Originally Posted by Cant Touch This
For example, does RISP recognize the difference between situation A: a runner on second with 2 outs in a 10-0 game vs. situation B: runners on second and third with 0 outs in a tie game in the bottom of the last?
There are stats that differentiate all of that if you want them. The problem is, the more qualifications you put on a certain situation the smaller the sample size and the greater the variation that would be considered within normal variation. So you might look at a certain situation, and a player might appear to be "clutch" yet it's well within expected normal variation (this is why I stay away from them because most people aren't go to understand the statistics enough to realize that fact). RISP is simple, understandable, the sample size is often large enough to minimize that variation, it incorporates mostly "clutch" situations, and over a long enough timeframe the makeup of those situations will be similar for all players. If you prefer something else, that is out there too. Yet you have to realize sample size plays a major role. If a guy OPS's 200 points better in a specific situation that could actually be well within normal variation and not prove he is "clutch" in any way.