Stat Guy tracking
  •  What is NBA trak?
  •  PROFITS standings
  •  WAMMERS rankings

NBA data links
  •  Hoops Stats
  •  82 Games
  •  Doug's Stats

Scoreboard
  •  Previous scores
  •  Statistics
  •  Odds
  •  Teams
  •  Matchups
  •  Transactions
  •  NL standings
  •  NL schedule
  •  NL statistics
  •  NL rotations
  •  NL injuries
  •  NL previews
  •  NL game capsules
  •  AL standings
  •  AL schedule
  •  AL statistics
  •  AL rotations
  •  AL injuries
  •  AL previews
  •  AL game capsules
  •  Pitching report cards
  •  Probable pitchers
  •  Teams





Tuesday, December 20, 2005

NBA Trak, 12/20

Raptors 92, Magic 90
Magic 100 EFF - Raptors 87. Why? Eight more three-pointers. The great equalizer. Crappy game, nonetheless.
Celtics 109, Warriors 98
Warriors lose despite combined 74 points from Davis, Richardson and Murphy.
Bobcats 106, Kings 103, OT
When a pick'em game goes OT for the second-straight night, you got to like what your system is telling you.
Pistons 106, Grizzlies 104
You'd like the Grizzlies' chances to beat the Pistons if you knew they'd shoot 51.8% from the field but Memphis was done in by a 26-10 advantage at the foul line for Detroit.
Sonics 111, Wizards 101
Alonzo Jamison outscored 10-4 by Reggie Evans.

Updated rankings and predictions are here.

FULL DEFENSIVE RANKINGS ARE HERE
Yesterday, I finished incorporating on-court/off-court statistics into my defensive database. The information, which I am taking from www.82games.com, will be updated periodically but the latest update will always be included in the player rankings.

Again, I have taken a three-prong approach to address the issue of man-on-man defense as part of a player-rating system:

1. First, each player is assigned the team defensive rate (EFF per minute)
2. The team rate is then adjusted according to counterpart statistics. Each day, I enter player efficiencies from box scores and assign an opposing player as his counterpart. The system looks at the efficiency his counterpart posted and compares it to the expected efficiency based on season averages and minutes played. The difference, positive or negative, is that player's defensive efficiency for that game. The sum of all of these games are then compared to league averages and the team defensive rate is then adjusted upwards or downwards.
3. The player's team defensive rate is adjusted a second time using on-court/off-court statistics. If a team allows more points than its average when a player is on the floor, it has a negative impact on his rating. It the team allows fewer points, then the player's rating goes up.

So with those explanations out of the way, here are my initial man-on-man rankings:

DEFENSIVE MAN-ON-MAN RANKINGS

finley,michael

6.70

bowen,bruce

6.13

howard,juwan

5.68

stoudamire,damon

5.64

garnett,kevin

5.02

francis,steve

4.94

jackson,stephen

4.80

artest,ron

4.60

battier,shane

4.58

billups,chauncey

4.52


The name that will jump out to most people here is Damon Stoudemire. Clearly, Stoudemire, who has a terrible defensive reputation, benefits from the terrific team defense orchestrated by Czar Mike Fratello. But go to www.hoopsstats.com and look at the numbers posted by point guards against the Grizzlies. Stoudemire has been a part of that.

Of course, blocks, steals, fouls and defensive rebounds - all measurables - are components of a player's defensive value. These are and have been included in the TPR ratings, which were missing only individual defense to be truly comprehensive. But to rank the best and worst defensive players strictly for their all-around defensive contribution, defensive TPR needs to be combined with these plays made.

To accomplish this, I developed a formula which blends defensive TPR with the other categories weighted by value per possession, which is a method taken from John Hollinger's excellent Pro Basketball Forecast. With that, here are the best and worst defensive players in the NBA so far this season:

BEST DEFENDERS

WORST DEFENDERS

garnett,kevin

11.2

williams,marvin

-10.5

marion,shawn

10.6

arenas,gilbert

-6.1

artest,ron

9.8

bibby,mike

-5.5

ming,yao

9.3

james,mike

-5.0

gasol,pau

9.1

graham,joey

-4.7

bowen,bruce

9.1

turkoglu,hedo

-3.5

duncan,tim

9.0

simmons,bobby

-2.6

battier,shane

8.8

jones,damon

-2.4

finley,michael

8.7

davis,ricky

-2.1

francis,steve

8.7

stojakovic,peja

-1.8

wade,dwyane

8.4

rose,jalen

-1.8

howard,juwan

8.4

calderon,jose

-1.8

jackson,stephen

8.1

ridnour,luke

-1.6

camby,marcus

8.1

dixon,juan

-1.5

stoudamire,damon

7.9

brezec,primoz

-1.4

mcgrady,tracy

7.8

childress,josh

-1.3

mourning,alonzo

7.4

bosh,chris

-1.2

o'neal,jermaine

7.3

daniels,antonio

-0.8

wallace,rasheed

7.2

jeffries,jared

-0.8

kirilenko,andrei

7.2

lewis,rashard

-0.6


I didn't develop this system until around Thanksgiving. Many early-season box scores have not been entered. Thus, there is a fairly small sample size in the counterpart database. But the anamolies (Ben Wallace has very poor counterpart statistics) should smooth out as the database grows.

0 Comments:

Post a Comment

<< Home