Rox Talk

Since it is the offseason I have revisited some earlier research that I looked in particularly Game Score...see my post dated 7/30/2007. My previous research documented Aaron Cook's masterful 74 pitch complete game victory (Game 101). His game score was a measly 67 basically because he is not a power strike out pitcher which Bill James' Game Score is more tailored to. In an effort to "level" the playing field for "finesse" pitchers, I tried to come up with what I call Game Score +. My effort is described below:

Bill James devised the Game Score metric to determine the dominance of a pitcher in a game. This score is derived by:

1. Start with 50 points;
2. Add 1 point for each out recorded;
3. Add 2 points for each inning completed after the 4th;
4. Add 1 point for each strikeout;
5. Subtract 2 points for each hit allowed;
6. Subtract 4 points for each earned run allowed;
7. Subtract 2 points for each unearned run allowed; and
8. Subtract 1 point for each walk.

The highest Game Score for a 9 inning game (105) was recorded by Kerry Wood on May 6, 1998 when he struck out 20! In some circles this metric is also known as the "Ryan" since Nolan Ryan has had 31 starts with a game score equal or greater than a 90. In fact if you look at the top Game Score pitchers most of them are strikeout kings (with the likes of Ryan, Johnson, Koufax, Seaver, Gibson, and Clemens at the top). Based on the scoring above it is obvious that a high strikeout game will lead to the highest possible score for a 9 inning game. The maximum score possible is 114 points for a 27 strikeout perfect game. Meanwhile a perfect game of 27 groundball or flyout outs is only worth 87 points. These two extremes pretty much define "Power" (or dominance) and "Finesse" (or economy). James' method rewards the strikeout which like the Home Run is a bellwether of a pitcher and memorable games surrounding pitcher's typically is the 19 to 20 game strikeout performances.

My study turns to those unsung pitcher's who breezed through games, hardly breaking a sweat. When history looks back at the 90s it will probably be remembered by two (and baseball's last) 300 game winners (well besides Glavine). These two are at the polar opposites of the pitching spectrum. Clemens was obviously a power, dominant, strike out king while Maddux was the crafty, scholarly, economical pitcher.

Clemens 1984 - 2007
W L G IP H R ER HR BB SO ERA WHIP
354 184 709 4916.7 4185 1885 1707 363 1580 4672 3.12 1.173

Maddux 1986 - 2007
W L G IP H R ER HR BB SO ERA WHIP
347 214 711 4814.3 4522 1876 1665 332 969 3273 3.11 1.141

Gee talking about some similarities! About the only difference is that Maddux has about 600 less walks and of course about 1,400 less strikeouts. When Game Scores are considered Clemens had 13 starts with scores greater than 90 (high of 99) while Maddux had 4 starts greater than 90 (high of 96). Would like to try to find average game score of their 700 starts but that is beyond the stats I could find. My guess is that Clemens has a higher career game score average due to his greater amount of strikeouts. Great for Clemens and other strikeout pitchers but what about Maddux (based on career stats they basically look the same). Is there a way to equalize pitcher's who win games with outs and economy of pitches thrown? Is not Red Barrett's 58 pitch performance in August 1944 (Game Score of 84) not equal to Wood's performance (which by the way required 122 pitches!).

Bill James' Game Score is an esoteric stat that very few find all that intriguing. Its basis is another way to define a quality start. My premise is that for defining a pitcher's start it is a good tool and a good place to start (can it be tweaked meaningfully?). As mentioned above the highest possible Game Score for a 9 inning game is 114 which relates to a 27 strikeout performance. In a perfect perfect game this would equate to 81 pitches. Therefore to start any sort of equalization the Game Score + system relates to this magical 81. Next post I will get into the numbers.