Rich Lederer is the main voice behind Baseball Analysts, a blog which produces some interesting articles about baseball. There is no consistent tone, you might read an analysis of the National League East followed by a trip report from a tour of ballparks. (This blog can hardly be critical of that.) This is a blog I quite enjoy, and I should probably add it to my blogroll.
That said, I'm going to be a little hard on an article from last Tuesday in which Rich, using as a pivot point the election of Joe Gordon to the Baseball Hall of Fame, argues for the induction of Bobby Grich. Both played second base, and had pretty much comparable statistics. Rich argues that, if Gordon goes in, Grich, with slightly superior stats and little difference in fielding, deserves to go as well. (For my part, I'd have to think long and hard before putting either in the Hall, I consider both borderline cases.)
Rich goes through a comparison of the two, and conducts this study in a sabermetrically-correct way; that is to say, he takes context into account, understanding that counting numbers like hits and home runs have field and temporal effects, adjusting them using modern thinking. He looks at the tried-and-true rate stats like BA, OBP, and SLG, then looks at the adjusted OPS+.
He then takes defense into the analysis, fielding percentage and Gold Gloves figuring prominently. What follows is a look at some of the newer stats, Win Shares and WARP (Wins Above Replacement Player). In each case, Grich does a bit better than Gordon in both peak value and overall accomplishment.
Here's the problem: We read through this, and we become convinced that Grich probably was better than Gordon, and deserves more consideration for Cooperstown. The very weight of the evidence says so. But, the crux, a lot of the evidence is cross-correlated. Win Shares and WARP, as measures of the overall ballplayer, are going to come up with pretty similar results. All the offensive metrics mutually support one another - but they don't really add anything to the argument.
If I tell you that Bill Gates has more dollars than I do, we can conclude that he's richer. If I tell you that Bill Gates has more nickels than I do, that doesn't really tell us any more than the first statement.
It could be that Rich believes his readers are sensitive to the nuances of these statistics, that we know their strengths and weaknesses, so there is a distinction between the arguments. But I think that credits him a little too much; perhaps he's hoping to hit on each person's favorite stat to make his case. Either way, as I've seen in many such contentions, piling up more facts doesn't really make the case stronger.
That said, I'm going to be a little hard on an article from last Tuesday in which Rich, using as a pivot point the election of Joe Gordon to the Baseball Hall of Fame, argues for the induction of Bobby Grich. Both played second base, and had pretty much comparable statistics. Rich argues that, if Gordon goes in, Grich, with slightly superior stats and little difference in fielding, deserves to go as well. (For my part, I'd have to think long and hard before putting either in the Hall, I consider both borderline cases.)
Rich goes through a comparison of the two, and conducts this study in a sabermetrically-correct way; that is to say, he takes context into account, understanding that counting numbers like hits and home runs have field and temporal effects, adjusting them using modern thinking. He looks at the tried-and-true rate stats like BA, OBP, and SLG, then looks at the adjusted OPS+.
He then takes defense into the analysis, fielding percentage and Gold Gloves figuring prominently. What follows is a look at some of the newer stats, Win Shares and WARP (Wins Above Replacement Player). In each case, Grich does a bit better than Gordon in both peak value and overall accomplishment.
Here's the problem: We read through this, and we become convinced that Grich probably was better than Gordon, and deserves more consideration for Cooperstown. The very weight of the evidence says so. But, the crux, a lot of the evidence is cross-correlated. Win Shares and WARP, as measures of the overall ballplayer, are going to come up with pretty similar results. All the offensive metrics mutually support one another - but they don't really add anything to the argument.
If I tell you that Bill Gates has more dollars than I do, we can conclude that he's richer. If I tell you that Bill Gates has more nickels than I do, that doesn't really tell us any more than the first statement.
It could be that Rich believes his readers are sensitive to the nuances of these statistics, that we know their strengths and weaknesses, so there is a distinction between the arguments. But I think that credits him a little too much; perhaps he's hoping to hit on each person's favorite stat to make his case. Either way, as I've seen in many such contentions, piling up more facts doesn't really make the case stronger.
No comments:
Post a Comment