Wait for a better price by 1 × average true range of 20 (window 5, 80-day average)
A trend system's signal arrives after price has already moved, and taking it means paying dearly. Waiting a few days for a pullback against the signal's direction gets a better price without abandoning the trade — and the maximum window guarantees no good trend is lost to stubbornness.
Measured in forex — EUR/USD and GBP/USD · daily bars built from 15m · 1 pip spread (~0.009% per round trip) · no survivorship bias
- ✓invariance
- ?costs
- ?placebo
- ✗benchmark
- ?out of sample
- ✗multiple testing
At its worst the account was worth 0% less than its own best previous moment, and it spent 1,368 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 buying and holding
What would have happened to the money
What this result does NOT say
- One market, one universe
- Measured on 2 spot currency pairs. 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 2020-01-01 to 2026-06-26. 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 a 1.0 pip spread, crossed once. 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 98 trades, but only 69 independent market episodes: trades that overlap in time are not independent observations, and counting them as if they were inflates any result. It is the smaller number that governs the arithmetic. With 69 episodes, what the data supports is a range from -0.38% to +0.54% 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 | 0.30 | 2.00 | 69 |
| placebo | inconclusive | 0.17 | 2.00 | 69 |
| benchmark | failed | — | — | — |
| out of sample | inconclusive | — | — | — |
| multiple testing | failed | — | — | — |
- invariance98 signals across 4055 barsconcentration: +47.70% of the profit sits in the top 5% of trades — with the dates shuffled, +41.10% (fails above +50.00%)
- costsgross +0.078% · cost 0.008% · net +0.069% (t=0.30) · the range runs from -0.392% to +0.530%
- placeboactual +0.069% · placebo +0.030% · excess +0.039% ± 0.234% (t=0.17 against a threshold of 2.00, 69 real groups, 4,900 sham dates, draw error ±0.040%) — the status flips inside the placebo's own Monte Carlo error
- benchmarktechnique +0.07% · buy and hold (same horizon) +0.10% · excess -0.04%
- out of sampleasset half A: -0.096% (t=-0.34, 47 episodes) · asset half B: +0.241% (t=0.84, 42 episodes) · liquid half (>= US$ 0/day): +0.069% (t=0.30, 69 episodes) · period 1/4 (2020-04-08 a 2022-01-16): +0.248% (t=0.68, 19 episodes) · period 2/4 (2022-01-21 a 2024-03-11): +0.162% (t=0.23, 19 episodes) · period 3/4 (2024-03-26 a 2025-09-29): +0.386% (t=1.44, 17 episodes) · period 4/4 (2025-10-06 a 2026-06-17): -0.502% (t=-2.67, 15 episodes)
- multiple testing10 variation(s) tested · t=0.30 across 69 episodes (equivalent to t=0.29) · p≈0.7681 · false positives expected by chance ≈ 7.68
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
- ×1.00
- worst drawdown from the peak
- −0%
- days below the previous peak
- 1,368
- signals refused for lack of capital
- 0%
- paths where the account halved (out of 12)
- 0
Reproducibility
- period
- 2020-01-01 to 2026-06-26
- assets that traded
- 2
- variations tested before this one
- 10
- gross per trade
- +0.08%
- net per trade
- +0.07%
- exit rule
- the technique itself (held until the opposite signal)
- median duration bars
- 10
- mean duration bars
- 33.7
- max duration bars
- 217
- spread pips
- 1.0
- episode days
- 12
- seed
- 20260728
- waiting threshold
- 1.0
- waiting window
- 5
- slope period
- 80
- waiting ATR
- 20
Twelve Data forex (15m aggregated to 1d) · collected on 2026-07-25 · 2 assets · 4,055 bars · 2020-01-01 to 2026-06-26
Hypothesis, filed before the result
The same wait as the previous entry, now with a threshold: entry happens only if price improves on the signal price by a margin, and the 5-day maximum window still applies. THE SOURCE'S THRESHOLD IS IN POINTS AND DOES NOT CROSS MARKETS, and it declares the problem: «a different range of thresholds was used for each market because the threshold is defined as a point value. It may be worth trying this with a percentage, but back-adjusted futures prices can be distorted when the data goes back many years, so care is needed when using percentages». Its grids are 0.03125 to 0.50 on bonds (16 values), 0.5 to 10.0 on the S&P (20) and 1 to 20 on gold (20). A point does not exist in spot crypto nor in a currency pair. RENDERED AS A MULTIPLE OF THE 20-DAY AVERAGE TRUE RANGE, and the choice is declared as OURS: it is scale-free, it transfers between the two markets, and the average true range is already a quantity of this corpus — the `stops` family publishes targets and stops in it. The source's caveat against percentages is about back-adjusted futures, which is not our data, but the average true range is the quantity the chapter itself uses to size a move. THE N CHARGED IS OUR GRID'S, not its — it enters as a SWEEP and control 6 charges the true number. Family-level prediction, recorded before measuring: (1) NONE survives the corpus-wide Benjamini-Hochberg; (2) the FRACTIONAL ENTRY and the two PYRAMIDS ON PROFIT will move the mean by LESS THAN THE ERROR BAR and will NARROW the deviation — it is the mechanism the protocol has already measured twice, at breakeven (median from −2.75% to 0.00%, mean unchanged) and at partial profit taking (−0.023% ± 0.518%, deviation from 19.9% to 12.4%): position management moves the SHAPE and is read as if it had moved the CENTRE; (3) the EQUITY CURVE GATE fails, and for lack of autocorrelation — this is the family's most falsifiable prediction, because the number already exists in the corpus: the source states that «first one must find out whether returns are positively autocorrelated», and `confirmacoes.py` publishes, over 742,175 crypto days, that daily direction is a coin flip WITH NO MEMORY; (4) AVERAGING DOWN will be the worst in the family and will die at CONTROL 1, concentration, because adding to losses fattens the tail — and the source itself already rejects it in 3 of 4 markets; (5) the gravedigger in crypto will be CONTROL 1 and not the benchmark, which is the correction of the prediction refuted in part 76 and what part 74 had already established: in crypto, what pays a lot pays it concentrated; (6) the 80-day BASE will do better than any layer built on top of it. MEASURED IN FOREX (EUR/USD and GBP/USD, daily aggregated from 15m), and not in crypto. This is a pre-registered REPLICATION — not a new discovery —, and the multiple-testing count treats it as such. And there is one difference that matters to this family: waiting for a better price is measured in multiples of the asset's own true range, so it adapts to the market by construction — but a pullback that happens every week in crypto may be rare in a currency pair, and the 5-day window is the same in both.
