View Single Post
Old 11-24-2014, 01:41 PM   #6
RchW
Hall Of Famer
 
RchW's Avatar
 
Join Date: Jul 2004
Location: Toronto ON by way of Glasgow UK
Posts: 15,629
Quote:
Originally Posted by FourSixThree View Post
This would simply be a tool for managers and it would really only apply to simulations. I've read a lot about proving or disproving "clutch". There are some arguments I agree with and some that I don't. I'm not saying that some players perform better in pressure situations, but I do think some players perform worse. The most celebrated argument claims that you can't use past performances to predict how a player will perform in the future given similar circumstances.
You are somewhat misrepresenting the argument here. Statistical analysis regarding clutch performance is not generally seen as reliable often due to small sample size or too much variability. Other statistical outputs where sufficient data is available and where the reliability meets certain criteria are absolutely used to predict future performance. I can guarantee that such analysis along with detailed skills anaysis is used to develop contract offers for FA and to predict how and when a prospect may be ready to move up on the depth chart.


Quote:
Originally Posted by FourSixThree View Post
I'm not sure I agree with this because all other aspects of baseball refute this argument. If Player X has a history of hammering left-handed pitchers, don't managers use this to predict how well said player might perform against a southpaw?
Of course they do. That's what platoon splits are for. Nothing to do with clutch though.

Quote:
Originally Posted by FourSixThree View Post
If Player Y has performed exceptionally well against Pitcher Z in past matchups, won't a manager be more likely to use this player in a tough situation against this pitcher?
Yes but statistically there is probably no evidence that the performance will continue to be good. As "The Book" says take 1500 random PA over 30 PA every time. I don't have a problem with this anyway. Human nature is irrational.

Quote:
Originally Posted by FourSixThree View Post
I think a "clutch" rating could work similarly in this respect. If Player Z has historically performed well late in the game/in close games/in tied games/etc, perhaps this player's "clutch" rating would be a tic or two higher than a player who hasn't done as well. During a simulation, the computer controlled manager would then be more inclined to call on that player.
As we said earlier how many PA would you consider enough to rely on any given performance? If a marginal player gets hot for a month and gets several "clutch" hits would his clutch rating be legitimately higher than an 8-10 year veteran with 5000 PA? Would you not have to consider how that performance came about. Who was the opposition, when in the season, the specific pitcher etc. I'd be more inclined to trust a less spectacular but consistent veteran than the flavor of the week.
__________________
Cheers

RichW

If you’re looking for a good cause to donate money to please consider a Donation to Parkinson’s Canada. It may help me have a better future and if not me, someone else. Thanks.

“Conservatism consists of exactly one proposition …There must be in-groups whom the law protects but does not bind, alongside out-groups whom the law binds but does not protect.” Frank Wilhoit
RchW is offline   Reply With Quote