That is a good point. I probably should run tests with larger data sets. I just didn't want to start with 500,000 records. I will try a set much larger then 10 records. Hopefully with a larger set, I will see some more meaningful numbers.

You are also right that the coverage profiler is not totally useless (I tend to over exaggerate). In using the profiler with my staff yesterday, we did conclude that its usefulness was limited in the example that I give, but it will be of great interest in all of the other areas that you pointed out.

Thanks
Dean