Originally Posted by BadFundamentals
But again, (as with the Baseball Prospectus) study when comparing a RELATIVELY MINOR (to OPS) attribute (like productive outs) to WINNING, OPS should be held constant. He makes no mention of holding OPS constant. He just tosses out the .283 correlation and dismisses it. Not surprising that there isn't a stronger correlation - OPS smothers it.
I have no idea what this means, and I've had a good amount of statistical training. It seems like he's saying that if you take a team's POP and line it up with wins, there isn't any correlation. It has nothing to do with OPS. OPS wouldn't be involved at all. If I looked an independent
variable and found that it only correlated by a factor of .283, I'd throw it out as well.