What I did was turn on AFR correction, which gets it very close cruising around using the auto tune.

I have a bunch of failsafes, AFR/PSI, Oil Pressure, Coolant temp, Oil Temp, Over Boost, knock, Fuel Pressure, etc. With an LED light triggered in my face along with massive timing pull or fuel/spark cut. That saves you from looking at 20 gauges while doing WOT by yourself with a laptop.

Log every WOT pull. View the log manually afterwards looking at RPM vs PSI vs AFR. Compare that to your AFR table, Mine is 11-11.5AFR everywhere above spooling PSI. So at low RPM and Low PSI I may have 13.5-12.5AFR to get a bit of extra spool.

Look at the AFR vs targeted AFR vs amount auto tune adjusted to get to what it thinks the AFR you need is. Manually add or subtract if auto-tune didn't add or subtract enough. I guess mine is set relatively passively, as I would want it to be. I'm accustomed to manually adjusting the tune, but a small percentage change and highlighted cells is very easy to compare fuel cell vs graph.

As I'm autotuning, I don't constantly re-save my tune. I noticed you said you did. I may be wrong but I believe it gives a larger sample size for a particular cell when it isn't re-saved so often. :shrug: I drive for quite a while before saving the tune.

I don't think that is the case. I save my tune quite a bit as well. It clears out the chart that shows actual changes but leaves the chart that shows cell weighting so I assume the sample size and weighting remains untouched.

I haven't noticed a difference either way. Only thing is if you do save the tune you can re-upload the changed file into VE analyze which also gives a predictive change. Otherwise it's going off the first tune before any modification. I usually save each tune as something different, but not after each pull, after each session which usually includes a couple pulls and maybe 30minutes to two hours driving while I'm focusing on a specific area..