You Are Here > Basketball-Reference.com > BBR Blog > NBA and College Basketball Analysis

SITE NEWS: We are moving all of our site and company news into a single blog for Sports-Reference.com. We'll tag all Basketball-Reference content, so you can quickly and easily find the content you want.

Also, our existing Basketball-Reference blog rss feed will be redirected to the new site's feed.

Basketball-Reference.com // Sports Reference

For more from Neil, check out his new work at BasketballProspectus.com.

Which Teams Are Allocating Their Possessions Efficiently?

Posted by Neil Paine on January 3, 2011

Among the players in their most common lineup, which teams divvy up possession usage most efficiently?

To answer that question, let's use a method I introduced here. Just like that old post, this one is going to lean heavily on the concept of "skill curves", which says that a player's offensive efficiency drops as he shoulders more and more of a team's possessions. I realize this isn't always the case for all players -- but as a very general rule it holds, so let's pretend for a moment that this simple model does in fact explain the fundamental usage-efficiency tradeoff in basketball. Under those rules, a player using 18% (or fewer) of team possessions while on the court would see his efficiency change by 1.65 points of offensive rating for every 1% change in usage, a player using 18-23% would see a change of 1.24 pts of ORtg for every 1% of usage change, and a player using 23% or more would see a change of 0.82 pts per 1% change in usage.

To find every team's most common lineup, I gathered data from 82games.com, and scaled the sum of those players' season-long possession usages to equal 100%. I found their predicted lineup efficiency based on actual ORtgs and usage patterns, and also found the optimal distribution of possessions that would maximize offensive efficiency according to the rules above. The teams with the smallest difference between their actual usage pattern and the optimal pattern can be considered to be efficiently allocating their possessions.

Here are the teams, sorted by the squared difference between their actual and optimal usage patterns:

