Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 11/03/84 (WLS Mods); site fisher.UUCP
Path: utzoo!watmath!clyde!burl!ulysses!allegra!princeton!astrovax!fisher!david
From: david@fisher.UUCP (David Rubin)
Newsgroups: net.sport.baseball
Subject: Re: Unusual Occurences in the NL East (Ozzie Smith's Value)
Message-ID: <589@fisher.UUCP>
Date: Wed, 24-Apr-85 11:01:54 EST
Article-I.D.: fisher.589
Posted: Wed Apr 24 11:01:54 1985
Date-Received: Thu, 25-Apr-85 04:00:48 EST
References: <652@spuxll.UUCP>
Organization: Princeton University Department of Statistics
Lines: 22

> Whitey says Ozzie's glove is worth 100 rbi's. Maybe. But does it save
> 40+ games? 
 
Almost certainly not.  Using Bill James' Pythagorean approximation to
winning percentage (% = r1^2/(r1^2+r2^2) where r1 is runs scored and
r2 is opponents' runs scored), we get a guess of 10-14 games won by
Ozzie's glove (using Herzog's exaggeration of its value).  While it is
not 40 games, if it WERE true, Smith would be worth more than Sutter
and his 40+ saves.  Of his 42 saves, it would be a CONSERVATIVE guess
that 3/4 of those leads could have been held by other pitchers on the
Cards' staff (notably Allen, who had saved 25 games with a last place
team the year before he came to the Cards), leaving Sutter valued at
no more than 10 games.  Furthermore, we'd be comparing a typical year
for Smith with a career best for Sutter; Sutter's expected value is
somewhat less.

Conclusion: if Smith really prevents 100 runs, he's worth more than
Sutter (or Hernandez); it is the premise, not the inference, which begs
for closer examination.

					David Rubin
			{allegra|astrovax|princeton}!fisher!david