WBO2 AFR Wild Variation

Background:
16197427 $0D BJYK bin.
Vortec L31 with carb mani and adapter plate

I have disable spark retard and closed loop modifications.
Closed loop was causing the BLM and INT to constantly add fuel, I assume due to a bad O2 sensor, which I will eventually get around to replacing, but as I have a WB and can tell if it is rich/lean from that I just disabled it for now.

I was getting issues from the knock sensor or esc module that was causing the PCM to pull virtually all spark advance. again, disabled it to get around the issue until I figure out exactly what it is, but the engine is not knocking or pinging, so I assume it is simply an issue with knock sensor. I have heard of flat tappet knock sensors not playing nice with roller cam engines.

i digress, the reason for me posting are wild variations of AFR at part throttle. it would appear that it is primarily at the points where there is overlap of the idle ve table and open throttle ve table. in my bin the values at RPM vs MAP values do have some variation between tables.

looking at my log, specifically the std dev of WBO2 values, the std dev of all these overlapping values near 2.0 which is a crazy amount of variation. values exclusive to the open throttle table appear to be all less than one and in some cases near zero, which is what I would expect.

my next step was going to be to match the values between tables to rule this out being the problem, but I seem to recall that there is a method of disabling the idle table by qualifiers. can someone tell me which qualifier needs to be changed to keep the PCM from entering that Idle VE table?

Any ideas of something else that could be causing this? log attached and bin.