FAILED

40-day average by the direction of the line · entry only in high volatility (2.5σ)

Not every trade a trend system signals is worth taking. Waiting for the market to show signs of activity — a move that is large relative to its recent normal — before entering avoids getting into still markets, where price has no direction and losses are small and frequent.

Measured in forex — EUR/USD and GBP/USD · daily bars built from 15m · 1 pip spread (~0.009% per round trip) · no survivorship bias

Net per trade

+0.38%

after fees

The fee is charged on both legs: every trade pays to open and pays to close.

Died at

multiple testing

chance alone would already produce a result like this

Worst drawdown

−0%

926 days underwater

How far the account fell below its own best previous moment.

Timeframe

1 day

The time grid this was measured on. The same technique on a coarser or finer grid is a different measurement, and can earn a different verdict.

invariance
costs
placebo
benchmark
out of sample
multiple testing

Equity curve

Risking 1.0% per trade, marked to market every day — not only when a trade closes. The line at 1.0 is the capital it started with.

At its worst the account was worth 0% less than its own best previous moment, and it spent 926 days below that peak.

The technique against chance

Same number of trades, same holding time, same assets — only the dates were drawn at random.

The average trade and the typical one

The mean sits above the median: the lottery signature — the typical trade loses and a rare handful pays for everything.

What was measured

We coded the rule exactly as it is described and let it trade on its own from 2020-01-01 to 2026-06-26. It found trades in 2 coins, 72 in total. Each one enters and exits at the price that existed on that day — the program never sees what comes next, which is the most common mistake people make testing strategies at home. Wherever the description was ambiguous we took the reading least favourable to the technique; whatever was left open is stated in the hypothesis, at the foot of this page.

How many of those trades actually count

Only 42 of the 72. Cryptocurrencies rise and fall almost in unison, so the same rule fires on dozens of coins on the same day — and that is one bet repeated, not dozens of different bets. Counting them all separately is the trick that makes a bad test look impressive.

What it paid, before any deductions

+0.39% per trade. This is the number the technique claims, and the only one on this page that has not yet been through a single control.

Why it did not pass — it died here · how many versions were tested

We counted how many versions of the same idea were tested before this one was published. There were many. Publishing only the best of many versions is the same as flipping a coin repeatedly and reporting only the heads — and the result was not good enough to survive that discount.

What would have happened to the money

Risking 1.0% of capital per trade — the most widely taught rule — across the whole period: The capital would have ended at 1.00× what it started with. At its worst the account was worth 0% less than its own best previous moment, and it spent 926 days below that peak. In none of the 12 paths tested did it fall to less than half of what it started with.

What this result does NOT say

For an edge to be assertable here it would have to reach 0.65% per trade — that is the size that survives this archive's multiple-testing correction, and it rises as the archive grows. So FAILED means “we found nothing above that size”, and never “it cannot possibly work”. The instrument is far more sensitive than that: on synthetic data, with a clean effect, it separates from 0.09% upwards. The distance between the two numbers is the price of a real market and the price of publishing many claims. The difference matters, and it is the rule of this house: the card confronts the claim, never the person who made it.

What this card does not measure

Every verdict holds for the conditions it was measured under. These are this card's — and outside them the result does not apply.

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 72 trades, but only 42 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 42 episodes, what the data supports is a range from -0.29% to +1.07% 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	1.12	2.02	42
placebo	inconclusive	0.99	2.02	42
benchmark	passed	—	—	—
out of sample	inconclusive	—	—	—
multiple testing	failed	—	—	—

invariance72 signals across 4055 barsconcentration: +27.23% of the profit sits in the top 5% of trades — with the dates shuffled, +36.19% (fails above +50.00%)
costsgross +0.386% · cost 0.008% · net +0.378% (t=1.12) · the range runs from -0.303% to +1.059%
placeboactual +0.378% · placebo +0.043% · excess +0.335% ± 0.339% (t=0.99 against a threshold of 2.02, 42 real groups, 3,600 sham dates, draw error ±0.039%)
benchmarktechnique +0.38% · buy and hold (same horizon) +0.09% · excess +0.29%
out of sampleasset half A: +0.657% (t=1.38, 30 episodes) · asset half B: +0.141% (t=0.37, 31 episodes) · liquid half (>= US$ 0/day): +0.378% (t=1.12, 42 episodes) · period 1/4 (2020-02-27 a 2021-09-16): +0.197% (t=0.35, 10 episodes) · period 2/4 (2021-09-23 a 2023-06-08): +1.195% (t=1.58, 12 episodes) · period 3/4 (2023-06-15 a 2024-08-01): -0.062% (t=-0.14, 10 episodes) · period 4/4 (2024-08-02 a 2026-05-14): +0.181% (t=0.28, 13 episodes)
multiple testing56 variation(s) tested · t=1.12 across 42 episodes (equivalent to t=1.09) · p≈0.2767 · false positives expected by chance ≈ 15.50