2010-11 Stats In Lineup Optimized
Pos Player Tm Type G Min Poss PProd ORtg %Pos tradeoff usg1 eff1 usg2 eff2 diff^2
1 Darren Collison IND mid 29 841 378.9 395.3 104.3 22.9 1.24 21.3 106.2 20.3 107.5
2 Mike Dunleavy IND lo 31 922 304.8 336.2 110.3 16.8 1.65 15.7 112.1 16.9 110.1
3 Danny Granger IND hi 31 1139 595.9 600.9 100.8 26.5 0.82 24.8 102.3 24.5 102.5
4 Josh McRoberts IND lo 30 656 204.1 217.8 106.7 15.8 1.65 14.7 108.4 15.3 107.5
5 Roy Hibbert IND hi 32 918 455.0 452.7 99.5 25.1 0.82 23.5 100.9 23.0 101.3
100.0 105.2 100.0 105.3 3.1
Pos Player Tm Type G Min Poss PProd ORtg %Pos tradeoff usg1 eff1 usg2 eff2 diff^2
1 Carlos Arroyo MIA lo 34 767 201.0 232.8 115.8 13.8 1.65 12.0 118.9 13.5 116.3
2 Dwyane Wade MIA hi 33 1195 703.0 791.5 112.6 31.0 0.82 26.9 116.0 26.8 116.1
3 LeBron James MIA hi 35 1312 789.5 892.6 113.1 31.7 0.82 27.5 116.5 27.4 116.6
4 Chris Bosh MIA mid 35 1240 539.8 617.8 114.5 22.9 1.24 19.9 118.2 19.7 118.4
5 Zydrunas Ilgauskas MIA lo 35 623 186.2 204.1 109.6 15.7 1.65 13.7 113.1 12.6 114.8
100.0 116.5 100.0 116.6 3.6
Pos Player Tm Type G Min Poss PProd ORtg %Pos tradeoff usg1 eff1 usg2 eff2 diff^2
1 Brandon Jennings MIL hi 25 887 441.7 454.6 102.9 26.4 0.82 25.7 103.5 27.1 102.4
2 John Salmons MIL mid 30 1060 408.0 402.8 98.7 20.4 1.24 19.9 99.4 17.7 102.0
3 Luc Mbah a Moute MIL lo 30 754 173.7 174.6 100.5 12.2 1.65 11.9 101.1 12.3 100.4
4 Drew Gooden MIL hi 20 492 215.5 217.3 100.8 23.2 0.82 22.6 101.4 24.2 100.0
5 Andrew Bogut MIL mid 25 890 341.9 345.0 100.9 20.4 1.24 19.9 101.6 18.6 103.1
100.0 101.5 100.0 101.6 10.6
Pos Player Tm Type G Min Poss PProd ORtg %Pos tradeoff usg1 eff1 usg2 eff2 diff^2
1 Rodney Stuckey DET hi 29 940 451.7 500.4 110.8 25.8 0.82 27.1 109.8 29.7 107.6
2 Richard Hamilton DET hi 30 796 376.9 389.4 103.3 25.5 0.82 26.7 102.3 25.0 103.7
3 Tayshaun Prince DET mid 33 1067 406.6 445.0 109.4 20.5 1.24 21.5 108.2 20.9 108.9
4 Jason Maxiell DET lo 27 504 125.5 127.3 101.4 13.4 1.65 14.0 100.4 12.3 103.3
5 Ben Wallace DET lo 32 777 149.4 158.3 106.0 10.3 1.65 10.8 105.2 12.1 103.0
100.0 105.7 100.0 105.8 15.0
Pos Player Tm Type G Min Poss PProd ORtg %Pos tradeoff usg1 eff1 usg2 eff2 diff^2
1 Stephen Curry GSW hi 25 831 424.1 472.9 111.5 25.8 0.82 24.2 112.8 26.9 110.6
2 Monta Ellis GSW hi 33 1350 735.3 799.0 108.7 27.5 0.82 25.9 110.0 26.1 109.9
3 Dorell Wright GSW mid 33 1285 464.6 503.2 108.3 18.3 1.24 17.2 109.6 17.2 109.6
4 David Lee GSW mid 25 924 391.9 408.4 104.2 21.4 1.24 20.1 105.8 17.1 109.5
5 Andris Biedrins GSW lo 25 695 183.9 199.2 108.3 13.4 1.65 12.6 109.6 12.7 109.4
100.0 109.7 100.0 109.9 16.4
Pos Player Tm Type G Min Poss PProd ORtg %Pos tradeoff usg1 eff1 usg2 eff2 diff^2
1 Rajon Rondo BOS mid 21 804 343.3 369.8 107.7 22.6 1.24 21.4 109.1 19.6 111.4
2 Ray Allen BOS mid 32 1174 438.3 502.2 114.6 19.7 1.24 18.7 115.8 21.0 113.1
3 Paul Pierce BOS mid 32 1138 484.3 563.4 116.3 22.5 1.24 21.3 117.8 23.0 115.7
4 Kevin Garnett BOS mid 30 949 374.4 418.2 111.7 20.9 1.24 19.8 113.0 20.3 112.3
5 Glen Davis BOS mid 32 965 360.0 369.1 102.5 19.7 1.24 18.7 103.8 16.1 107.0
