12:1 is perfectly acceptable around peak torque when the car is only seeing 1-2 degrees of advance. Keep in mind o2 sensors lag rpm by 200-300rpm, especially at the low end where exhaust flow rates are low. So the car isn't really running 12:1 at 4000rpm. A simple steady-state dyno run will prove that.

It should hit around 12:1 by 4000rpm, then drop into the mid 11s and then drop into the mid 10s in the last 1500rpm of the rev range.

 

You make some good points, but nothing that's new to me. I think I forget where I'm posting sometimes. I never see 1-2 degrees advance. I don't think I ever drop below 6, on DSMlink or AEM. I've never seen an O2 sensor lag that far, but I also run mine at the test pipe. Combine the fact that dyno sensors are further down the pipe, and most tend to respond slowly because they don't replace them often enough with race gas use, and you are most likely correct though.

It's also important to remember that you don't change the AFR just because the timing advance is lower at peak torque. The timing advance is lower for a reason. Higher charge density resulting in reduced propagation delay, slower piston speed, etc. Of course I'm refering to timing that is set up properly, not the anti-bankrupcy bull**** the factory programs in. If you get best torque at 4k rpm at 5 degrees, and best torque at 8k rpm at 18 degrees, the "timing" isn't really any different as far as peak cylinder pressure lining up with the "ideal" mechanical point in the cycle goes. So you really aren't running "less" timing at 4k as far as AFR requirements go.