Fuelman, a test on only two engines is nearly meaninless. There can be enough tolerance differences between the two to account for any differences. To be a valid test, you would have to run identical tests using at least 10 engines, preferably a 100, under each condition and then noted any differences or trends. While your test are interesting, they prove nothing unfortunately.
Rodney,
Those were just a couple of the experiments, tests, whatever you want to call them. The single common thread to all the ABC type 2-stroke running and breaking in I've done is that hot is good, cold is not.
For example, with the OS 32SX that was run slobbery rich for break in, (and had about 500 rpm less than the other engine broken in at the same time by the correct method) all was good when I replaced the piston and sleeve much later on and the performance was right with the other one after I broke it in correctly. So tolerances and everything else aside, that is what i experienced. Even the tolerances are so good now days with CNC, I feel that is a minimum influence. I run these little model engines for a living (mostly fuel testing and comparison) but sometimes I do have the luxery of playing like I did with the OS 32's and when I teach break in techniques to car guys.