-5)). Engine 1.1 has powered the sandbox since Aug 18, i.e. inside this window; no cross-engine PnL comparisons are claimed.",
"testing_next": [
"Monthly #2 (September): the same three series over Sep 1-30, cut on Oct 1 plus the maturity lag, so the first month-over-month comparison runs against a FULL prior month instead of a partial one.",
"Monthly Config Watch #1: the same August window through the config search engine (w200), published as the fourth monthly report of this issue.",
"Quarterly Stability Index: the next identical-prompt run lands around Oct 8 and is the first cross-quarter point on model stability."
],
"related_research": [
{
"title": "Monthly Calibration #1",
"url": "https://marketmania.ai/research/reports/calibration-monthly-2026-08.pdf"
},
{
"title": "Monthly Model Watch #1",
"url": "https://marketmania.ai/research/reports/model-watch-monthly-2026-08.pdf"
},
{
"title": "Consensus Watch #4",
"url": "https://marketmania.ai/research/reports/consensus-watch-2026-08-24.pdf"
}
],
"related_research_note": "The three monthly reports publish together as one issue each month; each links straight to the others' PDF and to the last weekly issue of the same series. Direct links are the posting rule from wave 2 on.",
"research_to_date": {
"this_report": {
"loo_scored_observations": 16089
},
"platform": {
"as_of_cutoff": "2026-09-01",
"resolved_forecasts": 39073,
"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": 7,
"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": 19
},
"source": "monthly_metrics_2026-08.json",
"control_reproduces_wave3_pin": null,
"pin_note": "Research-to-date counter: directional forecasts resolved inside the FH gate (1h/4h/1d) since Jul 11, counted at the issue cutoff (Mon Sep 1, 2026, 16:00 UTC) -- the D2 definition pinned in the weekly series; this issue prints 39,073. The pack carries no prior-cutoff pin for a monthly window (the prior-cutoff pin is null by construction), so the control this issue is the weekly reproduction: the W4 block of this pack reproduces the published weekly #4 on 7 of 7 checks. Counter deltas against pre-wave-3 issues remain definitional.",
"control_w4_reproduces_wave4": {
"pins_source": "weekly_metrics_2026-08-24.json (the published wave-4 pack)",
"note": "W4 of this run is the wave-4 weekly window with the wave-4 frozen cutoff, so every pin must reproduce exactly. A mismatch means the window or the cutoff moved -- non-fatal here, but the monthly numbers must not be published until it is explained.",
"checks": {
"counters.mature_directional": {
"pin": 4576,
"got": 4576,
"match": true,
"tolerance": 0
},
"calibration.field.hit_rate": {
"pin": 0.4294,
"got": 0.4294,
"match": true,
"tolerance": 5e-05
},
"calibration.field.gap_pp": {
"pin": 19.6,
"got": 19.6,
"match": true,
"tolerance": 0.05
},
"consensus.loo_pooled.lift_pp": {
"pin": -6.0,
"got": -6.0,
"match": true,
"tolerance": 0.05
},
"model_watch.leaderboard.rows[0].model": {
"pin": "deepseek-v4-pro",
"got": "deepseek-v4-pro",
"match": true,
"tolerance": 0
},
"model_watch.leaderboard.rows[0].hit_rate": {
"pin": 0.4458,
