ONON forecast model
and what it got wrong.
On Holding is one of 25 stocks GioModel models. This page is the public record: the inputs, the method, the evidence gates, the validation metrics, and past forecasts graded against what actually happened. Current frozen ranges are the only part held for members.
Company overview
- Company
- ONON On Holding
- Sector
- —
- Market cap
- —
- Model status
- 3 / 8 evidence gates passed
How the ONON forecast is generated
An interval model: it estimates a distribution of prices at 5, 10 and 21 sessions ahead from a frozen history, then reports the 10th, 50th and 90th percentiles. Selection, calibration and terminal evaluation happen in that order with a strict 21-session embargo between them, so the block used to score the model was never used to choose it.
- 01Freeze the history
A point-in-time panel of completed ONON sessions is fixed and hashed — this release froze on Aug 5, 2026. Nothing after that date can influence the numbers, which is what makes a later grade meaningful.
- 02Build inputs that could have existed
Every feature is computed from a strictly prior window and tested for prefix invariance, so a value never changes when later data arrives. The admitted inputs are listed below.
- 03Fit and score through time
Position-range model v1 candidates are compared inside purged, embargoed walk-forward splits across 46 evaluation origins from Sep 10, 2024 to Jun 29, 2026. The block used to score a model is never the block used to pick it.
- 04Publish, then grade in public
The release states which of the nine gates it passed and which it did not. Thirty days later the frozen numbers appear on the public track record and are graded on the exact trading session the horizon lands on.
What data actually goes in
Not everything a model could use is admitted. Each input is marked with how far it has actually been taken — admitted means it feeds the published forecast; anything else is disclosed rather than quietly implied.
| Category | Input | Status |
|---|---|---|
| Fundamental | Revenue growth | PLANNED |
| Fundamental | Margins | PLANNED |
| Fundamental | Earnings | PLANNED |
| Fundamental | Valuation | PLANNED |
| Technical | Momentum | CONSTRUCTED |
| Technical | Moving averages | CONSTRUCTED |
| Technical | Volatility and volume | CONSTRUCTED |
| Market | Interest rates | PLANNED |
| Market | Sector trends | CONSTRUCTED |
| Market | Market regime | CONSTRUCTED |
| Portfolio | Correlation | PORTFOLIO LAYER |
| Portfolio | Position size | LIVE ARITHMETIC |
| Portfolio | Risk contribution | PORTFOLIO LAYER |
Validation metrics
Walk-forward results were inspected for candidate selection. They are historical development evidence, not an independent terminal holdout or post-protocol forward evidence. Overlapping horizons are not independent observations.
- Evaluation origins
- 46
- 80% band coverage
- 84.8%
- Mean IS vs baseline
- +0.4% lower is better
| Sessions | Coverage | Mean IS | Baseline IS |
|---|---|---|---|
| 5 | 80.4% (37/46) | 0.2173 | 0.2124 |
| 10 | 89.1% (41/46) | 0.2974 | 0.2971 |
| 21 | 84.8% (39/46) | 0.5006 | 0.5016 |
Historical interval comparisons only. Overlapping outcomes are not independent evidence; nominal coverage is 80% and lower interval score is better.
Evidence gates: 3 of 8
The v4 interval-only release passes data quality, date integrity and challenger discipline: 3 of 8. The sample-size gate fails on the shortest listing history in the universe, and cross-verification, walk-forward coverage and the naive-baseline comparison remain open; ranges are descriptive and unapproved.
Every gate below is reproduced from ONON's own frozen release artifact — the rule as written before the results were seen, the measured value, and the verdict. Gate numbers are local to this protocol: each model family predeclared its own set, so a G3 here is not a G3 on another model's page. Compare the protocols side by side.
This model is graded out of 8, not out of a single site-wide number. The position-forecast family dropped its direction gate on 15 August 2026 because the product stopped publishing a direction call, so it is graded out of eight. Families that still make a directional claim keep the gate that tests it, and are still graded out of nine.
