Bitcoin after a confirmed Weekly Higher Low
Performance after a weekly pivot low confirms above the previous confirmed pivot low.
median forward return
Context before conviction.
After removing overlapping forward windows, N=14 independent observations remain from 31 raw events. Their 26W median return was +26.98% with a 64.3% win rate and -18.16% median adverse excursion.
Outcome by horizon
Primary non-overlapping results are shown beside the complete raw event population.
| Horizon | Population | Sample | Median | Mean | Win rate | 25th percentile | 75th percentile | Median MAE |
|---|---|---|---|---|---|---|---|---|
| 4W | Independent | N=32 | +4.0% | +3.3% | 62.5% | -8.3% | +9.3% | -6.2% |
| 4W | Raw events | N=32 | +4.0% | +3.3% | 62.5% | -8.3% | +9.3% | -6.2% |
| 8W | Independent | N=29 | +8.0% | +10.9% | 55.2% | -18.6% | +26.1% | -10.3% |
| 8W | Raw events | N=32 | +5.5% | +9.0% | 53.1% | -18.6% | +18.8% | -11.3% |
| 13W | Independent | N=22 | +13.8% | +19.2% | 63.6% | -10.8% | +38.7% | -17.0% |
| 13W | Raw events | N=32 | +9.2% | +24.3% | 56.3% | -17.3% | +42.6% | -18.8% |
| 26W | Independent | N=14 | +27.0% | +42.3% | 64.3% | -20.8% | +58.8% | -18.2% |
| 26W | Raw events | N=31 | +25.6% | +50.8% | 58.1% | -25.4% | +60.1% | -26.1% |
| 52W | Independent | N=9 | +67.6% | +124.2% | 66.7% | -46.5% | +142.2% | -51.5% |
| 52W | Raw events | N=30 | +67.2% | +105.0% | 70% | -30.7% | +152.1% | -31.7% |
Distribution and drawdown
Inspect the stored event-level outcomes behind the aggregate table.
Temporal validation
Chronological 70/30 holdouts and expanding-window folds use only independent observations. Later outcomes never influence event selection or thresholds.
3 of 3 expanding folds retained the training median direction.
| Horizon | Independent N | Train median | Test median | Direction | Stable folds | Assessment |
|---|---|---|---|---|---|---|
| 4W | N=32 | +4.4% | +3.7% | Stable | 2/3 · 66.7% | Consistent |
| 8W | N=29 | +10.5% | -0.4% | Changed | 1/3 · 33.3% | Unstable |
| 13W | N=22 | +18.0% | +7.9% | Stable | 1/3 · 33.3% | Mixed |
| 26W | N=14 | +28.4% | +25.6% | Stable | 3/3 · 100% | Consistent |
| 52W | N=9 | +71.8% | +67.6% | Stable | 2/2 · 100% | Consistent |
| Fold | Train through | Future test window | Train / test N | Train median | Test median | Direction |
|---|---|---|---|---|---|---|
| 1 | 2020-07-20 | 2021-06-14 → 2022-11-07 | 7 / 3 | +62.6% | +28.4% | Stable |
| 2 | 2022-11-07 | 2023-06-05 → 2024-02-19 | 10 / 2 | +32.3% | +29.7% | Stable |
| 3 | 2024-02-19 | 2024-09-30 → 2025-07-14 | 12 / 2 | +32.3% | +0.9% | Stable |
Sensitivity analysis
One detector setting changes at a time around the published configuration. Returns never determine which configuration is canonical.
100% of 2 comparable alternatives retained the canonical median direction.
| Configuration | Parameters | Raw N | Independent N | Median | Win rate | Direction |
|---|---|---|---|---|---|---|
| Pivot strength 2 | Pivotstrength 2 | N=39 | N=15 | +33.5% | 66.7% | Stable |
| Canonical · pivot strength 3Published | Pivotstrength 3 | N=31 | N=14 | +27.0% | 64.3% | Baseline |
| Pivot strength 4 | Pivotstrength 4 | N=23 | N=14 | +26.7% | 64.3% | Stable |
This is a robustness check, not model selection. Each row changes one assumption while every other detector rule remains fixed.
Cross-exchange benchmark
The published detector is replayed independently on each venue over the shared completed-candle window. Exchange populations are never mixed.
Both eligible venue samples retain the same independent median direction.
Coinbase remains the canonical research source. Bitstamp is an external robustness check; its candles, events, and forward-return population remain separate.
Strict weekly HL · 3 left bars · 3 right bars
The signal is evaluated only with information available at the event close. Future observations are used solely to calculate outcomes.
The candidate low must be strictly below three completed weekly bars on its left and the first completed bar on its right.
Its low must remain above the previous strict pivot low that was already confirmed with three right-side bars.
A candidate is invalidated if either of the next two completed weeks trades at or below its low.
The research event is timestamped only after all three right-side bars close; forward returns begin at that confirmation close.
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 26W 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.