Jump to content
Brewer Fanatic

9/24/08 Pirates (Maholm) at Brewers (Sabathia): 7:05 PM CDT

wOOgiE22

By wOOgiE22,
in Archived Game Threads

  • Replies 225
  • Created
  • Last Reply
steve267268 Draft Eligible

steve267268

Verified Member
Posted
Posted

Small sample size is fine for things like "clutch". But the stat geeks try to make too much out of their numbers. No one is trying to predict what will happen next, and no one should because anything can happen. If a player goes something like 35 for 91 in situations you could reasonably define as clutch situations then what's wrong with saying that they were very clutch in those situations? It doesn't mean that player will perform that way tomorrow or next week. That's what the stat guys don't get. You can't try to predict things. Nothing is an accurate predictor of anything. The players on the field are human, not robots.

 

And i mean geek in the most affectionate way possible, as i consider myself one.. just not the stat sector of geekville.

Link to comment
Share on other sites
Blancofan1 Draft Eligible

Blancofan1

Verified Member
Posted
Posted
hurricanecrush[/b]]I think I'm going to start cheering FOR the Mets. It seems like the teams I root for in September always suck.
hurricane--I know this is asking a lot, but could you please root for the Mets tomorrow, and then switch your allegiance to the Cubs for the final 3 games.
Everything I've ever known, I've learned from Brewerfan.net....Seriously though
Link to comment
Share on other sites
Ennder Carolina Mudcats

Ennder

Verified Member
Posted
Posted
Small sample size is fine for things like "clutch". But the stat geeks try to make too much out of their numbers. No one is trying to predict what will happen next, and no one should because anything can happen. If a player goes something like 35 for 91 in situations you could reasonably define as clutch situations then what's wrong with saying that they were very clutch in those situations? It doesn't mean that player will perform that way tomorrow or next week. That's what the stat guys don't get. You can't try to predict things. Nothing is an accurate predictor of anything. The players on the field are human, not robots.

 

And i mean geek in the most affectionate way possible, as i consider myself one.. just not the stat sector of geekville.

You are looking at stats in completely the wrong way though. I play poker which has given me a pretty strong background in 'random stats'. You can make a -EV(expected value) play in poker and it might work out 40% of the time but it will also fail 60% of the time and eventually cost you money. Over 10,000 hands of poker that -EV move might be profitable for you because of pure dumb luck but making the move again is a mistake every time you make it, eventually it will catch up with you.

 

Play percentages in baseball based on small samples is the same thing, regardless of what has actually happened eventually it will catch up to you and you will go -EV and lose more games than you should have.

Link to comment
Share on other sites
owbc Carolina Mudcats

owbc

Community Moderator
Posted
Posted

I also agree that the bigger the sample size, the more likely players are going to regress toward their regular batting average. That's probability like we all learned in high school when we flipped a coin 100 times.

The other issue is how do you define "clutch"? There's too much ambiguity for my liking.

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

Go to topic listing
The Twins Daily Caretaker Fund
The Brewer Fanatic Caretaker Fund

You all care about this site. The next step is caring for it. We’re asking you to caretake this site so it can remain the premier Brewers community on the internet. Included with caretaking is ad-free browsing of Brewer Fanatic.

Become a Caretaker
×
×
  • Create New...