I'd add that over the years I have found that a properly maintained dyno with an operator that understands how to set it up and run it consistantly should never demonstrate anything like a 15% error factor.
If you want to get a feel for the effect that your tweaks are having, then you need to do one change at a time and then do a pull. Last week we did a series of 19 pulls to get my car where I wanted it, but watching how each change affected the power curve really gave me a feel for the process. The results we achieved wer right where we calculated the engine would be, so I am confident that the error factor was minimal.
Rick
