Low Opportunity Score by Market Structure
Compares forward Bitcoin outcomes after low Opportunity Score entries when price structure was intact versus confirmed broken.
Low score, two structural contexts
A score at or below 40 starts the observation. Intact includes intact, neutral, or improving structure; Broken requires confirmed structural deterioration at that close.
| Horizon | Structure | Sample | Median | Mean | Win rate | Median MAE | Median MFE | Drawdown | Follow-through |
|---|---|---|---|---|---|---|---|---|---|
| 7D | INTACT | 17 independent · 18 raw | -3.5% | -2.3% | 35.3% | -6.4% | +3.4% | -8.4% | 17.6% |
| 7D | BROKEN | 11 independent · 15 raw | +5.4% | +2.7% | 63.6% | -3.0% | +6.3% | -6.6% | 63.6% |
| 30D | INTACT | 9 independent · 18 raw | -3.2% | -3.9% | 44.4% | -11.6% | +7.5% | -15.9% | 44.4% |
| 30D | BROKEN | 7 independent · 15 raw | -2.4% | -4.7% | 42.9% | -7.2% | +12.5% | -20.4% | 42.9% |
| 90D | INTACT | 5 independent · 17 raw | +11.6% | +8.6% | 60% | -23.7% | +15.5% | -24.4% | 60% |
| 90D | BROKEN | 3 independent · 15 raw | -12.5% | -14.4% | 33.3% | -25.9% | +7.1% | -30.8% | 33.3% |
| 180D | INTACT | 2 independent · 15 raw | +38.8% | +38.8% | 100% | -10.5% | +60.5% | -19.3% | 100% |
| 180D | BROKEN | 3 independent · 13 raw | -29.1% | -7.0% | 33.3% | -44.8% | +7.1% | -48.5% | 33.3% |
Follow-through requires a positive endpoint return and at least +5% maximum favorable excursion. Drawdown is the median maximum peak-to-trough decline within the window. Results are descriptive, not forecasts.
Outcome by horizon
Pooled low-score outcomes are retained here for audit; the structure-group table above is the primary comparison.
| Horizon | Population | Sample | Median | Mean | Win rate | 25th percentile | 75th percentile | Median MAE |
|---|---|---|---|---|---|---|---|---|
| 7D | Independent | N=28 | -0.4% | -0.4% | 46.4% | -4.9% | +5.5% | -5.5% |
| 7D | Raw events | N=33 | +0.7% | +0.3% | 51.5% | -4.9% | +6.0% | -4.2% |
| 30D | Independent | N=16 | -2.8% | -4.2% | 43.8% | -13.0% | +4.4% | -11.2% |
| 30D | Raw events | N=33 | +0.2% | -0.0% | 51.5% | -11.4% | +9.8% | -11.2% |
| 90D | Independent | N=8 | +2.5% | -0.0% | 50% | -14.4% | +15.2% | -24.4% |
| 90D | Raw events | N=32 | +10.4% | +13.7% | 59.4% | -13.3% | +38.8% | -14.8% |
| 180D | Independent | N=5 | +31.7% | +11.3% | 60% | -29.1% | +45.9% | -17.2% |
| 180D | Raw events | N=28 | +22.7% | +14.3% | 60.7% | -17.9% | +43.3% | -17.1% |
Sensitivity analysis
One detector setting changes at a time around the published configuration. Returns never determine which configuration is canonical.
100% of 8 comparable alternatives retained the canonical median direction.
