Mean reversion study

Bitcoin returns after a 30% drawdown

Forward returns from the first daily close 30% below the trailing all-time high.

VERY LOW · N=5 ACTIVE Last event 2025-11-21 Model 1.1.0
HOW TO READ THIS

Context before conviction.

After removing overlapping forward windows, N=5 independent observations remain from 7 raw events. Their 180D median return was +47.77% with a 60% win rate and -30.66% median adverse excursion.

A 5-event independent sample can produce extreme win rates by chance. Treat this result as exploratory.
OVERLAP CONTROL5 of 7 completed 180D observations retained · 28.6% 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
30DIndependentN=7+2.1%+24.4%57.1%-11.6%+42.8%-10.5%
30DRaw eventsN=7+2.1%+24.4%57.1%-11.6%+42.8%-10.5%
90DIndependentN=6-4.3%+22.9%50%-21.5%+24.2%-30.4%
90DRaw eventsN=7+7.9%+78.7%57.1%-19.8%+108.0%-30.2%
180DIndependentN=5+47.8%+129.7%60%-11.3%+70.5%-30.7%
180DRaw eventsN=7+47.8%+110.8%57.1%-23.7%+126.0%-30.7%
365DIndependentN=3+99.3%+157.1%66.7%+34.5%+250.8%-41.9%
365DRaw eventsN=6+34.5%+93.5%50%-55.3%+192.2%-54.5%
OUTCOME GEOMETRY

Distribution and drawdown

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

FORWARD RETURN · 180DDistribution
N=5
Most populated range-36.0% to +117.3%4 observations
MAXIMUM ADVERSE EXCURSION · 180DDrawdown by event
NON-OVERLAPPING
0.0%-11.6%-23.1%-34.7%-46.3%
21 Nov 2025-30.7% MAEReturn -11.3%MFE +13.2%
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
30DN=7Insufficient Data
90DN=6Insufficient Data
180DN=5Insufficient Data
365DN=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 · 180DInsufficient 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
20% drawdownThresholdpct 20N=17N=9+30.8%66.7%Stable
30% drawdownPublishedThresholdpct 30N=7N=5+47.8%60%Baseline
40% drawdownThresholdpct 40N=4N=4-24.7%25%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 · 180DInsufficient 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=5
Completed candles4062
Raw / independent N7 / 5
Independent median+47.8%
Win rate60%
COMPARISON SOURCEBITSTAMPBTC/USD
VERY LOW · N=5
Completed candles4062
Raw / independent N7 / 5
Independent median+47.8%
Win rate60%

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

EVENT DEFINITION

Drawdown crosses below −30% · one event per ATH cycle

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

STUDY RULES

Drawdown is measured from the highest prior daily close.

Only the first −30% crossing in an all-time-high cycle is counted.

A new cycle begins after a new all-time closing high.

Forward returns do not assume staged entries.

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 180D 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.