100.0 112.0 100.0 112.2 18.5
Pos Player Tm Type G Min Poss PProd ORtg %Pos tradeoff usg1 eff1 usg2 eff2 diff^2
1 Mike Bibby ATL lo 36 1069 319.1 361.8 113.4 15.9 1.65 15.5 114.0 15.7 113.7
2 Joe Johnson ATL hi 27 983 470.0 493.5 105.0 25.4 0.82 24.9 105.5 23.1 107.0
3 Marvin Williams ATL lo 30 899 268.7 304.7 113.4 15.9 1.65 15.5 114.0 15.7 113.8
4 Josh Smith ATL hi 36 1229 557.7 577.6 103.6 24.1 0.82 23.6 104.0 21.6 105.7
5 Al Horford ATL mid 36 1222 481.2 584.5 121.5 21.0 1.24 20.5 122.1 24.0 117.7
100.0 111.2 100.0 111.4 20.1
Pos Player Tm Type G Min Poss PProd ORtg %Pos tradeoff usg1 eff1 usg2 eff2 diff^2
1 Andre Miller POR mid 33 1078 456.1 505.4 110.8 22.9 1.24 22.8 111.0 21.9 112.1
2 Brandon Roy POR mid 23 813 336.2 359.0 106.8 22.4 1.24 22.2 107.0 20.0 109.7
3 Nicolas Batum POR lo 33 926 296.2 330.9 111.7 17.3 1.65 17.2 111.9 16.8 112.6
4 LaMarcus Aldridge POR hi 34 1304 574.0 635.7 110.7 23.8 0.82 23.7 110.9 27.5 107.7
5 Marcus Camby POR lo 31 876 230.3 246.6 107.1 14.2 1.65 14.1 107.2 13.8 107.8
100.0 109.7 100.0 109.9 21.0
Pos Player Tm Type G Min Poss PProd ORtg %Pos tradeoff usg1 eff1 usg2 eff2 diff^2
1 Russell Westbrook OKC hi 35 1243 786.9 847.5 107.7 32.6 0.82 30.6 109.4 26.9 112.4
2 Thabo Sefolosha OKC lo 35 1035 198.6 218.4 110.0 9.9 1.65 9.3 111.0 10.9 108.2
3 Kevin Durant OKC hi 31 1206 697.0 782.1 112.2 29.7 0.82 27.9 113.7 28.2 113.5
4 Jeff Green OKC mid 28 1057 386.8 422.4 109.2 18.8 1.24 17.7 110.7 17.1 111.4
5 Serge Ibaka OKC lo 35 910 275.2 331.0 120.3 15.6 1.65 14.6 121.8 16.9 118.1
100.0 112.8 100.0 113.0 21.9
Pos Player Tm Type G Min Poss PProd ORtg %Pos tradeoff usg1 eff1 usg2 eff2 diff^2
1 Derrick Rose CHI hi 31 1181 718.0 782.4 109.0 31.5 0.82 31.5 108.9 31.4 109.0
2 Keith Bogans CHI lo 32 586 124.7 123.1 98.7 11.0 1.65 11.0 98.7 10.2 100.0
3 Luol Deng CHI mid 32 1246 483.8 523.0 108.1 20.1 1.24 20.1 108.1 20.1 108.1
4 Taj Gibson CHI mid 30 686 247.6 245.5 99.2 18.7 1.24 18.7 99.2 15.8 102.7
5 Joakim Noah CHI mid 24 883 317.9 366.9 115.4 18.6 1.24 18.7 115.4 22.4 110.8
100.0 107.0 100.0 107.3 22.8
Pos Player Tm Type G Min Poss PProd ORtg %Pos tradeoff usg1 eff1 usg2 eff2 diff^2
1 Devin Harris NJN hi 32 1013 528.1 579.2 109.7 27.9 0.82 27.2 110.3 29.4 108.4
2 Anthony Morrow NJN lo 25 853 248.3 279.2 112.4 15.6 1.65 15.2 113.1 16.3 111.1
3 Travis Outlaw NJN lo 34 1088 339.9 332.5 97.8 16.7 1.65 16.3 98.5 12.5 104.8
4 Kris Humphries NJN lo 34 846 250.8 283.3 113.0 15.8 1.65 15.4 113.6 16.7 111.6
5 Brook Lopez NJN hi 34 1185 590.9 611.9 103.6 26.6 0.82 26.0 104.1 25.1 104.8
100.0 107.7 100.0 108.0 23.1
Pos Player Tm Type G Min Poss PProd ORtg %Pos tradeoff usg1 eff1 usg2 eff2 diff^2
1 Jrue Holiday PHI hi 33 1142 501.9 521.8 104.0 23.0 0.82 23.4 103.7 24.8 102.5
2 Jodie Meeks PHI lo 25 617 191.0 209.3 109.6 16.2 1.65 16.5 109.1 16.5 109.2
3 Andre Iguodala PHI mid 25 940 351.9 378.8 107.7 19.6 1.24 19.9 107.3 20.2 107.0
4 Elton Brand PHI mid 32 1092 424.4 482.1 113.6 20.4 1.24 20.7 113.2 22.9 110.4
5 Spencer Hawes PHI mid 33 682 251.4 243.5 96.8 19.3 1.24 19.6 96.5 15.6 101.4
