Scalping and liquidity hunting

What the strategy actually does

Liquidity hunting targets predictable forced flow: clusters of stops and liquidation levels that must transact regardless of price. The trader positions ahead of that flow and exits into it, often within seconds.

The edge per trade is tiny and the trade count is enormous. Profitability depends entirely on execution quality — queue position, latency, and the ability to cancel and re-enter faster than the market moves.

• Thousands of fills, minuscule average result per fill.
• Edge is execution, not opinion.
• Any delay between leader and follower consumes a large share of the per-trade margin.
• Fee sensitivity is extreme relative to slower archetypes.

Why replication degrades it

A mirrored order is by definition later than the leader's. For a strategy whose entire edge is measured in basis points captured over seconds, that delay is not a rounding error — it can be the whole edge.

The honest conclusion is that a great scalper is often a poor allocation target. Scoring can identify them accurately and still exclude or heavily de-weight them, because the score describes the leader's results, not the follower's.

What is used instead

Archetypes with holding periods long enough that a few seconds of delay is immaterial receive the bulk of allocation. That is a deliberate trade of headline score for realized fidelity.

Past performance is not indicative of future results. Perpetual futures are leveraged instruments and carry a substantial risk of loss, including the loss of your entire position.