Directional bias and net exposure

Isolation versus economics

Sub-accounts prevent the exchange from netting a long against a short, which is what keeps per-leader results attributable and margin contained. Economically, though, if six leaders are long the same major and one is short it, your portfolio is net long that major.

Both facts are true at once, and confusing them leads to the mistaken belief that isolation is a hedge. It is not a hedge; it is a measurement and containment mechanism.

• Gross exposure: the sum of all position notionals, which drives fees and margin use.
• Net exposure: long minus short, which drives directional sensitivity.
• Market concentration: how much of the net sits in one perp.
• Crowding: how many leaders hold the same view simultaneously.

Why the tilt appears

Leaders are selected on score, not on the direction of their current book. In a strong bull regime most high-scoring accounts will be long, so the basket inherits a long bias without anyone choosing one.

That is not a malfunction — it reflects where profitable behaviour has been. It does mean drawdowns concentrate when that direction reverses, which is exactly the scenario diversification is weakest against.

What limits the tilt

Per-leader notional ceilings and independent leverage caps bound how large any single contribution to the tilt can become, and archetype spread reduces the chance that every leader takes the same side. Neither mechanism guarantees a balanced book.

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.