The most criticised feature of any trend-following architecture is the wide initial stop. Sixteen volatility units of adverse room strikes the untrained eye as recklessness — surely risk control means cutting losses close?

Papers I and II supply the reread. If milestones in price are appointments in time, then a stop denominated in volatility units is not primarily a distance concession. It is a time grant. Adverse travel of sixteen units at any given efficiency requires a calculable minimum number of bars — at even near-perfect hostile efficiency, no fewer than thirty-two bars of continuously adverse movement before the position can be killed. The stop is a guarantee that the market must argue against the thesis, coherently and at length, before it is allowed to end the conversation.

That is the hidden symmetry of a well-built exit architecture. Death below and immortality above are denominated in the same unit — volatility-time — and the wide stop’s true function is revealed: it keeps the lead-time window open long enough for the likelihood-ratio test of Paper II to finish. A tight stop does not merely risk less; it asks less — it forecloses the very boundaries that distinguish truth from noise, and converts a trend system into a coin-flip system while leaving the branding intact.

None of this licenses recklessness. The width is paired with sizing law — risk per position is fixed as a fraction of capital through the stop distance, so a wider stop means a smaller position, never a larger loss. What the width buys, at constant risk, is time. And time, Paper I showed, is the one input truth requires and luck cannot supply.

The practitioners’ summary: judge a stop not by how close it sits, but by how many bars of hostile argument it demands before surrender — and whether that number exceeds the time your thesis needs to prove itself.