"got": 0.4458,
"match": true,
"tolerance": 5e-05
},
"research_to_date.canonical_value_at_cutoff": {
"pin": 38385,
"got": 38385,
"match": true,
"tolerance": 0
}
},
"all_match": true,
"n_checks": 7,
"n_match": 7
}
},
"site_entry": {
"slug": "consensus-watch-monthly-2026-08",
"series": "monthly",
"seriesLabel": "Monthly · Consensus",
"date": "September 4, 2026",
"title": "Monthly Consensus Watch #1",
"summary": "40,979 mature forecasts over one calendar month asked whether agreeing with the crowd makes an LLM market call safer. Herding ran at 98.7% of leave-one-out scoreable calls and the pooled monthly lift came in at +1.9pp: agree-hit 45.5% against 43.6% on disagreement, on a disagree bucket of n=211 and a 46.3% field base. The margin curve read 45.9% / 45.1% / 45.8% / 45.1% from 3 peers to full unanimity, while the four weeks inside the month ran lifts of -22.8pp to +17.1pp. Field accuracy moved 45.0% -> 46.3% (+1.4pp) against the partial July, with 5 of 7 models improving.",
"stats": [
"16,089 peer-judged calls",
"98.7% ran with the herd",
"lift +1.9pp vs disagree"
],
"manifest_key_finding": "Herding ran at 98.7% of leave-one-out scoreable calls across August and the pooled lift came in at +1.9pp: agree-hit 45.5% against 43.6% on disagreement (disagree bucket n=211), against a 46.3% field base, while the four weeks inside the month ran -22.8pp to +17.1pp.",
"files": {
"pdf": "/research/reports/consensus-watch-monthly-2026-08.pdf",
"md": "/research/reports/consensus-watch-monthly-2026-08.md",
"json": "/research/reports/consensus-watch-monthly-2026-08.json"
}
},
"social": {
"telegram": {
"photo_caption_html": "🧭 Monthly Consensus Watch #1 — monthly series (window Aug 1–31)\n\nResearch question: if a model's forecast matches the consensus of the other models, is it more reliable?\n\n16,089 leave-one-out scored calls over one month, 7 model lines:\n• Herding ran at 98.7% — agree-hit 45.5% vs 43.6% when a model left the crowd (field base 46.3%, disagree n=211)\n• Margin curve: 3-peer 45.9%, 4-peer 45.1%, 5-peer 45.8%, unanimity 45.1%\n• Week by week the lift ran -22.8pp to +17.1pp — the month pools four weeks that disagreed\n• Market check: field accuracy 45.0% -> 46.3% (+1.4pp) vs the partial July, 5 of 7 models improved; avg |1d| move 1.54% -> 1.98%\n\nNote: prior window = PARTIAL July (forecasts since Jul 11); two model lines are lineage splices\n\n📄 Full report (PDF) · all research\n\n#cryptotrading #LLM #AI #crypto #trading #consensus #research #Monthly",
"document_caption_html": "📄 Monthly Consensus Watch #1 — full report (PDF).\nWeb copy: consensus-watch-monthly-2026-08.pdf · machine-readable: JSON · all research\n\n#cryptotrading #LLM #AI #crypto #trading #consensus #research #Monthly",
"caption_len": 1011,
"document_caption_len": 434,
"limit": 1024
},
"x": {
"main": "Monthly Consensus Watch #1 is out.\n\n16,089 peer-judged LLM calls over August: 98.7% ran with the herd, agreeing scored 45.5% vs 43.6% for the 211 that left it (+1.9pp).\n\nFull report: https://marketmania.ai/research/reports/consensus-watch-monthly-2026-08.pdf\n\n#AI #Crypto #Research #Consensus",
