{ "slug": "consensus-watch-2026-08-10", "series": "weekly", "series_label": "WEEKLY · CONSENSUS", "issue": 2, "title": "Consensus Watch #2", "language": "en", "window_start": "2026-08-10T00:00:00+00:00", "window_end": "2026-08-17T00:00:00+00:00", "window_text": "Aug 10-16, 2026 UTC", "cutoff": "2026-08-17T16:00:00+00:00", "generated_at": "2026-08-19T07:14:48.804827+00:00", "methodology": "v1.1 (2026-08-10)", "methodology_hash": "e66c7e8c864a2233", "source": "weekly_metrics_2026-08-10.json", "research_question": "if a model's forecast matches the consensus of the other models, is the forecast more reliable?", "method": "leave-one-out strict majority, >=3 directional peers", "base_field_hit": { "n": 3670, "hit_rate": 0.4379 }, "market_state": { "panel": "market_state_6p2", "week": { "start": "2026-08-10", "end": "2026-08-17" }, "prev_week": { "start": "2026-08-03", "end": "2026-08-10" }, "generated_utc": "2026-08-19T06:26:44+00:00", "settings_module": "marketmania.settings", "current": { "per_symbol": { "BTC": { "net_pct": -3.08, "days": 7, "quote_vol_usd": 4676300109.0 }, "ETH": { "net_pct": -1.81, "days": 7, "quote_vol_usd": 1799344934.0 }, "SOL": { "net_pct": -2.16, "days": 7, "quote_vol_usd": 545506557.0 }, "BNB": { "net_pct": 0.13, "days": 7, "quote_vol_usd": 373245161.0 }, "XRP": { "net_pct": -3.52, "days": 7, "quote_vol_usd": 357704108.0 } }, "btc_net_pct": -3.08, "realized_vol_ann_pct": 10.03, "vol_n_returns": 7, "avg_daily_range_pct": 1.63, "hivol_days": 0, "hivol_days_of": 7, "volume_top5_usd": 7752100869.0, "volume_symbols_counted": 5, "avg_pairwise_corr": 0.52, "corr_pairs": 10 }, "previous": { "per_symbol": { "BTC": { "net_pct": 2.09, "days": 7, "quote_vol_usd": 4914474500.0 }, "ETH": { "net_pct": 1.34, "days": 7, "quote_vol_usd": 2141079318.0 }, "SOL": { "net_pct": 3.57, "days": 7, "quote_vol_usd": 680571970.0 }, "BNB": { "net_pct": 2.34, "days": 7, "quote_vol_usd": 388062382.0 }, "XRP": { "net_pct": -5.24, "days": 7, "quote_vol_usd": 382877404.0 } }, "btc_net_pct": 2.09, "realized_vol_ann_pct": 11.42, "vol_n_returns": 7, "avg_daily_range_pct": 1.64, "hivol_days": 0, "hivol_days_of": 7, "volume_top5_usd": 8507065574.0, "volume_symbols_counted": 5, "avg_pairwise_corr": 0.4, "corr_pairs": 10 }, "delta": { "btc_net_pct": -5.17, "realized_vol_ann_pct": -1.39, "avg_daily_range_pct": -0.01, "volume_top5_usd_pct": -8.9, "avg_pairwise_corr": 0.12 }, "_debug": { "models": { "Candle": "datalake.Candle", "MetricSnapshot": "datalake.MetricSnapshot", "EtfFlow": "datalake.EtfFlow" }, "candle_fields": [ "close", "exchange", "high", "id", "ingested_at", "low", "market_type", "open", "quote_volume", "source", "symbol", "timeframe", "ts_open", "volume" ], "market_types": [ "commodity", "crypto_futures", "crypto_spot", "forex", "stock" ], "timeframes": [ "1M", "1d", "1h", "1w", "4h" ], "picked": { "market_type": "crypto_spot", "timeframe": "1d", "ohlc": [ "open", "high", "low", "close" ], "qv": "quote_volume", "ts": "ts_open" }, "symbol_map": { "BTC": "BTCUSDT", "ETH": "ETHUSDT", "SOL": "SOLUSDT", "BNB": "BNBUSDT", "XRP": "XRPUSDT" }, "funding_trace": "Traceback (most recent call last):\n File \"/tmp/w73_market_state.py\", line 202, in \n out[\"current\"][\"funding_btc\"] = funding_block(W0, W1)\n ~~~~~~~~~~~~~^^^^^^^^\n File \"/tmp/w73_market_state.py\", line 190, in funding_block\n raise RuntimeError(\"funding field mapping failed: \" + json.dumps(dbg))\nRuntimeError: funding field mapping failed: {\"fields\": [\"exchange\", \"id\", \"market_type\", \"metric\", \"raw\", \"symbol\", \"ts\", \"value_num\"], \"ts\": \"ts\", \"val\": null, \"sym\": \"symbol\", \"symbols_seen\": [\"1000BONKUSDT\", \"1000FLOKIUSDT\", \"1000PEPEUSDT\", \"1000SHIBUSDT\", \"1INCHUSDT\", \"AAVEUSDT\", \"ADAUSDT\", \"ALGOUSDT\", \"ANKRUSDT\", \"APEUSDT\", \"APTUSDT\", \"ARBUSDT\", \"ATOMUSDT\", \"AVAXUSDT\", \"AXSUSDT\", \"BATUSDT\", \"BCHUSDT\", \"BLURUSDT\", \"BNBUSDT\", \"BONKUSDT\", \"BTCUSDT\", \"CAKEUSDT\", \"CELOUSDT\", \"CFXUSDT\", \"CHZUSDT\", \"COMPUSDT\", \"CRVUSDT\", \"DASHUSDT\", \"DOGEUSDT\", \"DOTUSDT\"], \"symbol_used\": \"BTCUSDT\"}\n", "etf_trace": "Traceback (most recent call last):\n File \"/tmp/w73_market_state.py\", line 246, in \n out[\"current\"][\"etf_flow\"] = etf_block(W0, W1)\n ~~~~~~~~~^^^^^^^^\n File \"/tmp/w73_market_state.py\", line 217, in etf_block\n raise RuntimeError(\"etf field mapping failed: \" + json.dumps(dbg))\nRuntimeError: etf field mapping failed: {\"fields\": {\"id\": \"BigAutoField\", \"asset\": \"CharField\", \"fund\": \"CharField\", \"flow_date\": \"DateField\", \"flow_musd\": \"DecimalField\", \"source\": \"CharField\", \"raw\": \"JSONField\", \"first_seen\": \"DateTimeField\", \"last_seen\": \"DateTimeField\"}, \"date\": \"flow_date\", \"asset\": \"asset\", \"val\": null}\n" }, "_errors": [ "funding: RuntimeError('funding field mapping failed: {\"fields\": [\"exchange\", \"id\", \"market_type\", \"metric\", \"raw\", \"symbol\", \"ts\", \"value_num\"], \"ts\": \"ts\", \"val\": null, \"sym\": \"symbol\", \"symbols_seen\": [\"1000BONKUSDT\", \"1000FLOKIUSDT\", \"1000PEPEUSDT\", \"1000SHIBUSDT\", \"1INCHUSDT\", \"AAVEUSDT\", \"ADAUSDT\", \"ALGOUSDT\", \"ANKRUSDT\", \"APEUSDT\", \"APTUSDT\", \"ARBUSDT\", \"ATOMUSDT\", \"AVAXUSDT\", \"AXSUSDT\", \"BATUSDT\", \"BCHUSDT\", \"BLURUSDT\", \"BNBUSDT\", \"BONKUSDT\", \"BTCUSDT\", \"CAKEUSDT\", \"CELOUSDT\", \"CFXUSDT\", \"CHZUSDT\", \"COMPUSDT\", \"CRVUSDT\", \"DASHUSDT\", \"DOGEUSDT\", \"DOTUSDT\"], \"symbol_used\": \"BTCUSDT\"}')", "etf: RuntimeError('etf field mapping failed: {\"fields\": {\"id\": \"BigAutoField\", \"asset\": \"CharField\", \"fund\": \"CharField\", \"flow_date\": \"DateField\", \"flow_musd\": \"DecimalField\", \"source\": \"CharField\", \"raw\": \"JSONField\", \"first_seen\": \"DateTimeField\", \"last_seen\": \"DateTimeField\"}, \"date\": \"flow_date\", \"asset\": \"asset\", \"val\": null}')" ] }, "tables": { "pooled": { "agree": { "n": 2923, "hit_rate": 0.4307, "wilson_95": [ 0.4129, 0.4488 ] }, "disagree": { "n": 6, "hit_rate": 0.5, "wilson_95": [ 0.1876, 0.8124 ] }, "lift_pp": -6.9, "thin_or_tied": 741, "total_scoreable": 2929, "caveat": "lift_pp rests on disagree n=6; Wilson 95% CI [18.8%, 81.2%] swallows the agree bucket's CI [41.3%, 44.9%]; descriptive, one week only, not a durability claim" }, "by_cell": [ { "cell": "fh_1h_tf_1h", "fh": "1h", "tf": "1h", "n_subjects": 2346, "agree": { "n": 1865, "hit_rate": 0.4504, "wilson_95": [ 0.4279, 0.4731 ] }, "disagree": { "n": 6, "hit_rate": 0.5, "wilson_95": [ 0.1876, 0.8124 ] }, "lift_pp": -5.0, "thin_or_tied": 