The question: did the process move, or did one observation sting? Establish the baseline Use a stable reference window and compare like with like. Small-cap open orders should not share a control range with large-cap mid-day orders. Segment enough to make the baseline meaningful, but not so much that every segment is too sparse. Track process outputs Useful execution outputs include implementation shortfall, arrival-to-fill delay, reject rate, partial-fill rate, venue adverse selection, and realized spread capture. PnL alone is too delayed and too confounded. Look for patterns, not drama Investigate a point outside the action range, a run of same-side…
Read the full lesson
Sign up free: one personalized lesson every day, matched to your role and goals.
Already have an account? Sign in