"main_tco_len": 240,
"reply": "Lift by week, then pooled:\nAug 3-9: -22.8pp\nAug 10-16: -6.9pp\nAug 17-23: +17.1pp\nAug 24-30: -6.0pp\nMonth: +1.9pp. Margin curve 45.9 / 45.1 / 45.8 / 45.1, field base 46.3.\nData: https://marketmania.ai/research/reports/consensus-watch-monthly-2026-08.json",
"reply_tco_len": 200,
"limit_tco": 280,
"replies_max": 1
}
},
"monthly_window": {
"start": "2026-08-01T00:00:00+00:00",
"end": "2026-09-01T00:00:00+00:00",
"text": "Aug 1-31, 2026 UTC",
"cutoff": "2026-09-01T16:00:00+00:00",
"cutoff_text": "Mon Sep 1, 2026, 16:00 UTC",
"days": 31,
"label": "Aug 1-31"
},
"prior_window": {
"start": "2026-07-01T00:00:00+00:00",
"end": "2026-08-01T00:00:00+00:00",
"text": "Jul 11-31, 2026 (partial month; forecasts since Jul 11)",
"printed_as": "Jul 11-31 (partial)",
"cutoff_text": "Sat Aug 1, 2026, 16:00 UTC",
"partial": true,
"note": "July is partial: forecasts since 2026-07-11",
"applicable": true,
"not_applicable_reason": null,
"source": "month.prior_window -- window A of the same pull, matured at the prior cutoff with the same gate, maturity and hit rule",
"effective_start": "2026-07-15T18:01:00+00:00",
"counters": {
"forecasts_total": 46285,
"ok_in_gate": 42134,
"mature": 41985,
"mature_directional": 19541,
"mature_sideways": 22444
},
"definitions": {
"maturity": "horizon expired AND trade closed by the PRIOR cutoff (the same rule the current window uses at its own cutoff)",
"hit": "direction: exit tp1/tp2 -> hit, sl -> miss, expiry -> sign of gross pnl",
"fh_gate": [
"1h",
"4h",
"1d"
],
"min_n": 10,
"lineage": "rows spliced by the generalised 7-head lineage map before any aggregate",
"note": "identical definitions to the current window; only the window and the cutoff differ"
}
},
"weeks_index": {
"W1": {
"label": "Aug 3-9",
"issue": 1,
"published_slug_suffix": [
"2026-08-03T00:00:00Z",
"2026-08-10T00:00:00Z"
],
"bounds": [
"2026-08-03T00:00:00Z",
"2026-08-10T00:00:00Z"
],
"regime": "calm"
},
"W2": {
"label": "Aug 10-16",
"issue": 2,
"published_slug_suffix": [
"2026-08-10T00:00:00Z",
"2026-08-17T00:00:00Z"
],
"bounds": [
"2026-08-10T00:00:00Z",
"2026-08-17T00:00:00Z"
],
"regime": "calm"
},
"W3": {
"label": "Aug 17-23",
"issue": 3,
"published_slug_suffix": [
"2026-08-17T00:00:00Z",
"2026-08-24T00:00:00Z"
],
"bounds": [
"2026-08-17T00:00:00Z",
"2026-08-24T00:00:00Z"
],
"regime": "volatile"
},
"W4": {
"label": "Aug 24-30",
"issue": 4,
"published_slug_suffix": [
"2026-08-24T00:00:00Z",
"2026-08-31T00:00:00Z"
],
"bounds": [
"2026-08-24T00:00:00Z",
"2026-08-31T00:00:00Z"
],
"regime": "volatile"
}
},
"roster_rule": "7 models tracked (current line-up; earlier versions folded into their successors' lineage)",
"lineage_labels": {
"labels": {
"claude-fable-5": "claude-fable-5",
"claude-opus-5": "claude-opus-5",
"deepseek-v4-pro": "deepseek-v4-pro",
"gemini-3.1-pro": "gemini-3.1-pro",
"gpt-5.6-sol": "gpt-5.6-sol",
"grok-4.6": "grok-4.6 (incl. 4.5-era, Aug 1-24)",
"qwen-3.8-max": "qwen-3.8-max (incl. 3.7-era, Aug 1-5)"
},
"spliced_in_window": [
"grok-4.6",
"qwen-3.8-max"
],