475, "by_margin": { "2": { "n": 9, "hit_rate": 0.6667 }, "3": { "n": 272, "hit_rate": 0.5 }, "4": { "n": 350, "hit_rate": 0.38 }, "5": { "n": 492, "hit_rate": 0.4024 }, "6": { "n": 742, "hit_rate": 0.4946 } } }, { "cell": "fh_4h_tf_4h", "fh": "4h", "tf": "4h", "n_subjects": 577, "agree": { "n": 435, "hit_rate": 0.4092, "wilson_95": [ 0.364, 0.456 ] }, "disagree": { "n": 0, "hit_rate": null, "wilson_95": [ null, null ] }, "lift_pp": null, "thin_or_tied": 142, "by_margin": { "3": { "n": 88, "hit_rate": 0.4659 }, "4": { "n": 50, "hit_rate": 0.3 }, "5": { "n": 108, "hit_rate": 0.5741 }, "6": { "n": 189, "hit_rate": 0.3175 } } }, { "cell": "fh_4h_tf_1h", "fh": "4h", "tf": "1h", "n_subjects": 553, "agree": { "n": 463, "hit_rate": 0.4082, "wilson_95": [ 0.3644, 0.4536 ] }, "disagree": { "n": 0, "hit_rate": null, "wilson_95": [ null, null ] }, "lift_pp": null, "thin_or_tied": 90, "by_margin": { "3": { "n": 52, "hit_rate": 0.4231 }, "4": { "n": 110, "hit_rate": 0.4455 }, "5": { "n": 126, "hit_rate": 0.3571 }, "6": { "n": 175, "hit_rate": 0.4171 } } }, { "cell": "fh_1d_tf_1d", "fh": "1d", "tf": "1d", "n_subjects": 99, "agree": { "n": 88, "hit_rate": 0.3409, "wilson_95": [ 0.2504, 0.4447 ] }, "disagree": { "n": 0, "hit_rate": null, "wilson_95": [ null, null ] }, "lift_pp": null, "thin_or_tied": 11, "by_margin": { "3": { "n": 12, "hit_rate": 1.0 }, "4": { "n": 5, "hit_rate": 1.0 }, "5": { "n": 36, "hit_rate": 0.1944 }, "6": { "n": 35, "hit_rate": 0.1714 } } }, { "cell": "fh_1d_tf_4h", "fh": "1d", "tf": "4h", "n_subjects": 95, "agree": { "n": 72, "hit_rate": 0.3056, "wilson_95": [ 0.2113, 0.4195 ] }, "disagree": { "n": 0, "hit_rate": null, "wilson_95": [ null, null ] }, "lift_pp": null, "thin_or_tied": 23, "by_margin": { "3": { "n": 4, "hit_rate": 1.0 }, "4": { "n": 10, "hit_rate": 0.5 }, "5": { "n": 30, "hit_rate": 0.4333 }, "6": { "n": 28, "hit_rate": 0.0 } } } ], "by_margin": [ { "margin": 2, "n": 9, "hit_rate": 0.6667 }, { "margin": 3, "n": 428, "hit_rate": 0.5023 }, { "margin": 4, "n": 525, "hit_rate": 0.3943 }, { "margin": 5, "n": 792, "hit_rate": 0.4104 }, { "margin": 6, "n": 1169, "hit_rate": 0.4328 } ], "per_model": [ { "model": "claude-fable-5", "agree": { "n": 401, "hit_rate": 0.419, "wilson_95": [ 0.3717, 0.4678 ] }, "disagree": { "n": 0, "hit_rate": null, "wilson_95": [ null, null ] }, "thin_or_tied": 71 }, { "model": "claude-opus-5", "agree": { "n": 334, "hit_rate": 0.4132, "wilson_95": [ 0.3616, 0.4667 ] }, "disagree": { "n": 0, "hit_rate": null, "wilson_95": [ null, null ] }, "thin_or_tied": 27 }, { "model": "deepseek-v4-pro", "agree": { "n": 371, "hit_rate": 0.4205, "wilson_95": [ 0.3713, 0.4713 ] }, "disagree": { "n": 2, "hit_rate": 0.0, "wilson_95": [ 0.0, 0.6576 ] }, "thin_or_tied": 81 }, { "model": "gemini-3.1-pro", "agree": { "n": 406, "hit_rate": 0.4458, "wilson_95": [ 0.3982, 0.4944 ] }, "disagree": { "n": 4, "hit_rate": 0.75, "wilson_95": [ 0.3006, 0.9544 ] }, "thin_or_tied": 145 }, { "model": "gpt-5.6-sol", "agree": { "n": 477, "hit_rate": 0.4088, "wilson_95": [ 0.3656, 0.4535 ] }, "disagree": { "n": 0, "hit_rate": null, "wilson_95": [ null, null ] }, "thin_or_tied": 159 }, { "model": "grok-4.5", "agree": { "n": 482, "hit_rate": 0.4544, "wilson_95": [ 0.4104, 0.499 ] }, "disagree": { "n": 0, "hit_rate": null, "wilson_95": [ null, null ] }, "thin_or_tied": 138 }, { "model": "qwen-3.8-max", "agree": { "n": 452, "hit_rate": 0.4469, "wilson_95": [ 0.4017, 0.493 ] }, "disagree": { "n": 0, "hit_rate": null, "wilson_95": [ null, null ] }, "thin_or_tied": 120 } ], "herd_cards": [ { "slot": "2026-08-11T12:01:00+00:00", "symbol": "BNB", "fh": "4h", "tf": "1h", "side": "long", "n_models": 7, "mean_conf": 71, "hit_rate": 0.0 }, { "slot": "2026-08-12T12:01:00+00:00", "symbol": "SOL", "fh": "1h", "tf": "1h", "side": "long", "n_models": 7, "mean_conf": 69, "hit_rate": 0.0 }, { "slot": "2026-08-11T13:01:00+00:00", "symbol": "BNB", "fh": "1h", "tf": "1h", "side": "long", "n_models": 7, "mean_conf": 68.1, "hit_rate": 0.0 }, { "slot": "2026-08-11T22:01:00+00:00", "symbol": "SOL", "fh": "1h", "tf": "1h", "side": "long", "n_models": 7, "mean_conf": 68, "hit_rate": 0.0 }, { "slot": "2026-08-12T12:01:00+00:00", "symbol": "ETH", "fh": "4h", "tf": "4h", "side": "long", "n_models": 7, "mean_conf": 67.9, "hit_rate": 0.0 } ], "issue1_margins": [ { "margin": 3, "hit_rate": 0.305 }, { "margin": 4, "hit_rate": 0.433 }, { "margin": 5, "hit_rate": 0.499 }, { "margin": 6, "hit_rate": 0.336 } ], "counters": { "forecasts_total": 9380, "ok_in_gate": 9307, "out_of_gate_1w": 69, "invalid": 4, "mature": 9307, "mature_directional": 3670, "mature_sideways": 5637, "pending_next_issue": 0, "pending_by_fh": {}, "late_closes": 0, "uptime": [ { "fh": "1h", "tf": "1h", "slots_seen": 168, "slots_expected": 168 }, { "fh": "4h", "tf": "4h", "slots_seen": 42, "slots_expected": 42 }, { "fh": "4h", "tf": "1h", "slots_seen": 42, "slots_expected": 42 }, { "fh": "1d", "tf": "1d", "slots_seen": 7, "slots_expected": 7 }, { "fh": "1d", "tf": "4h", "slots_seen": 7, "slots_expected": 7 } ], "regime_days": { "2026-08-10": { "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" } } } }, "herd_note": "299 cells had 6+ directional models on one side; unanimous_cells_ge6 per weekly metrics", "key_finding": { "label": "KEY FINDING · OBSERVATION (one weekly window)", "text": "The margin curve inverted: the thinnest majorities (3 peers) were the only tier to beat the field, and near-unanimity stayed below base for a second week.", "evidence_level": "observation" }, "key_findings": [ "OBSERVATION -- Herding intensified, and still did not pay. Of 2,929 LOO-scoreable directional calls, 2,923 (99.8%) sided with peers' majority (issue #1: 99.4%); agree-hit was 43.1% vs. 50.0% for the 6 disagreers, a pooled lift of -6.9pp on n=6 with a Wilson 95% CI of [18.8%, 81.2%] that swallows the agree bucket's own CI [41.3%, 44.9%]. Descriptive, one week, not a claim of a durable edge.", "The margin curve flipped upside-down vs. issue #1: 3 peers 50.2% (was 30.5%), 4 peers 39.4% (was 43.3%), 5 peers 41.0% (was 49.9%), full unanimity 43.3% (was 33.6%). Only the thinnest majority beat the 43.8% base; two weeks, two opposite shapes.", "Daily-horizon herds were the week's disaster: 1d/1d margin-5/6 hit 19.4%/17.1%; 1d/4h near-unanimous went 0 for 28.", "Herd failures were real: 299 cells with 6+ on one side; the 5 worst -- all 7 models long at ~70 mean confidence -- all resolved 0%, in a week where 'sideways' was 60.6% of mature forecasts. All five landed on Aug 11-12." ], "practical_implications": [ "Last week unanimity behaved like a crowding