Book of the Month: Stat One
In talking about Stat One by Craig Messmer, I am really talking about an entire genre of books. Particularly in baseball, there is an entire class of books in which an author invents a superstat and proceeds to rank the greatest players ever to play the game. Most famous of these, because it also does so many other things, is Bill James’ The New Historical Baseball Abstract, in which Win Shares are integral to his player ratings. Stat One is representative of the genre at large, so I’d like to give it a bit of consideration here in January.
Messmer’s book came out in 2008, and his statistical invention fit into a long line of statistics based on the number of bases advanced divided by the number of outs made. Messmer’s version is called Offensive Production and Efficiency Average (P/E), and it varies from earlier stats like Tom Boswell’s total average because of its production measure. To compute it, you take [2(Net Runs) + Complete Bases]/Plate Appearances. Net Runs = Runs Scored + RBI – Home Runs. Complete Bases = Total Bases +BB + HBP + SB – Caught Stealing. Net Runs are Messmer’s production factor, complete bases are the measure of efficiency. As his complete bases factor tends to be about twice as large as the net runs factor, Messmer doubles net runs in order to make them roughly equal in his P/E average. At this point, we already have a serious flaw. The only justification of the doubling is to make the numbers look more even. It has no real logical grounding or any basis in the way baseball works. It simply makes the stat look better.
From laying out his stat, Messmer moves on to ranking the best players at each position. But he does not rank them by his stat. His stat, which is so boldly claimed to be stat one, is only a small part of his overall ranking. If Messmer does not trust his stat’s results, why should anyone else? Regardless, Messmer does interestingly break players down into 5 categories. The Category 5’s are the best ever to play, the Category 4’s are Hall of Famers a slight cut below. The Category 3’s are the Hall of Fame dividing line: Some make the Hall, some don’t, and the difference between the two probably has more to do with the voters than the players. Category 2’s played a while but were nothing special, and Category 1’s are the classic players just called up for a cup of coffee. This basic system I like. I think it captures a good way of looking at players. I wish he had better justifications of who fit in each category, but that does not challenge the basic insight.
Finally, what is the payoff? The best player ever is Babe Ruth. Shocking, I know, but it helps give the stat some grounding in the real world. Interestingly, Joe DiMaggio comes in third, just behind Lou Gehrig. This is one of the classic examples of Messmer ignoring his own stat. His stat places Ted Williams third and DiMaggio sixth. For some reason, Messmer argues for the opposite. If you happen to see his top 25 as ranked by his stat, they are almost all from high offense eras like the 1920’s, 1930’s, and 1990’s. Without any corrections for era, park, league, etc., his stat gives more misleading information than it should. Nevertheless, if you enjoy debates about the best players ever, Stat One is an entertaining read. Messmer writes well enough to hold a reader’s attention, and his work should spark some further thought. As part of its genre, it is distinctly one of the best.