# Weekly Calibration #9

**WEEKLY · CALIBRATION** · October 7, 2026 · MarketMania Research · Weekly series

Window: **Sep 28-Oct 4, 2026 UTC** · Cutoff: **Mon Oct 5, 2026, 16:00 UTC** · Methodology **v1.1 (2026-08-10)**, hash `e66c7e8c864a2233`

PDF: https://marketmania.ai/research/reports/weekly-calibration-2026-09-28.pdf · Open data (JSON): https://marketmania.ai/research/reports/weekly-calibration-2026-09-28.json

> Research question: *"when a model says 70, does it hit 70% of the time?"*

## Research Snapshot

| RESEARCH SNAPSHOT |  |  |  |
|---|---|---|---|
| **4,331** directional calls scored | **7** models (stable lineup) | of **9,286** mature forecasts | window **Sep 28-Oct 4, 2026 UTC** |
| hit = direction rule: tp1/tp2->hit, sl->miss, expiry->sign of gross pnl | FH gate **1h / 4h / 1d** | base field hit **49.6%** vs **62.5** stated | cutoff **Mon Oct 5, 2026, 16:00 UTC** |
| market: BTC net **+2.44%** (prior +4.06%) | ann vol **21.1%** (was 52.2%) | TOP5 volume **$17.96B**, -20% w/w | pairwise corr **0.57** (was 0.71) |

## Key finding

> **KEY FINDING · OBSERVATION (one weekly window)**
>
> ### The field's overconfidence gap moved +16.1pp -> +12.9pp week over week, 5 of 7 model lines narrowed, and field Brier went 0.2751 -> 0.2698.

## TL;DR

- **OBSERVATION -- The gap read +12.9pp this week.** Directional calls hit 49.6% against 62.5 stated mean confidence (issue #8: +16.1pp on 46.8% vs 63.0). Field Brier is 0.2698 (was 0.2751) and 0 of 7 model lines scored below 0.25, the score of an uninformative always-50% predictor. One window, descriptive.
- **5 of 7 models narrowed their gap.** Gap moves ran from deepseek-v4-pro's -7.7pp (+17.0pp -> +9.3pp) to gemini-3.1-pro's +1.5pp (+16.2pp -> +17.7pp); 5 of 7 model lines narrowed week over week. grok-4.6 is both the lowest-Brier line (Brier 0.2566) and the hit-rate leader (52.7%).
- **The high-confidence buckets, N>=10 only.** deepseek-v4-pro 70-80 56.2% (n=16), gemini-3.1-pro 70-80 46.4% (n=209), gemini-3.1-pro 80-100 54.5% (n=22), gpt-5.6-sol 70-80 47.2% (n=216). Cells below N=10 are marked insufficient and rank nothing.
- **Trading, kept separate from every calibration table.** 0 of 7 lines finished net-positive; net PnL ran -$30.94 (grok-4.6, best) to -$91.49 (gemini-3.1-pro, worst), field -$408.24 against issue #8's -$315.68.

### Week-over-week chart (see PDF for the grouped bar chart)

Overconfidence gap in percentage points (mean stated confidence minus hit-rate), week over week, sorted by this week's gap. Short names: opus-5 = claude-opus-5, fable-5 = claude-fable-5, gemini-3.1 = gemini-3.1-pro, gpt-5.6 = gpt-5.6-sol, qwen-3.8 = qwen-3.8-max, deepseek = deepseek-v4-pro. The grok column compares two pure grok-4.6 weeks: the 4.5 -> 4.6 flip sits five windows back. "Was" values are the numbers published in issue #8 for the back-to-back window Sep 21-27.

## 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. With issue #9 the series has nine points on every model's gap -- enough to see movement, not enough to claim a trend: calibration and market regime move together, and durability claims belong in the Monthly series, not here.

## 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. The grok 4.5 -> 4.6 flip (Aug 24, 2026 09:22 UTC) sits five windows back (issue #4): both weeks of this issue are pure grok-4.6, with no 4.5-era rows in either window, so the grok week-over-week row is the fourth pure-4.6 against pure-4.6 comparison of the series.