100.0 105.8 100.0 106.1 23.1
Pos Player Tm Type G Min Poss PProd ORtg %Pos tradeoff usg1 eff1 usg2 eff2 diff^2
1 Raymond Felton NYK hi 33 1287 623.2 687.1 110.2 24.1 0.82 23.0 111.2 23.2 110.9
2 Landry Fields NYK lo 33 1038 288.1 336.0 116.6 13.8 1.65 13.2 117.7 14.4 115.6
3 Wilson Chandler NYK mid 33 1143 454.1 524.1 115.4 19.8 1.24 18.8 116.6 19.4 115.8
4 Danilo Gallinari NYK lo 33 1146 386.2 477.2 123.5 16.8 1.65 16.0 124.8 18.0 121.5
5 Amare Stoudemire NYK hi 33 1243 762.1 821.4 107.8 30.5 0.82 29.1 109.0 24.9 112.4
100.0 114.6 100.0 114.8 23.3
Pos Player Tm Type G Min Poss PProd ORtg %Pos tradeoff usg1 eff1 usg2 eff2 diff^2
1 Daniel Gibson CLE mid 31 912 372.4 401.9 107.9 21.1 1.24 20.5 108.6 21.2 107.8
2 Mo Williams CLE hi 27 837 457.1 449.2 98.3 28.2 0.82 27.4 98.9 24.2 101.6
3 Anthony Parker CLE lo 34 1041 310.9 311.2 100.1 15.4 1.65 15.0 100.8 13.3 103.6
4 Antawn Jamison CLE hi 31 924 420.2 431.4 102.7 23.5 0.82 22.8 103.2 24.5 101.9
5 Anderson Varejao CLE lo 30 959 271.4 307.5 113.3 14.6 1.65 14.2 114.0 16.9 109.6
100.0 104.3 100.0 104.6 23.8
Pos Player Tm Type G Min Poss PProd ORtg %Pos tradeoff usg1 eff1 usg2 eff2 diff^2
1 John Wall WAS hi 20 726 354.3 355.5 100.3 25.2 0.82 25.2 100.3 25.8 99.8
2 Kirk Hinrich WAS lo 30 1016 326.2 360.7 110.6 16.6 1.65 16.6 110.5 18.0 108.2
3 Al Thornton WAS lo 28 707 239.3 245.9 102.8 17.4 1.65 17.5 102.7 16.1 105.1
4 Andray Blatche WAS hi 27 961 460.7 430.1 93.4 24.7 0.82 24.8 93.3 21.3 96.2
5 JaVale McGee WAS lo 31 844 259.7 297.7 114.6 15.9 1.65 15.9 114.5 18.9 109.7
100.0 102.9 100.0 103.3 25.0
Pos Player Tm Type G Min Poss PProd ORtg %Pos tradeoff usg1 eff1 usg2 eff2 diff^2
1 Mike Conley MEM mid 34 1220 485.8 521.5 107.3 20.5 1.24 19.4 108.7 18.8 109.4
2 O.J. Mayo MEM mid 34 996 403.1 411.4 102.1 20.8 1.24 19.7 103.4 16.8 107.0
3 Rudy Gay MEM mid 32 1284 570.2 620.3 108.8 22.9 1.24 21.6 110.3 20.6 111.6
4 Zach Randolph MEM hi 30 1088 514.3 569.4 110.7 24.3 0.82 23.0 111.8 27.0 108.5
5 Marc Gasol MEM lo 33 1101 365.4 414.5 113.4 17.1 1.65 16.2 114.9 16.8 113.9
100.0 109.7 100.0 110.0 26.3
Pos Player Tm Type G Min Poss PProd ORtg %Pos tradeoff usg1 eff1 usg2 eff2 diff^2
1 Kyle Lowry HOU mid 29 945 354.6 382.4 107.8 18.9 1.24 19.1 107.6 16.5 110.9
2 Kevin Martin HOU hi 33 1031 558.7 668.8 119.7 27.3 0.82 27.6 119.5 31.4 116.4
3 Shane Battier HOU lo 33 1001 259.4 289.2 111.5 13.1 1.65 13.2 111.3 12.9 111.8
4 Luis Scola HOU hi 33 1079 555.3 595.3 107.2 26.0 0.82 26.2 107.0 23.1 109.6
5 Chuck Hayes HOU lo 31 668 182.3 221.1 121.3 13.8 1.65 13.9 121.1 16.2 117.3
100.0 113.1 100.0 113.5 36.5
Pos Player Tm Type G Min Poss PProd ORtg %Pos tradeoff usg1 eff1 usg2 eff2 diff^2
1 Derek Fisher LAL lo 34 911 240.5 245.6 102.1 13.6 1.65 13.2 102.8 11.0 106.4
2 Kobe Bryant LAL hi 34 1120 741.2 817.5 110.3 34.1 0.82 33.1 111.2 30.5 113.3
3 Ron Artest LAL lo 34 910 256.4 267.4 104.3 14.5 1.65 14.1 105.1 12.2 108.2
4 Lamar Odom LAL mid 34 1200 438.5 521.2 118.9 18.8 1.24 18.3 119.6 21.8 115.2
5 Pau Gasol LAL mid 34 1288 551.5 668.7 121.2 22.1 1.24 21.4 122.1 24.4 118.3
100.0 113.1 100.0 113.5 36.9
