FanPost

Do Passes Per Possession Matter?

San Antonio just won the championship, now suddenly statistics where they lead other teams seem particularly relevant, like this one:

Team Passes Per Game Time Of Possession Passes Per Minute




Spurs 330.4 18.1

18.25414365

Bobcats 334.3 18.9 17.68783069
Hawks 318.2 18 17.67777778
76ers 305.1 17.3 17.63583815
Clippers 303 17.3 17.51445087
Lakers 306.5 17.6 17.41477273
Wizards 317.1 18.3 17.32786885
Bucks 320.5 18.8 17.04787234
Mavericks 309.7 18.6 16.65053763
Jazz 320.5 19.3 16.60621762
Rockets 292.7 17.8 16.44382022
Bulls 320 19.5 16.41025641
Timberwolves 303.5 18.6 16.3172043
Cavaliers 301.5 18.5 16.2972973
Celtics 300.7 18.5 16.25405405
Pacers 302.7 19 15.93157895
Grizzlies 307 19.3 15.90673575
Heat 292.7 18.5 15.82162162
Trail Blazers 295.5 18.8 15.71808511
Nets 292 18.8 15.53191489
Raptors 299.7 19.4 15.44845361
Suns 278 18.3 15.19125683
Magic 286.4 19.1 14.9947644
Thunder 268.3 18 14.90555556
Nuggets 258 17.6 14.65909091
Knicks 279.3 19.2 14.546875
Pistons 259.6 18.3 14.18579235
Pelicans 272.3 19.3 14.10880829
Kings 266.5 19 14.02631579
Warriors 243.3 17.9 13.59217877

Nevermind that PHI, LAL, and MIL, are in the top 8; the Spurs lead by a significant amount so it's probably important.

If only there was a way to see if these statistics actually had an impact on team's success in the 2013-2014 NBA season. But wait, there is:

TEAM Passes Possession Passes Per Possession Wins Hollinger Offensive Effeciency TeamRankings Offensive Efficiency
Spurs 330.4 18.1 18.25414365 62 108.2 1.081
Bobcats 334.3 18.9 17.68783069 43 101.2 1.002
Hawks 318.2 18 17.67777778 38 103.4 1.024
76ers 305.1 17.3 17.63583815 19 96.8 0.962
Clippers 303 17.3 17.51445087 57 109.4 1.084
Lakers 306.5 17.6 17.41477273 27 101.9 1.01
Wizards 317.1 18.3 17.32786885 44 103.3 1.022
Bucks 320.5 18.8 17.04787234 15 100.2 0.994
Mavericks 309.7 18.6 16.65053763 49 109 1.077
Jazz 320.5 19.3 16.60621762 25 100.6 1.002
Rockets 292.7 17.8 16.44382022 54 108.6 1.073
Bulls 320 19.5 16.41025641 48 99.7 0.989
Timberwolves 303.5 18.6 16.3172043 40 105.6 1.053
Cavaliers 301.5 18.5 16.2972973 33 101.3 1.005
Celtics 300.7 18.5 16.25405405 25 99.7 0.993
Pacers 302.7 19 15.93157895 56 101.5 1.006
Grizzlies 307 19.3 15.90673575 50 103.3 1.024
Heat 292.7 18.5 15.82162162 54 109 1.081
Trail Blazers 295.5 18.8 15.71808511 54 108.3 1.071
Nets 292 18.8 15.53191489 44 104.4 1.036
Raptors 299.7 19.4 15.44845361 48 105.8 1.048
Suns 278 18.3 15.19125683 48 107.1 1.059
Magic 286.4 19.1 14.9947644 23 99.3 0.985
Thunder 268.3 18 14.90555556 59 108.1 1.063
Nuggets 258 17.6 14.65909091 36 103.3 1.024
Knicks 279.3 19.2 14.546875 37 105.4 1.046
Pistons 259.6 18.3 14.18579235 29 102.9 1.023
Pelicans 272.3 19.3 14.10880829 34 104.7 1.036
Kings 266.5 19 14.02631579 28 102.9 1.021
Warriors 243.3 17.9 13.59217877 51 105.3 1.043




41




Avg Wins TOP 15 38.6




Avg Wins Bottom 15 43.4
Correlation to Team Wins (Pearson) 0.0080678567 -0.0771049505 0.0359800858 Correlation PPP to Offensive Metric -0.1149396964 -0.0956359537

Considering there is no correlation, the answer to whether passes per possession matters, looks like a resounding no.

How a team plays basketball, and how well a team plays basketball, are separate things. Style does not necessarily connect to performance, and certainly not in regards to this statistic.

The question appropriate to this statistic is not if altering it will alter team performance, but rather if it affects the valuation, and the resources (salary), paid for the same performance. If that style, of less ball movement, results in the performance being undervalued, teams should desire to hold onto assets that have that stylistic difference, of less passes per possession.