## Calibration by model (sorted by Brier, best first)

| Model | N | Cov. | Hit rate | Mean conf | Gap pp | Brier | 95% CI |
|---|---|---|---|---|---|---|---|
| grok-4.6 | 431 | 32.6% | 52.7% | 60.4 | +7.7 | 0.2566 | 47.9%-57.3% |
| deepseek-v4-pro | 687 | 51.6% | 50.9% | 60.3 | +9.3 | 0.2594 | 47.2%-54.7% |
| qwen-3.8-max | 711 | 53.5% | 48.9% | 59.4 | +10.5 | 0.2608 | 45.3%-52.6% |
| claude-opus-5 | 534 | 40.2% | 50.7% | 61.1 | +10.3 | 0.2617 | 46.5%-55.0% |
| claude-fable-5 | 623 | 46.8% | 49.9% | 61.0 | +11.1 | 0.2630 | 46.0%-53.8% |
| gemini-3.1-pro | 694 | 52.2% | 49.0% | 66.6 | +17.7 | 0.2858 | 45.3%-52.7% |
| gpt-5.6-sol | 651 | 49.5% | 46.5% | 67.9 | +21.3 | 0.2951 | 42.7%-50.4% |
| **Field (all models)** | **4,331** | - | **49.6%** | **62.5** | **+12.9** | **0.2698** | - |

N = directional calls scored this window. Cov. = coverage, scored / mature calls available to that model. Gap pp = mean stated confidence minus hit-rate (positive = overconfident). 95% CI is the descriptive Wilson interval on hit-rate. The grok 4.5 -> 4.6 flip (Aug 24, 2026 09:22 UTC) sits five windows back (issue #4): both weeks of this issue are pure grok-4.6, with no 4.5-era rows in either window, so the grok week-over-week row is the fourth pure-4.6 against pure-4.6 comparison of the series.

## Calibration curve: hit-rate by stated-confidence bucket

| Model | 50-60 | 60-70 | 70-80 | 80-100 |
|---|---|---|---|---|
| claude-fable-5 | 47.3% (201) | 51.3% (421) | 0.0% (1) * | n/a (0) |
| claude-opus-5 | 53.4% (103) | 50.1% (431) | n/a (0) | n/a (0) |
| deepseek-v4-pro | 49.3% (270) | 51.9% (401) | 56.2% (16) | n/a (0) |
| gemini-3.1-pro | 60.0% (5) * | 49.8% (458) | 46.4% (209) | 54.5% (22) |
| gpt-5.6-sol | 50.0% (6) * | 46.2% (429) | 47.2% (216) | n/a (0) |
| grok-4.6 | 50.7% (138) | 53.9% (291) | 0.0% (2) * | n/a (0) |
| qwen-3.8-max | 46.4% (345) | 50.8% (352) | 75.0% (4) * | n/a (0) |

n in parentheses. * = N<10 (insufficient) -- no conclusions drawn from these cells. "n/a (0)" = no calls landed in that bucket this week. Sub-50 bucket (n greater than 0 only): qwen-3.8-max 60.0% (n=10). Mid-scale ranking: 4 of the 5 models with N>=10 in both 50-60 and 60-70 out-hit 50-60 from 60-70 this week (issue #8: 4 of 5).

## Hit-rate by forecast horizon (FH)

| Model | 1h | 4h | 1d |
|---|---|---|---|
| claude-fable-5 | 50.2% (434) | 48.6% (177) | 58.3% (12) |
| claude-opus-5 | 51.5% (375) | 49.3% (150) | 44.4% (9) * |
| deepseek-v4-pro | 50.3% (459) | 49.5% (192) | 66.7% (36) |
| gemini-3.1-pro | 49.7% (477) | 47.4% (192) | 48.0% (25) |
| gpt-5.6-sol | 46.2% (429) | 46.2% (199) | 56.5% (23) |
| grok-4.6 | 52.8% (301) | 52.0% (123) | 57.1% (7) * |
| qwen-3.8-max | 48.6% (469) | 48.4% (213) | 58.6% (29) |

n in parentheses; * = N<10 (insufficient). 5 of 7 model lines hit better on 1d than on 1h this window (issue #8: 2 of 7).

## Trading result (a different question)

A different question: is any of this profitable to trade. "Accurate" and "profitable" are not the same thing, and per methodology this section is never mixed into the prediction/calibration metrics above.

| Model | Trades | WR | Net PnL | Gross PnL | Max DD |
|---|---|---|---|---|---|
| grok-4.6 | 431 | 45.5% | -$30.94 | +$12.16 | -$55.18 |
| deepseek-v4-pro | 687 | 43.4% | -$38.00 | +$30.70 | -$63.69 |
| qwen-3.8-max | 711 | 41.6% | -$49.80 | +$21.30 | -$77.40 |
| claude-opus-5 | 534 | 40.6% | -$56.43 | -$3.03 | -$71.19 |
| claude-fable-5 | 623 | 41.6% | -$60.76 | +$1.54 | -$71.65 |
| gpt-5.6-sol | 651 | 37.6% | -$80.82 | -$15.72 | -$100.20 |
| gemini-3.1-pro | 694 | 38.3% | -$91.49 | -$22.09 | -$109.15 |

0 of 7 model lines finished net-positive and 4 gross-positive this week (issue #8: 0 of 7 net-positive, 6 gross-positive). Ranked best-to-worst net PnL. Per methodology this table is never combined with the calibration tables above; a week's PnL is regime as much as skill.

