Do liquidity sweeps predict bitcoin’s next move?
Not beyond chance, in our data. Running a published test protocol on bitcoin from January 2020 to October 2026, the gap between “accepted” and “rejected” sweeps of monthly, weekly and round-number levels was the same at real levels as at random price points — 0.117 versus 0.114 of a typical daily range over the next 24 hours. The level itself added nothing.
The claim
Liquidity sweeps are a staple of price-action trading: price runs through an obvious level, triggers the stops resting beyond it, and then either holds on the far side (acceptance) or snaps back (rejection) — and the reaction is supposed to tell you where it goes next. Rehan writes this up as a pre-registered protocol for bitcoin, with exact definitions of levels, sweeps, acceptance and rejection, and placebo controls, but without results.
What we tested
We ran the protocol on bitcoin perpetual futures from January 2020 to October 2026, with a spot-market replica from 2023. Levels were monthly and weekly pivots, round $5,000 prices and seven-day extremes. A sweep was a one-minute move through a level, and the next four hourly closes decided acceptance or rejection. The placebo ran the same pipeline at price points the same distance away that were not levels, taken from days with similar volatility.
Result
After acceptance bitcoin did better over the next 24 hours than after rejection — but by the same margin at random price points as at real levels: 0.117 against 0.114 of a typical daily range. That margin is generic four-hour momentum and appears at any price mark. Effects larger than about 0.3 of a daily range are ruled out; the spot replica and the period since 2024 did not even keep the sign. A strategy that followed acceptances and faded rejections for 24 hours did worse than its own placebo version.
Next-24-hour return after acceptance minus after rejection, in units of the average daily range
| Real levels | Random price points | |
|---|---|---|
| Acceptance minus rejection | +0.117 | +0.114 |
| Part attributable to the level | +0.003 | — |
| Follow acceptance, fade rejection (24 hours) | Below its placebo | — |
What it means
A sweep followed by four hours on the far side of a level is four hours of trend; the level does not make it special. The pattern feels predictive because, after the fact, it is a description of momentum.
What would change the verdict
Sweeps of visible order-book walls — the part of the protocol that needs a long order-book history — remain untested. Many more events, for example intraday levels across many coins, would make smaller effects measurable.
This note re-tests a claim made by the cited authors on our own crypto market data; the summary of their method is ours. Tested · published · last reviewed . Research, not investment advice.