About any longer length of time, There may be not often a statistically substantial autocorrelation in superior frequency returns. If there was, then the above would be relevant which might dampen the result.$begingroup$ For an alternative with rate $C$, the P$&$L, with respect to modifications from the underlying asset rate $S$ and volatility $sig