Yeah and...
I might regret saying this but... Couldent we test these theories? Basically if we take someone who has kr on shifts and then do various runs with torque management on and off wouldent that settle this? Heck ill do it if this is still such a hot issue. Wouldent take me more then 20 mins to put a small post together.
As you can see, a frame caught mid-shift. The timing is at ZERO... this represents torque management. Notice there's NO KR during this torque management. Your infantile position on this is put to rest right here.
As for the blips of KR, they happen every time, at the same RPM, with the same intensity. This is FALSE KNOCK. There was no actual ping/detonation/knock etc, etc, etc.