{
"slug": "calibration-monthly-2026-08",
"series": "monthly",
"series_label": "MONTHLY · CALIBRATION",
"issue": 1,
"title": "Monthly Calibration #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",
"hit_rule": "direction: exit tp1/tp2 -> hit, sl -> miss, expiry -> sign of gross pnl",
"bibtex_key": "mm_calibration_monthly_2026m08",
"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": {
"calibration_main": [
{
"model": "claude-opus-5",
"model_label": "claude-opus-5",
"legacy": false,
"is_field": false,
"n": 2281,
"coverage": 0.3889,
"hit_rate": 0.4779,
"mean_conf": 61.0,
"gap_pp": 13.2,
"brier": 0.2677,
"wilson_95": {
"lo": 0.4574,
"hi": 0.4984
}
},
{
"model": "claude-fable-5",
"model_label": "claude-fable-5",
"legacy": false,
"is_field": false,
"n": 2812,
"coverage": 0.48,
"hit_rate": 0.4673,
"mean_conf": 60.6,
"gap_pp": 13.9,
"brier": 0.2689,
"wilson_95": {
"lo": 0.4489,
"hi": 0.4858
}
},
{
"model": "grok-4.6",
"model_label": "grok-4.6 (incl. 4.5-era, Aug 1-24)",
"legacy": false,
"is_field": false,
"n": 2708,
"coverage": 0.4621,
"hit_rate": 0.4527,
"mean_conf": 60.3,
"gap_pp": 15.1,
"brier": 0.271,
"wilson_95": {
"lo": 0.4341,
"hi": 0.4715
}
},
{
"model": "qwen-3.8-max",
"model_label": "qwen-3.8-max (incl. 3.7-era, Aug 1-5)",
"legacy": false,
"is_field": false,
"n": 3001,
"coverage": 0.5166,
"hit_rate": 0.4565,
"mean_conf": 60.3,
"gap_pp": 14.7,
"brier": 0.273,
"wilson_95": {
"lo": 0.4388,
"hi": 0.4744
}
},
{
"model": "deepseek-v4-pro",
"model_label": "deepseek-v4-pro",
"legacy": false,
"is_field": false,
"n": 2578,
"coverage": 0.4397,
"hit_rate": 0.4554,
"mean_conf": 60.9,
"gap_pp": 15.4,
"brier": 0.2752,
"wilson_95": {
"lo": 0.4362,
"hi": 0.4747
}
},
{
"model": "gemini-3.1-pro",
"model_label": "gemini-3.1-pro",
"legacy": false,
"is_field": false,
"n": 3033,
"coverage": 0.5177,
"hit_rate": 0.4761,
"mean_conf": 66.4,
"gap_pp": 18.8,
"brier": 0.2881,
"wilson_95": {
"lo": 0.4584,
"hi": 0.4939
}
},
{
"model": "gpt-5.6-sol",
"model_label": "gpt-5.6-sol",
"legacy": false,
"is_field": false,
"n": 2970,
"coverage": 0.5064,
"hit_rate": 0.4586,
"mean_conf": 67.7,
"gap_pp": 21.8,
"brier": 0.3005,
"wilson_95": {
"lo": 0.4407,
"hi": 0.4765
}
},
{
"model": "Field (all models)",
"legacy": false,
"is_field": true,
"n": 19383,
"coverage": null,
"hit_rate": 0.4633,
"mean_conf": 62.6,
"gap_pp": 16.3,
"brier": 0.2784,
"wilson_95": {
"lo": null,
"hi": null
}
}
],
"buckets": [
{
"model": "claude-opus-5",
"bucket": "50-60",
"n": 528,
"hit_rate": 0.4943,
"mean_conf": 57.9,
"insufficient": false,
"no_data": false
},
{
"model": "claude-opus-5",
"bucket": "60-70",
"n": 1753,
"hit_rate": 0.4729,
"mean_conf": 61.9,
"insufficient": false,
"no_data": false
},
{
"model": "claude-opus-5",
"bucket": "70-80",
"n": 0,
"hit_rate": null,
"mean_conf": null,
"insufficient": false,
"no_data": true
},
{
"model": "claude-opus-5",
"bucket": "80-100",
"n": 0,
"hit_rate": null,
"mean_conf": null,
"insufficient": false,
"no_data": true
},
{
"model": "claude-fable-5",
"bucket": "50-60",
"n": 900,
"hit_rate": 0.4611,
"mean_conf": 56.6,
"insufficient": false,
"no_data": false
},
{
"model": "claude-fable-5",
"bucket": "60-70",
"n": 1912,
"hit_rate": 0.4702,
"mean_conf": 62.5,
"insufficient": false,
"no_data": false
},
{
"model": "claude-fable-5",
"bucket": "70-80",
"n": 0,
"hit_rate": null,
"mean_conf": null,
"insufficient": false,
"no_data": true
},
{
"model": "claude-fable-5",
"bucket": "80-100",
"n": 0,
"hit_rate": null,
"mean_conf": null,
"insufficient": false,
"no_data": true
},
{
"model": "grok-4.6",
"bucket": "50-60",
"n": 1324,
"hit_rate": 0.4434,
"mean_conf": 57.5,
"insufficient": false,
"no_data": false
},
{
"model": "grok-4.6",
"bucket": "60-70",
"n": 1370,
"hit_rate": 0.4613,
"mean_conf": 62.9,
