FAILED

Colby's 2-day rules (exit as the negation of the entry)

Shortening the directional movement calculation to two days, and accepting as a buy signal both a positive directional reading and a rising average index, turns a filter indicator into a complete system — one that stays profitable for decades and returns more than buying and holding.

Measured in crypto — Binance spot · 540 pairs, delisted ones included · 0.2% per round trip

Net per trade

−0.26%

after fees

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

Died at

invariance

the result is not physically plausible

Worst drawdown

−92%

1,787 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 92% less than its own best previous moment, and it spent 1,787 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 2017-08-17 to 2026-07-28. It found trades in 540 coins, 190,594 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 467 of the 190,594. 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

-0.06% per trade. Look at the sign: the average is already negative before any fee is taken out. The controls below still hold, but here they are not knocking down an edge — they are confirming there never was one.

Why it did not pass — it died here · where the result came from

We look at where the result came from: whether it is spread across the trades or packed into a handful of them. It concentrated. That is the shape of a lottery: the typical trade loses, and a rare handful pays for everything. Anyone following the rule for real would face a long run of losses before any gain, and would almost certainly quit before it arrived. To give the measure: 51% of the profit came from the top 5% of trades, and the same technique with its dates shuffled would concentrate 46%. The control fails above 50%.

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 0.41× what it started with. At its worst the account was worth 92% less than its own best previous moment, and it spent 1,787 days below that peak. In 8 of the 12 paths tested, the account fell to less than half of what it started with.

The account would not fit every signal

73% of the trades had to be turned down: by the time they appeared, the money was already tied up in other positions. That matters because the average return per trade the technique claims is computed over trades that nobody could have taken all of.

What this result does NOT say

For an edge to be assertable here it would have to reach 4.54% 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 0.69% 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 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 190,594 trades, but only 467 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 467 episodes, what the data supports is a range from -0.51% to +0.39% 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	failed	—	—	—
costs	inconclusive	-1.14	1.97	467
placebo	failed	-0.46	1.97	467
benchmark	failed	—	—	—
out of sample	inconclusive	—	—	—
multiple testing	inconclusive	—	—	—

invariance51% of the gross profit comes from 9529 trades (5% of the total) — lotteryconcentration: +51.14% of the profit sits in the top 5% of trades — with the dates shuffled, +46.15% (fails above +50.00%)
costsgross -0.060% · cost 0.200% · net -0.260% (t=-1.14) · the range runs from -0.710% to +0.190%
placeboactual -0.260% · placebo -0.138% · excess -0.122% ± 0.266% (t=-0.46 against a threshold of 1.97, 467 real groups, 9,529,700 sham dates, draw error ±0.005%)
benchmarktechnique -0.26% · buy and hold (same horizon) -0.11% · excess -0.15%
out of sampleasset half A: -0.218% (t=-0.64, 465 episodes) · asset half B: -0.306% (t=-1.00, 466 episodes) · liquid half (>= US$ 2,066,613/day): -0.095% (t=-0.42, 467 episodes) · illiquid half: -0.451% (t=-1.96, 433 episodes) · period 1/4 (2017-08-21 a 2022-04-30): +0.600% (t=1.52, 245 episodes) · period 2/4 (2022-05-01 a 2023-12-19): -0.462% (t=-1.37, 87 episodes) · period 3/4 (2023-12-20 a 2025-04-27): -0.365% (t=-0.97, 71 episodes) · period 4/4 (2025-04-28 a 2026-07-27): -0.812% (t=-2.66, 66 episodes)
multiple testing2 variation(s) tested · t=-1.14 across 467 episodes (equivalent to t=-1.13) · p≈0.2575 · false positives expected by chance ≈ 0.51

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: ×0.41
worst drawdown from the peak: −92%
days below the previous peak: 1,787
signals refused for lack of capital: 73%
paths where the account halved (out of 12): 8

