|
On a certain level, basketball is purely a game of lineup combinations. The teamwork and chemistry required of the five-man unit lies at the very heart of the game, making it fundamentally different (and, I'd argue, better) than a sport composed mostly of one-on-one interactions like baseball. Here's what I wrote on the subject in May 2010:
"In basketball perhaps more than any other sport, the concept of team-building--creating a cohesive group that fits together and may be greater than the sum of its parts--is phenomenally important. In baseball, a sport dominated by one-on-one matchups, not a whole lot of consideration has to be made for how teammates work together; to make a great team, you basically grab the 25 best players you can, throw them together, and watch them produce. But in basketball, teammates have to work together while simultaneously 'competing' for touches & shots. Throw together a baseball lineup of nine guys who each create 100 runs, you'll probably score 900 runs; throw together a basketball lineup of five 20 PPG scorers, you probably won't score 100 PPG. There's no upper limit on the number of runs the baseball lineup can produce, but there is an upper limit to the points the basketball lineup scores, because teams are limited by a finite number of minutes in a game, and as a result, lineups are limited by a finite number of touches & shots to be allocated to the individual players."
"That's why a stat like Possession% (the percentage of team possessions a player uses while on the floor) is important in looking at how the pieces of a team fit together. A lineup of All-Stars would be interesting, but perhaps a less-talented lineup with one 26 percent usage guy, two 20 percent guys, an 18 percent guy, and a 16 percent guy would be even better if the All-Stars are not happy with the way they fit together or are unable to operate at peak efficiency in lesser roles, while the less talented lineup features players who are all at their optimal usage levels. The whole of the latter would be greater than the sum of the former's parts."
As we saw with the 2011 Miami Heat, there will be diminishing statistical returns if you throw multiple high-usage stars into the same 5-man group and ask them to share one basketball. Much like the rules of physics break down within the singularity of a black hole, the normal tradeoff between usage and efficiency breaks down under the extreme conditions of such a star-studded lineup, making predictions go haywire.
In fact, any time a player is forced to deviate significantly from his established role and/or usage pattern, it's tough to say how his efficiency will react. Some players have managed to increase their usage and their efficiency at the same time, while others find it difficult to maintain their points/possession when taking on a larger role in the offense. Weirder still, some players (including two of Miami's Big Three a year ago, LeBron James & Chris Bosh) will drop their usage but see a decline in efficiency as well. Because of these irregularities, it might be better to assemble a team whose players have to deviate from their established usage rates as little as possible, because it will be easier to predict how they will perform when tossed into a lineup together.
In the spirit of that line of thinking, I decided to construct the best hypothetical 5-man units of 2012 where no player had to increase or decrease their usage rate. In other words, I picked groups of players whose total usage adds up to 100%. I did this by selecting 5-man units at random based on Basketball Value's position designations, and keeping 11,458 unique groups whose combined usage happened to sum to 100%. This is not a complete list--obviously there are more than 11,450 possible combinations when picking groups of five from a pool of 335 players--but it does represent a good cross-section of the possible lineups out there.
For each group, I then used Dean Oliver's player Offensive and Defensive Ratings to predict composite Offensive/Defensive efficiency ratings (and therefore efficiency margins). The groups with the best margins were considered to be the best "fits":
Rk PG SG SF PF C ORtg DRtg Net
----------------------------------------------------------------------------------------
1 Paul C. Brewer Korver Love C. Andersen 118.5 100.2 18.2
2 Rose S. Curry E. Davis B. Wright T. Chandler 116.2 99.8 16.4
3 Ridnour Harden Dudley B. Wright Love 117.3 102.2 15.1
4 Jack Gee Korver James T. Chandler 116.9 101.8 15.1
5 Lucas Dragic R. Anderson B. Wright T. Chandler 115.2 100.3 15.0
6 Chalmers McGrady L. James Novak J. Hill 114.3 100.9 13.4
7 Burks M. Miller T. Young Love C. Andersen 113.6 100.3 13.3
8 Paul Lowry Novak McRoberts Howard 113.4 100.1 13.2
9 Meeks J. Johnson Millsap Duncan T. Chandler 113.1 99.8 13.2
10 Duhon Wade R. Anderson P. Gasol Ibaka 113.9 100.8 13.1
11 Dragic Harden L. Allen P. Gasol Gortat 114.2 101.1 13.0
12 Paul T. Allen Pietrus P. Gasol Brand 111.8 98.8 13.0
13 Cook L. Williams Leonard James K. Thomas 112.8 99.9 12.9
14 Curry Harden R. Anderson Leuer Mozgov 116.8 103.9 12.9
15 Paul Kev. Martin Pierce Vucevic J. Anthony 113.3 100.5 12.9
16 C. Watson Rush L. James Cunningham Pachulia 113.5 100.8 12.7
17 Paul Childress Durant Leuer Dalembert 115.5 102.8 12.7
18 Bibby L. Williams T. Harris R. Anderson T. Chandler 114.4 101.8 12.5
19 Lowry Redick L. James R. Evans Ibaka 113.7 101.2 12.5
20 Rondo Hill Batum Aldridge D. Jordan 112.7 100.2 12.4
Players who appear frequently in the top 1% of lineups might be considered to be "good fit" players, capable of working with a variety of different teammates and producing good results:
Player Pos #
---------------------------
LeBron James PF/SF 21
Chris Paul PG 15
Kevin Love C/PF 15
Tyson Chandler C 14
Brandan Wright C/PF 13
Ryan Anderson PF/SF 12
James Harden SG 10
Kyle Korver SF/SG 10
Chris Andersen C/PF 9
Danilo Gallinari PF/SF 9
Derrick Rose PG 9
Steve Novak PF/SF 9
Lavoy Allen PF/SF 8
Louis Williams PG/SG 8
Nicolas Batum SF 8
Pau Gasol C/PF 8
Stephen Curry PG/SG 8
Gerald Wallace PF/SF 7
Joakim Noah C/PF 7
Kyle Lowry PG/SG 7
Thaddeus Young PF/SF 7
The top of the list resembles any ranking of the best, most efficient players in the game--James, Paul, and Love are all present, for instance. However, other players might be good examples of the "fit" phenomenon: in a vacuum, Chris Andersen is no more than a solid role player, but his combination of strong defense, high efficiency, and a 15% usage rate allows him to slot in along other stars and make small but essential contributions. Andersen might be only the 43rd-most productive player in the league on a per-possession basis (using a metric based on Oliver's stats), but he might be the 10th-best "fit" player because his profile is easier to mesh with in a lineup than a higher-usage player with more star power.
