Typical refers to the fact that since you don't like sabermetric analysis you insult those involved in the field. Why else would it be in the same paragraph where I defend/explain the process? But hey, I understand, it's over your head so the only way you walk away feeling good is to put it down.
And, no, it isn't a scientific fact. Rate stats are nothing more than probabilites based on previous performance, and runs-based metrics are just a way of trying to show the context free value of a players performance. That doesn't mean the principles of the scientific process cannot be employed. Hypothesis, data, analysis, test, rinse, repeat.
No, RISP doesn't account for score, but C&L does (close and late). C&L situations are as follows:
-After the 6th inning
-Up by no more than 1
-Down by no more than 2
I know, that doesn't account for every situation that could be considered clutch. That said, I trust it more than your or my ability to remember a players performance in each and every "clutch" situation for an entire season. It is entirely more likely that those split stats show a clearer picture than your memory does. You remember the dramatic walk-off homers/hits for a long time, but the anti-climactic outs fade from your mind rather quickly.