I get this asked my opionion alot on this matter, so I figured why not have it documented:

2 years of NBA and CBB games analyzed.

1-4 are the least key numbers (fouls obviously)

Games landing on 1 is the fewest for obvious reasons, 3-4 very close in regards to %.

7-8 are key if could call any of the findings "key #s", even though the actual %s in comparison to other #s are not much higher to even deem them key in terms of great value.