A friend asked me to comment on this article, which was published by Michael Lewis in the New York Times. It is related to this article, published on by Howard Bryant on ESPN.com.
The first article discusses how teams use statistics to assess performances. There are definitely methods to evaluate effectiveness beyond points, rebounds, steals, blocks, and assists. Plus/Minus, Player Efficiency Ratings, Points-Created, and Win-Shares all offer a more balanced approach. Billy Beane decided to eschew batting average in favour of on-base plus slugging percentage and other statistics, which was catalogued in Michael Lewis’ book Moneyball.
Given the success that the Oakland Athletics experienced with a limited budget, Beane and Lewis experienced sudden fame. The term Moneyball became part of the vocabulary of sport to describe any situation where a team relies heavily on statistics. Other General Managers refused to deal with Beane because they felt that he knew something which they didn’t. A film was in production.
Now, the public perception of Moneyball is different and the film has been cancelled. First of all, it will likely be revealed that the Oakland clubhouse from 1999 to 2003 was the one of the most juiced in the majors. Drafting low-ceiling college players pays off when they increase their potential artificially. Secondly, even during their peak, they never won a playoff series (maybe if Jason Giambi had slid into home plate). Thirdly, after the initial crop of stars left, Beane has been unable to replace them.
Is an overreliance on statistics to blame? The Boston Red Sox have won two World Series since 2004 and they also use statistics. However, they also draft elite talent to stock their minor league system who can be called up (Dustin Pedroia, Jacoby Ellesbury) or traded to acquire needed players (Josh Beckett, Victor Martinez).
But like the balance between quals and quants in financial analysis, there is not a single correct solution. A junior gold company may have a great chart but not enough gold in its mines. As New York Yankees General Manager Brian Cashman said: “You need someone with a high degree of common sense that surrounds himself with a strong team to run an efficient business. You have to have a blend, and he’ll gravitate toward the best possible solutions.”
Inbreeding of ideas and conservatism are to blame. Taking one idea to an extreme is problematic. According to the Howard Bryant article, some believe that the Moneyball emphasis on walks leads hitters to become more concerned with seeing pitches than hitting pitches. It’s like if a coach told Allen Iverson that he had to pass first because that’s what point guards do.
Red Auerbach was one of the most open-minded coaches ever. Auerbach didn’t care how Bob Cousy made the pass, as long as it got there. He never discussed statistics with Bill Russell, only winning percentage. Those were the best approaches for those particular players.
Michael Lewis wrote about the Houston Rockets and their use of statistics. According to the Houston front office, Shane Battier is one of the most effective players in the Association in terms of how he plays defence and contributes to the team offence. Trevor Ariza is a good signing and I am sure that his high effective shooting percentage and tipped passes were part of the reason that he was signed. However, the Rockets will go nowhere without someone who can take over the game, since it appears that Yao Ming and Tracy McGrady will not play much next year.
When you are allocating the mid-level exception or signing complimentary pieces, Moneyball is a good idea. But eventually, they will need a player who can create his own shot. The Triangle Offence worked well for the Los Angeles Lakers’ second unit but in the playoffs, the team had to relied on the weak-side two-man game or isolations for Kobe Bryant to score.
Using Moneyball in basketball is not a new idea. I’m not sure if the Oklahoma Thunder still employ the idea under Sam Presti but they are in a good situation with a promising young roster and plenty of cap room. Still, they needed to draft an elite player such as Kevin Durant to lead their team.
Any coach should define their own set of statistics that suit their systems and the skill set of their players. Aside from statistics which track points created and chemistry among different combinations of players, these criteria should be revised and updated with each season. After the season, there should be some sort of evaluation to determine if the outcome validates the process.
Interesting stats for basketball include: fast-breaks initiated, moving the ball to the open man, effective shooting percentage, turnovers caused by dribbling or passing in a rushed manner due to overdribbling, ball reversal speed, forcing a pass on an odd-person transition, (1) forcing the dribble and (2) forcing the dribble to the left, tipped passes, help and rotation speed, hedging/shocking the ball screen, boxing out, and low shot-clock situations.
Not all of these situations can be recorded with hard data (such as a shot chart); others are more subjective (like defensive communication). To play up-tempo basketball, there are decision making criteria to watch (initiating the fast break, moving the ball to an open man) but they are mood if the player cannot execute basic skills at high speed. The A’s have guys who can take pitches but not necessarily the bat speed or the eye to hit a cutter into the gap with runners on base and two outs. They reached they have reached their ceiling for 2009 and their record and attendance ($) shows.
The goal is not to do the best you can with less but be the best. The Oakland Athletics and Houston Rockets are not the models to follow but the Boston Red Sox and the Los Angeles Lakers.