The Sonics Play Moneyball: Part Two - The Outsider
Kevin Pelton, SUPERSONICS.COM | January 27, 2005
What would you do to follow your passion? What risks would you take? What sacrifices would you make?

This is the second installment of a four-part series detailing how the Sonics have used statistical analysis to their advantage. If you missed them, check out the other three installments:
  • Part One
  • Part Three: The Front Office
  • Part Four: The Coaching Staff

    Also see the following associated articles:
  • Statistical Analysis Primer
  • Dean Oliver's Individual Ratings
  • Dean Oliver's Four Factors
  • Dean Oliver quit his job.

    At the age of 34, Oliver gave up his comfortable work as an environmental engineer for Environ in order to try to make a living out of what had formerly largely been a hobby, statistical analysis of the NBA. But Oliver's journey to the NBA began long before he went out looking for a new job in January 2004.

    For the entirety of his adult life through quitting his job, Oliver had balanced work and basketball, taking advantage of his love and aptitude for math and science in both work and play.

    While an undergrad at Caltech, he played point guard and later served as an assistant coach. Division I experience it was not, but the experience would prove invaluable when Oliver began working with basketball lifers. It was while at Caltech that Oliver developed many of his methods, beginning from his critical discovery - the importance of possessions to evaluating teams who play at different speeds on a level playing field and determining the value of various events on the court.

    Oliver described the balancing act best in the introduction to the unpublished NBA annual, Basketball Hoopla, he wrote at the tender age of 19:

    "There were many physics quizzes that had notes written on them like: 'Is the three-pointer hurting the offenses of the NBA? Check development since '80.' In the back of one of those blue test booklets used to take a math final, there are workings of a study to determine why Kareem Abdul-Jabbar has stayed in the league the past five years despite low rebound and blocked shot totals."

    Oliver earned his PhD at North Carolina, the perfect place for a basketball junkie to further his education on and off the court. Oliver met legendary Tar Heels coach Dean Smith - who, he'd already learned, had been using possessions to evaluate his team for decades - and worked as an area scout for the college scouting service operated by Lakers assistant coach Bill Bertka.

    Oliver has achieved his dream of working for an NBA team.
    courtesy Dean Oliver
    After Oliver settled into his career as an environmental engineer, he could hardly shake basketball. So he wrote about it online, at and his personal Web site, the Journal of Basketball Studies. He consulted for several teams, most notably and most frequently the Sonics. He helped a budding statistical analysis community by generously sharing his knowledge in a basketball statistics discussion group online.

    Oliver enjoyed his work. It was fulfilling and useful. But it wasn't basketball.

    So he quit.

    "I talked to a lot of people who had started their own business and friends who had done some crazy things - they convinced me that taking the chance is worth it and, even if things went horribly, life wouldn't be over," Oliver says. "It was a long decision, but easy to make after I gathered all the information."

    The NBA's Moneyball movement still being in its infancy, Oliver couldn't exactly walk into an NBA front office and demand a job. Getting permanent employment would require meeting a lot of people and selling himself, a process made easier by the fact that, in the fall of 2003, he published his first book, Basketball On Paper - something Oliver refers to as "a giant business card."

    Oliver attended the NBA's Pre-Draft Camp in Chicago, meeting with several executives around the league. (While also noticing a Georgia alumnus named Damien Wilkins whom he would mention to the Sonics as someone worth signing for their summer-league team.) He did the same at the Long Beach and Rocky Mountain Revue summer leagues.

    There was plenty of interest from NBA teams, but far less money. In the end, all roads led to the Sonics, the team Oliver most closely identified with because of his consulting work and his mother's and step-father's residence in the Seattle area.

    The Friday before the Sonics began their 2004 training camp, when they were finalizing deals for training-camp invitees, they inked a 34-year-old free agent, a move that would go unreported in the agate type of the newspaper but stood the potential of having more long-term impact than anyone else they could have signed.

    That explains how Dean Oliver got here, to a position with Seattle the team terms an "experiment". (The Sonics and Oliver agreed to a one-year position as a paid consultant with the understanding that, if things went well, Oliver would ascend to a more formal position in the future.) What it doesn't explain is how Dean Oliver became Dean Oliver, and why he would want to work for an NBA team at all.

