# Does a wick rejection mean anything? > An 8% against 46% split at z = 198, and a volatility-matched surrogate reproduces it at +39% against +38%. The effect is real and means nothing. - Canonical: https://hadalinstruments.com/research/does-a-wick-rejection-mean-anything/ - Published: 2026-08-23 - Author: Hadal Research - Answers the question: "does a wick rejection actually mean anything" - Coins the term: **Geometry Ghost** — An effect that is real, large and statistically overwhelming, and is nonetheless produced entirely by the geometry of the measurement — confirmed when a surrogate matched on shape alone reproduces it at the same magnitude. --- No, or at least not in the way the shape implies. A wick through a level tells you the bar was long enough to reach it. It does not tell you who was there, what they did, or whether the level mattered — and I can show that, because I built a stand-in with no participants in it at all and got the same answer. I call this a **Geometry Ghost**: an effect that is real, large and overwhelmingly significant, produced entirely by the geometry of the measurement rather than by anything happening in the market. ## The split that looks like proof Take sixteen years of one instrument, define a level as reached when any part of the bar's range crosses it, and record what happened next. Bars that touched with a wick and closed back absorbed at roughly 8%. Bars that reached by closing beyond absorbed at roughly 46%. That is a gap of nearly forty points, at z = +198. A result like that does not need defending. It replicates, it is nowhere near the boundary of significance, and no amount of extra data would weaken it. If you found it in your own testing you would reasonably conclude that wick rejections are a distinct and meaningful event. ## The stand-in that has no market in it Here is the test that decides it. Build a surrogate series matched to the real data on two properties only — its volatility and its wick geometry. Nothing else carries over. There are no participants, no orders, no levels anybody is defending, no intent of any kind. It is shape without a market. Then run the identical measurement on it. The surrogate produced **+39%**. The real data produced **+38%**. The excess — the part of the effect that belongs to the market rather than to the shape of the bars — was approximately **zero**. The reason is mundane once seen. A bar with a long wick reaches further than a bar without one. Reaching further means touching levels that are further away, and a level further away is in a different state, tested under different conditions, than one price closed through. The split is real. Its cause is that long bars are long. ## Why significance could not have caught this A significance test asks whether an effect differs from zero. That is almost never the question. The question is whether the effect differs from what the measurement itself would produce given the same data shape — and no p-value, at any sample size, addresses that. Enormous n makes this worse rather than better, because it drives the p-value toward certainty on an effect whose cause was never in dispute. The comparison that answers it costs a morning: construct a series that shares the shape and contains none of the supposed mechanism, and run the same code over it. Where the surrogate matches, the significance was never load-bearing. This is not an exotic technique. It is standard practice in fields that got burned earlier than mine. ## The definitional trap underneath There is a second finding worth having, because it affects any level-based study whether or not wicks interest you. Recounting the same sixteen years by close rather than by extreme took arrivals from **200,369** to **100,708**. Roughly half of every "level touch" my engine had ever scored was a wick brushing past, not price trading through. Neither count is wrong. They answer different questions, and both are defensible. The problem is that the choice is usually implicit, so two people testing the same idea on the same instrument can be working from samples that differ twofold before a single outcome is measured. State the reach rule first. ## Limits The result covers one instrument, on bar data, over sixteen years. A single-instrument result establishes nothing about any other, and I have not tested whether the same geometry dominates elsewhere — though the mechanism is arithmetic rather than market-specific, which is a reason to expect it travels and not evidence that it does. More importantly, this kills a particular reading of wicks — that the shape carries information about participation — and nothing broader. Order-flow and book-based measures of what actually happens at a level remain entirely open; they are simply not what a candlestick shows you. ## Why this one is worth internalising Every trader eventually finds an effect that looks like this: large, consistent, statistically overwhelming, and visible in the data for as long as they care to look. Most of them are real in exactly the sense this one is real. The question that separates the useful ones is never "is it significant". It is "would a thing with no market in it