Pos Player Tm Type G Min Poss PProd ORtg %Pos tradeoff usg1 eff1 usg2 eff2 diff^2
1 Deron Williams UTA hi 34 1287 687.7 814.7 118.5 28.3 0.82 27.9 118.8 32.2 115.3
2 Raja Bell UTA lo 30 897 215.0 230.6 107.3 12.7 1.65 12.5 107.6 12.0 108.5
3 Andrei Kirilenko UTA lo 32 1044 344.6 374.2 108.6 17.5 1.65 17.3 109.0 14.8 113.1
4 Paul Millsap UTA mid 34 1178 469.0 557.5 118.9 21.1 1.24 20.8 119.2 22.8 116.8
5 Al Jefferson UTA mid 34 1208 495.3 530.3 107.1 21.7 1.24 21.4 107.4 18.3 111.3
100.0 113.3 100.0 113.8 38.1
Pos Player Tm Type G Min Poss PProd ORtg %Pos tradeoff usg1 eff1 usg2 eff2 diff^2
1 Jameer Nelson ORL hi 28 851 394.0 445.2 113.0 24.2 0.82 22.6 114.3 27.9 109.9
2 Vince Carter ORL mid 22 664 286.3 316.2 110.5 22.5 1.24 21.1 112.3 20.8 112.6
3 Quentin Richardson ORL lo 27 601 165.3 170.9 103.4 14.4 1.65 13.4 104.9 12.2 107.0
4 Rashard Lewis ORL lo 25 810 270.7 277.4 102.5 17.5 1.65 16.3 104.3 13.4 109.1
5 Dwight Howard ORL hi 31 1096 593.7 628.7 105.9 28.3 0.82 26.5 107.4 25.7 108.1
100.0 109.1 100.0 109.5 38.6
Pos Player Tm Type G Min Poss PProd ORtg %Pos tradeoff usg1 eff1 usg2 eff2 diff^2
1 Steve Nash PHO hi 30 976 512.3 625.4 122.1 26.7 0.82 26.2 122.5 30.7 118.8
2 Jason Richardson PHO hi 25 796 365.0 416.2 114.0 23.3 0.82 22.9 114.4 24.1 113.4
3 Grant Hill PHO mid 32 975 389.6 461.7 118.5 20.3 1.24 20.0 119.0 20.3 118.6
4 Hedo Turkoglu PHO lo 25 630 210.5 234.1 111.2 17.0 1.65 16.7 111.7 13.9 116.4
5 Channing Frye PHO lo 32 982 278.9 295.8 106.1 14.5 1.65 14.2 106.5 11.0 111.7
100.0 115.9 100.0 116.3 39.6
Pos Player Tm Type G Min Poss PProd ORtg %Pos tradeoff usg1 eff1 usg2 eff2 diff^2
1 Tony Parker SAS hi 33 1096 546.7 623.1 114.0 25.7 0.82 23.0 116.2 24.3 115.2
2 Manu Ginobili SAS hi 33 1041 517.3 609.6 117.8 25.6 0.82 23.0 120.0 26.5 117.0
3 Richard Jefferson SAS lo 33 1045 328.1 390.8 119.1 16.2 1.65 14.5 121.8 15.3 120.5
4 DeJuan Blair SAS mid 33 672 265.3 271.0 102.2 20.3 1.24 18.2 104.8 13.0 111.2
5 Tim Duncan SAS hi 33 955 439.9 484.0 110.0 23.7 0.82 21.3 112.0 20.9 112.4
100.0 114.9 100.0 115.4 42.7
Pos Player Tm Type G Min Poss PProd ORtg %Pos tradeoff usg1 eff1 usg2 eff2 diff^2
1 Jarrett Jack TOR mid 13 347 157.3 153.7 97.7 23.0 1.24 22.8 98.0 18.8 102.9
2 DeMar DeRozan TOR mid 33 1066 397.3 433.9 109.2 18.9 1.24 18.7 109.4 21.4 106.1
3 Linas Kleiza TOR mid 31 809 338.8 327.0 96.5 21.2 1.24 21.0 96.8 17.4 101.2
4 Reggie Evans TOR lo 15 409 95.1 101.4 106.6 11.8 1.65 11.7 106.7 14.0 102.8
5 Andrea Bargnani TOR hi 27 938 481.0 504.8 104.9 26.0 0.82 25.8 105.1 28.3 103.0
100.0 102.7 100.0 103.3 48.2
Pos Player Tm Type G Min Poss PProd ORtg %Pos tradeoff usg1 eff1 usg2 eff2 diff^2
1 Chauncey Billups DEN mid 27 898 407.5 470.7 115.5 22.7 1.24 22.6 115.6 23.0 115.2
2 Arron Afflalo DEN lo 32 1134 311.8 368.1 118.0 13.7 1.65 13.7 118.1 16.3 113.8
3 Carmelo Anthony DEN hi 25 875 552.3 578.3 104.7 31.6 0.82 31.4 104.8 26.6 108.8
4 Shelden Williams DEN lo 30 576 168.1 171.5 102.0 14.6 1.65 14.5 102.1 11.9 106.4
5 Nene Hilario DEN lo 28 882 314.4 410.2 130.5 17.8 1.65 17.8 130.6 22.2 123.3
100.0 113.3 100.0 114.0 56.4
