The View From Wisconsin
Just a random set of rants from a Sports Fan from Wisconsin.
Saturday, April 09, 2005
Adjusting Hall Of Fame Standards
One of my favorite books happens to be written by Bill James - and it's not his Historical Baseball Abstract (though I have both editions of that tome). It would be his work regarding the Hall of Fame, titled Whatever Happened To The Hall of Fame? (or The Politics of Glory).
Anyone who knows James also knows that he has a history of using a shorthand "point" system for approximating the value of a player. In fact, during his Abstracts of the 1980's, he used the Approximate Value method to rate players and teams. He came up with his "ultimate" version of AV in the aforementioned book, the Hall of Fame Standards method.
Basically, players are awarded points for reaching career milestones - 1 point if you're a career .300 hitter, etc. The total number of points would be expressed as the "percentage" of Hall of Fame Standards that a player or pitcher meets. James' intent was that the "average" Hall of Famer would score 50 points on the list.
The list isn't perfect, though. There are three major shortcomings that James probably didn't think through completely when he first drew up the list - which was actually back in 1983, when he first posited about what it would take for players to make the Hall of Fame. The shortcomings are:
For fielders, instead of awarding points for just the player's primary fielding position, you count the number of games played at each position by the player during his career. Divide that number by the total number of games played, and then multiply the result by the "award factor" for that position.
For example, Carlton Fisk played in 2,499 career games. 2,226 of those were at catcher, 27 were at first base, 4 were at third base, and 41 were in left field. James awarded 20 points for games played as a catcher, 1 point for first basemen, 14 for second basemen, 13 for third basemen, 16 for shortstops, 3 for left fielders, 12 for center fielders, 6 for right fielders - and zero for everything else. Thus, Fisk's points for his positions are [20 x (2226/2499)] + (27/2499) + [13 x (4/2499)] + [3 x (41/2499)] = 17.9 points. That gives Fisk a HOFS score of 52.3; Baseball Reference lists him with a HOFS of 49.1, so he actually does a little better with the credit for his games played while not behind the plate.
What about the player who really gets hosed by HOFS - Ernie Banks? BR has him listed at 46.1; using this method, he earns a solid 8.0 points: (1259/2528) + [13 x (69/2528)] + [12 x (1125/2258)] + [3 x (23/2528)]. His new HOFS score is 49.8.
As to what to do about relievers: this isn't as difficult as it may seem. First of all, with now a handful of relievers in the HOF, it's actually easier to determine what the standards are. Relief Pitchers are defined here as a pitcher who had fewer starts than relief appearances (or, more than half of his appearances were in relief). An additional requirement is that a pitcher must average less than 4.5 innings per appearance. If a pitcher meets these requirements, instead of awarding points for wins over 100 (as in rule 1 of the HOF Standards method), you do the following:
These new standards help Bruce Sutter immensely: he's listed with a 17.0 HOFS score on BR, but his adjusted score is now 48.4 with the new rules. Some other players under the new rules:
Anyone who knows James also knows that he has a history of using a shorthand "point" system for approximating the value of a player. In fact, during his Abstracts of the 1980's, he used the Approximate Value method to rate players and teams. He came up with his "ultimate" version of AV in the aforementioned book, the Hall of Fame Standards method.
Basically, players are awarded points for reaching career milestones - 1 point if you're a career .300 hitter, etc. The total number of points would be expressed as the "percentage" of Hall of Fame Standards that a player or pitcher meets. James' intent was that the "average" Hall of Famer would score 50 points on the list.
The list isn't perfect, though. There are three major shortcomings that James probably didn't think through completely when he first drew up the list - which was actually back in 1983, when he first posited about what it would take for players to make the Hall of Fame. The shortcomings are:
- In reality, the average Hall of Famer scores 49.4 (49.0 for pitchers) on the list;
- Players who played a significant number of games at more than one position (Ernie Banks, Pete Rose, Stan Musial) are actually penalized because of how James awards points for positions; and
- Relief pitchers get short-sheeted because of their lack of non-save decisions.
For fielders, instead of awarding points for just the player's primary fielding position, you count the number of games played at each position by the player during his career. Divide that number by the total number of games played, and then multiply the result by the "award factor" for that position.
For example, Carlton Fisk played in 2,499 career games. 2,226 of those were at catcher, 27 were at first base, 4 were at third base, and 41 were in left field. James awarded 20 points for games played as a catcher, 1 point for first basemen, 14 for second basemen, 13 for third basemen, 16 for shortstops, 3 for left fielders, 12 for center fielders, 6 for right fielders - and zero for everything else. Thus, Fisk's points for his positions are [20 x (2226/2499)] + (27/2499) + [13 x (4/2499)] + [3 x (41/2499)] = 17.9 points. That gives Fisk a HOFS score of 52.3; Baseball Reference lists him with a HOFS of 49.1, so he actually does a little better with the credit for his games played while not behind the plate.
What about the player who really gets hosed by HOFS - Ernie Banks? BR has him listed at 46.1; using this method, he earns a solid 8.0 points: (1259/2528) + [13 x (69/2528)] + [12 x (1125/2258)] + [3 x (23/2528)]. His new HOFS score is 49.8.
As to what to do about relievers: this isn't as difficult as it may seem. First of all, with now a handful of relievers in the HOF, it's actually easier to determine what the standards are. Relief Pitchers are defined here as a pitcher who had fewer starts than relief appearances (or, more than half of his appearances were in relief). An additional requirement is that a pitcher must average less than 4.5 innings per appearance. If a pitcher meets these requirements, instead of awarding points for wins over 100 (as in rule 1 of the HOF Standards method), you do the following:
- Figure the pitcher's career relief points total: 3 x SV + 2 x W - 2 x L. (Because blown saves were not kept as an official statistic over the years, you can't really consider them in determining HOF Standards.)
- Divide the total by two.
- Award one point for every 10 points over 100, up to a limit of 25 points.
These new standards help Bruce Sutter immensely: he's listed with a 17.0 HOFS score on BR, but his adjusted score is now 48.4 with the new rules. Some other players under the new rules:
- Christy Mathewson, 87.2 (84.0 according to BR)
- Walther Johnson, 86.5 (82.0)
- Willie Mays, 84.5 (75)
- Babe Ruth, 84.0 (78.6; as a pitcher: 36.7 vs. 34.0)
- Cy Young, 82.7 (82.0)
- Barry Bonds, 77.7 (77.5)
- Roger Clemens, 75.5 (70.0)
- Wade Boggs, 64.1 (57.5)
- Cal Ripken, 63.1 (58.3)
- Roberto Alomar, 61.9 (56.8)
- Alex Rodriguez, 59.4 (54.1)
- Pete Rose, 59.4 (54.9)
- Dennis Eckerlsey, 56.9 (34.0)
- Bert Blyleven, 54.2 (50.0)
- Ryne Sandberg, 50.3 (42.7)
- Ron Santo, 50.0 (40.9)
- Lee Smith, 45.4 (13.0)
- Bill Mazeroski, 20.5 (15.9)
- Satchel Paige, 11.6 (0.0)
- Mike Munoz, 0.0 (0.0)