Liquidity Sweeps / Stop Hunts as Signals

The “liquidity sweep” or “stop hunt” idea holds that price deliberately pushes through clusters of stop orders just above or below obvious swing points, then reverses — and that this sweep can be used as a predictive entry trigger.

The honest, narrow claim is this: there is no peer-reviewed, out-of-sample published evidence that liquidity sweeps predict the subsequent direction after costs. That is not a claim of fraud, and it is not a denial that stop clusters exist or that price sometimes moves through them. It is a statement about the evidence base.

The central difficulty is that wicks beyond swing points are common in noisy data. In any liquid market, price routinely pokes past a recent high or low and then does something — sometimes reverse, sometimes continue. Highlighting the instances that preceded a reversal is selection after the fact. To establish an edge you would need to show, out-of-sample and with realistic costs, that entering after a sweep beats entering at random. That demonstration is what is missing.

The underlying observation — that stop orders pool near obvious levels — is real and well understood. The unsupported part is the predictive claim that a sweep tells you which way price goes next.

The broad technical-analysis literature (Park & Irwin 2007; Lo, Mamaysky & Wang 2000) finds mixed-to-weak results for rule-based methods after costs. Sweep-based entries have not faced comparable testing, and the burden of proof rests with whoever claims an edge.