I am hoping someone has some experience or ideas in this arena that they would care to share...In the world of electronic trading, how does one measure latency performance?I'm not talking about the actual measurement of latency, but rather how to quantify the effect latency has on P&L. Often latency is a major factor: shave 10 microseconds and P&L will go up, lose 10 micros and the strategy is worthless. Other times a strategy is not sensitive to latency at all: 10 microseconds either way makes no discernible difference.However, since there are so many factors which can affect a strategy's performance, such as current market conditions, parameters/inputs, a trader's 'gut feel', etc, it is not trivial to attribute some portion of P&L to latency.So what approaches do you take?Any and all advice is greatly appreciated.