Conversely, here are the worst "fit" lineups, which are generally groups of poor players who are having terrible efficiency seasons:
Rk PG SG SF PF C ORtg DRtg Net
----------------------------------------------------------------------------------------
1 Higgins Jo. Crawford Outlaw Samuels Magloire 80.2 108.1 -27.9
2 Higgins Selby Daniels Gooden Magloire 81.7 104.5 -22.8
3 Fredette Higgins Daye Vesely Mullens 87.6 109.7 -22.1
4 Udrih Higgins Aminu Outlaw Mullens 87.9 109.5 -21.6
5 Telfair Higgins Garcia Odom Diaw 85.9 107.3 -21.4
6 D. Gibson Higgins Daniels S. Jackson Samuels 85.7 106.1 -20.4
7 K. Walker Carroll Prince Daye Hickson 89.1 109.5 -20.3
8 Higgins Pondexter Daye Speights Samuels 85.7 105.8 -20.1
9 K. Walker Selby R. Williams Sha. WilliamsFrye 88.6 108.6 -20.1
10 Je. Pargo Goudelock Gr. Hill Daye Hickson 86.8 106.5 -19.8
11 Selby Higgins S. Jackson Vesely Brand 85.1 104.5 -19.4
12 Ja. Pargo M. Daniels Maggette Daye Vesely 85.2 104.6 -19.4
13 Selby Terry Outlaw Blatche L. Sanders 85.2 104.3 -19.1
14 S. Brown Carroll Wafer Daye Diaw 88.9 108.0 -19.0
15 Knight Ja. Pargo Outlaw Blatche Hickson 87.9 106.9 -19.0
16 Knight Wilkins Maggette Blatche Amundson 87.3 106.2 -18.9
17 Je. Pargo Higgins Fields Hickson Bogut 86.6 105.5 -18.8
18 Higgins Stephenson Garcia Daye Garnett 85.6 104.4 -18.8
19 Higgins Telfair Jeffries Daye C. Smith 86.1 104.8 -18.7
20 Higgins DeRozan World Peace Forbes Mbah a Moute 87.2 105.7 -18.5
Just as players who show up often in good lineups might be considered "good fits", here's the all-"bad fit" team:
Player Pos #
------------------------------
Cory Higgins PG/SG 32
Austin Daye PF/SF 23
Jeremy Pargo PG/SG 23
Travis Outlaw PF/SF 17
Josh Selby PG/SG 15
Sebastian Telfair PG/SG 15
Toney Douglas PG/SG 12
Andray Blatche C/PF 11
Boris Diaw C/PF 11
Brandon Knight PG/SG 11
Louis Amundson C/PF 11
Matt Carroll SF/SG 11
Stephen Jackson PF/SF 11
Gary Forbes PF/SF 10
Tristan Thompson C/PF 10
Reggie Williams SF/SG 9
Samardo Samuels C/PF 8
Corey Maggette SF 7
Jamaal Magloire C/PF 7
Jan Vesely C/PF 7
Jordan Crawford PG/SG 7
Mehmet Okur C/PF 7
The anti-Andersen, Cory Higgins is the undisputed king of "bad fit", thanks to brutal efficiency marks (77.0 ORtg, 114.3 DRtg) and a medium-usage profile (22.7 percent) that finds him frequently being the not-so-missing piece for a lineup that needs to get to 100%.
This is not to say that the construction of these lineups is an exact science (in fact, it's about as far from it as possible). Usage is not an overly effective descriptor of a player's style, meaning two players who arrive at the same usage in different ways may in reality have dramatically different impacts on a given lineup. But it is more likely that a lineup operates according to statistical prediction if its players are allowed to stay in their customary roles, and that means teams seeking greater reliability from their acquisitions should think about targeting players whose usage profiles fit in well with their existing roster.
Neil Paine is an author of Basketball Prospectus.
You can contact Neil by clicking here or click here to see Neil's other articles.
| 
Start/Sit Projections (02/17)
0 comments have been left for this article. 