Pos Player Tm Type G Min Poss PProd ORtg %Pos tradeoff usg1 eff1 usg2 eff2 diff^2
1 D.J. Augustin CHA mid 31 1063 376.7 448.7 119.1 18.7 1.24 17.7 120.4 24.4 112.1
2 Stephen Jackson CHA hi 30 1073 531.8 528.2 99.3 26.2 0.82 24.7 100.5 23.7 101.4
3 Gerald Wallace CHA mid 26 1013 410.3 415.1 101.2 21.4 1.24 20.2 102.6 18.5 104.8
4 Boris Diaw CHA lo 31 1112 367.4 371.2 101.0 17.4 1.65 16.5 102.6 14.5 105.8
5 Nazr Mohammed CHA mid 30 527 221.8 224.6 101.3 22.2 1.24 21.0 102.8 18.9 105.3
100.0 105.3 100.0 106.0 57.5
Pos Player Tm Type G Min Poss PProd ORtg %Pos tradeoff usg1 eff1 usg2 eff2 diff^2
1 Beno Udrih SAC mid 31 975 351.1 407.5 116.0 18.6 1.24 19.0 115.5 25.4 107.6
2 Tyreke Evans SAC hi 29 1064 536.4 492.0 91.7 26.0 0.82 26.7 91.2 22.4 94.7
3 Omri Casspi SAC lo 29 639 209.5 228.4 109.0 16.9 1.65 17.3 108.3 18.4 106.5
4 Carl Landry SAC mid 31 856 340.1 358.6 105.4 20.5 1.24 21.0 104.8 22.1 103.5
5 Samuel Dalembert SAC lo 29 579 174.7 155.6 89.0 15.6 1.65 16.0 88.4 11.7 95.4
100.0 101.2 100.0 102.2 79.0
Pos Player Tm Type G Min Poss PProd ORtg %Pos tradeoff usg1 eff1 usg2 eff2 diff^2
1 Eric Bledsoe LAC mid 34 892 320.1 299.5 93.6 18.7 1.24 17.9 94.6 12.5 101.3
2 Eric Gordon LAC hi 32 1204 629.1 712.2 113.2 27.3 0.82 26.1 114.2 30.2 110.8
3 Al-Farouq Aminu LAC mid 33 592 215.9 208.0 96.3 19.0 1.24 18.2 97.4 13.8 102.9
4 Blake Griffin LAC hi 34 1260 658.3 732.4 111.3 27.3 0.82 26.1 112.2 29.1 109.8
5 DeAndre Jordan LAC lo 34 808 190.4 215.8 113.3 12.3 1.65 11.8 114.2 14.5 109.7
100.0 107.1 100.0 108.1 83.0
Pos Player Tm Type G Min Poss PProd ORtg %Pos tradeoff usg1 eff1 usg2 eff2 diff^2
1 Jason Kidd DAL mid 33 1097 380.7 406.5 106.8 18.3 1.24 19.1 105.8 16.0 109.6
2 DeShawn Stevenson DAL lo 30 466 136.6 165.5 121.1 15.5 1.65 16.1 120.1 17.2 118.3
3 Caron Butler DAL hi 29 867 390.1 387.8 99.4 23.7 0.82 24.7 98.6 17.6 104.4
4 Dirk Nowitzki DAL hi 29 1026 510.3 599.2 117.4 26.2 0.82 27.3 116.5 29.9 114.5
5 Tyson Chandler DAL lo 32 892 206.9 275.8 133.3 12.2 1.65 12.7 132.5 19.3 121.7
100.0 112.7 100.0 114.0 110.2
Pos Player Tm Type G Min Poss PProd ORtg %Pos tradeoff usg1 eff1 usg2 eff2 diff^2
1 Luke Ridnour MIN lo 29 844 298.3 349.6 117.2 17.4 1.65 16.7 118.3 17.8 116.5
2 Wesley Johnson MIN lo 34 964 289.4 309.2 106.9 14.8 1.65 14.2 107.8 13.4 109.2
3 Michael Beasley MIN hi 32 1107 602.4 627.0 104.1 26.8 0.82 25.8 104.9 23.7 106.6
4 Kevin Love MIN hi 34 1228 604.9 725.5 119.9 24.2 0.82 23.3 120.7 32.0 113.5
5 Darko Milicic MIN mid 32 773 325.3 305.8 94.0 20.7 1.24 19.9 95.0 13.1 103.4
100.0 109.3 100.0 110.5 128.3
Pos Player Tm Type G Min Poss PProd ORtg %Pos tradeoff usg1 eff1 usg2 eff2 diff^2
1 Chris Paul NOH hi 34 1186 530.6 661.6 124.7 23.9 0.82 23.9 124.7 34.3 116.1
2 Marco Belinelli NOH lo 34 965 318.9 333.2 104.5 17.6 1.65 17.6 104.5 13.9 110.7
3 Trevor Ariza NOH lo 34 1143 381.9 350.0 91.6 17.8 1.65 17.8 91.6 10.0 104.4
4 David West NOH hi 33 1123 518.0 576.8 111.4 24.6 0.82 24.6 111.4 26.6 109.8
5 Emeka Okafor NOH lo 34 1075 323.5 361.4 111.7 16.1 1.65 16.1 111.7 15.3 113.0
100.0 109.9 100.0 112.0 187.6