Reproducibility

period: 2017-08-17 to 2026-07-28
assets that traded: 540
variations tested before this one: 2
gross per trade: −0.06%
net per trade: −0.26%
exit rule: the technique itself (held until the opposite signal)
median duration bars: 2
mean duration bars: 3.9
max duration bars: 62
fee per leg: 0.001
episode days: 7
seed: 20260728
period: 2
reading: mirror

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

Goes long when the 2-day PDI exceeds the 2-day MDI OR when the 2-day average directional index exceeds its own 2-day smoothing; flattens on the inverse condition; sells by the opposite rules; ALL ORDERS AT THE CLOSE — the first technique in this corpus that executes exactly where the engine measures. The two printed conditions overlap: with one true and the other false, the entry rule and the exit rule fire on the same bar. It is not executable without a convention, and two of them close it — exit as the negation of the entry (always in the market) and literal exit with precedence (out of the market when the conditions disagree). Both will be measured: they coincide on 47.74% of the bars, so the choice is not cosmetic. The second condition is blind to direction, and the claim of 72 years of profit rests in part on it. Family-level prediction, recorded before measuring: (1) the gravedigger will be the INVARIANT in crypto and the BENCHMARK in forex — the two auditable techniques are always-in-the-market reversal systems, the same mechanics as `adaptativos`, where the invariant killed 7 of 10 in crypto and the benchmark 6 of 10 in forex; (2) Hochheimer's variant WITH CONFIRMATION will survive cost better than the immediate one, because it turns over half as much (485 reversals against 998 over 13,607 bars); (3) the MIRROR reading of the 2-day rules will do worse than the LITERAL one, because it stays in the market 100% of the time and on 32.8% of the bars it buys merely because the average directional index is rising — a condition blind to direction, followed by −0.332% over five bars; (4) none survives the family's Benjamini-Hochberg. And it is on record that a BH over 4 claims is weak by construction: the family is small because the section is small, not because anything was left out.

filed on 2026-08-03, before the number existed

The original, as it was filed

Entra comprado quando o PDI de 2 dias supera o MDI de 2 dias OU quando o índice direcional médio de 2 dias supera a própria suavização de 2 dias; zera na condição inversa; vendas pelas regras opostas; TODAS AS ORDENS NO FECHAMENTO — a primeira técnica deste corpus que executa exatamente onde o motor mede. ⚠️ As duas condições impressas se sobrepõem: com uma verdadeira e a outra falsa, a regra de entrada e a de saída disparam na mesma vela. Não é executável sem uma convenção, e há duas que fecham — a saída como negação da entrada (sempre no mercado) e a saída literal com precedência (fora do mercado quando as condições discordam). As duas serão medidas: elas coincidem em 47,74% das velas, então a escolha não é cosmética. ⚠️ A segunda condição é cega para a direção, e a alegação de 72 anos de lucro repousa em parte sobre ela. Previsão da família, registrada antes de medir: (1) o coveiro será o INVARIANTE em cripto e o BENCHMARK em forex — as duas técnicas auditáveis são sistemas de reversão sempre no mercado, a mesma mecânica de `adaptativos`, onde o invariante matou 7 de 10 em cripto e o benchmark 6 de 10 em forex; (2) a variante de Hochheimer COM CONFIRMAÇÃO sobreviverá ao custo melhor que a imediata, porque gira metade (485 inversões contra 998 em 13.607 velas); (3) a leitura ESPELHO das regras de 2 dias irá pior que a LITERAL, porque fica no mercado 100% do tempo e em 32,8% das velas compra só porque o índice direcional médio está subindo — condição cega para direção, seguida de −0,332% em cinco velas; (4) nenhuma sobrevive ao Benjamini-Hochberg da família. ⚠️ E fica registrado que o BH sobre 4 alegações é fraco por construção: a família é pequena porque a seção é pequena, não porque alguma coisa foi deixada de fora.

Pre-registration exists to keep prediction apart from rationalisation: written after the number, every hypothesis is right.

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.

2026-08-03FAILEDopen ↗
this measurement →FAILED