If scouts and GMs are biased against players lacking in this statistic (that does not correlate to team winrate) these players end up undervalued, just like if their body type is judged as less than ideal, like being short for their position, or if they were drafted late in the second round or undrafted. If a player fit all three categories he would probably cost far, far less than his performance, and effect on team winrate, would otherwise cost. Especially if there were also other factors, like casual observation of their defense appearing far worse than actual statistical evidence suggests.

In a league that dramatically overpays top picks after 4+ years of terrible performance; if you have a player whose style no GM wants to pay much for, whose physical attributes lead to them being stereotyped as something other than their performance indicates, and that performance is at a very high level, a high level which you cannot replace without significant difficultly, you should probably keep that player. And certainly not lose them for nothing.

Editing in a look at individual point guards, also added offensive efficiency stats to the team statistics, they are missing in individual as their addition is a little more time intensive than I can manage at the moment. Also, to clarify why wins were used initially; it provided me with a frame of reference to correlation to winrate for all other team statistics, not just ones that can be isolated to offense, to have some frame of reference for what correlation values would be high enough to suggest meaningful correlation.

For Individual players I used http://espn.go.com/nba/statistics/player/_/stat/scoring-per-game/sort/avgMinutes/position/point-guards and took every point guard over 25 minutes, to reach Collison, and then replaced Sessions and Augustin as they played for multiple teams with Mo Williams and Norris Cole as they played 24.6+ minutes and retained the sample size.

Trying to analyze the impact of this statistic in regards to individual players has numerous issues, and without considering all players in the league (who would be expected to have the same result as analyzing by team) it is not capable of being strong analytical evidence. GP, min, etc, aren't standardized. Plus the statistic itself has its own issues.

With that massive caveat, and keeping in my mind correlation to PER may be meaningless as it may be a correlation to some factor of itself in the PER formula:

Player GP MIN per game Time of Possession (min) Passes per game Passes Per Possession PER Team WINS
Alec Burks (UTA) 78 28.2 2.7 36.7 13.5925925926 15.82 25
Avery Bradley (BOS) 60 31.2 2.1 29.6 14.0952380952 12.8 25
Brandon Jennings (DET) 80 34.4 7 59.9 8.5571428571 15.67 29
Brandon Knight (MIL) 72 33.5 6.2 59.4 9.5806451613 16.59 15
Chris Paul (LAC) 62 35.3 7.1 71.5 10.0704225352 25.98 57
Damian Lillard (POR) 82 36 6.9 60.8 8.8115942029 18.69 54
Darren Collison (LAC) 80 26.1 3.6 41.5 11.5277777778 16.21 57
Deron Williams (BKN) 64 32.3 6.1 60.6 9.9344262295 17.69 44
George Hill (IND) 76 32.2 5.3 56.3 10.6226415094 13.42 56
Isaiah Thomas (SAC) 71 34.9 7.6 59 7.7631578947 20.54 28
Jameer Nelson (ORL) 68 32.2 5.9 53.6 9.0847457627 13.89 23
Jarrett Jack (CLE) 80 28.3 3.9 44.2 11.3333333333 11.57 33
Jeff Teague (ATL) 78 32.4 6 55.6 9.2666666667 17.16 38
Jeremy Lin (HOU) 71 29.1 3.6 41.7 11.5833333333 14.31 54
John Wall (WAS) 81 36.6 7.8 70.1 8.9871794872 19.61 44
Jose Calderon (DAL) 81 30.6 4.7 61.8 13.1489361702 15.25 49
Kemba Walker (CHA) 72 36 7.5 77.3 10.3066666667 16.82 43
Kyle Lowry (TOR) 78 36.4 6.8 68.2 10.0294117647 20.2 48
Kyrie Irving (CLE) 71 35.3 6.2 56.1 9.0483870968 20.14 33
Mario Chalmers (MIA) 73 30 4.3 52.2 12.1395348837 14.06 54
Michael Carter-Williams (PHI) 70 34.6 6.4 62.6 9.78125 15.59 19
Mike Conley (MEM) 73 33.7 7.3 65.4 8.9589041096 20.07 50
Mo Williams (POR) 74 24.9 4.2 41 9.7619047619 11.87 54
Norris Cole (MIA) 82 24.8 3.4 39.5 11.6176470588 8.8 54
Raymond Felton (NYK) 65 31.2 5.4 55.8 10.3333333333 12.9 37
Reggie Jackson (OKC) 80 28.7 4.7 42.2 8.9787234043 15.43 59
Ricky Rubio (MIN) 82 32.4 6.7 69.4 10.3582089552 15.35 40
Shaun Livingston (BKN) 75 26.1 3.8 42 11.0526315789 14.52 44
Stephen Curry (GSW) 77 36.7 6.7 53 7.9104477612 24.13 51
Tony Parker (SAS) 68 29.5 6 55.4 9.2333333333 19 62
Trey Burke (UTA) 70 32.4 6.1 62.3 10.2131147541 12.64 25
Ty Lawson (DEN) 62 36.1 6.7 56.7 8.4626865672 19.02 36
Correlation (Pearson) to Passes Per Possession




-0.5246975206 -0.0017069015

The team statistic is more interesting to me, as passes per minute of team possession, while passes per minute of individual possession seems to have more problems. The ratio a player's touch ends in a shot, pass, or turnover has different value to different players, an inefficient scorer should ideally pass more per minute of individual possession, just like an efficient scorer should ideally shoot more, thereby passing less, per minute of individual possession.