"cell_form": {
"grok-4.6": "grok-4.6 (spliced)",
"qwen-3.8-max": "qwen-3.8-max (spliced)"
},
"families": [
"grok",
"qwen"
]
},
"controls": {
"w4_reproduces_wave4": {
"pins_source": "weekly_metrics_2026-08-24.json (the published wave-4 pack)",
"note": "W4 of this run is the wave-4 weekly window with the wave-4 frozen cutoff, so every pin must reproduce exactly. A mismatch means the window or the cutoff moved -- non-fatal here, but the monthly numbers must not be published until it is explained.",
"checks": {
"counters.mature_directional": {
"pin": 4576,
"got": 4576,
"match": true,
"tolerance": 0
},
"calibration.field.hit_rate": {
"pin": 0.4294,
"got": 0.4294,
"match": true,
"tolerance": 5e-05
},
"calibration.field.gap_pp": {
"pin": 19.6,
"got": 19.6,
"match": true,
"tolerance": 0.05
},
"consensus.loo_pooled.lift_pp": {
"pin": -6.0,
"got": -6.0,
"match": true,
"tolerance": 0.05
},
"model_watch.leaderboard.rows[0].model": {
"pin": "deepseek-v4-pro",
"got": "deepseek-v4-pro",
"match": true,
"tolerance": 0
},
"model_watch.leaderboard.rows[0].hit_rate": {
"pin": 0.4458,
"got": 0.4458,
"match": true,
"tolerance": 5e-05
},
"research_to_date.canonical_value_at_cutoff": {
"pin": 38385,
"got": 38385,
"match": true,
"tolerance": 0
}
},
"all_match": true,
"n_checks": 7,
"n_match": 7
},
"month_vs_weeks": {
"month": 19383,
"sum_weeks": 17418,
"per_week": {
"W1": 4042,
"W2": 3670,
"W3": 5130,
"W4": 4576
},
"remainder_days": [
"2026-08-01",
"2026-08-02",
"2026-08-31"
],
"remainder_days_n": 3,
"remainder_mature_directional_month_pull": 1965,
"remainder_by_day": {
"2026-08-02": 756,
"2026-08-01": 521,
"2026-08-31": 688
},
"sum_weeks_plus_remainder": 19383,
"month_minus_sum_weeks_plus_remainder": 0,
"month_pull_by_week": {
"W1": 4042,
"W2": 3670,
"W3": 5130,
"W4": 4576
},
"note": "These WILL differ, and the difference is expected: each weekly pack is computed at its own frozen cutoff (week_end + 16h), while the MONTH pack is computed at 2026-09-01T16:00Z. A forecast still pending at its week's cutoff (trade not yet closed) is immature in the weekly pack and mature in the MONTH pack, so month >= sum_weeks + remainder. remainder_days_n counts the calendar days of August outside the four Mon-Sun weeks (Aug 1, Aug 2, Aug 31). month_pull_by_week is the same month pull split by slot week at the MONTH cutoff -- compare THAT with per_week to see the cutoff effect alone."
},
"lineage_splice_counts": {
"MONTH": {
"claude-opus-4.8": 0,
"grok-4.5": 4520,
"qwen-3.7-max": 870
},
"W1": {
"claude-opus-4.8": 0,
"grok-4.5": 1310,
"qwen-3.7-max": 480
},
"W2": {
"claude-opus-4.8": 0,
"grok-4.5": 1340,
"qwen-3.7-max": 0
},
"W3": {
"claude-opus-4.8": 0,
"grok-4.5": 1370,
"qwen-3.7-max": 0
},
"W4": {
"claude-opus-4.8": 0,
"grok-4.5": 110,
"qwen-3.7-max": 0
}
},
"lineage_splice_detail": {
"MONTH": {
"total_window": 5390,
"total_prev_window": 19532,
"labels": {
"claude-fable-5": "claude-fable-5",
"claude-opus-5": "claude-opus-5",
"deepseek-v4-pro": "deepseek-v4-pro",
"gemini-3.1-pro": "gemini-3.1-pro",
"gpt-5.6-sol": "gpt-5.6-sol",
"grok-4.6": "grok-4.6 (incl. 4.5-era, Aug 1-24)",