## Platform note: TP/SL calibration

> **Context, not a finding.** Since Aug 13 the public sandbox can trade with model-specific TP/SL multipliers learned from this same weekly history (v1); since Aug 19, v2 adds per-ticker and confidence-bucket (50-70 / 70-100) resolution. 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 v2. The toggle lives at marketmania.ai/indices.

## Market check: an up week on lower volatility

> **MARKET CHECK · OBSERVATION (two adjacent calendar weeks)**
>
> OBSERVATION — mean |1d move| 1.83% -> 0.87% (-52% rel), BTC realized vol 35.8% -> 32.2% (ann., hourly); field directional accuracy 46.8% -> 49.6% (+2.8pp), 5 of 7 models improved, sim win-rate up for 5 of 7, field sim PnL -$315.68 -> -$408.24 (adjacent calendar weeks Sep 21-27 vs Sep 28-Oct 4; descriptive, one pair of weeks, not a claim).
>
> Robustness: raw price-sign accuracy 49.2% -> 49.0% — the opposite direction to the trade-based hit rule.

| Measure | Sep 21-27 (week A) | Sep 28-Oct 4 (week B) | Change |
|---|---|---|---|
| Mean \|1d move\| (5 assets) | 1.83% | 0.87% | -52% rel |
| BTC realized vol (ann., hourly) | 35.8% | 32.2% | -3.6pp |
| Field directional accuracy | 46.8% | 49.6% | +2.8pp |
| Models improving hit-rate | -- | 5 of 7 | -- |
| Raw price-sign accuracy | 49.2% | 49.0% | -0.2pp |
| Field sim win-rate | 40.0% | 41.0% | +1.1pp |
| Field sim net PnL | -$315.68 | -$408.24 | -- |

Week A is exactly the issue-#8 window, so this slice and every week-over-week column in this report compare the same two back-to-back weeks. Descriptive, one pair of weeks; the hit rule is methodology v1.1 (trade-based) and the raw price-sign row is the robustness check (hourly closes at :00 against slots at :01). BTC realized vol here is the hourly-return series from the alive slice; the snapshot row's figure is the audited daily-candle estimate -- different estimators, both reported as measured. Lineage splice ACTIVE: grok-4.5 rows are aggregated into grok-4.6 (flip 2026-08-24T09:22:00+00:00, five windows back, inside issue #4's window). Week A (2026-09-21..2026-09-27) is pure grok-4.6 (it reproduces the published issue #8); week B is pure grok-4.6.

## Practical implications

- Stated confidence is still a ranking hint at best, not a probability: the field's gap is +12.9pp this week against +16.1pp in issue #8.
- Gap and regime move together. The field hit 49.6% against 46.8% in issue #8 and stated mean confidence read 62.5 against 63.0, so the gap came out at +12.9pp: a field that hits 49.6% instead of 46.8% closes or opens a confidence gap without a single stated number changing.
- Trading direction this week (0 of 7 lines net-positive) is a regime read, not a strategy result; the same lines printed the opposite sign inside seven weeks.

## Limitations

- 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 4,331 scored calls sit inside one market regime. This window ran 2 trend days of 7 (issue #8: 2 of 7); every comparison with issue #8 is a comparison across regimes as well as across weeks. Nine week-over-week points cannot separate drift from regime; no durability claim is made.
- Models report confidence at discrete levels, not a continuous scale; the buckets reflect those natural breakpoints. 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 (wave 9): (1) rolling-1w forecasts emit daily since Aug 22, 2026 and rolling-1M daily since Aug 23, 2026, so inside this window (Sep 28-Oct 4) daily coverage is FULL, as it was in issues #4, #5, #6, #7 and #8: the 1w series has slots on 7 of 7 days and the 1M series on 7 of 7 days; both sit outside this report's FH gate (1h/4h/1d) and appear only in the exclusions counter (1w 490, 1M 486). (2) The grok line flipped 4.5 -> 4.6 at Aug 24, 2026 09:22 UTC, five windows back (inside the issue-#4 window): this window carries 1,470 grok rows and none from the 4.5 era, and neither does week A (the issue-#8 window), so every grok number in this issue and in issue #8 is a pure grok-4.6 line -- the grok week-over-week row is the fourth pure-4.6 against pure-4.6 comparison of the series. (3) qwen-3.8-max succeeded qwen-3.7-max on Aug 5, 2026 (fully before this window).
- Market-state row is computed on a single-exchange (binance) daily candle series, as in issue #8 after the audit that found the raw table mixes two exchanges; issue #8 values are as published.
- Research-to-date counter: directional forecasts resolved inside the FH gate (1h/4h/1d) since Jul 11, counted at the issue cutoff (Oct 5 16:00 UTC) -- the definition pinned in issue #3 and carried by issues #4, #5, #6, #7 and #8, which printed 57,929 at the Sep 28 16:00 cutoff. The pack reproduces that pin at the previous cutoff (control OK), so this issue's 62,260 is an additive step under one definition; counter deltas against issue #7 and earlier remain definitional.

