1. Welcome! Please take a few seconds to create your free account to post threads, make some friends, remove a few ads while surfing and much more. ClutchFans has been bringing fans together to talk Houston Sports since 1996. Join us!

Assist Quality, the hidden perspective, Lin and Rockets

Discussion in 'Houston Rockets: Game Action & Roster Moves' started by sutton, Mar 11, 2014.

  1. sutton

    sutton Member

    Joined:
    Feb 20, 2014
    Messages:
    288
    Likes Received:
    15
    I wrote this during all-star break, posted at hupu and realgm.

    -------

    Assist Quality, the hidden perspective, Lin and Rockets


    2012 SportVU released an article on “Potential Assist Locations”, which shows Lin’s potential assists contain 36% at rim opportunities, the most among PG.

    [​IMG]

    [​IMG]



    As you can see from the diagram, such passes make big difference for shots at rim, not so much for long 2 and 3. However we knew that most 3pt shots were only taking when you have a open look.

    Back then, there is no statistics publicly available for further quantitative analysis. Until this season NBA.com introduces SportVU passing tracking statistics.

    2013-2014 season until ASB, among 100 most assisters, the median Assist Efficiency (AE) is 1.20.

    Assist Efficiency (AE) = [Point created by assist per game] / [Assist Opportunities per game] (AO)

    Assist Opportunity means passes result in shots, result in shooting foul not included.

    Lin’s AE = 10.9/7.9 = 1.38, is 3rd among league, 1st among PGs, only 5 PGs ranked among TOP25 AE. ( Parson 1.28 Harden 1.17 Beverley 1.17)

    No doubt Lin has high AE because of close to rim passes, which comes at a price: turnovers, was it worth the effort?

    If we adjust Lin’s AE to 1.20, then

    10.9/1.20 – 7.9 = 1.18 possessions saved (this value is 3rd among league)

    In other words, Lin’s AE at 1.38 plus 1.18 TOs equals to AE at 1.20 (7.9+1.18). Call it AE compensation 1(AE_C1)

    Then Lin’s AST/TO ratio is 4.5/(2.7-1.18), which is close to 3:1 with AE at 1.20, very decent.

    However, for a player like Harden, many of his TOs come from aggressive scoring (euro step, ISO etc.), which reflects on FT shooting.

    Harden makes 8.8 free throws per game at 85%, if we convert it to 56% TS, then

    FT*FT%/0.56*2-FT/2 = 2.27 possessions

    In other words, if Harden shots 8.8 free throws while TO 2.27 times, net result is decent 56% TS. Call it AE compensation 2 (AE_C2)

    Combine AE_C1 and AE_C2, the highest among league are:

    1st Durant AE_C1+C2 = 3.35, 2nd LBJ=2.45

    3rd LIN=1.18+0.89=2.07, Harden at 4th = -0.3+2.27=1.97

    Surprisingly, Lin's compensation value at 3rd despite far low usage rate, it shows Lin has rare combination of both aggressive assist and aggressive scoring. This might be added to the explanations why Lin could pull off wins despite 6 TOs per game during his Linsanity run.

    Using C1,C2 to adjust AST/TO ratio,

    For Lin, 4.5/(2.7-2.27)= 7.1, which is 3rd among PGs, only Chris Paul 12.98, Marshall 9.81 are higher.

    [​IMG]

    Rondo’s stats is at small sample size this season, he was much better, however about one third of his assist always comes from long 2, probably the most among PG. This doesn't mean that he can’t adjust to Huston’s offense scheme. But it will come at a price.

    Because current advance player rating formula such as PER has no reflection on passing efficiency, one can argue player has “high risk high reward” passing style like Lin and parson among others are undervalued, by 1.18 and 0.45 (AE_C1) respectively.
     
    4 people like this.
  2. AvgJoe

    AvgJoe Contributing Member

    Joined:
    Apr 8, 2009
    Messages:
    3,637
    Likes Received:
    392
    It also helps when your team is one of the best offensive teams in the league. Good read, need sometime to chew though.
     
  3. justbuckets

    justbuckets Member

    Joined:
    Oct 14, 2013
    Messages:
    522
    Likes Received:
    7
    Good info and analysis, could use much better formatting for ease of reading though.
     
  4. larsv8

    larsv8 Contributing Member

    Joined:
    Nov 11, 2007
    Messages:
    21,663
    Likes Received:
    13,913
    Excluding assist opportunities that result in foul shots breaks the whole model.

    Getting to the line is the best result in basketball.
     
  5. Sports2012

    Sports2012 Member

    Joined:
    Dec 11, 2012
    Messages:
    1,505
    Likes Received:
    16
    Great new perspective to facilitate more complete assessment of player performance, especially PGs'. May help settle a lot of disputes on why Lin plays like he's been.
     
  6. Joe Joe

    Joe Joe Go Stros!
    Supporting Member

    Joined:
    May 3, 1999
    Messages:
    23,839
    Likes Received:
    13,895
    I have been using "points created/(assist opportunities + turnovers)" as a measure of play making this year in 2 or 3 posts concerning Lin, Rondo, and Teague. While free throws is a big problem, it is still a lot better than assist/turnovers which has the same problem, but does not take into account the quality of the assist.
     
  7. d12babymamas

    d12babymamas Member

    Joined:
    Oct 10, 2013
    Messages:
    1,608
    Likes Received:
    34
    wow that was like deciphering hieroglyphics
     
  8. Stats

    Stats Member

    Joined:
    Nov 13, 2013
    Messages:
    284
    Likes Received:
    8
    But then passing to Harden is like having a cheat code on the model.
     
