Variable-period moving average (5–30)
The size of the price changes can govern the average's period: when the market turns more volatile, a shorter period follows price closely; when the changes are small or volatility is steady, the period is lengthened to slow the line down. The same average then serves both regimes, instead of forcing a choice between them.
Measured in crypto — Binance spot · 540 pairs, delisted ones included · 0.2% per round trip
- ✓invariance
- ?costs
- ?placebo
- ✓benchmark
- ✗out of sample
- ✓multiple testing
At its worst the account was worth 72% less than its own best previous moment, and it spent 1,861 days below that peak.
What was measured
How many of those trades actually count
What it paid, before any deductions
Why it did not pass — it died here · does it come back in the pieces?
What would have happened to the money
The account would not fit every signal
What this result does NOT say
- One market, one universe
- Measured on 540 spot cryptocurrency pairs, delisted ones included. It says nothing about futures, equities or indices, nor about how the same technique behaves in another market.
- One window of time, not every window
- The measured period runs from 2017-08-17 to 2026-07-28. A market moves through regimes, and a technique can work in one and fail in another — the card measures the regimes that fit inside this window, not the ones still to come.
- One cost structure
- The cost charged is 0.10% per leg, in and out. Anyone paying more than that gets a worse result, and anyone paying less gets a better one — the verdict holds for this fee.
- One exit rule
- The trade was closed by: the technique itself (held until the opposite signal). The same entry measured with a different exit is a different strategy, and can earn a different verdict — it happens in this archive.
- The number of trades is not the sample size
- There are 432,440 trades, but only 29 independent market episodes: a single move fires the technique across dozens of assets at once, and counting those as separate observations inflates any result. It is the smaller number that governs the arithmetic. With 29 episodes, what the data supports is a range from -0.07% to +0.21% per trade — the published average is the centre of it, not the exact measurement.
- Daily bars
- Measured at the daily close. Nothing here measures what happens inside the day, and an intraday technique is not auditable with this data.
Numbers and reproducibility
The six controls
| control | status | t | threshold | episodes |
|---|---|---|---|---|
| invariance | passed | — | — | — |
| costs | inconclusive | -1.83 | 2.05 | 29 |
| placebo | inconclusive | 0.92 | 2.05 | 29 |
| benchmark | passed | — | — | — |
| out of sample | failed | — | — | — |
| multiple testing | passed | — | — | — |
- invariance432440 signals across 750934 barsconcentration: +39.82% of the profit sits in the top 5% of trades — with the dates shuffled, +39.21% (fails above +50.00%)
- costsgross +0.072% · cost 0.200% · net -0.128% (t=-1.83) · the range runs from -0.272% to +0.015%
- placeboactual -0.128% · placebo -0.194% · excess +0.065% ± 0.071% (t=0.92 against a threshold of 2.05, 29 real groups, 21,622,000 sham dates, draw error ±0.002%)
- benchmarktechnique -0.13% · buy and hold (same horizon) -0.16% · excess +0.03%
- out of sampleasset half A: -0.127% (t=-0.62, 29 episodes) · asset half B: -0.129% (t=-2.28, 29 episodes) · liquid half (>= US$ 2,066,613/day): -0.093% (t=-1.28, 29 episodes) · illiquid half: -0.169% (t=-0.84, 28 episodes) · period 1/4 (2017-09-17 a 2022-05-06): -0.008% (t=-0.07, 16 episodes) · period 2/4 (2022-05-07 a 2023-12-02): -0.252% (6 episodes — too small, does not count) · period 3/4 (2023-12-03 a 2025-04-12): +0.155% (5 episodes — too small, does not count) · period 4/4 (2025-04-13 a 2026-07-27): -0.409% (5 episodes — too small, does not count)
- multiple testing1 variation(s) tested · t=-1.83 across 29 episodes (equivalent to t=-1.76) · p≈0.0792 · false positives expected by chance ≈ 0.08
Equity — outside the six controls, and here is why
The t of the trade series is invariant to bet size: 0.5%, 1% and 3% agree to the sixth decimal. Nothing here moves the verdict — it moves what the account would have lived through.
- risking 1.0% per trade
- ×0.30
- worst drawdown from the peak
- −72%
- days below the previous peak
- 1,861
- signals refused for lack of capital
- 77%
- paths where the account halved (out of 12)
- 12
Reproducibility
- period
- 2017-08-17 to 2026-07-28
- assets that traded
- 540
- variations tested before this one
- 1
- gross per trade
- +0.07%
- net per trade
- −0.13%
- exit rule
- the technique itself (held until the opposite signal)
- median duration bars
- 1
- mean duration bars
- 1.7
- max duration bars
- 31
- fee per leg
- 0.001
- episode days
- 112
- seed
- 20260728
- fast period, variable
- 5
- slow period, variable
- 30
- volatility window
- 30
- lower bound
- 0.25
- upper bound
- 1.75
Binance spot klines (delisted pairs included) · collected from 2026-07-27 23:31 to 2026-08-03 10:49 · 540 assets · 750,934 bars · 2017-08-17 to 2026-07-28
Hypothesis, filed before the result
Specified (5 to 30 periods, limits at ±0.25 and ±1.75 standard deviations, linear interpolation given by formula). No signal rule in the chapter: direction of the line, the auditor's choice, declared. Family prediction, filed before measuring: (1) none of the adaptive averages survives the family's Benjamini-Hochberg in crypto; (2) the control that kills the most will be COST, not the benchmark — unlike the chart-pattern census, where the benchmark was the gravedigger, because these are always-in-the-market systems and they turn over a lot; (3) each adaptive average will have HIGHER turnover than its fixed-period counterpart already in the archive, and will die more at cost than it does. The mechanism is in the source itself: Table 17.1 reports a profit factor 'even before costs' and states that success is 'inversely related to the average number of trades' (KAMA 159 trades, factor 1.53; VIDYA 443, factor 1.17). That is a cost story told as a quality story. If I am wrong and one survives cost with higher turnover, the chapter's thesis gains evidence it did not present.
filed on 2026-07-29, before the number existed
The original, as it was filed
Especificada (5 a 30 períodos, limites em ±0,25 e ±1,75 desvios padrão, interpolação linear dada por fórmula). ⚠️ Sem regra de sinal no capítulo: direção da linha, escolha do auditor, declarada. Previsão da família, registrada antes de medir: (1) nenhuma das adaptativas sobrevive ao Benjamini-Hochberg da família em cripto; (2) o controle que mais mata será o CUSTO, e não o benchmark — diferente do censo de padrões gráficos, onde o benchmark foi o coveiro, porque estas são sistemas sempre-no-mercado e giram muito; (3) cada adaptativa terá giro MAIOR que a sua contraparte de período fixo já no corpus, e morrerá mais no custo do que ela. O mecanismo está na própria fonte: a Tabela 17.1 relata fator de lucro 'even before costs' e afirma que o sucesso é 'inversely related to the average number of trades' (KAMA 159 operações, fator 1,53; VIDYA 443, fator 1,17). Isso é uma história de custo contada como história de qualidade. Se eu estiver errado e alguma sobreviver ao custo com giro maior, a tese do capítulo ganha uma evidência que ele não apresentou.
Pre-registration exists to keep prediction apart from rationalisation: written after the number, every hypothesis is right.
The same technique in the other market
The verdict held in the other market too, on independent data.
- cryptoFAILED← this one
- forexFAILEDopen that card
Earlier audits of the same technique
Each variation an author teaches enters as its own test, so that whatever might work in the strategy gets covered. The verdict held in all of them.