"insufficient": false,
"no_data": false
},
{
"model": "grok-4.6",
"bucket": "70-80",
"n": 14,
"hit_rate": 0.5,
"mean_conf": 71.1,
"insufficient": false,
"no_data": false
},
{
"model": "grok-4.6",
"bucket": "80-100",
"n": 0,
"hit_rate": null,
"mean_conf": null,
"insufficient": false,
"no_data": true
},
{
"model": "qwen-3.8-max",
"bucket": "50-60",
"n": 1306,
"hit_rate": 0.4763,
"mean_conf": 56.7,
"insufficient": false,
"no_data": false
},
{
"model": "qwen-3.8-max",
"bucket": "60-70",
"n": 1488,
"hit_rate": 0.4442,
"mean_conf": 62.4,
"insufficient": false,
"no_data": false
},
{
"model": "qwen-3.8-max",
"bucket": "70-80",
"n": 161,
"hit_rate": 0.441,
"mean_conf": 73.8,
"insufficient": false,
"no_data": false
},
{
"model": "qwen-3.8-max",
"bucket": "80-100",
"n": 5,
"hit_rate": 0.2,
"mean_conf": 82.8,
"insufficient": true,
"no_data": false
},
{
"model": "deepseek-v4-pro",
"bucket": "50-60",
"n": 853,
"hit_rate": 0.4467,
"mean_conf": 56.6,
"insufficient": false,
"no_data": false
},
{
"model": "deepseek-v4-pro",
"bucket": "60-70",
"n": 1566,
"hit_rate": 0.4636,
"mean_conf": 62.5,
"insufficient": false,
"no_data": false
},
{
"model": "deepseek-v4-pro",
"bucket": "70-80",
"n": 140,
"hit_rate": 0.4,
"mean_conf": 71.9,
"insufficient": false,
"no_data": false
},
{
"model": "deepseek-v4-pro",
"bucket": "80-100",
"n": 4,
"hit_rate": 0.25,
"mean_conf": 80,
"insufficient": true,
"no_data": false
},
{
"model": "gemini-3.1-pro",
"bucket": "50-60",
"n": 25,
"hit_rate": 0.48,
"mean_conf": 55,
"insufficient": false,
"no_data": false
},
{
"model": "gemini-3.1-pro",
"bucket": "60-70",
"n": 1953,
"hit_rate": 0.4803,
"mean_conf": 63.0,
"insufficient": false,
"no_data": false
},
{
"model": "gemini-3.1-pro",
"bucket": "70-80",
"n": 1028,
"hit_rate": 0.465,
"mean_conf": 72.6,
"insufficient": false,
"no_data": false
},
{
"model": "gemini-3.1-pro",
"bucket": "80-100",
"n": 27,
"hit_rate": 0.5926,
"mean_conf": 80.7,
"insufficient": false,
"no_data": false
},
{
"model": "gpt-5.6-sol",
"bucket": "50-60",
"n": 16,
"hit_rate": 0.375,
"mean_conf": 58.6,
"insufficient": false,
"no_data": false
},
{
"model": "gpt-5.6-sol",
"bucket": "60-70",
"n": 2078,
"hit_rate": 0.4808,
"mean_conf": 65.4,
"insufficient": false,
"no_data": false
},
{
"model": "gpt-5.6-sol",
"bucket": "70-80",
"n": 870,
"hit_rate": 0.4069,
"mean_conf": 73.3,
"insufficient": false,
"no_data": false
},
{
"model": "gpt-5.6-sol",
"bucket": "80-100",
"n": 6,
"hit_rate": 0.5,
"mean_conf": 81,
"insufficient": true,
"no_data": false
}
],
"sub50": [
{
"model": "claude-opus-5",
"n": 0,
"hit_rate": null
},
{
"model": "claude-fable-5",
"n": 0,
"hit_rate": null
},
{
"model": "grok-4.6",
"n": 0,
"hit_rate": null
},
{
"model": "qwen-3.8-max",
"n": 41,
"hit_rate": 0.3659
},
{
"model": "deepseek-v4-pro",
"n": 15,
"hit_rate": 0.6667
},
{
"model": "gemini-3.1-pro",
"n": 0,
"hit_rate": null
},
{
"model": "gpt-5.6-sol",
"n": 0,
"hit_rate": null
}
],
"by_fh": [
{
"model": "claude-opus-5",
"fh": "1h",
"n": 1357,
"hit_rate": 0.4724,
"flag": null
},
{
"model": "claude-opus-5",
"fh": "4h",
"n": 761,
"hit_rate": 0.477,
"flag": null
},
{
"model": "claude-opus-5",
"fh": "1d",
"n": 163,
"hit_rate": 0.5276,
"flag": null
},
{
"model": "claude-fable-5",
"fh": "1h",
"n": 1776,
"hit_rate": 0.4651,
"flag": null
},
{
"model": "claude-fable-5",
"fh": "4h",
"n": 861,
"hit_rate": 0.4623,
"flag": null
},
{
"model": "claude-fable-5",
"fh": "1d",
"n": 175,
"hit_rate": 0.5143,
"flag": null
},
{
"model": "grok-4.6",
"fh": "1h",
"n": 1713,
"hit_rate": 0.4513,
"flag": null
},
{
"model": "grok-4.6",
"fh": "4h",
"n": 851,
"hit_rate": 0.4512,
"flag": null
},
{
"model": "grok-4.6",
"fh": "1d",
"n": 144,
"hit_rate": 0.4792,
"flag": null
},
{
"model": "qwen-3.8-max",
"fh": "1h",
