# Do round numbers matter in forex? > A 13-month window gave z = minus 2.59. Run the same test across twelve windows and the mean is plus 0.08, none significant, some pointing the other way. - Canonical: https://hadalinstruments.com/research/do-round-numbers-matter-in-forex/ - Published: 2026-08-23 - Author: Hadal Research - Answers the question: "do round numbers matter in forex trading" - Coins the term: **Window Luck** — An effect that appears significant on the window it was discovered in and scatters around zero when the identical test is run across other windows of the same history — diagnosed by replication across windows, never by any property of the original result. --- On the evidence I have, no — and this one is worth reading carefully, because the original finding was mine, it was strong, and I spent months treating it as the most promising structural feature on the board before it dissolved. The test that killed it is one anybody can run, and it is not a significance test. I call the failure mode **Window Luck**: an effect that looks real on the stretch of history where it was found, and scatters around zero when the same test is run on other stretches of the same data. ## The finding I had Round-50 levels swept measurably more often than other levels. Measured over a thirteen-month window: **z = −2.59**. That is a conventional, respectable, publishable result. It was also a-priori — you can identify a round number before anything happens at it, which is exactly what a tradable structural feature requires and what almost nothing else I had managed. When my engine's other candidates were being eliminated one by one, this was the survivor I recorded as still live. ## The test that killed it Run the identical test across **twelve** roughly sixteen-month windows of the same sixteen-year history. - Mean z across windows: **+0.08** - Windows reaching significance: **0 of 12** - Individual windows: some as high as **+2.5** The last line is the one that settles it. Plus 2.5 is the *opposite* direction — round levels sweeping *less* than others. If the effect were real and merely noisy, windows would disagree about magnitude while agreeing about sign. These disagree about sign. That is what noise looks like when you slice it twelve ways. The original z of −2.59 was one of those twelve slices, found first and believed because it was found first. ## Why this is not a sample-size objection The natural response is that the windows were too short. They were longer. Roughly sixteen months each, against the thirteen months that produced the original result. Every replication had strictly more data than the finding it was checking. And the failure mode is diagnostic. A power problem gives you wide confidence intervals around a consistent centre — the effect is there, you just cannot pin it down. This gave a scattered centre. No quantity of additional history repairs that, because the disagreement is *between* windows rather than *within* them. ## The general lesson, which is worth more than the finding Almost every trading idea is discovered on a particular stretch of data. That is not a flaw; it is how discovery works. The flaw is stopping there. A significance test tells you whether the effect differs from zero *on the data you found it in*. It cannot tell you whether you would have found it had you looked somewhere else, and that is the question that decides whether an idea survives contact with next year. The check costs almost nothing: split your history into windows, run the same test on each, and look at the sign before you look at the p-values. If the sign scatters, you have a property of your sample. Nothing about the original result — not its magnitude, not its p-value, not how sensible the mechanism sounds — distinguishes the two cases in advance. Only replication does. ## Limits One instrument, sixteen years, bar data, round-50 levels specifically. A single-instrument result establishes nothing about any other, and I have not tested whether round levels behave differently elsewhere. The scope is also narrower than "levels do not matter". This tests a claim about the *price scale* — that certain numeric values are special before anything happens at them. It does not test whether levels with recent history behave differently, and that separate effect is real: a level's most recent outcome predicts its next one with an odds ratio near 3.5, out of sample. Structure exists. It is sequential, not positional, and it is a general property of price series rather than evidence for anything about where round numbers sit. --- ## Claims examined ### Claim 01 — canonical: https://hadalinstruments.com/refutations/#claim-eedadc78 > "Price reacts at round numbers and 50-levels — everyone watches them, so they matter." — our reading: Unproven I found this effect myself and believed it. On a thirteen-month window, round-50 levels swept measurably more often, at z = minus 2.59 — a result that would pass any conventional significance bar and which I recorded as the one a-priori structural feature the engine had. Replicating the identical test across twelve roughly sixteen-month windows of the same history gave a mean z of plus 0.08, zero of twelve significant, and a scatter that included windows at plus 2.5. The sign is not stable. An effect whose direction depends on which stretch of history you sampled is a property of the sample. **What is true:** Tested across twelve separate windows of