02-09-2009, 03:45 PM
<!--quoteo(post=16815:date=Feb 9 2009, 02:20 PM:name=BT)-->QUOTE (BT @ Feb 9 2009, 02:20 PM) <{POST_SNAPBACK}><!--quotec--><!--quoteo(post=16809:date=Feb 9 2009, 12:56 PM:name=leonardsipes)--><div class='quotetop'>QUOTE (leonardsipes @ Feb 9 2009, 12:56 PM) <{POST_SNAPBACK}><!--quotec-->The Clutch argument exemplifies why it seems stat people do not watch the games. How can they deny something that you observe and that common sense suggests exists? It does not make sense that all people react the same to a given situation. In our everyday lives we see it, in football we see it, but in baseball it does not exist.
You can not assume that statistical analysis, no matter how sloppy, is better than observation. The stat folks have not even taken the time to properly collect the data. Clutch at bats do not equal at bats w/ RISP or even close and late. They may also be using the wrong baseline. Clutch ABs are usually against better than average pitchers, so it is probably wrong to compare clutch averages to career averages. Making the assumption that just because a result falls withing the range that could be random, means it must be random. They do not establish normal clutchness or the range of clutchness.<!--QuoteEnd--><!--QuoteEEnd-->
I guess I'm not following you Leonard. Doesn't KB's examples show why stat people think "clutchiness" might be a myth? Again, people will remember Jackson's clutch home runs. They will forget the unclutch strikeouts. Stats don't have that bias.
<!--QuoteEnd--></div><!--QuoteEEnd-->
Statistical analysis is biased. It is not like finding the area under a curve. It is done by people who assume something is or is not the case and try to prove it. In scientific research it is kind of an accepted fact. Why do you think every time a study that comes out showing one result another study showing the opposite result comes out. How many years did it take to prove smoking was bad?
I listed three valid reasons why statistical analysis suggesting clutch does not exist may be flawed, and somehow, it is refuted by anecdotal evidence about Reggie Jackson.
Brock, my post said it seems stat people don't watch the games, because they refuse to admit things they observe happening, really occur unless shown by stats.
You can not assume that statistical analysis, no matter how sloppy, is better than observation. The stat folks have not even taken the time to properly collect the data. Clutch at bats do not equal at bats w/ RISP or even close and late. They may also be using the wrong baseline. Clutch ABs are usually against better than average pitchers, so it is probably wrong to compare clutch averages to career averages. Making the assumption that just because a result falls withing the range that could be random, means it must be random. They do not establish normal clutchness or the range of clutchness.<!--QuoteEnd--><!--QuoteEEnd-->
I guess I'm not following you Leonard. Doesn't KB's examples show why stat people think "clutchiness" might be a myth? Again, people will remember Jackson's clutch home runs. They will forget the unclutch strikeouts. Stats don't have that bias.
<!--QuoteEnd--></div><!--QuoteEEnd-->
Statistical analysis is biased. It is not like finding the area under a curve. It is done by people who assume something is or is not the case and try to prove it. In scientific research it is kind of an accepted fact. Why do you think every time a study that comes out showing one result another study showing the opposite result comes out. How many years did it take to prove smoking was bad?
I listed three valid reasons why statistical analysis suggesting clutch does not exist may be flawed, and somehow, it is refuted by anecdotal evidence about Reggie Jackson.
Brock, my post said it seems stat people don't watch the games, because they refuse to admit things they observe happening, really occur unless shown by stats.
I like you guys a lot.