I have done a lot of tuning and testing in the way that Mike has outlined. Here are my suggestions:
Do a series of at least 5 runs in a short space of time with no changes and look at the variance in the data. This will help you identify real changes from "noise".
Do a number of runs, one after another with a particular setup and use the mean time. I used 5 runs.
Be wary of inferring performance changes from data collected on different days. You may be surprised how much temperature, barometric pressure, humidity, wind speed and wind direction can affect timed runs. I tried to test one change per day.
Some changes may produce more torque in one part of the rev range and less in another so do the tests over the rev range that is most meaningful to you.
76 Suzuki GT500
2011 Jaguar XKR
A little government and a little luck are necessary in life, but only a fool trusts either of them. - P.J. O'Rourke
Last edited by alfaparticle; 07-14-2019 at 07:51 PM.