A/D Oscillator (0.30/30/70, tied to the previous close)
Where the market opens and where it closes, within the day's range, reveals the implied direction of that session's trading: opening at the low and closing at the high is buying; the reverse is selling. Because each day is measured against its own range, the index needs no recalibration when price or volatility shift levels — it solves on its own the problem that defeats fixed-scale indicators.
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 100% less than its own best previous moment, and it spent 2,282 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 · against randomly drawn dates
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 510 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 5,986 trades, but only 53 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 53 episodes, what the data supports is a range from -18.36% to +12.27% 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 | -0.43 | 2.01 | 53 |
| placebo | failed | -0.28 | 2.01 | 53 |
| benchmark | failed | — | — | — |
| out of sample | inconclusive | — | — | — |
| multiple testing | inconclusive | — | — | — |
- invariance5986 signals across 750934 barsconcentration: +41.07% of the profit sits in the top 5% of trades — with the dates shuffled, +54.73% (fails above +50.00%)
- costsgross -3.044% · cost 0.200% · net -3.244% (t=-0.43) · the range reaches +12.071%
- placeboactual -3.244% · placebo -1.061% · excess -2.184% ± 7.716% (t=-0.28 against a threshold of 2.01, 53 real groups, 299,300 sham dates, draw error ±0.306%)
- benchmarktechnique -3.24% · buy and hold (same horizon) +11.37% · excess -14.62%
- out of sampleasset half A: -5.071% (t=-0.62, 50 episodes) · asset half B: -1.580% (t=-0.21, 53 episodes) · liquid half (>= US$ 2,147,260/day): -8.158% (t=-0.82, 52 episodes) · illiquid half: +3.132% (t=0.63, 50 episodes) · period 1/4 (2017-08-27 a 2022-02-20): -12.656% (t=-0.88, 27 episodes) · period 2/4 (2022-02-21 a 2024-03-01): -11.105% (t=-1.97, 13 episodes) · period 4/4 (2024-12-19 a 2026-07-02): +10.713% (t=3.40, 10 episodes) · period 3/4 (2024-03-02 a 2024-12-18): -0.203% (6 episodes — too small, does not count)
- multiple testing1 variation(s) tested · t=-0.43 across 53 episodes (equivalent to t=-0.42) · p≈0.6780 · false positives expected by chance ≈ 0.68
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
- ×0.00
- worst drawdown from the peak
- −100%
- days below the previous peak
- 2,282
- signals refused for lack of capital
- 91%
- paths where the account halved (out of 12)
- 12
Reproducibility
- period
- 2017-08-17 to 2026-07-28
- assets that traded
- 510
- variations tested before this one
- 1
- gross per trade
- −3.04%
- net per trade
- −3.24%
- exit rule
- the technique itself (held until the opposite signal)
- median duration bars
- 62
- mean duration bars
- 104.6
- max duration bars
- 1329
- fee per leg
- 0.001
- episode days
- 62
- seed
- 20260728
- A/D smoothing
- 0.3
- oversold
- 30.0
- overbought
- 70.0
- tied to previous close
- 1.0
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
Buying force = high − open; selling force = close − low; the daily index is the sum of the two over twice the range. Zones of 80/20 on the raw series and 70/30 on the smoothed one (constant 0.30). The source describes a DEFECT of its own and the remedy: on a gap day the index inverts its sign without price having inverted, and it corrects this by replacing the high or the low with the previous close, in the manner of the true range. Both readings will be measured — without and with the correction —, because the source presents both. Family prediction, filed before measuring: (1) the control that kills the most will be COST, and not invariance — unlike the `adaptativos` family, because these are discrete-signal oscillators, they turn over less and are not always in the market; (2) the DIVERGENCE techniques will come out mostly INCONCLUSIVE, because they require two aligned peaks and fire rarely; (3) none survives the family's Benjamini-Hochberg. I record that prediction (1) is the opposite of the one I made in `adaptativos` and that was confirmed there — if I get it wrong again in the same direction, that is a sign I am misreading the mechanism of cost, not the technique.
filed on 2026-07-31, before the number existed
The original, as it was filed
Força compradora = máxima − abertura; vendedora = fechamento − mínima; o índice diário é a soma das duas sobre o dobro da amplitude. Zonas 80/20 no cru e 70/30 no suavizado (constante 0,30). ⚠️ A fonte descreve um DEFEITO próprio e o remédio: em dia de gap o índice inverte o sinal sem que o preço tenha invertido, e ela corrige substituindo máxima ou mínima pelo fechamento anterior, à moda do true range. As duas leituras serão medidas — sem e com a correção —, porque a fonte apresenta as duas. Previsão da família, registrada antes de medir: (1) o controle que mais mata será o CUSTO, e não o invariante — diferente da família `adaptativos`, porque estes são osciladores de sinal discreto, giram menos e não são sempre-no-mercado; (2) as técnicas de DIVERGÊNCIA sairão majoritariamente INCONCLUSIVAS, porque exigem dois picos alinhados e disparam pouco; (3) nenhuma sobrevive ao Benjamini-Hochberg da família. ⚠️ Registro que a previsão (1) é o oposto da que fiz em `adaptativos` e que se confirmou lá — se eu errar de novo na mesma direção, é sinal de que estou lendo mal o mecanismo do custo, e não a técnica.
Pre-registration exists to keep prediction apart from rationalisation: written after the number, every hypothesis is right.