Thursday, November 20, 2008
Outlier Seasons
(originally posted on the Bill James Online forums, in response to user evanecurb's call for player's who had "great years that came from nowhere and never happened again")
Well, it's an interesting question to try to quantify. I figured that looking at Win Share discrepancies would do the trick. However, that ends up being a little tricky.
First, I looked for any person (since 1950, to make it easier) who had only one season with more than 20 Win Shares and no other seasons with more than 12. This rules out someone like Matt Nokes, whose 1987 season was actually only 20 Win Shares, and who also had 17 Win Shares in 1988. Also, this method ends up ignoring a lot of the names that you mentioned.
So I tried it a different way, looking for people whose max Win Shares year was much higher than his next-highest Win Shares year. Doing it this way, I do end up seeing the names that you've mentioned, but much lower on the list than you might think (or not even on the list at all). For example, Brady Anderson's 1996 was insane, and I say that as an Orioles fan. however, in Win Shares, it's a different story. His Win Shares for 1996 was 28; however, in 1992, he actually had 29 Win Shares (and 26 in '97). all that to say, he was a highly productive player for much more than that one season. (Luis Gonzalez is in pretty much the same boat)
With all that said, the interesting names of note I found are (there are many more than this):
Billy Grabarkewitz, 3B, LAD 1970 - 29 WS, no more than 5 for the rest of his 7 year career
Mark Fidrych 1976 (not a big surprise) - 27 WS, no more than 7 the rest of his career
Dick Ellsworth, 1963 - 32 WS, with his next highest being 13
Kevin Mitchell, 1989 - 38 WS, with his next highest being 20
Willie McGee, 1985 - 35 WS, with his next highest being 21
Norm Cash fits in well - 1961, 42 WS; next highest was 27
Cito Gaston, 1970 - 24 WS; next highest is 10
Rick Wilkins, 1993 - 28 & 14
Howard Johnson, 1989 - 38 & 25
Rick Cerone, 1980 - 21 & 9
Joe Torre, 1971 - 41 & 29
Chris Hoiles, 1993 - 26 & 14
Ken Caminiti, 1996 - 38 & 26
Craig Worthington, 1989 - 20 & 8
Dwight Gooden, 1985 - 33 & 18
With Hoiles, Worthington, and Brady, there are a lot of early-to-mid 90s Orioles on the list.
I'm not sure how convinced I am in saying that people like Joe Torre or Howard Johnson or Ken Caminiti who had multiple good years can be considered "outliers". Again, it depends on how you want to classify an "outlier". is it someone who had one outrageously good year despite multiple other solid-to-very good years? or is it someone who had only one very good year and only mediocre years otherwise? it is fun to consider some of those crazy seasons, though.
oh, and one more, since I had originally ignored active players (assuming that they still have a chance to have another great season)... Adrian Beltre's contract year of 2004 gave him 37 WS... his next highest year was 2000, when he had 22 WS (was he playing for arbitration money or something then)... maybe this coming year will be another great year, though, since it's another contract year...
Well, it's an interesting question to try to quantify. I figured that looking at Win Share discrepancies would do the trick. However, that ends up being a little tricky.
First, I looked for any person (since 1950, to make it easier) who had only one season with more than 20 Win Shares and no other seasons with more than 12. This rules out someone like Matt Nokes, whose 1987 season was actually only 20 Win Shares, and who also had 17 Win Shares in 1988. Also, this method ends up ignoring a lot of the names that you mentioned.
So I tried it a different way, looking for people whose max Win Shares year was much higher than his next-highest Win Shares year. Doing it this way, I do end up seeing the names that you've mentioned, but much lower on the list than you might think (or not even on the list at all). For example, Brady Anderson's 1996 was insane, and I say that as an Orioles fan. however, in Win Shares, it's a different story. His Win Shares for 1996 was 28; however, in 1992, he actually had 29 Win Shares (and 26 in '97). all that to say, he was a highly productive player for much more than that one season. (Luis Gonzalez is in pretty much the same boat)
With all that said, the interesting names of note I found are (there are many more than this):
Billy Grabarkewitz, 3B, LAD 1970 - 29 WS, no more than 5 for the rest of his 7 year career
Mark Fidrych 1976 (not a big surprise) - 27 WS, no more than 7 the rest of his career
Dick Ellsworth, 1963 - 32 WS, with his next highest being 13
Kevin Mitchell, 1989 - 38 WS, with his next highest being 20
Willie McGee, 1985 - 35 WS, with his next highest being 21
Norm Cash fits in well - 1961, 42 WS; next highest was 27
Cito Gaston, 1970 - 24 WS; next highest is 10
Rick Wilkins, 1993 - 28 & 14
Howard Johnson, 1989 - 38 & 25
Rick Cerone, 1980 - 21 & 9
Joe Torre, 1971 - 41 & 29
Chris Hoiles, 1993 - 26 & 14
Ken Caminiti, 1996 - 38 & 26
Craig Worthington, 1989 - 20 & 8
Dwight Gooden, 1985 - 33 & 18
With Hoiles, Worthington, and Brady, there are a lot of early-to-mid 90s Orioles on the list.
I'm not sure how convinced I am in saying that people like Joe Torre or Howard Johnson or Ken Caminiti who had multiple good years can be considered "outliers". Again, it depends on how you want to classify an "outlier". is it someone who had one outrageously good year despite multiple other solid-to-very good years? or is it someone who had only one very good year and only mediocre years otherwise? it is fun to consider some of those crazy seasons, though.
oh, and one more, since I had originally ignored active players (assuming that they still have a chance to have another great season)... Adrian Beltre's contract year of 2004 gave him 37 WS... his next highest year was 2000, when he had 22 WS (was he playing for arbitration money or something then)... maybe this coming year will be another great year, though, since it's another contract year...
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment