Monday, November 10, 2008

NBA Stat Geek

So, there's no secret to the fact that I'm a math geek. There's also no secret to the fact that I'm a big sports guy. And, there's no secret that there's a human nature to attempt to combine the things you like.

Having said that, after a good bit of work on it, I've developed my own mathematical formula for ranking NBA teams. I'll leave the specifics of how I formulate scores to myself, but I'll share a little about how I wanted to develop them:

  • reward team that can impose their will during stretches of a game on both offense and defense
  • punish teams that get dominated during stretches of a game on both offense and defense
  • give teams credit for winning games
  • punish teams for losing games they should have won


  • And, that's it. I debated a few other factors, like margin of victory, or superstar presence, but decided those would be less unbiased and less helpful in predicting the best teams in the league. I'm hoping to keep up with it, and keep publishing my rankings periodically.

    So, here they are, my first rankings, to share with you:

    Rank | Team | score
    1. Boston 2.72
    2. Utah 2.37
    3. LA Lakers 2.2
    4. Atlanta 1.87
    5. New York 1.87
    6. Houston 1.78
    7. Phoenix 1.26
    8. Orlando 1.21
    9. Toronto 1.21
    10. Miami 1.04
    11. Indiana 0.94
    12. Cleveland 0.92
    13. Detroit 0.88
    14. Denver 0.71
    15. Chicago 0.64
    16. Golden State 0.5
    17. Portland 0.38
    18. Minnesota 0.38
    19. New Jersey 0.28
    20. San Antonio 0.08
    21. Philadelphia -0.12
    22. Sacramento -0.36
    23. New Orleans -0.45
    24. Dallas -0.45
    25. Charlotte -0.45
    26. Milwaukee -0.69
    27. Memphis -0.69
    28. Washington -0.78
    29. LA Clippers -1.64
    30. Oklahoma City -2.61

    No comments: