One last note for the OP:
while i think your description of bfs vs cfs is decent, i think it's not written neutrally at all, instead clearly cfs biased (BFS can be laggy, may be higher quadrant but that doesn't mean it's faster).
I think to be fair you should put a more neutral opinion of either scheduler, and for those who really "want to know" there's actually a fairly decent writeup that can be used as a reference
found here.
I would consider the conclusion in this article to be very accurate:
"The results indicate that CFS outperformed BFS with minimizing turnaround time but that BFS outperformed CFS for minimizing latency. This indicates that BFS is better for interactive tasks that block on I/O or user input and that CFS is better for batch processing that is CPU bound."
AKA how you use your phone is going to make the biggest difference (do you run tons of stuff at once and truly multi-task or do you usually just work in the current app and switch as your needs switch?)
IIRC, the default cynaogen kernel is BFS (or at least was, not kept up if it got changed), whereas I believe the default Google releases are CFS.