Without the time to bring the value of time of team possession into the data at the moment, and instead just quickly comparing to Passes Per Minute (regardless of team possession) towards those same point guards is more interesting, and of note, as it relates to the Kings, Thomas ranks higher here than Collison. As the correlation across all players is virtually 0, and the addition or elimination of two or three players at extremes like Paul/Curry would swing it tremendously, it may be an artifact of sample size, but seems interesting. Also, those players TS% is added to show its weight, correlated to those same statistics, to provide frame of reference. Again, PER is possibly meaningless in regards to correlation,as the data may end up being correlated with itself (as a result of what is included in the PER formula):

Player Passes Per Minute RANK(Passes Per Minute)) PER Team WINS TS%
Alec Burks (UTA) 1.3014184397 31 15.82 25 0.547
Avery Bradley (BOS) 0.9487179487 32 12.8 25 0.51
Brandon Jennings (DET) 1.7412790698 15 15.67 29 0.486
Brandon Knight (MIL) 1.7731343284 13 16.59 15 0.523
Chris Paul (LAC) 2.0254957507 3 25.98 57 0.58
Damian Lillard (POR) 1.6888888889 19 18.69 54 0.568
Darren Collison (LAC) 1.5900383142 24 16.21 57 0.575
Deron Williams (BKN) 1.8761609907 9 17.69 44 0.564
George Hill (IND) 1.748447205 14 13.42 56 0.563
Isaiah Thomas (SAC) 1.6905444126 18 20.54 28 0.574
Jameer Nelson (ORL) 1.6645962733 20 13.89 23 0.515
Jarrett Jack (CLE) 1.5618374558 27 11.57 33 0.498
Jeff Teague (ATL) 1.7160493827 17 17.16 38 0.541
Jeremy Lin (HOU) 1.4329896907 30 14.31 54 0.572
John Wall (WAS) 1.9153005464 7 19.61 44 0.524
Jose Calderon (DAL) 2.0196078431 4 15.25 49 0.596
Kemba Walker (CHA) 2.1472222222 1 16.82 43 0.499
Kyle Lowry (TOR) 1.8736263736 10 20.2 48 0.567
Kyrie Irving (CLE) 1.5892351275 25 20.14 33 0.533
Mario Chalmers (MIA) 1.74 16 14.06 54 0.567
Michael Carter-Williams (PHI) 1.8092485549 11 15.59 19 0.48
Mike Conley (MEM) 1.940652819 5 20.07 50 0.545
Mo Williams (POR) 1.6465863454 21 11.87 54 0.507
Norris Cole (MIA) 1.5927419355 23 8.8 54 0.497
Raymond Felton (NYK) 1.7884615385 12 12.9 37 0.476
Reggie Jackson (OKC) 1.4703832753 28 15.43 59 0.527
Ricky Rubio (MIN) 2.1419753086 2 15.35 40 0.491
Shaun Livingston (BKN) 1.6091954023 22 14.52 44 0.551
Stephen Curry (GSW) 1.4441416894 29 24.13 51 0.61
Tony Parker (SAS) 1.8779661017 8 19 62 0.555
Trey Burke (UTA) 1.9228395062 6 12.64 25 0.473
Ty Lawson (DEN) 1.5706371191 26 19.02 36 0.553
Correlation (Pearson) to Passes Per Minute

0.2446972605 0.1588342159 -0.0832968591
CORRELATION (Pearson) to TS%

0.5790483228 0.5079571962

(This is a FanPost from a member of the Sactown Royalty community. The views expressed come from the member, and not Sactown Royalty staff.)

X
Log In Sign Up

forgot?
Log In Sign Up

Forgot password?

We'll email you a reset link.

If you signed up using a 3rd party account like Facebook or Twitter, please login with it instead.

Forgot password?

Try another email?

Almost done,

By becoming a registered user, you are also agreeing to our Terms and confirming that you have read our Privacy Policy.

Join Sactown Royalty

You must be a member of Sactown Royalty to participate.

We have our own Community Guidelines at Sactown Royalty. You should read them.

Join Sactown Royalty

You must be a member of Sactown Royalty to participate.

We have our own Community Guidelines at Sactown Royalty. You should read them.

Spinner.vc97ec6e

Authenticating

Great!

Choose an available username to complete sign up.

In order to provide our users with a better overall experience, we ask for more information from Facebook when using it to login so that we can learn more about our audience and provide you with the best possible experience. We do not store specific user data and the sharing of it is not required to login with Facebook.

tracking_pixel_9347_tracker