Many caveats:

  • The lineup usages (and, thus, efficiencies) are not exact, but simply what we'd predict from their season averages. Despite playing together in the team's most frequently-used 5-man unit, they also accumulated some of their stats apart from each other.
  • Even though we're almost half of the way into the 2010-11 season, players' numbers are not necessarily indicative of true skill -- especially when it comes to outliers like Tyreke Evans (91.7 ORtg). Instead of making tactical decisions based on a half-season's sample, true talent estimates from multiple seasons would be even more beneficial for an exercise like this.
  • It's not clear whether a player like Tyson Chandler (or even Chris Paul) could actually increase his usage at the theoretical tradeoff rate. The rate could be much steeper than predicted for certain specific types of players.
  • Likewise, it's not clear how much control a coach has over possession allocation beyond running plays for specific players. My unsubstantiated guess is that a large amount of possession usage comes "in the flow of the offense" and therefore would be difficult to alter intentionally. It could even be that a 5-man unit's allocation is relatively set in stone -- that is, perhaps players already use roughly the ratio of possessions that is efficient for the offense.
  • Finally, as David Lewin pointed out, "players often affect team efficiency differently than you might expect from their offensive rating and possession rate".

Having said that, it's still interesting to think about different ways to maximize lineup combinations. If a team's offense is struggling and there may be a better way to distribute possessions within their top lineup, it could be worth their while to experiment with different usage patterns.

