That's a very good point Bill, I don't know how much knock is needed to force the PCM to use the low timing table. The much lower timing on the low octane table probably explains why I did not see a decrease in gas mileage on 87 octane, while typically, other members have in the past. I guess the question to ask is, how much knock in how short of a time period causes the car to use the low octane table?
Maybe this experiment would have been more insightful if I had left the tables alone for awhile and logged the commanded timing advance, and then been able to see how much the car takes before dropping to the low octane table. What worries me is that the KR spiked significantly higher during high load conditions, and did not seem to switch tables due to these spikes. One could potentially get a very damaging spike of knock before the computer could react.