  9. Amiga

    Amiga I get vaunted sacred revelations from social media
    Supporting Member

    Joined:
    Sep 18, 2008
    Messages:
    21,710
    Likes Received:
    18,486
    You lost me from before you posted.
     
  10. shortfuse3

    shortfuse3 Member

    Joined:
    Aug 7, 2008
    Messages:
    3,912
    Likes Received:
    59
    I'm sorry OP but your post is completely flawed and skewed.

    You're taking two sets of data and claiming they're related:

    1) Passes to players close to the rim
    2) Shot percentage based on distance to the rim

    Just because you pass to a player closest to the rim doesn't mean that it will have the best positive result. Otherwise, the entire game of basketball would be decided based on who could pass to the player closest to the rim as much as possible.

    Also, passes to players close to the rim are risky passes, and you aren't taking into account the turnovers involved in these types of passes.
     
    2 people like this.
  11. steady

    steady Member

    Joined:
    Dec 14, 2012
    Messages:
    1,329
    Likes Received:
    38
    OP or anyone. could someone explain the AE compensation values a little more?

    I get general point that assists for shots close to the basket have the greatest impact in terms of made baskets, but also carry the risk of more turnovers. So the AE compensation factor I think is meant to compensate for that in some way??
     
  12. phantoman

    phantoman Contributing Member

    Joined:
    Oct 11, 2007
    Messages:
    1,546
    Likes Received:
    273
    I think you bring up a good point - the comparison of the shots after and FG% does skew things. However they are in terms some what related. If the purpose of the anlaysis is to find the quality of a pass - ideally there are 3 catagories.

    1. getting it to the highest fg% on the court. (inside the paint)
    2. getting it the pass to the person with a good look or high value shot
    3. getting it to the person that finishes the best (ie: harden/dwight on a pick & roll).

    The analysis does need to be broken down further but for the OP purpose - he is looking at our offensive scheme - 3's and at the rim.
     
  13. t3ddt3ch

    t3ddt3ch Member

    Joined:
    Nov 4, 2010
    Messages:
    193
    Likes Received:
    6
    Are you kidding? How are they not related? Besides getting the ball in the hoop, the offensive goal is to get as close to the rim as you can. Why? Because they are higher percentage shots...see Shaq, Yao, Dwight...<insert center here> Would you rather pass to a 35% 3 point shooter or a 50% center? The math is basic. After a 50 shots who has more points?
     
  14. haoafu

    haoafu Contributing Member

    Joined:
    Jun 29, 2006
    Messages:
    2,021
    Likes Received:
    56
    Good analysis. It does account for turnovers, and the lack of assist to foul shots is ok because it's the same exclusion for all players(Including it may increase Lin's assist efficiency stats).
     
  15. Amiga

    Amiga I get vaunted sacred revelations from social media
    Supporting Member

    Joined:
    Sep 18, 2008
    Messages:
    21,710
    Likes Received:
    18,486
    I think this is unnecessarily complicated (since I can't understand it :)). Assists Quality/Eff/Effectiveness should be quite simple. How many points is generated (counting FT made as well) per pass minus how many points is lost per TO as a result of the pass attempt.
     
  16. Joe Joe

    Joe Joe Go Stros!
    Supporting Member

    Joined:
    May 3, 1999
    Messages:
    23,839
    Likes Received:
    13,895
    I'm not the OP, but my method of "points created/(assist opportunities + turnovers)" overcompensates for turnovers as all turnovers are assumed to be from passes that would be assist opportunities.

    On passing to players closest players to rim, this is the Rockets first choice. Passing to players with open lane to rim is second choice. Passing to players with open three point shot is third choice. This is oversimplified, but Rockets want best shot attempts and those tend to be at the rim or open three pointers. Rockets get 73.8% of their points off of points in the paint (PITP) and from three pointers.
     
  17. Easy

    Easy Boban Only Fan
    Supporting Member

    Joined:
    Jul 23, 2002
    Messages:
    35,060
    Likes Received:
    23,967
    Points generated is skewed by the skills of your teammates. If you are playing with a bunch of good offensive players, your assist number would be higher than if you played with bad offensive players. AE is trying to level the field.

    I do think that not counting passes that result in free throws is a big downer.

    Anyway, we know that Morey has been tracking assist quality from Day 1. He once said the McGrady was the second best quality non-PG passer behind LeBron.
     
  18. TheGM

    TheGM Contributing Member

    Joined:
    May 1, 2006
    Messages:
    1,248
    Likes Received:
    74
    Its basic math yet you don't realize it goes against your point. The 3 point shooter would have more points.
     
  19. khyberjones

    khyberjones Member

    Joined:
    Jan 11, 2014
    Messages:
    555
    Likes Received:
    27
    He also makes passes that result in FT attempts. Those don't recorded as assists but are just as effective.
     
  20. Amiga

    Amiga I get vaunted sacred revelations from social media
    Supporting Member

    Joined:
    Sep 18, 2008
    Messages:
    21,710
    Likes Received:
    18,486
    Definitely!, but that's exactly what make a smart PG a smart PG. Know who and where and when to pass. At the end of the days, the net points you generate is all that count to YOUR team. So pass wisely.
     

Share This Page

  • About ClutchFans

    Since 1996, ClutchFans has been loud and proud covering the Houston Rockets, helping set an industry standard for team fan sites. The forums have been a home for Houston sports fans as well as basketball fanatics around the globe.

  • Support ClutchFans!

    If you find that ClutchFans is a valuable resource for you, please consider becoming a Supporting Member. Supporting Members can upload photos and attachments directly to their posts, customize their user title and more. Gold Supporters see zero ads!


    Upgrade Now