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: 926
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: 56
gross per trade: +0.39%
net per trade: +0.38%
exit rule: the technique itself (held until the opposite signal)
median duration bars: 26
mean duration bars: 32.2
max duration bars: 120
spread pips: 1.0
episode days: 31
seed: 20260728
slope period: 40
volatility regime filter: bullish
volatility filter threshold: 2.5
volatility filter window: 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 layer in the opposite regime: the base only enters on a day where the day's move is LARGER than k times the standard deviation of the returns of the previous 20 days. IT IS THE CONTROL PAIR of the low filter and exists for that: if picking the regime is what improves things, the two cannot improve together. The source claims the low one is better with two numbers and no error bar (information ratio 1.143 against 0.867, in a single market). The threshold is a search optimum too, and over a larger grid: «test entries were 0.25 to 3.0 in steps of 0.05» — 56 values. AND THE SOURCE SWAPS THE NAMES OF THE TWO FILTERS between the two halves of the section: step 3 on p. 862 names the filter by the POSITION OF THE THRESHOLD («enter if volatility is above the low filter threshold»), and p. 864 names it by the REGIME IT ADMITS («the high-volatility filter, which allows trades only above its threshold»). The one on p. 864 governs, which is the one of the tables. Family-level prediction, recorded before measuring: (1) NONE survives the corpus-wide Benjamini-Hochberg — a filter that only decides WHEN NOT TO ENTER does not create an effect, it redistributes one; (2) the gravedigger will be the BENCHMARK, and not the placebo, and this is where this family parts from `fibonacci`: there the claim was that one specific level matters, and the placebo is what asks that; here the claim is that REMOVING TRADES IMPROVES THE RESULT, and what asks that is the control that compares against doing nothing — the protocol already records that a filter almost always cuts `n` without separating anything; (3) the LOW volatility filter and the HIGH one will give results IN THE SAME DIRECTION, despite the source claiming the low one is superior with two numbers and no error bar (information ratio 1.143 against 0.867) — if both improve, what improves is cutting trades, not picking a regime; (4) the RESET rule will have the family's best `t`, being the only layer that gives exposure back instead of taking it away, and even so it will not pass the BH; (5) the 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 especially to this family: the threshold is measured in standard deviations of the asset itself, so it adapts to the market by construction — but what counts as «a busy day» in forex is a move an order of magnitude smaller than in crypto, and the 10σ exit, already extreme in a fat-tailed market, may be unreachable in one with thinner tails.

filed on 2026-08-04, before the number existed

The original, as it was filed

A mesma camada no regime oposto: a base só entra num dia em que o movimento do dia for MAIOR que k vezes o desvio-padrão dos retornos dos 20 dias anteriores. ⚠️ É O PAR DE CONTROLE do filtro de baixa e existe por isso: se escolher o regime é o que melhora, os dois não podem melhorar juntos. A fonte alega que o baixo é melhor com dois números e nenhuma barra de erro (razão de informação 1,143 contra 0,867, num único mercado). ⚠️ Limiar também ótimo de busca, e com grade maior: «entradas de teste foram 0,25 a 3,0 em passos de 0,05» — 56 valores. ⚠️ E A FONTE TROCA OS NOMES DOS DOIS FILTROS entre as duas metades da seção: o passo 3 da p. 862 nomeia o filtro pela POSIÇÃO DO LIMIAR («entre se a volatilidade estiver acima do limiar do filtro baixo»), e a p. 864 o nomeia pelo REGIME QUE ADMITE («o filtro de volatilidade alta, que permite operações apenas acima do seu limiar»). Vale a da p. 864, que é a das tabelas. Previsão da família, registrada antes de medir: (1) NENHUMA sobrevive ao Benjamini-Hochberg do corpus — um filtro que só decide QUANDO NÃO ENTRAR não cria efeito, redistribui; (2) o coveiro será o BENCHMARK, e não o placebo, e é aqui que esta família se separa de `fibonacci`: lá a alegação era que um nível específico importa, e o placebo é quem pergunta isso; aqui a alegação é que REMOVER OPERAÇÕES MELHORA O RESULTADO, e quem pergunta isso é o controle que compara com não fazer nada — o protocolo já registra que filtro quase sempre corta `n` sem separar nada; (3) o filtro de volatilidade BAIXA e o de ALTA darão resultados NA MESMA DIREÇÃO, apesar de a fonte alegar superioridade do baixo com dois números e nenhuma barra de erro (razão de informação 1,143 contra 0,867) — se os dois melhoram, o que melhora é cortar operação, não escolher regime; (4) a regra de RESET terá o melhor `t` da família, por ser a única camada que devolve exposição em vez de tirar, e ainda assim não passará no BH; (5) a BASE 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 especialmente a esta família: o limiar é medido em desvios-padrão do próprio ativo, então ele se adapta ao mercado por construção — mas o que conta como «um dia agitado» em forex é um movimento uma ordem de grandeza menor que em cripto, e a saída de 10σ, que já é extrema num mercado de caudas gordas, pode ser inalcançável num de caudas menores.

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.

This claim has been audited once — there is no history to compare against.