"n": 1887,
"hit_rate": 0.4542,
"flag": null
},
{
"model": "qwen-3.8-max",
"fh": "4h",
"n": 933,
"hit_rate": 0.4598,
"flag": null
},
{
"model": "qwen-3.8-max",
"fh": "1d",
"n": 181,
"hit_rate": 0.4641,
"flag": null
},
{
"model": "deepseek-v4-pro",
"fh": "1h",
"n": 1576,
"hit_rate": 0.4505,
"flag": null
},
{
"model": "deepseek-v4-pro",
"fh": "4h",
"n": 851,
"hit_rate": 0.463,
"flag": null
},
{
"model": "deepseek-v4-pro",
"fh": "1d",
"n": 151,
"hit_rate": 0.4636,
"flag": null
},
{
"model": "gemini-3.1-pro",
"fh": "1h",
"n": 1951,
"hit_rate": 0.4741,
"flag": null
},
{
"model": "gemini-3.1-pro",
"fh": "4h",
"n": 912,
"hit_rate": 0.466,
"flag": null
},
{
"model": "gemini-3.1-pro",
"fh": "1d",
"n": 170,
"hit_rate": 0.5529,
"flag": null
},
{
"model": "gpt-5.6-sol",
"fh": "1h",
"n": 1850,
"hit_rate": 0.4622,
"flag": null
},
{
"model": "gpt-5.6-sol",
"fh": "4h",
"n": 956,
"hit_rate": 0.4508,
"flag": null
},
{
"model": "gpt-5.6-sol",
"fh": "1d",
"n": 164,
"hit_rate": 0.4634,
"flag": null
}
],
"trading": [
{
"model": "claude-opus-5",
"n_trades": 2281,
"wr": 0.3792,
"pnl_net_usd": -25.2,
"pnl_gross_usd": 202.9,
"max_dd_usd": -160.92
},
{
"model": "claude-fable-5",
"n_trades": 2812,
"wr": 0.3617,
"pnl_net_usd": -96.55,
"pnl_gross_usd": 184.65,
"max_dd_usd": -210.11
},
{
"model": "gemini-3.1-pro",
"n_trades": 3033,
"wr": 0.3627,
"pnl_net_usd": -98.64,
"pnl_gross_usd": 204.66,
"max_dd_usd": -197.07
},
{
"model": "deepseek-v4-pro",
"n_trades": 2578,
"wr": 0.3658,
"pnl_net_usd": -176.06,
"pnl_gross_usd": 81.74,
"max_dd_usd": -182.97
},
{
"model": "qwen-3.8-max",
"n_trades": 3001,
"wr": 0.3505,
"pnl_net_usd": -198.93,
"pnl_gross_usd": 101.17,
"max_dd_usd": -231.65
},
{
"model": "gpt-5.6-sol",
"n_trades": 2970,
"wr": 0.3525,
"pnl_net_usd": -215.58,
"pnl_gross_usd": 81.42,
"max_dd_usd": -235.74
},
{
"model": "grok-4.6",
"n_trades": 2708,
"wr": 0.3353,
"pnl_net_usd": -243.84,
"pnl_gross_usd": 26.96,
"max_dd_usd": -247.38
}
],
"gap_month_over_month": [
{
"model": "claude-opus-5",
"gap_pp_prior_window": 15.6,
"gap_pp_monthly1": 13.2,
"hit_rate_prior_window": 0.4492,
"mean_conf_prior_window": 60.5,
"n_prior_window": 3448,
"delta_pp": -2.4
},
{
"model": "claude-fable-5",
"gap_pp_prior_window": 15.9,
"gap_pp_monthly1": 13.9,
"hit_rate_prior_window": 0.445,
"mean_conf_prior_window": 60.4,
"n_prior_window": 2690,
"delta_pp": -2.0
},
{
"model": "qwen-3.8-max",
"gap_pp_prior_window": 18.7,
"gap_pp_monthly1": 14.7,
"hit_rate_prior_window": 0.4568,
"mean_conf_prior_window": 64.4,
"n_prior_window": 3371,
"delta_pp": -4.0
},
{
"model": "grok-4.6",
"gap_pp_prior_window": 14.2,
"gap_pp_monthly1": 15.1,
"hit_rate_prior_window": 0.453,
"mean_conf_prior_window": 59.5,
"n_prior_window": 2265,
"delta_pp": 0.9
},
{
"model": "deepseek-v4-pro",
"gap_pp_prior_window": 16.6,
"gap_pp_monthly1": 15.4,
"hit_rate_prior_window": 0.4505,
"mean_conf_prior_window": 61.6,
"n_prior_window": 1869,
"delta_pp": -1.2
},
{
"model": "gemini-3.1-pro",
"gap_pp_prior_window": 20.0,
"gap_pp_monthly1": 18.8,
"hit_rate_prior_window": 0.4511,
"mean_conf_prior_window": 65.1,
"n_prior_window": 2813,
"delta_pp": -1.2
},
{
"model": "gpt-5.6-sol",
"gap_pp_prior_window": 22.1,
"gap_pp_monthly1": 21.8,
"hit_rate_prior_window": 0.4431,
"mean_conf_prior_window": 66.4,
"n_prior_window": 3085,
"delta_pp": -0.3
},
{
"model": "Field",
"gap_pp_prior_window": 17.8,
"gap_pp_monthly1": 16.3,
"hit_rate_prior_window": 0.4498,
"mean_conf_prior_window": 62.8,
"n_prior_window": 19541,
"delta_pp": -1.5
}
],
"ok_rate_by_model": {
"claude-opus-5": 1.0,
"claude-fable-5": 0.9989,
"grok-4.6": 1.0,
"qwen-3.8-max": 0.9906,
"deepseek-v4-pro": 0.9997,
"gemini-3.1-pro": 0.9992,
"gpt-5.6-sol": 1.0
},
"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."
