I hear you entirely -- Too much information often confuses more than it helps, but here you go: Our repeatability in the wind tunnel is very good. The wind tunnel test setup is fairly dialed and the balance gives us results with standard deviations around 1/100 of a lb (.01 lbs). This is after the wind speed is corrected to standard atmospheric conditions. Of course, once you pull a product out, get a rider back on the bike, etc, etc. you have introduced several variables. I got data within .03 lbs between different tests of the bottle (not back to back)...back to back tests were within .01 of a lb. These variations are usually more due to the rider slightly shifting in the wind tunnel. However, testing without an athlete would be completely useless.