signal; this week it merely matched base while the whole tier ordering flipped. The practical read is not 'fade the herd' -- it is that NO consensus tier has yet shown a stable meaning, so a consensus filter should not carry weight in a config on two weeks of evidence.", "Treat any agree-vs-disagree lift as regime-dependent description, not strategy: this week's -6.9pp rests on 6 disagreeing calls whose CI swallows the agree bucket's entirely.", "Consensus Watch is a living series: each issue adds one more week to the LOO comparison and one more point on the margin curve. Two points cannot separate shape from noise; several issues can." ], "limitations": [ "Single week (Aug 10-16, 2026 UTC). Every finding is descriptive for this window only.", "Observations are not independent: overlapping symbol/FH/TF cells hour after hour; Wilson intervals are descriptive, not inferential.", "Correlational, not causal: shared inputs mean shared blind spots; LOO removes self-match bias, not this confound.", "The pooled lift rests on 6 disagreeing calls, all in FH 1h; per-model disagree counts run 0-4.", "741 calls (20.2% of the 3,670 mature directional universe) were thin or tied and are excluded from every consensus table." ], "citation": { "bibtex_key": "mm_consensus_watch_2026w33", "title": "Consensus Watch #2: does agreeing with the crowd make an LLM's market call safer?", "author": "MarketMania Research", "year": 2026, "month": "August", "day": 19, "url": "https://marketmania.ai/research/reports/consensus-watch-2026-08-10.pdf" }, "living_series_note": "Issue #2. Consensus Watch is a living weekly comparison; each issue appends one more week of leave-one-out agreement-vs-hit data. Issue #1's open question -- does full unanimity keep underperforming? -- closed as 'yes, but barely'; the sharper result is the inverted margin curve. Engine 1.1 powers the sandbox since Aug 18 (after this window closed); no cross-engine PnL comparisons are claimed.", "testing_next": [ "Next issue: does the margin-curve inversion persist -- or is tier ordering week-to-week noise? Three points make a trend.", "Monthly test (Sep 2): the agreement curve across regimes at monthly n." ], "related_research": [ { "title": "Weekly Calibration #2", "url": "https://marketmania.ai/research/reports/weekly-calibration-2026-08-10.pdf" }, { "title": "Weekly Model Watch #2", "url": "https://marketmania.ai/research/reports/model-watch-2026-08-10.pdf" }, { "title": "Config Watch #2", "url": "https://marketmania.ai/research/reports/config-watch-2026-08-12.pdf" }, { "title": "Consensus Watch #1", "url": "https://marketmania.ai/research/reports/consensus-watch-2026-08-03.pdf" } ], "related_research_note": "The three weekly reports publish together as one issue each week; Config Watch follows on its own cycle. Direct links are the posting rule from this wave on.", "research_to_date": { "this_report": { "loo_scored_observations": 2929 }, "platform": { "as_of_cutoff": "2026-08-17", "resolved_forecasts": 23458, "resolved_forecasts_since": "2026-07-11", "models_tracked": 9, "models_tracked_detail": "7 current frontier + 2 archived legacy generations", "assets": 5, "forecast_horizons": 5, "cadence": "hourly", "published_research_reports": 9 }, "source": "platform_counters_2026-08-17" } }