}
},
"gap_drift": {
"rows": [
{
"model": "claude-opus-5",
"gap_pp_by_week": [
20.0,
19.2,
2.2,
18.8
],
"month_gap_pp": 13.2,
"gap_slope_pp_per_week": -2.06,
"gap_range_pp": 17.8,
"weeks_narrowing": 2,
"direction": "narrowing",
"direction_reason": "OLS slope over weeks 1..4"
},
{
"model": "claude-fable-5",
"gap_pp_by_week": [
19.2,
18.1,
4.7,
17.9
],
"month_gap_pp": 13.9,
"gap_slope_pp_per_week": -1.73,
"gap_range_pp": 14.5,
"weeks_narrowing": 2,
"direction": "narrowing",
"direction_reason": "OLS slope over weeks 1..4"
},
{
"model": "qwen-3.8-max",
"gap_pp_by_week": [
18.3,
14.1,
7.4,
16.9
],
"month_gap_pp": 14.7,
"gap_slope_pp_per_week": -1.09,
"gap_range_pp": 10.9,
"weeks_narrowing": 2,
"direction": "narrowing",
"direction_reason": "OLS slope over weeks 1..4"
},
{
"model": "grok-4.6",
"gap_pp_by_week": [
18.3,
15.8,
9.4,
20.3
],
"month_gap_pp": 15.1,
"gap_slope_pp_per_week": -0.04,
"gap_range_pp": 10.9,
"weeks_narrowing": 2,
"direction": "flat",
"direction_reason": "abs(slope) < 0.5 pp/week"
},
{
"model": "deepseek-v4-pro",
"gap_pp_by_week": [
20.3,
18.2,
9.4,
15.9
],
"month_gap_pp": 15.4,
"gap_slope_pp_per_week": -2.2,
"gap_range_pp": 10.9,
"weeks_narrowing": 2,
"direction": "narrowing",
"direction_reason": "OLS slope over weeks 1..4"
},
{
"model": "gemini-3.1-pro",
"gap_pp_by_week": [
22.2,
18.4,
13.1,
22.5
],
"month_gap_pp": 18.8,
"gap_slope_pp_per_week": -0.44,
"gap_range_pp": 9.4,
"weeks_narrowing": 2,
"direction": "flat",
"direction_reason": "abs(slope) < 0.5 pp/week"
},
{
"model": "gpt-5.6-sol",
"gap_pp_by_week": [
24.4,
24.0,
13.9,
24.8
],
"month_gap_pp": 21.8,
"gap_slope_pp_per_week": -0.89,
"gap_range_pp": 10.9,
"weeks_narrowing": 2,
"direction": "narrowing",
"direction_reason": "OLS slope over weeks 1..4"
},
{
"model": "Field (all models)",
"is_field": true,
"gap_pp_by_week": [
20.4,
18.3,
8.6,
19.6
],
"month_gap_pp": 16.3,
"gap_slope_pp_per_week": -1.21,
"gap_range_pp": 11.8,
"weeks_narrowing": 2,
"direction": "narrowing"
}
],
"slope_rule": "flat if |gap_slope_pp_per_week| < 0.5, else widening (slope > 0) / narrowing (slope < 0)",
"flat_rule_pp_per_week": 0.5,
"source": "weekly packs calibration.per_model / calibration.field (frozen cutoffs)",
"field_by_week": {
"W1": {
"n": 4042,
"mean_conf": 62.8,
"hit_rate": 0.4243,
"gap_pp": 20.4,
"brier": 0.2908
},
"W2": {
"n": 3670,
"mean_conf": 62.0,
"hit_rate": 0.4379,
"gap_pp": 18.3,
"brier": 0.2836
},
"W3": {
"n": 5130,
"mean_conf": 62.7,
"hit_rate": 0.5405,
"gap_pp": 8.6,
"brier": 0.2559
},
"W4": {
"n": 4576,
"mean_conf": 62.6,
"hit_rate": 0.4294,
"gap_pp": 19.6,
"brier": 0.2884
}
},
"weeks": {
"W1": "Aug 3-9",
"W2": "Aug 10-16",
"W3": "Aug 17-23",
"W4": "Aug 24-30"
}
},
"bucket_drift": {
"rows": [
{
"bucket": "50-60",
"week": "W1",
"n": 985,
"mean_conf": 57.0,
"hit_rate": 0.4376
},
{
"bucket": "50-60",
"week": "W2",
"n": 1222,
"mean_conf": 56.9,
"hit_rate": 0.4427
},
{
"bucket": "50-60",
"week": "W3",
"n": 1182,
"mean_conf": 57.2,
"hit_rate": 0.5025
},
{
"bucket": "50-60",
"week": "W4",
"n": 1079,
"mean_conf": 57.0,
"hit_rate": 0.4551
},
{
"bucket": "60-70",
"week": "W1",
"n": 2528,
"mean_conf": 63.2,
"hit_rate": 0.4288
},
{
"bucket": "60-70",
"week": "W2",
"n": 2021,
"mean_conf": 63.1,
"hit_rate": 0.4374
},
{
"bucket": "60-70",
"week": "W3",
"n": 3411,
"mean_conf": 63.0,
"hit_rate": 0.555
},
{
"bucket": "60-70",
"week": "W4",
"n": 2994,
"mean_conf": 62.9,
"hit_rate": 0.4259
},
{
"bucket": "70-80",
"week": "W1",
"n": 498,
"mean_conf": 72.8,
"hit_rate": 0.3795
},
{
"bucket": "70-80",
"week": "W2",
"n": 406,
"mean_conf": 73.0,
"hit_rate": 0.4212
},
{
"bucket": "70-80",
"week": "W3",
"n": 518,
"mean_conf": 72.8,
"hit_rate": 0.5309
},
{
"bucket": "70-80",
"week": "W4",
"n": 485,
"mean_conf": 73.0,
"hit_rate": 0.3918
},
{
"bucket": "80-100",
"week": "W1",
"n": 9,
"mean_conf": 81.2,
"hit_rate": 0.3333
},
{
"bucket": "80-100",
"week": "W2",
"n": 5,
"mean_conf": 80.4,
"hit_rate": 0.2
},
{
"bucket": "80-100",
"week": "W3",
"n": 13,
"mean_conf": 81.2,
"hit_rate": 0.8462
},
{
"bucket": "80-100",
"week": "W4",
"n": 12,
"mean_conf": 80.6,
"hit_rate": 0.4167
}
],