"qwen-3.8-max": "qwen-3.8-max (incl. 3.7-era, Aug 1-5)"
},
"per_head": {
"claude-fable-5": {
"legacy_ids_seen_in_window": [],
"n_rows_remapped": {},
"first_legacy_slot": null,
"last_legacy_slot": null,
"first_successor_slot": "2026-08-01T00:01:00+00:00"
},
"claude-opus-5": {
"legacy_ids_seen_in_window": [],
"n_rows_remapped": {
"claude-opus-4.8": 0
},
"first_legacy_slot": null,
"last_legacy_slot": null,
"first_successor_slot": "2026-08-01T00:01:00+00:00"
},
"deepseek-v4-pro": {
"legacy_ids_seen_in_window": [],
"n_rows_remapped": {},
"first_legacy_slot": null,
"last_legacy_slot": null,
"first_successor_slot": "2026-08-01T00:01:00+00:00"
},
"gemini-3.1-pro": {
"legacy_ids_seen_in_window": [],
"n_rows_remapped": {},
"first_legacy_slot": null,
"last_legacy_slot": null,
"first_successor_slot": "2026-08-01T00:01:00+00:00"
},
"gpt-5.6-sol": {
"legacy_ids_seen_in_window": [],
"n_rows_remapped": {},
"first_legacy_slot": null,
"last_legacy_slot": null,
"first_successor_slot": "2026-08-01T00:01:00+00:00"
},
"grok-4.6": {
"legacy_ids_seen_in_window": [
"grok-4.5"
],
"n_rows_remapped": {
"grok-4.5": 4520
},
"first_legacy_slot": "2026-08-01T00:01:00+00:00",
"last_legacy_slot": "2026-08-24T09:01:00+00:00",
"first_successor_slot": "2026-08-24T10:01:00+00:00"
},
"qwen-3.8-max": {
"legacy_ids_seen_in_window": [
"qwen-3.7-max"
],
"n_rows_remapped": {
"qwen-3.7-max": 870
},
"first_legacy_slot": "2026-08-01T00:01:00+00:00",
"last_legacy_slot": "2026-08-05T09:01:00+00:00",
"first_successor_slot": "2026-08-05T10:01:00+00:00"
}
}
},
"W1": {
"total_window": 1790,
"total_prev_window": 3373,
"labels": {
"claude-fable-5": "claude-fable-5",
"claude-opus-5": "claude-opus-5",
"deepseek-v4-pro": "deepseek-v4-pro",
"gemini-3.1-pro": "gemini-3.1-pro",
"gpt-5.6-sol": "gpt-5.6-sol",
"grok-4.6": "grok-4.6 (incl. 4.5-era, Aug 3-9)",
"qwen-3.8-max": "qwen-3.8-max (incl. 3.7-era, Aug 3-5)"
},
"per_head": {
"claude-fable-5": {
"legacy_ids_seen_in_window": [],
"n_rows_remapped": {},
"first_legacy_slot": null,
"last_legacy_slot": null,
"first_successor_slot": "2026-08-03T00:01:00+00:00"
},
"claude-opus-5": {
"legacy_ids_seen_in_window": [],
"n_rows_remapped": {
"claude-opus-4.8": 0
},
"first_legacy_slot": null,
"last_legacy_slot": null,
"first_successor_slot": "2026-08-03T00:01:00+00:00"
},
"deepseek-v4-pro": {
"legacy_ids_seen_in_window": [],
"n_rows_remapped": {},
"first_legacy_slot": null,
"last_legacy_slot": null,
"first_successor_slot": "2026-08-03T00:01:00+00:00"
},
"gemini-3.1-pro": {
"legacy_ids_seen_in_window": [],
"n_rows_remapped": {},
"first_legacy_slot": null,
"last_legacy_slot": null,
"first_successor_slot": "2026-08-03T00:01:00+00:00"
},
"gpt-5.6-sol": {
"legacy_ids_seen_in_window": [],
"n_rows_remapped": {},
"first_legacy_slot": null,
"last_legacy_slot": null,
"first_successor_slot": "2026-08-03T00:01:00+00:00"
},
"grok-4.6": {
"legacy_ids_seen_in_window": [
"grok-4.5"
],
"n_rows_remapped": {
"grok-4.5": 1310
},
"first_legacy_slot": "2026-08-03T00:01:00+00:00",
"last_legacy_slot": "2026-08-09T23:01:00+00:00",
"first_successor_slot": null
},
"qwen-3.8-max": {
