Jason Kidd does not make his teammates better. Defense doesn’t necessarily win championships. Not only is Kobe Bryant not the next Jordan, he’s not even the best guard in the NBA.
All of these statements go against the conventional wisdom of basketball writers and fans across the NBA. But author John Hollander says he has the numbers to prove all these statements correct. With his new book Pro Basketball Prospectus, Hollander wants to change the way we look at basketball — in the same way that statistical analysis has transformed baseball.
It has been 25 years since a Kansas writer named Bill James put out a 68-page mimeographed packet called the Baseball Abstract and began a statistical revolution in the national pastime. It took a while for James’ idea to permeate baseball culture, but every time you see a mention of pitcher run support, park effects, or OPS, you are seeing the influence of James and those who followed him.
Keith Law, a writer for the analytical annual Baseball Prospectus, was hired as a consultant by the Toronto Blue Jays. Billy Beane, who runs the Oakland A’s with strategies based on the work of James and Baseball Prospectus, is now widely-acclaimed as baseball’s top executive. Even James has now joined the establishment, hired this month as “senior advisor, baseball operations” for the Boston Red Sox.
But for all the changes that James and his followers have wrought in the world of baseball, there has never been a similar revolution in any other sport. Now the publishers of Baseball Prospectus are trying a brand extension by enlisting Hollander, the Portland (now Atlanta) based writer of a basketball website called alleyoop.com. That website was the incubator for most of the basic ideas that are published in print for the first time in the Pro Basketball Prospectus.
Basketball is probably the sport that, next to baseball, keeps the most statistics. But basketball fans looking for statistical analysis have only found an occasional article focusing on some all-encompassing master statistic that claims to represent all of a player’s contributions without any proof that the formula actually reflects how much a player contributes towards wins.
Perhaps the biggest breakthrough Hollander has made is the idea of “pace.” In baseball, a major new concept of the statistical revolution was park effects. Each team gets 27 outs to score each game, but the size and altitude of the playing fields effect player statistics. Just ask any hitter who has played in Colorado.
Hollander realized the opposite is true in basketball: the playing field is always the same size, but the amount of possessions that give an opportunity to score changes depending on team strategy. A game with a fast-tempo team like Dallas will have many more possessions than a game with a slow-tempo team like Miami.
The few previous attempts at forming an all-in-one basketball statistic concentrated on how many points, rebounds, or assists a player got per minute, with 48 minutes in a game. But different teams play at different paces, providing more shot opportunities to both themselves and their opponents. Just watch a Dallas game and then a New York Knicks game to see the difference.
For example, conventional statistics show that Sacramento and Houston both had middle-of-the-pack defenses last season. Each gave up the same number of points per game, about 97. But according to Hollander’s statistics, Sacramento is the sixth best defense in the league, while Houston is the worst… because Sacramento gave up those 97 points in games with the most scoring opportunities of any team in the NBA, while Houston gave up those 97 points in games with the second-least scoring opportunities of any team in the NBA (only the unwatchable Miami Heat played at a slower tempo).
The book goes further in the articles that accompany each team section. Bill James has said that he began analyzing baseball to find out if conventional wisdom was true. Did a veteran catcher do a better job of running a pitching staff? Was there such a thing as a clutch player? Was sacrifice bunting a good idea? Pro Basketball Prospectus takes the same approach to tackle the similar questions about the NBA.
For example, last year we heard over and over that New Jersey turned things around because Jason Kidd makes his teammates better. So Hollander analyzes each of Jason Kidd’s teammates to see if they performed better after playing with Kidd. Then for good measure he measures the influence of Michael Jordan by checking his teammates before and after both of his comebacks, and he measures the influence of John Stockton by comparing every player who has ever been traded to Utah, before and after. The result shows that great players make their teams better because they are great, not because they have an effect on their new teammates.
Another issue addressed by Hollander is the canard that defense wins championships. It turns out that over the past 14 years, the average rank for an NBA champion in Offensive Efficiency was fifth — and the average rank in Defensive Efficiency was also fifth. Don’t like Hollander’s new-fangled stats? Take a look at field goal percentage and opponents field goal percentage back to 1970 and once again, the average rank of an NBA champion is fifth in both. Defense is no more important than offense.
Pro Basketball Prospectus poses a number of other basketball questions, and then tries to answer them. Does it make sense to take draft big men late in the lottery? Do players play better with consistent minutes? Are mid-career performance jumps a fluke? Fans of Jon Barry and Elden Campbell aren’t going to want to read that one.
Hollander combines this analysis with the same biting humor that has made Baseball Prospectus a good read as well as a statistical bible. Steve Francis was bothered all season by migraines, “perhaps induced by watching Glen Rice try to play defense.” Shawn Kemp has gone downhill “thanks to a unique training regimen that combined the worst vices of Henry VIII and William S. Burroughs.” About Keith Van Horn: “There have been guys with bad hair, and there have been guys with bad socks, but never has someone combined the two as sublimely as Van Horn has.”
It will take time for Hollander’s ideas to be accepted by NBA executives and fans alike, but he is often proven correct. Before the season, you might have disagreed with his assertion that Tracy McGrady, not Kobe Bryant, is the best guard in the NBA — and that the Lakers need Shaq much more than they need Kobe. Check the current NBA standings, however, and Hollander’s analysis proves prophetic.
It will be a long time before Hollander’s ideas create an NBA version of Billy Beane, but you can learn a lot more about the NBA right now with this first edition of the book. After all, don’t you want to know why the greatest rebounder of all time was not Bill Russell or Wilt Chamberlain, but Dennis Rodman?
A shorter version of this review appeared in the Boston Phoenix.Powered by Sidelines