    After all, the comparison that often pops up in reference to Oliver from his fellow NBA analysts is the "Bill James of basketball." It's an accurate description to the extent that both are figureheads for the statistical revolution in their respective sports, but the similarities break down after that. James considered himself a writer who used statistics in his work; Oliver generally considers himself an analyst who writes to spread his ideas. James didn't go to work for a baseball team (the Boston Red Sox) until he'd been an icon amongst fans for nearly two decades; Oliver quit his job so he could go work for a basketball team.

    Ironically, it was James' own struggle to become an insider that convinced Oliver. While James came to relish his role as outsider, using it to his advantage by exposing the many mistakes made by baseball's front offices, Oliver took a look at what happened to James and decided he'd do whatever he could to get inside a front office.

    Oliver wasn't the only impressionable basketball fan who read James' Abstracts in the 1980s and wondered whether the same logic could be applied to basketball; far from it. With varying success, people like Bob Bellotti, Dave Heeren and Martin Manley created player-rating formulas, wrote books and were touted as the Bill James of Basketball. None of them can claim, however, to have matched Oliver's longevity or influence.

    The reason for that, it could be argued, is that Oliver realized very early on in his dorm-room calculations that the linear weights formulas favored by most basketball analysts even now just didn't work as well as James' Runs Created, amongst others, had summed up baseball (at least hitting). How much of that was Oliver's practical experience playing, coaching and scouting the game?

    "As a coach and a scout, I didnít care who was the best," says Oliver. "I wanted to know what plays I could run with certain players and when, what players could do to become better, what their strengths were and how to enhance those strengths. Linear weights addressed none of this."

    Oliver went further than his peers dared. Like James, he decided to rate players the same way he rated teams. In baseball, that meant runs - hence, runs created. In basketball, it meant points per possession. To call Oliver's work simple would be a misstatement; his offensive rating formula alone takes up seven pages in the appendix of Basketball On Paper. The logic behind the ratings, however, was simple: How much of his team's offense or defense - measured in possessions - was a player responsible for? How efficient was he with those possessions? (More on Oliver's Individual Ratings)

    "From James' work, I quickly did a lot of baseball stuff, recalculating things he did to get the concepts in my head," says Oliver. "I started work on basketball pretty seriously in late '84 or early '85, just trying to adapt James' methods. It was in late '86/early '87 that I broke from trying to adapt his work and took a completely different approach. That approach is what took me where I am now."

    Even in that framework, Oliver always viewed rating players as more of a necessary evil than a starting point. More than any of his peers, Oliver has considered coaching issues, helping him work with the Sonics coaching staff on a more subtle and profound level than saying, "Smith is good. He should be playing more."

    At its heart, Oliver's work, like James', is founded on a simple premise. How do teams win? He most accurately addressed this point with the creation of the Four Factors system, which breaks down offense and defense into the four most critical components - shooting, rebounding, free throws and turnovers. (More on the Four Factors)

    "It was impossible to ape James," Michael Lewis wrote in Moneyball. "The whole point of James was: don't be an ape! Think for yourself along rational lines."

    Dean Oliver is no ape, a fact further proven in 2004 when James endorsed the paperback edition of Basketball On Paper.

    "There are a lot of math guys who just rush from the numbers to the conclusion," said James. "They'll tell you that Shaq is a real good player but his team would win a couple more games a year if he could hit a free throw. Dean is more than that; he's really struggling to understand the actual problem, rather than the statistical after-image of it. I learn a lot by reading him."

    It's now more than a year since Dean Oliver quit his job to take a chance on himself. There were tough moments along the way, but he wouldn't change a thing.

    "These days I am working so hard to do studies on holes in our team, on improved defensive measurement, on potential draftees, or on our next opponent that I don't really think about how I got here," says Oliver. "But I see my friends and they - the ones who really helped me feel comfortable with the decision - tell me that they're awed that I did it. After a brief moment of fear while thinking, "You're surprised and you helped convince me to do it?!?", I do then look back at the last year and realize, wow, I did it."

    This is the second installment of a multi-part series detailing how the Sonics have used statistical analysis. Next Wednesday, we take a look at how the Sonics front office has blended new statistics with traditional basketball management.