## Counters & lineage

| Counter | Value | Counter | Value |
|---|---|---|---|
| Forecasts total | 10,290 | Mature (scored pool) | 9,286 |
| OK in gate | 9,286 | -- of them directional | 4,331 |
| Out of gate (1w / 1M) | 490 / 486 | -- of them sideways | 4,955 |
| Invalid | 28 | Pending (next issue) | 0 |
| Late closes | 0 | Uptime, all grids | 266 / 266 slots |
| Source file | weekly_metrics_2026-09-28.json | Report cutoff | Mon Oct 5, 2026, 16:00 UTC |
| Generated at (pipeline) | 2026-10-07 06:16 UTC | Methodology | v1.1 (2026-08-10) · hash e66c7e8c864a2233 |

**THIS REPORT: 4,331 scored observations.** MARKETMANIA RESEARCH TO DATE (as of Oct 5, 2026 cutoff): 62,260 directional forecasts resolved since Jul 11 · 7 models tracked (current line-up; earlier versions folded into their successors' lineage) · 5 assets · 5 horizons · hourly · 44 published reports.

> **Issue #9.** Weekly Calibration is a living series. Issue #8's open questions -- "the gap moved +4.0pp while the field base moved -3.9pp -- the third issue running in which the two move nearly one-for-one in opposite directions; does that hold for a fourth?" and "The best-Brier line changed hands again, to qwen-3.8-max -- does it hold for a second issue, or change hands again?" -- read this issue as: field gap +16.1pp -> +12.9pp (5 of 7 lines narrowed) on a field base that moved +2.8pp, a fourth issue running in opposite directions, and the best-Brier line changed hands again: grok-4.6 at 0.2566 (issue #8: qwen-3.8-max at 0.2689). Also on the record: gpt-5.6-sol's 70-80 bucket read 47.2% (n=216) against its own 60-70 at 46.2% (n=429). The grok row is pure grok-4.6 this issue and was pure grok-4.6 in issue #8; the 4.5 -> 4.6 flip (Aug 24, 2026 09:22 UTC) sits five windows back. Engine 1.1 has powered the sandbox since Aug 18, i.e. before this window; no cross-engine PnL comparisons are claimed.

## What we're testing next

- Next issue: the gap moved -3.2pp while the field base moved +2.8pp -- the fourth issue running in which the two move nearly one-for-one in opposite directions; does that hold for a fifth?
- The best-Brier line changed hands again, to grok-4.6 -- does it hold for a second issue, or change hands again?
- Monthly series: is the overconfidence gap stable across market regimes (trend vs flat) at monthly n?

## Related research

| Report | Direct PDF link |
|---|---|
| Consensus Watch #9 | https://marketmania.ai/research/reports/consensus-watch-2026-09-28.pdf |
| Weekly Model Watch #9 | https://marketmania.ai/research/reports/model-watch-2026-09-28.pdf |
| Weekly Calibration #8 | https://marketmania.ai/research/reports/weekly-calibration-2026-09-21.pdf |

The three weekly reports publish together as one issue each week; each links straight to the others' PDF and to its own previous issue. Direct links are the posting rule from wave 2 on.

## Cite this report

```bibtex
@misc{mm_calibration_2026w40,
  title  = {Weekly Calibration #9: confidence vs. direction-hit, Sep 28-Oct 4 2026},
  author = {{MarketMania Research}},
  year   = {2026}, month = {October}, day = {7},
  url    = {https://marketmania.ai/research/reports/weekly-calibration-2026-09-28.pdf},
  note   = {Methodology v1.1, hash e66c7e8c864a2233; source weekly_metrics_2026-09-28.json}
}
```