"legacy_ids_seen_in_window": [
"qwen-3.7-max"
],
"n_rows_remapped": {
"qwen-3.7-max": 480
},
"first_legacy_slot": "2026-08-03T00:01:00+00:00",
"last_legacy_slot": "2026-08-05T09:01:00+00:00",
"first_successor_slot": "2026-08-05T10:01:00+00:00"
}
}
},
"W2": {
"total_window": 1340,
"total_prev_window": 1790,
"labels": {
"claude-fable-5": "claude-fable-5",
"claude-opus-5": "claude-opus-5",
"deepseek-v4-pro": "deepseek-v4-pro",
"gemini-3.1-pro": "gemini-3.1-pro",
"gpt-5.6-sol": "gpt-5.6-sol",
"grok-4.6": "grok-4.6 (incl. 4.5-era, Aug 10-16)",
"qwen-3.8-max": "qwen-3.8-max"
},
"per_head": {
"claude-fable-5": {
"legacy_ids_seen_in_window": [],
"n_rows_remapped": {},
"first_legacy_slot": null,
"last_legacy_slot": null,
"first_successor_slot": "2026-08-10T00:01:00+00:00"
},
"claude-opus-5": {
"legacy_ids_seen_in_window": [],
"n_rows_remapped": {
"claude-opus-4.8": 0
},
"first_legacy_slot": null,
"last_legacy_slot": null,
"first_successor_slot": "2026-08-10T00:01:00+00:00"
},
"deepseek-v4-pro": {
"legacy_ids_seen_in_window": [],
"n_rows_remapped": {},
"first_legacy_slot": null,
"last_legacy_slot": null,
"first_successor_slot": "2026-08-10T00:01:00+00:00"
},
"gemini-3.1-pro": {
"legacy_ids_seen_in_window": [],
"n_rows_remapped": {},
"first_legacy_slot": null,
"last_legacy_slot": null,
"first_successor_slot": "2026-08-10T00:01:00+00:00"
},
"gpt-5.6-sol": {
"legacy_ids_seen_in_window": [],
"n_rows_remapped": {},
"first_legacy_slot": null,
"last_legacy_slot": null,
"first_successor_slot": "2026-08-10T00:01:00+00:00"
},
"grok-4.6": {
"legacy_ids_seen_in_window": [
"grok-4.5"
],
"n_rows_remapped": {
"grok-4.5": 1340
},
"first_legacy_slot": "2026-08-10T00:01:00+00:00",
"last_legacy_slot": "2026-08-16T23:01:00+00:00",
"first_successor_slot": null
},
"qwen-3.8-max": {
"legacy_ids_seen_in_window": [],
"n_rows_remapped": {
"qwen-3.7-max": 0
},
"first_legacy_slot": null,
"last_legacy_slot": null,
"first_successor_slot": "2026-08-10T00:01:00+00:00"
}
}
},
"W3": {
"total_window": 1370,
"total_prev_window": 1340,
"labels": {
"claude-fable-5": "claude-fable-5",
"claude-opus-5": "claude-opus-5",
"deepseek-v4-pro": "deepseek-v4-pro",
"gemini-3.1-pro": "gemini-3.1-pro",
"gpt-5.6-sol": "gpt-5.6-sol",
"grok-4.6": "grok-4.6 (incl. 4.5-era, Aug 17-23)",
"qwen-3.8-max": "qwen-3.8-max"
},
"per_head": {
"claude-fable-5": {
"legacy_ids_seen_in_window": [],
"n_rows_remapped": {},
"first_legacy_slot": null,
"last_legacy_slot": null,
"first_successor_slot": "2026-08-17T00:01:00+00:00"
},
"claude-opus-5": {
"legacy_ids_seen_in_window": [],
"n_rows_remapped": {
"claude-opus-4.8": 0
},
"first_legacy_slot": null,
"last_legacy_slot": null,
"first_successor_slot": "2026-08-17T00:01:00+00:00"
},
"deepseek-v4-pro": {
"legacy_ids_seen_in_window": [],
"n_rows_remapped": {},
"first_legacy_slot": null,
"last_legacy_slot": null,
"first_successor_slot": "2026-08-17T00:01:00+00:00"
},
"gemini-3.1-pro": {
"legacy_ids_seen_in_window": [],
"n_rows_remapped": {},
"first_legacy_slot": null,
"last_legacy_slot": null,
"first_successor_slot": "2026-08-17T00:01:00+00:00"
},
"gpt-5.6-sol": {
