December 4, 2010

New NBA Ratings Dec 4

Filed under: Uncategorized — wwinston @ 11:18 am
Rank Team Off Def Overall
1 Boston Celtics 2.32 -5.80 8.11
2 Dallas Mavericks -1.38 -8.51 7.13
3 Miami Heat 2.34 -4.57 6.91
4 San Antonio Spurs 5.29 -1.46 6.75
9 Los Angeles Lakers 5.06 -1.31 6.37
6 Orlando Magic -1.78 -7.53 5.75
5 Utah Jazz 2.12 -3.41 5.53
7 New Orleans Hornets -3.17 -7.46 4.29
8 Denver Nuggets 6.27 2.98 3.29
15 Houston Rockets 4.50 2.52 1.98
11 Chicago Bulls -1.73 -3.00 1.27
10 Atlanta Hawks -0.42 -1.56 1.14
14 Portland Trail Blazers -3.78 -4.56 0.78
17 Phoenix Suns 10.27 9.75 0.52
16 Toronto Raptors 2.35 1.97 0.38
13 Charlotte Bobcats -4.71 -4.92 0.21
22 New York Knickerbockers 8.56 9.11 -0.56
20 Indiana Pacers -1.42 -0.75 -0.67
19 Milwaukee Bucks -7.22 -6.24 -0.98
12 Oklahoma City Thunder 5.22 6.23 -1.01
18 Memphis Grizzlies 1.27 3.27 -2.00
21 Philadelphia 76ers -3.67 -1.36 -2.30
23 New Jersey Nets -4.68 -0.46 -4.22
26 Washington Wizards -1.48 3.54 -5.01
25 Los Angeles Clippers -1.78 3.54 -5.32
24 Detroit Pistons -4.24 1.54 -5.78
28 Golden State Warriors -1.73 4.68 -6.41
27 Cleveland Cavaliers -4.78 2.18 -6.96
29 Minnesota Timberwolves 1.58 9.47 -7.89
30 Sacramento Kings -9.19 2.08 -11.28

6 Comments »

  1. Wayne,

    I realize I’ve been asking you a lot of questions on this, but I’m going to press my luck and ask another because I can’t find anyone that can give me the answer to this.

    It appears that 1 point in positive scoring differential is equal to approximately 3 extra wins over the course of a season.

    The problem I am having is that formula seems to break down badly at the extremes.

    For example, the worst team in the league is the Kings. They are -11.28. If they were playing on the road, that would be an extra 3.5 or -14.78.

    3 * 14.78 is more than 41 games which suggests they would have a zero percent chance of beating an average team on the road. Either something is amiss in my thinking or something is amiss in the value (or both).

    It appears the value of a point may start at 3, but slowly diminish as the probability approaches 100%. Unfortunately, I have no idea how to calculate what the values might be (at least approximately).

    Comment by Italian Stallion — December 6, 2010 @ 12:17 am

  2. Since you used to work for the Mavs, I was hoping you could tell me why they keep playing JJ Barea. There is not a single stat in the universe that ranks him anywhere above “terrible,” yet he continues to get almost 20 MPG. I assumed Cuban and company were fairly “progressive” and into statistical studies, so is there a reason Dallas continues to give Barea minutes when he ranks so poorly in every single category?

    Comment by JJB Hater — December 7, 2010 @ 6:34 pm

  3. Who else do they have? He is not perfect but the three guard lineup does well.

    Comment by wwinston — December 7, 2010 @ 11:32 pm

  4. Your point is well taken. 1 pt = 3 wins is for predicting season performance. If a team is say a 20 points favorite then there chance of losing can be computed in Excel by formula =NORMDIST(0,20,12,True), because margin of victory for favorite is normal with mean = 20 and sigma =12 in NBA (14 in NFL) . For more infor see my book Mathletics

    Comment by wwinston — December 7, 2010 @ 11:35 pm

  5. Thank you Wayne.

    I hope at some point you elaborate on the calculations you are doing to create these ratings (if you haven’t already) so I can update them daily if I need one.

    Comment by Italian Stallion — December 8, 2010 @ 1:55 am

  6. Again a combo of least squares and a “secret sauce”.

    Comment by wwinston — December 16, 2010 @ 5:52 pm

RSS feed for comments on this post. TrackBack URL

Leave a comment

Powered by WordPress