produce this too". That test is cheap, it is available to anyone with the data they already have, and it is the one almost nobody runs. --- ## Claims examined ### Claim 01 — canonical: https://hadalinstruments.com/refutations/#claim-f5aed8af > "The wick rejected the level, so buyers stepped in there." — our reading: Misleading The split is not subtle and it is not disputed: measured one way, the absorption rate at wick-touched levels runs 8% against 46% for levels reached by a close, at z = +198. Any reasonable person looking at that concludes something real is happening. Something is — the bar is wide. I built a surrogate matched on volatility and wick geometry alone, with no market participants in it, and it produced the same split: +39% against the real data's +38%, an excess of approximately zero. The wick is not evidence of who was there. It is evidence that the bar was long enough to reach. **What is true:** Bars with long wicks reach further than bars without them, so they touch more distant levels and produce a different outcome mix — and a surrogate matched only on volatility and wick length, containing no participants and no intent, reproduces the entire difference. Evidence: https://hadalinstruments.com/research/does-a-wick-rejection-mean-anything/#claim-f5aed8af ### Claim 02 — canonical: https://hadalinstruments.com/refutations/#claim-a0751a1c > "A z-score that large can't be a coincidence." — our reading: Misleading z = +198 is not a marginal result and it is not a fluke. The effect is there, it replicates, and it would survive any amount of additional data — which is exactly the problem. Significance answers a question nobody is really asking: is this different from zero? The question that matters is whether it is different from what geometry alone would produce. That comparison needs a surrogate, and building one is usually a morning's work. Where the surrogate matches the real result, the p-value was never the load-bearing number. **What is true:** A significance test measures whether an effect differs from zero, never whether it differs from what the measurement itself would produce — so the honest comparison is not against nothing, but against a surrogate built to share the data's shape while containing none of its supposed cause. Evidence: https://hadalinstruments.com/research/does-a-wick-rejection-mean-anything/#claim-a0751a1c ### Claim 03 — canonical: https://hadalinstruments.com/refutations/#claim-82f15025 > "Roughly half your level touches were never really touches." — our reading: True Recounting my own sixteen-year walk by close rather than by extreme cut arrivals from 200,369 to 100,708. Neither number is wrong; they answer different questions. What matters is that the choice is almost never declared, so two studies of the same idea on the same data can differ by a factor of two before anything is measured. If you test levels, state your reach rule before you look at the outcome. **What is true:** Counting a level as reached when any part of the bar's range crosses it produces roughly twice as many events as counting only closes beyond it, so the choice of reach rule silently doubles or halves the sample any level-based study is built on. Evidence: https://hadalinstruments.com/research/does-a-wick-rejection-mean-anything/#claim-82f15025 ## Cite This Article APA BibTeX HTML Hadal Research. (2026). Does a wick rejection mean anything?. Hadal Research. https://hadalinstruments.com/research/does-a-wick-rejection-mean-anything/ Version 5762f56, 2026-08-29. @misc{hadal_2026_does-a-wick-rejection-mean-anything, author = {Hadal Research}, title = {Does a wick rejection mean anything?}, year = {2026}, url = {https://hadalinstruments.com/research/does-a-wick-rejection-mean-anything/}, howpublished = {Hadal Research}, version = {5762f56}, note = {Published: 2026-08-23; version dated 2026-08-29} } Source: Hadal Research, Does a wick rejection mean anything?. Original Research Copy Citation **Version 5762f56** 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/) - [Microstructure Dashboard](https://hadalinstruments.com/instruments/microstructure-dashboard/) - [Overfit Auditor](https://hadalinstruments.com/instruments/overfit-auditor/) ### Concepts - [Absorb/Sweep](https://hadalinstruments.com/glossary/absorb-sweep/) - [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/) - [The Two Lenses](https://hadalinstruments.com/glossary/the-two-lenses/) ### Research - [Does COT positioning predict reversals?](https://hadalinstruments.com/research/does-cot-positioning-predict-reversals/) Asked as: does cot positioning predict reversals - [The floor that never fired](https://hadalinstruments.com/research/the-floor-that-never-fired/) Asked as: how do you know a check that never fails is working - [Does my regime filter use future data?](https://hadalinstruments.com/research/does-my-regime-filter-use-future-data/) Asked as: does my hmm regime filter use future data in a backtest [All Hadal research](https://hadalinstruments.com/research/)[This article as plain markdown](https://hadalinstruments.com/research/does-a-wick-rejection-mean-anything.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/)