"level": "field (all models pooled)",
"source": "month_pull_split_by_slot_week (cutoff 2026-09-01T16:00:00+00:00)"
},
"prior_window_calibration": {
"per_model": [
{
"model": "grok-4.6",
"n": 2265,
"n_mature_all": 4247,
"coverage": 0.5333,
"hit_rate": 0.453,
"wilson_95": [
0.4326,
0.4735
],
"mean_conf": 59.5,
"gap_pp": 14.2,
"brier": 0.2699,
"insufficient": false,
"buckets": [
{
"bucket": "50-60",
"n": 1400,
"mean_conf": 57.3,
"hit_rate": 0.4593
},
{
"bucket": "60-70",
"n": 849,
"mean_conf": 63.2,
"hit_rate": 0.4452
},
{
"bucket": "70-80",
"n": 10,
"mean_conf": 71.6,
"hit_rate": 0.4
},
{
"bucket": "80-100",
"n": 0,
"mean_conf": null,
"hit_rate": null
}
],
"by_fh": {
"1h": {
"n": 1430,
"hit_rate": 0.4748
},
"4h": {
"n": 730,
"hit_rate": 0.4438
},
"1d": {
"n": 105,
"hit_rate": 0.219
}
}
},
{
"model": "claude-opus-5",
"n": 3448,
"n_mature_all": 6298,
"coverage": 0.5475,
"hit_rate": 0.4492,
"wilson_95": [
0.4327,
0.4659
],
"mean_conf": 60.5,
"gap_pp": 15.6,
"brier": 0.2733,
"insufficient": false,
"buckets": [
{
"bucket": "50-60",
"n": 1149,
"mean_conf": 57.5,
"hit_rate": 0.4613
},
{
"bucket": "60-70",
"n": 2285,
"mean_conf": 62.0,
"hit_rate": 0.4446
},
{
"bucket": "70-80",
"n": 12,
"mean_conf": 71.2,
"hit_rate": 0.1667
},
{
"bucket": "80-100",
"n": 0,
"mean_conf": null,
"hit_rate": null
}
],
"by_fh": {
"1h": {
"n": 2177,
"hit_rate": 0.4644
},
"4h": {
"n": 1072,
"hit_rate": 0.4394
},
"1d": {
"n": 199,
"hit_rate": 0.3367
}
}
},
{
"model": "claude-fable-5",
"n": 2690,
"n_mature_all": 6310,
"coverage": 0.4263,
"hit_rate": 0.445,
"wilson_95": [
0.4263,
0.4638
],
"mean_conf": 60.4,
"gap_pp": 15.9,
"brier": 0.2746,
"insufficient": false,
"buckets": [
{
"bucket": "50-60",
"n": 976,
"mean_conf": 56.6,
"hit_rate": 0.4682
},
{
"bucket": "60-70",
"n": 1713,
"mean_conf": 62.6,
"hit_rate": 0.4314
},
{
"bucket": "70-80",
"n": 1,
"mean_conf": 72,
"hit_rate": 1.0
},
{
"bucket": "80-100",
"n": 0,
"mean_conf": null,
"hit_rate": null
}
],
"by_fh": {
"1h": {
"n": 1701,
"hit_rate": 0.4638
},
"4h": {
"n": 839,
"hit_rate": 0.4315
},
"1d": {
"n": 150,
"hit_rate": 0.3067
}
}
},
{
"model": "deepseek-v4-pro",
"n": 1869,
"n_mature_all": 6314,
"coverage": 0.296,
"hit_rate": 0.4505,
"wilson_95": [
0.4281,
0.4731
],
"mean_conf": 61.6,
"gap_pp": 16.6,
"brier": 0.2801,
"insufficient": false,
"buckets": [
{
"bucket": "50-60",
"n": 506,
"mean_conf": 55.3,
"hit_rate": 0.4565
},
{
"bucket": "60-70",
"n": 1102,
"mean_conf": 63.0,
"hit_rate": 0.4446
},
{
"bucket": "70-80",
"n": 210,
"mean_conf": 72.0,
"hit_rate": 0.4476
},
{
"bucket": "80-100",
"n": 11,
"mean_conf": 80.8,
"hit_rate": 0.6364
}
],
"by_fh": {
"1h": {
"n": 1204,
"hit_rate": 0.4659
},
"4h": {
"n": 589,
"hit_rate": 0.4414
},
"1d": {
"n": 76,
"hit_rate": 0.2763
}
}
},
{
"model": "gemini-3.1-pro",
"n": 2813,
"n_mature_all": 6196,
"coverage": 0.454,
"hit_rate": 0.4511,
"wilson_95": [
0.4328,
0.4696
],
"mean_conf": 65.1,
"gap_pp": 20.0,
"brier": 0.2896,
"insufficient": false,
"buckets": [
{
"bucket": "50-60",
"n": 22,
"mean_conf": 55,
"hit_rate": 0.2727
},
{
"bucket": "60-70",
"n": 2229,
"mean_conf": 63.4,
"hit_rate": 0.45
},
{
"bucket": "70-80",
"n": 559,
"mean_conf": 72.4,
"hit_rate": 0.4597
},
{
"bucket": "80-100",
"n": 3,
"mean_conf": 80,
"hit_rate": 1.0
}
],
"by_fh": {
"1h": {
"n": 1749,
"hit_rate": 0.4637
},
"4h": {
"n": 897,
"hit_rate": 0.4426
},
"1d": {
"n": 167,
"hit_rate": 0.3653
}
}
},
{
"model": "qwen-3.8-max",
"n": 3371,
"n_mature_all": 6305,
"coverage": 0.5347,
"hit_rate": 0.4568,
"wilson_95": [
0.4401,
0.4737
],
"mean_conf": 64.4,
"gap_pp": 18.7,
"brier": 0.291,
"insufficient": false,
"buckets": [
{
"bucket": "50-60",
"n": 260,
"mean_conf": 54.6,
"hit_rate": 0.55
},
{
"bucket": "60-70",
"n": 2030,
"mean_conf": 63.5,
"hit_rate": 0.4478
},
{
"bucket": "70-80",
"n": 829,
"mean_conf": 73.6,
"hit_rate": 0.4463
},
{
"bucket": "80-100",
"n": 41,
"mean_conf": 82.2,
"hit_rate": 0.5366
}
],
"by_fh": {
"1h": {
"n": 2147,
"hit_rate": 0.4681
},
"4h": {
"n": 1047,
"hit_rate": 0.4527
},
"1d": {
"n": 177,
"hit_rate": 0.3446
}
}
},
{
"model": "gpt-5.6-sol",
"n": 3085,
"n_mature_all": 6315,
"coverage": 0.4885,
"hit_rate": 0.4431,
