SNGPT and KS hand range %s

I was fooling around with SNGPT and, for one of my problems, got a calling range of:

22+,A2+,K9o+,K5s+,QTo+,Q9s+,JTo,J8s+,T7s+,97s+,86s +,76s,65s (34%)

This struck me as odd -- I wouldn't have guessed K8o was worse than 65s. I looked them up on the KS list and found K8o at #55 and 65s at #126.

So, I entered a range of 34% and had SNGPT convert it, getting:

22+,A2+,K5o+,K2s+,QTo+,Q8s+,J9s+.

I had always suspected that an ideal pushing range of 30% (or whatever) didn't always correspond to the top 30% of KS ranked hands. But, I had never noticed a case like this (never looked real hard). And, I have never seen it discussed. So, I thought I'd point it out. I don't know if it has any practical consequences.