filed on 2026-08-04, before the number existed
The original, as it was filed
A mesma espera da entrada anterior, agora com limiar: só se entra se o preço melhorar em relação ao do sinal por uma margem, e a janela máxima de 5 dias continua valendo. ⚠️ O LIMIAR DA FONTE É EM PONTOS E NÃO ATRAVESSA MERCADOS, e ela declara o problema: «uma faixa diferente de limiares foi usada para cada mercado porque o limiar é definido como um valor em pontos. Pode valer a pena tentar isto com um percentual, mas preços de futuros retroajustados podem ser distorcidos quando o dado recua muitos anos, então é preciso cuidado ao usar percentuais». As grades dela são 0,03125 a 0,50 nos títulos (16 valores), 0,5 a 10,0 no S&P (20) e 1 a 20 no ouro (20). Ponto não existe em cripto à vista nem em par de moeda. ⚠️ RENDERIZADO COMO MÚLTIPLO DA AMPLITUDE REAL DE 20, e a escolha está declarada como NOSSA: é escala-livre, transfere entre os dois mercados, e a amplitude real já é grandeza deste corpus — a família `stops` publica alvos e stops nela. A ressalva da fonte contra percentual é sobre futuros retroajustados, que não é o nosso dado, mas a amplitude real é a grandeza que o próprio capítulo usa para dimensionar movimento. ⚠️ O N COBRADO É O DA NOSSA GRADE, não o dela — entra como VARREDURA e o controle 6 cobra o número verdadeiro. Previsão da família, registrada antes de medir: (1) NENHUMA sobrevive ao Benjamini-Hochberg do corpus; (2) a ENTRADA FRACIONADA e as duas PIRÂMIDES SOBRE LUCRO moverão a média por MENOS QUE A BARRA DE ERRO e ESTREITARÃO o desvio — é o mecanismo que o protocolo já mediu duas vezes, no breakeven (mediana de −2,75% para 0,00%, média parada) e na realização parcial (−0,023% ± 0,518%, desvio de 19,9% para 12,4%): gestão de posição mexe no FORMATO e é lida como se tivesse mexido no CENTRO; (3) o GATE PELA CURVA DE PATRIMÔNIO reprova, e por falta de autocorrelação — esta é a previsão mais falsificável da família, porque o número já existe no corpus: a fonte declara que «primeiro é preciso descobrir se os retornos são positivamente autocorrelacionados», e `confirmacoes.py` publica, em 742.175 dias de cripto, que a direção diária é cara-ou-coroa SEM MEMÓRIA; (4) a MÉDIA PARA BAIXO será a pior da família e morrerá no CONTROLE 1, concentração, porque adicionar nas perdas engorda a cauda — e a própria fonte já a reprova em 3 de 4 mercados; (5) o coveiro em cripto será o CONTROLE 1 e não o benchmark, que é a correção da previsão refutada na parte 76 e o que a parte 74 já tinha estabelecido: em cripto, o que rende muito rende concentrado; (6) a BASE de 80 dias irá melhor que qualquer camada sobre ela. ⚠️ MEDIDO EM FOREX (EUR/USD e GBP/USD, diário agregado de 15m), e não em cripto. Isto é uma REPLICAÇÃO pré-registrada — não uma descoberta nova —, e a conta de múltiplos testes a trata como tal. ⚠️ E há uma diferença que importa a esta família: a espera por preço melhor é medida em múltiplos da amplitude real do próprio ativo, então ela se adapta ao mercado por construção — mas um repique que em cripto acontece toda semana pode ser raro num par de moedas, e a janela de 5 dias é a mesma nos dois.
Quotations from the source were translated from the Portuguese record and back into English — they are not the author's exact words.
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.
- forexFAILED← this one
- cryptoFAILEDopen that card
This claim has been audited once — there is no history to compare against.