"wilson_95": [
0.4257,
0.4607
],
"mean_conf": 66.4,
"gap_pp": 22.1,
"brier": 0.2984,
"insufficient": false,
"buckets": [
{
"bucket": "50-60",
"n": 61,
"mean_conf": 58.4,
"hit_rate": 0.3934
},
{
"bucket": "60-70",
"n": 2380,
"mean_conf": 64.9,
"hit_rate": 0.4471
},
{
"bucket": "70-80",
"n": 641,
"mean_conf": 73.0,
"hit_rate": 0.4321
},
{
"bucket": "80-100",
"n": 3,
"mean_conf": 81,
"hit_rate": 0.6667
}
],
"by_fh": {
"1h": {
"n": 1954,
"hit_rate": 0.4539
},
"4h": {
"n": 966,
"hit_rate": 0.441
},
"1d": {
"n": 165,
"hit_rate": 0.3273
}
}
}
],
"field": {
"n": 19541,
"hit_rate": 0.4498,
"wilson_95": [
0.4429,
0.4568
],
"mean_conf": 62.8,
"gap_pp": 17.8,
"brier": 0.2831,
"insufficient": false,
"buckets": [
{
"bucket": "50-60",
"n": 4374,
"mean_conf": 56.8,
"hit_rate": 0.465
},
{
"bucket": "60-70",
"n": 12588,
"mean_conf": 63.3,
"hit_rate": 0.4448
},
{
"bucket": "70-80",
"n": 2262,
"mean_conf": 73.0,
"hit_rate": 0.4443
},
{
"bucket": "80-100",
"n": 58,
"mean_conf": 81.8,
"hit_rate": 0.5862
}
]
}
},
"prior_window_trading": [
{
"model": "deepseek-v4-pro",
"trades": 1869,
"wr": 0.3216,
"net_pnl": -293.78,
"gross_pnl": -106.88,
"max_dd": -293.95,
"n_trades": 1869,
"pnl_net_usd": -293.78,
"pnl_gross_usd": -106.88,
"max_dd_usd": -293.95
},
{
"model": "grok-4.6",
"trades": 2265,
"wr": 0.3135,
"net_pnl": -376.69,
"gross_pnl": -150.19,
"max_dd": -377.25,
"n_trades": 2265,
"pnl_net_usd": -376.69,
"pnl_gross_usd": -150.19,
"max_dd_usd": -377.25
},
{
"model": "gemini-3.1-pro",
"trades": 2813,
"wr": 0.3185,
"net_pnl": -431.68,
"gross_pnl": -150.38,
"max_dd": -433.49,
"n_trades": 2813,
"pnl_net_usd": -431.68,
"pnl_gross_usd": -150.38,
"max_dd_usd": -433.49
},
{
"model": "claude-fable-5",
"trades": 2690,
"wr": 0.3204,
"net_pnl": -444.23,
"gross_pnl": -175.23,
"max_dd": -445.51,
"n_trades": 2690,
"pnl_net_usd": -444.23,
"pnl_gross_usd": -175.23,
"max_dd_usd": -445.51
},
{
"model": "qwen-3.8-max",
"trades": 3371,
"wr": 0.3228,
"net_pnl": -490.4,
"gross_pnl": -153.3,
"max_dd": -491.19,
"n_trades": 3371,
"pnl_net_usd": -490.4,
"pnl_gross_usd": -153.3,
"max_dd_usd": -491.19
},
{
"model": "gpt-5.6-sol",
"trades": 3085,
"wr": 0.317,
"net_pnl": -503.71,
"gross_pnl": -195.21,
"max_dd": -504.75,
"n_trades": 3085,
"pnl_net_usd": -503.71,
"pnl_gross_usd": -195.21,
"max_dd_usd": -504.75
},
{
"model": "claude-opus-5",
"trades": 3448,
"wr": 0.3176,
"net_pnl": -554.53,
"gross_pnl": -209.73,
"max_dd": -554.98,
"n_trades": 3448,
"pnl_net_usd": -554.53,
"pnl_gross_usd": -209.73,
"max_dd_usd": -554.98
}
],
"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": "Field gap (conf - hit), pp",
"weeks": {
"W1": {
"value": "+20.4",
"reading": "positive"
},
"W2": {
"value": "+18.3",
"reading": "positive"
},
"W3": {
"value": "+8.6",
"reading": "positive"
},
"W4": {
"value": "+19.6",
"reading": "positive"
}
},
"month": {
"value": "+16.3",
"reading": "positive"
},
"agree_count": 4,
"verdict": "Confirmed (4/4)"
},
{
"metric": "Lowest-Brier model",
"weeks": {
"W1": {
"value": "fable-5",
"reading": "claude-fable-5"
},
"W2": {
"value": "qwen-3.8",
"reading": "qwen-3.8-max"
},
"W3": {
"value": "opus-5",
"reading": "claude-opus-5"
},
"W4": {
"value": "deepseek",
"reading": "deepseek-v4-pro"
}
},
"month": {
"value": "opus-5",
"reading": "claude-opus-5"
},
"agree_count": 1,
"verdict": "Rejected (1/4)"
},
{
"metric": "Widest-gap model",
"weeks": {
"W1": {
"value": "gpt-5.6",
"reading": "gpt-5.6-sol"
},
"W2": {
"value": "gpt-5.6",
"reading": "gpt-5.6-sol"
},
"W3": {
"value": "gpt-5.6",
"reading": "gpt-5.6-sol"
},
"W4": {
"value": "gpt-5.6",
"reading": "gpt-5.6-sol"
}
},
"month": {
"value": "gpt-5.6",
"reading": "gpt-5.6-sol"
},
"agree_count": 4,
"verdict": "Confirmed (4/4)"
}
]
}
},
"key_finding": {
"label": "KEY FINDING · OBSERVATION (one monthly window)",
"text": "Over the month the field's overconfidence gap read +16.3pp against 62.6 stated mean confidence, the weekly gaps ran +8.6pp to +20.4pp inside it, and the field slope across the four weeks is -1.21 pp/week (narrowing).",
"evidence_level": "observation"
},
"key_findings": [