"legacy_ids_seen_in_window": [],
"n_rows_remapped": {},
"first_legacy_slot": null,
"last_legacy_slot": null,
"first_successor_slot": "2026-08-17T00:01:00+00:00"
},
"grok-4.6": {
"legacy_ids_seen_in_window": [
"grok-4.5"
],
"n_rows_remapped": {
"grok-4.5": 1370
},
"first_legacy_slot": "2026-08-17T00:01:00+00:00",
"last_legacy_slot": "2026-08-23T23:01:00+00:00",
"first_successor_slot": null
},
"qwen-3.8-max": {
"legacy_ids_seen_in_window": [],
"n_rows_remapped": {
"qwen-3.7-max": 0
},
"first_legacy_slot": null,
"last_legacy_slot": null,
"first_successor_slot": "2026-08-17T00:01:00+00:00"
}
}
},
"W4": {
"total_window": 110,
"total_prev_window": 1370,
"labels": {
"claude-fable-5": "claude-fable-5",
"claude-opus-5": "claude-opus-5",
"deepseek-v4-pro": "deepseek-v4-pro",
"gemini-3.1-pro": "gemini-3.1-pro",
"gpt-5.6-sol": "gpt-5.6-sol",
"grok-4.6": "grok-4.6 (incl. 4.5-era, Aug 24)",
"qwen-3.8-max": "qwen-3.8-max"
},
"per_head": {
"claude-fable-5": {
"legacy_ids_seen_in_window": [],
"n_rows_remapped": {},
"first_legacy_slot": null,
"last_legacy_slot": null,
"first_successor_slot": "2026-08-24T00:01:00+00:00"
},
"claude-opus-5": {
"legacy_ids_seen_in_window": [],
"n_rows_remapped": {
"claude-opus-4.8": 0
},
"first_legacy_slot": null,
"last_legacy_slot": null,
"first_successor_slot": "2026-08-24T00:01:00+00:00"
},
"deepseek-v4-pro": {
"legacy_ids_seen_in_window": [],
"n_rows_remapped": {},
"first_legacy_slot": null,
"last_legacy_slot": null,
"first_successor_slot": "2026-08-24T00:01:00+00:00"
},
"gemini-3.1-pro": {
"legacy_ids_seen_in_window": [],
"n_rows_remapped": {},
"first_legacy_slot": null,
"last_legacy_slot": null,
"first_successor_slot": "2026-08-24T00:01:00+00:00"
},
"gpt-5.6-sol": {
"legacy_ids_seen_in_window": [],
"n_rows_remapped": {},
"first_legacy_slot": null,
"last_legacy_slot": null,
"first_successor_slot": "2026-08-24T00:01:00+00:00"
},
"grok-4.6": {
"legacy_ids_seen_in_window": [
"grok-4.5"
],
"n_rows_remapped": {
"grok-4.5": 110
},
"first_legacy_slot": "2026-08-24T00:01:00+00:00",
"last_legacy_slot": "2026-08-24T09:01:00+00:00",
"first_successor_slot": "2026-08-24T10:01:00+00:00"
},
"qwen-3.8-max": {
"legacy_ids_seen_in_window": [],
"n_rows_remapped": {
"qwen-3.7-max": 0
},
"first_legacy_slot": null,
"last_legacy_slot": null,
"first_successor_slot": "2026-08-24T00:01:00+00:00"
}
}
}
},
"d2_at_cutoff": {
"window_key": "MONTH",
"cutoff": "2026-09-01T16:00:00+00:00",
"canonical_definition": "D2_directional_only",
"canonical_value_at_cutoff": 39073,
"since": "2026-07-11T00:00:00+00:00",
"d1_d3_at_cutoff": {
"D1_expired_ok_in_gate": 83113,
"D2_directional_only": 39073,
"D3_weekly_rule_mature": 82964
},
"marker": "RTD-D2-MONTHLY=39073",
"note": "this is the monthly RTD pin -- the number that goes into the Monthly plain boxes"
},
"prior_window_summary": {
"applicable": true,
"start": "2026-07-01T00:00:00+00:00",
"end": "2026-08-01T00:00:00+00:00",
"effective_start": "2026-07-15T18:01:00+00:00",
"cutoff": "2026-08-01T16:00:00+00:00",
"partial": true,
"note": "forecasts since 2026-07-11; candles since 2026-07-15",
"counters": {