ShareThis

9 Responses to “Which Teams Are Allocating Their Possessions Efficiently?”

  1. DSMok1 Says:

    Exceptional work!

  2. kkopi27 Says:

    Very cool stuff. And, lo and behold, there is a small negative correlation between diff^2 and that particular lineup's +/- of -.096.

  3. Sean Says:

    This is great stuff, but unless I'm missing something, it's still just an estimate (albeit a very fine-tuned one). Wouldn't we be able to analyze a basketball system much more accurately and applicably if stats such as touches per possession and time of possession (for each player) were recorded?

  4. Jason J Says:

    Nice post, Neil. I agree with your 4th caveat though. It's probable that the usage break down is pretty much an organic optimization in itself wherein players who are able to get more shots take more and those who aren't don't. Where the change would probably need to take place from a coaching standpoint is more in either overall system or in minute allocation. Very interesting to see.

  5. P Middy Says:

    Excellent post. Kind of sad that so many bad teams top the list. talent > efficiency.

  6. Coach R Says:

    Some teams do a good job, however it does not always reflect in wins/losses unfortunately..:( Top point guards are not always given credit when then they hold a good ratio, but when the teams fail.. well its inevitable.

  7. NickS Says:

    I realize that this is a casual study, but might it make sense to remove Offensive Rebounding from Ortg for these purposes?

    It seems like if a team decides to put the ball in a given player's hands more often that player will get get more FGs, Asts, and TOs, but won't necessarily get more offensive rebounds.

    So when the system spits out the result that people like

    Tyson Chandler
    Kevin Love
    DeAndre Jordan

    Should all increase their usage by 30-50% that might go away if their current Ortg wasn't boosted by their offensive rebounding.

  8. Anon Says:

    "It's not clear whether a player like Tyson Chandler (or even Chris Paul) could actually increase his usage at the theoretical tradeoff rate."

    Maybe, maybe not; but I think I'd rather see him take a couple more shots per game than Caron Butler, who hasn't exactly been lighting it up this season. Anytime you can give those shots to your big man who has been near flawless from the field this season (its not all just tips and put-backs either, and he's even hitting his free throws now too!), you help your team offensively. He doesn't have to be prime Shaq in the post, but running some more plays for him can't hurt.

    As for CP3 (who has been playing out of his mind), I think he SHOULD be increasing his usage. That Hornets team outside of him and David West is buns offensively. It looks like a down-south version of LeBron and the Cavs from past seasons.

    Paul needs to get outta there.

  9. Jesse Says:

    I'm surprised that the Heat and Bucks are that high, but for different reasons. I figured Heat would be calculated as underutilizing the Big 3, i.e. their skill curves can handle more, since they had done so in the past (skill curve has less steep slope on average). I'm also surprised that the Bucks are as high as they are, given that my "watching games" i.e. non-aggregated impression of Jennings is that he's using way too many possessions w/ his inefficient shot.