"**OBSERVATION -- The monthly gap read +16.3pp.** Directional calls hit 46.3% against 62.6 stated mean confidence over 19,383 calls. Field Brier is 0.2784 and 0 of 7 model lines scored below 0.25, the score of an uninformative always-50% predictor. One month, descriptive.",
"**The field gap slope across the four weeks is -1.21 pp/week (narrowing).** Across those weeks the field gap ran +20.4pp -> +18.3pp -> +8.6pp -> +19.6pp and the per-model slopes split 5 narrowing / 0 widening / 2 flat under the pack's own rule (|slope| < 0.5 pp/week = flat). claude-opus-5 is both the best-calibrated line (Brier 0.2677) and the hit-rate leader (47.8%).",
"**The widest weekly swing belongs to claude-opus-5.** Its weekly gap moved over a 17.8 pp range inside one month, against a field range of 11.8 pp. Gap moves against the prior window ran from -4.0pp for qwen-3.8-max (spliced) to +0.9pp for grok-4.6 (spliced); 6 of 7 lines sit below their prior-window gap.",
"**Trading, kept separate from every calibration table.** 0 of 7 lines finished net-positive over the month; net PnL ran -$25.20 (claude-opus-5, best) to -$243.84 (grok-4.6 (spliced), worst), field -$1054.79."
],
"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": "Every MarketMania forecast carries a model-stated confidence from 0 to 100 alongside its direction call. This report checks whether that number tracks reality: on a well-calibrated forecaster, calls made at 70% confidence should hit their direction about 70% of the time. The weekly series carries one point per model per issue. A calendar month carries four weekly points plus a pooled monthly figure and a prior-window baseline, which is what the slope table and the week-by-week confidence buckets below are computed from. Descriptive for this window.",
"how_to_read_this": "Gap pp = mean stated confidence minus hit-rate, in percentage points; positive means overconfident. Brier is the mean squared error of the stated probability against the realised outcome, so lower is better and 0.25 is what an uninformative always-50% predictor scores. Coverage = calls scored divided by the mature calls available to that model. Confidence buckets are the models' own natural breakpoints, not an arbitrary binning; cells below N=10 are marked insufficient and never used to rank anything. Prediction and trading metrics live in separate sections and are never combined. 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)). The monthly rows are one pull at the monthly cutoff (Mon Sep 1, 2026, 16:00 UTC); the weekly rows are the published issues at their own frozen cutoffs, so the two do not have to agree to the last decimal.",
"practical_implications": [
"Pooled over a whole month the field's gap is +16.3pp on 19,383 scored calls: stated confidence averaged 62.6 while the direction hit 46.3%.",
"The field gap spans 11.8 pp across the four weeks inside the month, against a pooled monthly gap of +16.3pp and a prior-window gap of +17.8pp; field mean stated confidence moved 62.8 -> 62.6 over the same pair of windows.",
"Trading this month finished 0 of 7 lines net-positive against 0 of 7 in the prior window; the pack labels two of the four weeks calm and two volatile."
],
"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.",
"Prediction metrics (this report) and trading metrics are kept in separate sections per methodology -- they are never combined into a single score. 95% Wilson CIs shown are descriptive, not inferential: observations inside one window are dependent, so read them as a range, not a formal coverage guarantee.",
"All 19,383 scored calls sit inside one calendar month, and four weekly points are four points: an OLS slope over them is a description of this month, not evidence of drift. 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. Models report confidence at discrete levels, not a continuous scale; the buckets reflect those natural breakpoints and cells below N=10 are marked insufficient.",
"Underlying price series are reconstructed from trade entry prices (median per symbol-slot), not an independent tick feed. 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."
