[Crex]

Why not take a second piece of the same steel as a "control" sample? Grind it to the same thickness and finish as the area on the blade you want to test, only with parallel sides. Heattreat at the same time under the same parameters (normalizing, clay coating, quenchant temps, etc) and do your testing on the "control". You should get more consistent and reliable data from the tests and thereby know more about what you think is going on with your HT.

Side note: Have to agree with the others on the grain depiction in the original pic. When it manifest that bold something got a little too hot. Bonus that steel should is still salvageable with some careful thermalcycling.

Never trusted the files for accuracy after the first couple of uses, too many human error variables that are effected on the process from day to day. Mental temperament being just one of them.
Example: Just prior to testing you manage to step in your neighbor's dog's deposit in your yard.... or the other neighbor's goat is in your garden eating the last of your prized okra.... or the boss just cancelled your long awaited vacation...
Any of these will affect just how hard you apply the files to the steel.....blown test results.