Inspired by yet another article on the concept of clutch players in baseball, I thought I'd make a few comments about what clutch actually means.
To most sabermetric guys, a clutch player is one who improves his level of play in important situations.
To me, a clutch player is one who performs well in important situations.
There's a subtle difference there. For the sabermetric definition, a clutch player must perform worse in regular situations than in important situations, otherwise there would be no improvement in the clutch.
With my definition, a player who plays consistently well no matter what the situation can still be considered clutch. A player who plays consistently well cannot be considered clutch using the first definition.
My problem with the new sabermetric definition is that a lousy player can be less lousy but still bad and be considered clutch, as long as he performs his best in the appropriate situations. This has little value to any fan watching at home or any manager in an important game. You don't go to your worst player and put him in because he's not quite as bad as usual in these situations. Instead, you look to the guy who is going to perform the best.
If an all star like Alex Rodriguez hits for a few points less of AVG or OPS in clutch situations than he does normally, do you want him at the plate when it counts? Do you put in Wilson Betemit who hits better than his usual in pressure situations? If A-Rod's bad is still better than Betemit's good, you play A-Rod.
Unfortunately, the idea of a clutch hitter actually performing slightly worse in clutch situations than normal is entirely unsatisfying to many in the sabermetric community and so they chose to alter their definition to make their analyses look more interesting.
Countless hours have been spent trying to come up with statistical evidence to prove that clutch hitting doesn't exist. The sabermetric community can't stand the notions that certain players are "clutch" and the constant use of that term by fans and other baseball personnel. There has been a deluge of "David Ortiz is clutch" talk in the popular media that gets under the skin of many people. I agree that common perception in this matter is likely incorrect, that guys we often think of as "clutch" probably aren't the best performing players in pressure situations.
For me personally, if you asked me to recall guys I thought were clutch based on performances I watched, I would think back to a few memorable Jim Leyritz hits back in the late 90s and my all time favorite clutch moment - Aaron Boone's solo walk off home run in the bottom of the 11th inning in the 2005 ALCS game 7 between the Yankees and the Red Sox.
Our gut instincts and recollections of clutch players are greatly biased due to a small sample size and while fun to recall, have little practical value.
Unfortunately, many sabermetric fans have the notion that a clutch player is one who can perform better than he normally does in a clutch situation. This notion has little to no value either.
The bottom line is that a true clutch player is the one who will perform better than other players in important situations, regardless of how well or poor he plays normally, and this almost always comes down to the player who generally plays best all season long. It's not very exciting from a fan's point of view and it's not very interesting to extreme statisticians, but it's the only perspective that has some actual value in my opinion.
