When Bill James' "Baseball Abstract" gave us reason to think about on base percentage and game-winning hits, Al McGuire was still coaching at Marquette and Magic Johnson was a high school senior. A generation later, a handful of NBA teams have hired or contracted stat wizards whose skills would make Billy Bean blush.
One of these is recent Nuggets hire Dean Oliver, formerly affiliated with the Seattle SuperSonics. Oliver, a member of the Association for Professional Basketball Research (basketball's equivalent of baseball's SABR), is the author of "Basketball on Paper." We interviewed him about the use of statistical analysis in the NBA.
SC: How many NBA teams employ a person such as yourself, either full time or as a consultant?
Dean Oliver: A handful. There are about five teams with people who do what I do. It involves pretty serious statistical analysis, and very few are qualified. Houston is the easiest example, in that they have a few quantitative analysts on staff. Everyone in the NBA in this field is pretty successful. It is a tough position to fill. It's not just about producing spreadsheets and information. A person must be both knowledgeable about the game, and technically competent, because the statistics must tell a clear story. Thus, it's about translation — one can't simply have a computer science background, or a basketball background, but both.
SC: Did this concept begin with Bill Bertka and his methods?
DO: Actually, it began with Dean Smith. His teams used statistical analysis of his players in the 1960s, and the 1950s. Bertka was my mentor as far as scouting is concerned. He used a detailed approach, and was tremendously helpful to me in that aspect. Bill didn't do too much innovation on the statistical side, but he was appreciative of how to manipulate the info and cull a good story from it. It's ultimately about how to help your team win, how do we beat the other guys? I worked for Bertka at Bertka Views. He was a very fine basketball man and a fine person.
SC: Since the discipline requires both a basketball and computer background, how will those in your profession be developed?
DO: I wish I knew. It's a hard balance. It's an easy balance for me because I grew up thinking about sports, and math and science. If I wrote a book report in school, I wanted to do it on something in Sports Illustrated. So I've had these things in my head my whole life.
There's this stereotype that those who play sports can't be doing well academically, and that math-science people can't do well in sports, but I knew plenty of athletes who didn't fit this example. A lot of people with a passion for sports realize they can't be professional athletes because it demands tremendous skill, but do not wish to allow their passion to go unfulfilled, may enter this field.
SC: How many of the players, if any, realize quantitative analysis is taking place?
DO: When I communicate with them, I don't talk about what I do. I don't think may are aware of it.
SC: What is your day like, say between games?
DO: Well, there are two areas, the coaching responsibility and the personnel responsibility. I use numbers to help up prepare for the next game, or next series of games. These responsibilities are not necessarily daily. It could involve detailed information prepared about the opposition between games, and personnel issues, trade issues, players on our radar.
For both aspects, I have to be sure that on the computer side I'm getting the information in the right format, and simplified. A lot of time is devoted to translating data into the right format, and checking it — making sure names are consistent, the difficult foreign names, and that the data is consistent. I look at how the data rise and fall, and what's happening with us. The status of the team also dictates what is worked on.
SC: Do you use the same, or different approaches, to evaluate players in high school, college, Europe, and the NBDL? Take a stat like turnovers, in college ball.
DO: Well, with the NCAA, we would look at how many of the turnovers were forced and how, and the degrees of turnovers, how much the stat varies. I go over numbers with an extremely fine-toothed comb, a lot more fine-toothed than most people would really ever want to think about. There is nothing simple about it, and I've been at it for 20 years. Now in Europe, a lot of players don't play enough for their stats to be meaningful, so we can't use them. We try to overcome this. When a guy plays a little, it's easier.
SC: Where does the software come from?
DO: There is no software for quantitative analysis of basketball. I wish there was. People have asked me about designing some. It is so much an interdisciplinary methodology, interweaving basketball, math, and science. There are a lot of pieces to it.
SC: So everything you do is in-house?
DO: Yes. There are a few basketball programs, a few different tools that allow for the collection of data, and they organize the data very nicely. But there is no software for analysis, for gleaning a story from the data.
SC: No predictive software?
DO: No. The words have to tell a story, to be translated from the data. The bar is set very high in his area.
SC: Do you feel you have equal tools as those who do this in the baseball world?
DO: I have read in the past about how basketball analysis is not as advanced as in baseball. I think we're closer than baseball's quantitative analysyts realize. Quantitative analysis of basketball is far more complex than in baseball. Baseball is primarily a question of the matchup of pitcher vs. batter. It is essentially about "can this guy hit, and which pitchers can he hit?" There is some data about fielding, and situations, but nothing like there is in basketball, which is much more of a team game. We look at how an individual's play affects our team, and how the numbers fit into winning games.
Baseball has so many more years of available statistics — we will never have that. We can't analyze games from the 1940s, because complete stats were not kept, and so many statistics (i.e. steals, blocked shots) were not tracked — some of the most important ones. But I'm grateful for the development that occurred in baseball, books like "Moneyball" helped explain the need for what I do in basketball. We'll never get as good as baseball with analysis of the past, but I feel on equal footing regarding present play.
I look at methods I used in the early-'90s, when I really thought that I was using innovative techniques, and where back then the numbers could tell a story that would be a magazine article, today I can use numbers to tell a much larger story, "a novel."
SC: Thank you for speaking with us, Dean.
DO: Thank you.
Leave a Comment