- G1At least 1,750 valid target sessions in the frozen panel. — open
Observed: not published in this release artifact
- G2No quarantined target OHLCV rows in the aligned panel. — passed
Observed: not published in this release artifact
- G3No duplicate session dates in the target series. — passed
Observed: not published in this release artifact
- G4At least 90% of owner-verified v52 workbook closes match the independently fetched closes within 0.5%. — open
Observed: not published in this release artifact
- G5Walk-forward P10-P90 coverage within [75%, 85%] for every horizon. — open
Observed: not published in this release artifact
- G6Walk-forward mean interval score no worse than the scaled random-walk baseline for every horizon. — open
Observed: not published in this release artifact
- G8Published candidate is whichever of unconditional vs regime-conditioned scored better out-of-sample; the loser is retained as challenger. — passed
Observed: not published in this release artifact
- G9At least 30 non-overlapping post-protocol outcomes spanning at least 90 calendar days. Cannot pass at release time by construction. — open
Observed: not published in this release artifact
Historical Evaluation Ledger — Development/OOS Evidence
These origins were evaluated during model development, before the public protocol began. They are out-of-sample within the development panel, but they were not frozen and published ahead of their outcome, so they cannot demonstrate what this model does on genuinely unseen data. Treat them as construction evidence.
The forward record — releases frozen and hashed before the market moved, then graded on their exact resolution session — is kept separately on the Public Forward Track Record. Nothing there has been disclosed yet: the first unlock is 2026-07-29.
Past evaluation origins for ONON, each graded against the realized move. Only origins older than 0 days appear; anything newer sits behind the same delay as the live release.
| Origin | Horizon | Forecast median | Realized | P10–P90 | Result |
|---|---|---|---|---|---|
| 2026-06-29 | 21 sessions | +0.34% | +6.79% | −17.70%–+18.45% | INSIDE INTERVAL |
| 2026-06-29 | 10 sessions | +0.40% | +6.34% | −11.17%–+12.11% | INSIDE INTERVAL |
| 2026-06-29 | 5 sessions | −0.65% | +3.33% | −8.96%–+9.02% | INSIDE INTERVAL |
| 2026-06-12 | 21 sessions | +0.06% | −2.95% | −17.32%–+19.07% | INSIDE INTERVAL |
| 2026-06-12 | 10 sessions | +0.26% | −8.55% | −13.36%–+12.12% | INSIDE INTERVAL |
| 2026-06-12 | 5 sessions | −0.13% | −6.34% | −8.43%–+8.36% | INSIDE INTERVAL |
| 2026-05-29 | 21 sessions | +0.22% | −14.19% | −16.87%–+19.53% | INSIDE INTERVAL |
| 2026-05-29 | 10 sessions | +0.36% | −5.64% | −11.62%–+13.00% | INSIDE INTERVAL |
| 2026-05-29 | 5 sessions | +0.49% | −9.61% | −8.24%–+9.03% | BELOW INTERVAL |
| 2026-05-14 | 21 sessions | +0.49% | +4.90% | −16.66%–+18.73% | INSIDE INTERVAL |
| 2026-05-14 | 10 sessions | +0.42% | +10.26% | −12.46%–+12.47% | INSIDE INTERVAL |
| 2026-05-14 | 5 sessions | −0.42% | +6.18% | −9.20%–+8.74% | INSIDE INTERVAL |
The Aug 5, 2026 ONON release is now public in full on the track record, graded on its exact resolution sessions. That delay is the whole product boundary: a forecast that is already thirty days old cannot be traded on profitably, but it can be checked — which is what makes the record worth reading. See every release and its grade.
NO-GO does not mean the research contains no information. It means GioModel refuses to convert incomplete evidence into a trading recommendation. The ONON model publishes its ranges, probabilities and diagnostics to members, and publishes every gate it has not passed to everyone — but it does not claim to have earned directional deployment, and the deployment gate is independent of payment. Nothing on this page is investment advice, a price target, or a recommendation to buy or sell On Holding. Read the full method and the research disclaimer.