},
"platform_note_tpsl": "**Context, not a finding.** Since Aug 13 the public sandbox can trade with model-specific TP/SL multipliers learned from this same history (v1); since Aug 19, v2 adds per-ticker and confidence-bucket (50-70 / 70-100) resolution. Both switches landed INSIDE this window, so the month spans three feature states. That feature consumes calibration history; it does not feed back into any table in this report, which measures stated confidence vs. direction-hit only. No effect size is claimed for v1 or v2. The toggle lives at marketmania.ai/indices.",
"living_series_note": "**Issue #1.** Monthly Calibration is a living monthly comparison; each issue appends one more calendar month of confidence-vs-hit data. This first issue sets the baseline: field gap +16.3pp on 19,383 scored calls, field Brier 0.2784, weekly gaps +20.4pp -> +18.3pp -> +8.6pp -> +19.6pp, field slope -1.21 pp/week (narrowing). 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 Consensus Watch #1",
"url": "https://marketmania.ai/research/reports/consensus-watch-monthly-2026-08.pdf"
},
{
"title": "Monthly Model Watch #1",
"url": "https://marketmania.ai/research/reports/model-watch-monthly-2026-08.pdf"
},
{
"title": "Weekly Calibration #4",
"url": "https://marketmania.ai/research/reports/weekly-calibration-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.",
"citation": {
"bibtex_key": "mm_calibration_monthly_2026m08",
"title": "Monthly Calibration #1: confidence vs. direction-hit, Aug 1-31 2026",
"author": "MarketMania Research",
"year": 2026,
"month": "September",
"day": 4,
"url": "https://marketmania.ai/research/reports/calibration-monthly-2026-08.pdf",
"note": "Methodology v1.1; window Aug 1-31, 2026 UTC; source monthly_metrics_2026-08.json"
},
"research_to_date": {
"this_report": {
"scored_observations": 19383
},
"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": 20
},
"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": "calibration-monthly-2026-08",
"series": "monthly",
"seriesLabel": "Monthly · Calibration",
"date": "September 4, 2026",
"title": "Monthly Calibration #1",
"summary": "19,383 directional calls over one calendar month asked whether stated confidence tracks reality. The field hit 46.3% against 62.6 stated mean confidence, an overconfidence gap of +16.3pp with field Brier at 0.2784. Inside the month the field gap ran +20.4pp -> +18.3pp -> +8.6pp -> +19.6pp across the four published weeks, an OLS slope of -1.21 pp per week (narrowing), and the per-model verdicts split 5 narrowing / 0 widening / 2 flat. claude-opus-5 is the best-calibrated line at 0.2677 and claude-opus-5 leads on hit-rate at 47.8%. Trading, kept in its own section, finished with 0 of 7 lines net-positive as field accuracy moved 45.0% -> 46.3% (+1.4pp) against the partial July.",
"stats": [
"19,383 directional calls",
"gap +16.3pp slope -1.21",
"0 of 7 net-positive"
],
"manifest_key_finding": "Over August the overconfidence gap read +16.3pp on 19,383 scored calls with field Brier 0.2784; inside the month the weekly gaps ran +20.4pp -> +18.3pp -> +8.6pp -> +19.6pp, an OLS slope of -1.21 pp per week (narrowing), and trading, kept separate, finished 0 of 7 lines net-positive.",
"files": {
"pdf": "/research/reports/calibration-monthly-2026-08.pdf",
"md": "/research/reports/calibration-monthly-2026-08.md",
"json": "/research/reports/calibration-monthly-2026-08.json"
}
},
"social": {
"telegram": {
"photo_caption_html": "🎯 Monthly Calibration #1 — monthly series (window Aug 1–31)\n\nResearch question: when a model says 70, does it hit 70% of the time?\n\n19,383 directional calls, 7 model lines, one calendar month:\n• Field hit 46.3% vs 62.6 stated — gap +16.3pp, Brier 0.2784\n• Drift inside the month: gap +20.4pp -> +18.3pp -> +8.6pp -> +19.6pp week by week, slope -1.21 pp/wk (narrowing)\n• Per-model: 5 narrowing / 0 widening / 2 flat\n• Best-calibrated: claude-opus-5 (Brier 0.2677); hit-rate leader claude-opus-5 at 47.8%\n• Market check: field 45.0% -> 46.3% (+1.4pp) vs the partial July; trading separate — 0 of 7 net-positive\n\nNote: prior window = PARTIAL July (forecasts since Jul 11); weekly gaps from the frozen weekly packs\n\n📄 Full report (PDF) · all research\n\n#cryptotrading #LLM #AI #crypto #trading #calibration #research #Monthly",
"document_caption_html": "📄 Monthly Calibration #1 — full report (PDF).\nWeb copy: calibration-monthly-2026-08.pdf · machine-readable: JSON · all research\n\n#cryptotrading #LLM #AI #crypto #trading #calibration #research #Monthly",
"caption_len": 1001,
"document_caption_len": 420,
"limit": 1024
},
"x": {
"main": "Monthly Calibration #1 is out.\n\n19,383 LLM calls over August: 46.3% hit vs 62.6 stated confidence — a gap of +16.3pp, drifting -1.21 pp/week across the four weeks inside the month.\n\nFull report: https://marketmania.ai/research/reports/calibration-monthly-2026-08.pdf\n\n#AI #Crypto #Research #Calibration",
"main_tco_len": 254,
"reply": "Gap W1 -> month (pp):\nfable-5: 19.2 -> 13.9\nopus-5: 20.0 -> 13.2\ndeepseek: 20.3 -> 15.4\ngemini-3.1: 22.2 -> 18.8\ngpt-5.6: 24.4 -> 21.8\ngrok-4.6: 18.3 -> 15.1\nqwen-3.8: 18.3 -> 14.7\nField +16.3pp, slope -1.21 pp/wk. v1.1.\nData: https://marketmania.ai/research/reports/calibration-monthly-2026-08.json",
"reply_tco_len": 250,
"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"
}
}