the same sixteen-year history, the round-level effect averages a z-score of plus 0.08 with none of the twelve reaching significance, and individual windows reach plus 2.5 in the direction opposite to the one the belief predicts. Evidence: https://hadalinstruments.com/research/do-round-numbers-matter-in-forex/#claim-eedadc78 ### Claim 02 — canonical: https://hadalinstruments.com/refutations/#claim-31fcb084 > "Your windows were too short — that's why it disappeared." — our reading: Misleading The objection would be right if the windows were shorter. They were longer — roughly sixteen months each against the original thirteen. More data per test, not less. And the failure mode is not that each window returned a weak result; it is that the windows disagreed with each other, including on direction. Power problems produce wide intervals around a consistent centre. This produced a scattered centre, which is a different diagnosis and is not fixed by adding history. **What is true:** Each replication window was longer than the thirteen-month period that produced the original result, so the effect vanished under strictly more data per test rather than less, and the disagreement is between windows rather than within any one of them. Evidence: https://hadalinstruments.com/research/do-round-numbers-matter-in-forex/#claim-31fcb084 ### Claim 03 — canonical: https://hadalinstruments.com/refutations/#claim-949bacaa > "So there is no structure at all in where price reacts." — our reading: False Something survives, and it would be dishonest to imply otherwise. What a level did most recently genuinely predicts its next outcome, with an odds ratio near 3.5 that holds on held-out data. The distinction is that this is about a level's recent history, not about where the level sits. A round number is a claim about the price scale — that certain values are special before anything happens at them. That claim is what died. Levels that were recently active behaving differently from levels that were not is a separate, weaker and more general phenomenon. **What is true:** Short-range recency at a level is a real and out-of-sample-valid effect — what a level did on its most recent visit predicts what it does next — so structure exists, it is simply sequential rather than a property of the level's position on the price scale. Evidence: https://hadalinstruments.com/research/do-round-numbers-matter-in-forex/#claim-949bacaa ## Cite This Article APA BibTeX HTML Hadal Research. (2026). Do round numbers matter in forex?. Hadal Research. https://hadalinstruments.com/research/do-round-numbers-matter-in-forex/ Version 3dc0ea0, 2026-08-23. @misc{hadal_2026_do-round-numbers-matter-in-forex, author = {Hadal Research}, title = {Do round numbers matter in forex?}, year = {2026}, url = {https://hadalinstruments.com/research/do-round-numbers-matter-in-forex/}, howpublished = {Hadal Research}, version = {3dc0ea0}, note = {Published: 2026-08-23; version dated 2026-08-23} } Source: Hadal Research, Do round numbers matter in forex?. Original Research Copy Citation **Version 3dc0ea0** identifies the commit that last changed this page in Hadal's content repository. That repository is not public, so the identifier does not resolve externally — it is published so a citation pins one specific state rather than a moving page. To obtain the exact version cited, use the [press and research route](https://hadalinstruments.com/press/). ## Explore further ### Instruments - [Data Forensics](https://hadalinstruments.com/instruments/data-forensics/) - [Epistemic Harness](https://hadalinstruments.com/instruments/epistemic-harness/) - [Overfit Auditor](https://hadalinstruments.com/instruments/overfit-auditor/) - [Reproducible Verdict Kernel](https://hadalinstruments.com/instruments/reproducible-verdict-kernel/) ### Concepts - [Backtest Overfitting](https://hadalinstruments.com/glossary/backtest-overfitting/) - [Data Snooping](https://hadalinstruments.com/glossary/data-snooping/) - [Effective Sample Size](https://hadalinstruments.com/glossary/effective-sample-size/) - [False Discovery Rate (FDR)](https://hadalinstruments.com/glossary/false-discovery-rate/) - [Probability of Backtest Overfitting (PBO)](https://hadalinstruments.com/glossary/probability-of-backtest-overfitting/) - [Walk-Forward Validation](https://hadalinstruments.com/glossary/walk-forward-validation/) ### Research - [Do order blocks predict anything?](https://hadalinstruments.com/research/do-order-blocks-predict-anything/) Asked as: do order blocks actually predict anything - [Does a wick rejection mean anything?](https://hadalinstruments.com/research/does-a-wick-rejection-mean-anything/) Asked as: does a wick rejection actually mean anything - [How do I test a backtest for overfitting?](https://hadalinstruments.com/research/how-to-test-a-backtest-for-overfitting/) Asked as: how to test a backtest for overfitting [All Hadal research](https://hadalinstruments.com/research/)[This article as plain markdown](https://hadalinstruments.com/research/do-round-numbers-matter-in-forex.md) --- ## Raw artifact — NOT PUBLISHED FOR THIS PAGE No downloadable artifact ships with this page. Eight published measurements do, each content-hashed so a reader can verify the figures independently. Where a measurement is published here without one, that is a gap rather than a policy, and it is stated rather than left to be noticed. [Measurements that ship their data](https://hadalinstruments.com/research/)