Where does the RTH close relative to the previous day's value area, given where it opened at 09:30?
The position of the open carries: below VA → below VA 69.1 % (n = 243) · inside VA → inside VA 36.2 % (n = 326) · above VA → above VA 68.6 % (n = 347). That is a conditional frequency, not a model — the table below is the entire result.
| Decision time | 09:30 New York — bars up to t−1 complete, the opening print and every tick of session t before 09:30 |
| Target | position of rc against the previous day's [rval, rvah] — below, inside, above |
| Data window | 2020-05-29 to 2026-08-14 · 1607 daily bars · last day 2026-08-14 |
| Walk-forward | Dev ≤ 2022 (definitions only), test 2023 / 2024 / 2025 / 2026, expanding with a yearly refit |
| Holdout | The PX holdout from 2026-05-01 concerns the PX theses on the tick store. This study is a walk-forward over 2023 to 2026 on daily data — explicitly approved by the user for regime rounds 1 and 2. Every test year is computed from data before it only; no fit ever sees its own test slice. |
Row = where the RTH opens at 09:30 against the previous day's value area, cells = where it closes. Majority is the largest cell of the row — exactly what a rule without further information would reach.
| Year | Open position | n | close below % | close inside % | close above % | Majority % |
|---|---|---|---|---|---|---|
| 2023 | below VA | 79 | 70.9 | 22.8 | 6.3 | 70.9 |
| 2023 | inside VA | 84 | 14.3 | 39.3 | 46.4 | 46.4 |
| 2023 | above VA | 89 | 16.9 | 15.7 | 67.4 | 67.4 |
| 2024 | below VA | 56 | 69.6 | 16.1 | 14.3 | 69.6 |
| 2024 | inside VA | 99 | 25.3 | 42.4 | 32.3 | 42.4 |
| 2024 | above VA | 100 | 13.0 | 17.0 | 70.0 | 70.0 |
| 2025 | below VA | 63 | 65.1 | 17.5 | 17.5 | 65.1 |
| 2025 | inside VA | 84 | 31.0 | 31.0 | 38.1 | 38.1 |
| 2025 | above VA | 105 | 16.2 | 19.0 | 64.8 | 64.8 |
| 2026 | below VA | 45 | 71.1 | 6.7 | 22.2 | 71.1 |
| 2026 | inside VA | 59 | 39.0 | 28.8 | 32.2 | 39.0 |
| 2026 | above VA | 53 | 13.2 | 11.3 | 75.5 | 75.5 |
| pooled | below VA | 243 | 69.1 | 16.9 | 14.0 | 69.1 |
| pooled | inside VA | 326 | 26.4 | 36.2 | 37.4 | 37.4 |
| pooled | above VA | 347 | 15.0 | 16.4 | 68.6 | 68.6 |
| Base Rate | the unconditional frequency in the training window, and at the same time the constant comparison forecast. |
| Baseline | the number a model has to beat. Three of them here: base rate, majority (always the more frequent class) and persistence. |
| Persistence | the forecast “today like yesterday” — the value of the same target on the previous day, turned into a rate. |
| BSS | Brier skill score: what percentage of the base rate constant's Brier score the model saves. 0 means equally good, negative means worse. |
| AUC | the probability that a random positive day is scored above a random negative one. 50 is a coin flip. |
| Brier | the mean squared error of the probability. Smaller is better. |
| Sharpness | on what percentage of the days the model says something clear (p ≥ 0.7 or ≤ 0.3) and how often it hits there. A calibrated model without sharpness is useless. |
| Value Area | the price band [rval, rvah] in which 70 % of the RTH volume traded. Always the previous day's here. |
| Trend Day | an RTH with a body ratio ≥ 0.6 — the body of the daily candle fills at least 60 % of its range. |
| Body Ratio | br = |rc − ro| / (rh − rl), computed on the RTH session. |
| relr | RTH range divided by the median of the 20 previous RTH ranges. relr ≥ 1 means: today is bigger than the typical one of the last four weeks. |
| Walk-Forward | every test year is computed from a model that has only seen data before it; the training window grows with every year. No fit sees its own test. |
The numbers in this report are recomputed on every publish — from data/daily-bars-*.json and data/open-features-*.json, which the extractor builds from the tick store.
Run: 1607 daily bars, 0.0 s.