"forecasts_total": 46285,
"ok_in_gate": 42134,
"mature": 41985,
"mature_directional": 19541,
"mature_sideways": 22444
},
"lineage_splice_counts": {
"claude-opus-4.8": 6315,
"grok-4.5": 6608,
"qwen-3.7-max": 6609
},
"field": {
"hit_rate": {
"month": 0.4633,
"prior": 0.4498,
"delta_pp": 1.4
},
"gap_pp": {
"month": 16.3,
"prior": 17.8,
"delta_pp": -1.5
},
"brier": {
"month": 0.2784,
"prior": 0.2831
},
"mean_conf": {
"month": 62.6,
"prior": 62.8
},
"n": {
"month": 19383,
"prior": 19541
}
},
"consensus": {
"lift_pp": {
"month": 1.9,
"prior": -10.6,
"delta_pp": 12.5
},
"n_scoreable": {
"month": 16089,
"prior": 16993
},
"herd_share_pct": {
"month": 98.7,
"prior": 97.4
}
},
"leader": {
"month": "claude-opus-5",
"month_hit_rate": 0.4779,
"prior": "qwen-3.8-max",
"prior_hit_rate": 0.4568
},
"per_model": {
"claude-fable-5": {
"hit_month": 0.4673,
"hit_prior": 0.445,
"hit_delta_pp": 2.2,
"gap_month": 13.9,
"gap_prior": 15.9,
"gap_delta_pp": -2.0,
"n_month": 2812,
"n_prior": 2690
},
"claude-opus-5": {
"hit_month": 0.4779,
"hit_prior": 0.4492,
"hit_delta_pp": 2.9,
"gap_month": 13.2,
"gap_prior": 15.6,
"gap_delta_pp": -2.4,
"n_month": 2281,
"n_prior": 3448
},
"deepseek-v4-pro": {
"hit_month": 0.4554,
"hit_prior": 0.4505,
"hit_delta_pp": 0.5,
"gap_month": 15.4,
"gap_prior": 16.6,
"gap_delta_pp": -1.2,
"n_month": 2578,
"n_prior": 1869
},
"gemini-3.1-pro": {
"hit_month": 0.4761,
"hit_prior": 0.4511,
"hit_delta_pp": 2.5,
"gap_month": 18.8,
"gap_prior": 20.0,
"gap_delta_pp": -1.2,
"n_month": 3033,
"n_prior": 2813
},
"gpt-5.6-sol": {
"hit_month": 0.4586,
"hit_prior": 0.4431,
"hit_delta_pp": 1.6,
"gap_month": 21.8,
"gap_prior": 22.1,
"gap_delta_pp": -0.3,
"n_month": 2970,
"n_prior": 3085
},
"grok-4.6": {
"hit_month": 0.4527,
"hit_prior": 0.453,
"hit_delta_pp": -0.0,
"gap_month": 15.1,
"gap_prior": 14.2,
"gap_delta_pp": 0.9,
"n_month": 2708,
"n_prior": 2265
},
"qwen-3.8-max": {
"hit_month": 0.4565,
"hit_prior": 0.4568,
"hit_delta_pp": -0.0,
"gap_month": 14.7,
"gap_prior": 18.7,
"gap_delta_pp": -4.0,
"n_month": 3001,
"n_prior": 3371
}
},
"per_symbol": {
"BNB": {
"hit_month": 0.4807,
"hit_prior": 0.4548,
"hit_delta_pp": 2.6,
"n_month": 3857,
"n_prior": 2720
},
"BTC": {
"hit_month": 0.4648,
"hit_prior": 0.4418,
"hit_delta_pp": 2.3,
"n_month": 4068,
"n_prior": 4169
},
"ETH": {
"hit_month": 0.4287,
"hit_prior": 0.4781,
"hit_delta_pp": -4.9,
"n_month": 3695,
"n_prior": 4564
},
"SOL": {
"hit_month": 0.4733,
"hit_prior": 0.4325,
"hit_delta_pp": 4.1,
"n_month": 3928,
"n_prior": 3443
},
"XRP": {
"hit_month": 0.4673,
"hit_prior": 0.4455,
"hit_delta_pp": 2.2,
"n_month": 3835,
"n_prior": 3053
}
},
"log_line": "prior_window: n_mature_directional=19541, effective_start=2026-07-15T18:01:00+00:00, field hit=0.4498 gap=17.8, loo lift=-10.6",
"note_definitions": "the prior block is window A of the MONTH pull, matured at the prior cutoff with the same rules; delta = month - prior in percentage points",
"per_symbol_retired_not_compared": [
"AVAX",
"DOGE",
"LINK",
"PEPE",
"TON"
]
},
"note": "controls are DESCRIPTIVE and never fatal; they are printed in the log as well"
}
}