Structure study

Bitcoin after a successful Weekly Bearish Retest

Performance after Bitcoin revisits a confirmed breakdown level and closes back below it.

VERY LOW · N=3 MONITORING Last event 2025-03-31 Model 1.1.0
HOW TO READ THIS

Context before conviction.

After removing overlapping forward windows, N=3 independent observations remain from 4 raw events. Their 26W median return was +36.2% with a 66.7% win rate and -9.66% median adverse excursion.

A 3-event independent sample can produce extreme win rates by chance. Treat this result as exploratory.
OVERLAP CONTROL3 of 4 completed 26W observations retained · 25% overlapping events excluded from the primary estimate.
FORWARD RETURNS

Outcome by horizon

Primary non-overlapping results are shown beside the complete raw event population.

HorizonPopulationSampleMedianMeanWin rate25th percentile75th percentileMedian MAE
4WIndependentN=4+19.3%+14.9%75%+7.8%+26.3%-6.2%
4WRaw eventsN=4+19.3%+14.9%75%+7.8%+26.3%-6.2%
8WIndependentN=4+31.3%+22.7%75%+17.6%+36.4%-6.2%
8WRaw eventsN=4+31.3%+22.7%75%+17.6%+36.4%-6.2%
13WIndependentN=4+43.3%+30.3%75%+15.4%+58.2%-6.2%
13WRaw eventsN=4+43.3%+30.3%75%+15.4%+58.2%-6.2%
26WIndependentN=3+36.2%+36.4%66.7%+12.9%+59.8%-9.7%
26WRaw eventsN=4+46.8%+41.6%75%+24.6%+63.8%-6.2%
52WIndependentN=3+20.4%+283.2%66.7%+0.2%+434.7%-27.2%
52WRaw eventsN=4+83.7%+249.1%75%+10.3%+322.5%-15.0%
OUTCOME GEOMETRY

Distribution and drawdown

Inspect the stored event-level outcomes behind the aggregate table.

FORWARD RETURN · 26WDistribution
N=3
Most populated range-10.4% to +20.9%1 observation
MAXIMUM ADVERSE EXCURSION · 26WDrawdown by event
NON-OVERLAPPING
0.0%-9.1%-18.2%-27.3%-36.4%
31 Mar 2025-9.7% MAEReturn +36.2%MFE +51.2%
OUT-OF-SAMPLE CHECK

Temporal validation

Chronological 70/30 holdouts and expanding-window folds use only independent observations. Later outcomes never influence event selection or thresholds.

At least 8 independent completed observations are required before a chronological holdout is published.
HorizonIndependent NTrain medianTest medianDirectionStable foldsAssessment
4WN=4Insufficient Data
8WN=4Insufficient Data
13WN=4Insufficient Data
26WN=3Insufficient Data
52WN=3Insufficient Data
PARAMETER ROBUSTNESS

Sensitivity analysis

One detector setting changes at a time around the published configuration. Returns never determine which configuration is canonical.

ASSESSMENT · 26WInsufficient Data

At least 8 independent observations are required for the canonical and compared variants.

INDEPENDENT MEDIAN RANGEEligible configurations only
SAMPLE RANGENon-overlapping observations
ConfigurationParametersRaw NIndependent NMedianWin rateDirection
Breakout buffer 0.5%Pivotstrength 3 · Retestendweek 8 · Reteststartweek 2 · Failurebufferpct 1 · Breakoutbufferpct 0.5 · Retesttolerancepct 1.5N=4N=3+36.2%66.7%Too small
Canonical detectorPublishedPivotstrength 3 · Retestendweek 8 · Reteststartweek 2 · Failurebufferpct 1 · Breakoutbufferpct 1 · Retesttolerancepct 1.5N=4N=3+36.2%66.7%Baseline
Breakout buffer 2%Pivotstrength 3 · Retestendweek 8 · Reteststartweek 2 · Failurebufferpct 1 · Breakoutbufferpct 2 · Retesttolerancepct 1.5N=5N=4+12.9%50%Too small
Pivot strength 2Pivotstrength 2 · Retestendweek 8 · Reteststartweek 2 · Failurebufferpct 1 · Breakoutbufferpct 1 · Retesttolerancepct 1.5N=4N=3-10.4%33.3%Too small
Pivot strength 4Pivotstrength 4 · Retestendweek 8 · Reteststartweek 2 · Failurebufferpct 1 · Breakoutbufferpct 1 · Retesttolerancepct 1.5N=4N=3+36.2%66.7%Too small
Retest tolerance 1%Pivotstrength 3 · Retestendweek 8 · Reteststartweek 2 · Failurebufferpct 1 · Breakoutbufferpct 1 · Retesttolerancepct 1N=3N=3+57.3%100%Too small
Retest tolerance 2%Pivotstrength 3 · Retestendweek 8 · Reteststartweek 2 · Failurebufferpct 1 · Breakoutbufferpct 1 · Retesttolerancepct 2N=4N=3+33.9%66.7%Too small
Failure buffer 0.5%Pivotstrength 3 · Retestendweek 8 · Reteststartweek 2 · Failurebufferpct 0.5 · Breakoutbufferpct 1 · Retesttolerancepct 1.5N=4N=3+36.2%66.7%Too small
Failure buffer 2%Pivotstrength 3 · Retestendweek 8 · Reteststartweek 2 · Failurebufferpct 2 · Breakoutbufferpct 1 · Retesttolerancepct 1.5N=5N=4+12.9%50%Too small
Retest window through week 6Pivotstrength 3 · Retestendweek 6 · Reteststartweek 2 · Failurebufferpct 1 · Breakoutbufferpct 1 · Retesttolerancepct 1.5N=3N=3+57.3%100%Too small
Retest window through week 10Pivotstrength 3 · Retestendweek 10 · Reteststartweek 2 · Failurebufferpct 1 · Breakoutbufferpct 1 · Retesttolerancepct 1.5N=4N=3+36.2%66.7%Too small

This is a robustness check, not model selection. Each row changes one assumption while every other detector rule remains fixed.

VENUE ROBUSTNESS

Cross-exchange benchmark

The published detector is replayed independently on each venue over the shared completed-candle window. Exchange populations are never mixed.

ASSESSMENT · 26WInsufficient Data

Both venues need at least 8 independent outcomes before direction is compared.

SHARED COMPLETED HISTORY2015-07-20 → 2026-09-01Identical calendar boundary for both detector replays
CANONICAL SOURCECOINBASEBTC-USD
VERY LOW · N=3
Completed candles4062
Raw / independent N4 / 3
Independent median+36.2%
Win rate66.7%
COMPARISON SOURCEBITSTAMPBTC/USD
VERY LOW · N=3
Completed candles4062
Raw / independent N4 / 3
Independent median+36.2%
Win rate66.7%

Coinbase remains the canonical research source. Bitstamp is an external robustness check; its candles, events, and forward-return population remain separate.

EVENT DEFINITION

Weeks 2–8 · high within −1.5% · close ≤ breakdown level

The signal is evaluated only with information available at the event close. Future observations are used solely to calculate outcomes.

STUDY RULES

The source breakdown must be a completed weekly close at least 1% below a confirmed pivot low.

A retest may qualify from the second through eighth completed week after breakdown.

The weekly high must come within 1.5% below the pivot level or trade through it, then close at or below the level.

A close 1% above the level fails the lifecycle before a later retest can qualify.

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.