| Configuration | Parameters | Raw N | Independent N | Median | Win rate | Direction |
|---|---|---|---|---|---|---|
| Canonical · local repair / 2-week broken-level reclaimPublished | Requireweeklyclose 1 · Atrbuffermultiplier 0 · Reclaimweeklycloses 2 · Breakdowndailycloses 2 · Structuralwindowdays 20 · Weeklyreclaimbufferpct 0.25 | N=32 | N=8 | +2.5% | 50% | Baseline |
| Structural support · prior 30D low | Requireweeklyclose 1 · Atrbuffermultiplier 0 · Reclaimweeklycloses 2 · Breakdowndailycloses 2 · Structuralwindowdays 30 · Weeklyreclaimbufferpct 0.25 | N=32 | N=8 | +2.5% | 50% | Stable |
| Structural support · prior 50D low | Requireweeklyclose 1 · Atrbuffermultiplier 0 · Reclaimweeklycloses 2 · Breakdowndailycloses 2 · Structuralwindowdays 50 · Weeklyreclaimbufferpct 0.25 | N=30 | N=8 | +2.5% | 50% | Stable |
| Breakdown · 1 daily close | Requireweeklyclose 0 · Atrbuffermultiplier 0 · Reclaimweeklycloses 2 · Breakdowndailycloses 1 · Structuralwindowdays 20 · Weeklyreclaimbufferpct 0.25 | N=32 | N=8 | +2.5% | 50% | Stable |
| Breakdown · 2 daily closes | Requireweeklyclose 0 · Atrbuffermultiplier 0 · Reclaimweeklycloses 2 · Breakdowndailycloses 2 · Structuralwindowdays 20 · Weeklyreclaimbufferpct 0.25 | N=32 | N=8 | +2.5% | 50% | Stable |
| Breakdown · weekly close | Requireweeklyclose 1 · Atrbuffermultiplier 0 · Reclaimweeklycloses 2 · Breakdowndailycloses 0 · Structuralwindowdays 20 · Weeklyreclaimbufferpct 0.25 | N=31 | N=8 | +2.5% | 50% | Stable |
| Breakdown · weekly close + 0.5 ATR | Requireweeklyclose 1 · Atrbuffermultiplier 0.5 · Reclaimweeklycloses 2 · Breakdowndailycloses 0 · Structuralwindowdays 20 · Weeklyreclaimbufferpct 0.25 | N=32 | N=8 | +2.5% | 50% | Stable |
| Reclaim · daily close | Requireweeklyclose 1 · Atrbuffermultiplier 0 · Reclaimweeklycloses 0 · Breakdowndailycloses 2 · Structuralwindowdays 20 · Weeklyreclaimbufferpct 0.25 | N=32 | N=8 | +2.5% | 50% | Stable |
| Broken-level reclaim · weekly close | Requireweeklyclose 1 · Atrbuffermultiplier 0 · Reclaimweeklycloses 1 · Breakdowndailycloses 2 · Structuralwindowdays 20 · Weeklyreclaimbufferpct 0.25 | N=32 | N=8 | +2.5% | 50% | Stable |
This is a robustness check, not model selection. Each row changes one assumption while every other detector rule remains fixed.
Opportunity Score ≤ 40 · first low-score close or structure-group change
The signal is evaluated only with information available at the event close. Future observations are used solely to calculate outcomes.
Each event uses the Opportunity Score and price-structure state stored at that completed daily close.
Intact-group observations include contemporaneous Intact, Neutral, and Improving states; Broken requires a weekly close below the prior 20-day low and at least two daily closes below that fixed support during the week.
An event is recorded on entry into a low-score spell and when the structure group changes during that spell; missing dates restart the spell.
Global chronological non-overlap is selected per horizon before outcomes are partitioned by structure group.
Sensitivity compares prior 20-day, 30-day, and 50-day lows; one-daily, two-daily, weekly, and weekly-plus-ATR breakdown confirmation; and daily, weekly, and two-weekly frozen broken-level confirmation.
Follow-through means a positive endpoint return and maximum favorable excursion of at least 5% within the completed forward window.
Drawdown is the median maximum peak-to-trough decline inside each completed forward window.
Primary confidence and headline statistics use the earliest chronological non-overlapping sample for each horizon; raw events remain reported for audit.
Temporal validation uses the independent sample only: a chronological 70/30 holdout plus expanding-window folds with at least two later observations per fold and no threshold tuning.
Sensitivity analysis changes one detector parameter at a time around the published canonical configuration, compares independent 90D outcomes, requires at least 8 observations per comparable variant, and never selects a parameter from performance.
Cross-exchange robustness replays the canonical detector independently on Coinbase BTC-USD and Bitstamp BTC/USD over their shared completed-daily history. Each source keeps its own signal and non-overlapping outcome population, with at least 8 observations required per benchmark.
Research, not a recommendation. Live market data can be delayed or temporarily unavailable. Historical results are displayed only when they have been computed and stored by the database research pipeline.