I believe it's a reporting thing with the app and the eCores are throwing things off. If you're running say 51x on the Pcores & 42 on the Ecores, not all cores are running at the top speed. My theory is that the benchmark expects all cores to be running at the same speed and hasn't been adapted for the big/little architecture. This is just a theory on my part but, depending on your CPU settings, it's the only explanation that makes sense to me.
Also note, I did this config, as many are running higher single/dual core clocks vs all core. So I hypothesize that the app is taking the reading from the first core and just assuming the others are in tow.