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!

A "net possessions created" stat (applied to Rockets)

Discussion in 'Houston Rockets: Game Action & Roster Moves' started by durvasa, Nov 14, 2007.

  1. Zfan

    Zfan Member

    Joined:
    Nov 7, 2007
    Messages:
    754
    Likes Received:
    8
    What if neglect the missing FG?
    I assume FG attempt is nothing to do with possession.
    Making a FG means losing a possession. Missing a FG gives a loose ball, which may result in losing a possession, or keeping a possession if you got an offensvie rebound. Since the OR factor is considered already (0.7*OR), i suggest to neglect the missing FG.
     
  2. wnes

    wnes Contributing Member

    Joined:
    Feb 19, 2003
    Messages:
    8,196
    Likes Received:
    19
    LOL ... no. It means a possession is not wasted.
     
  3. foo82

    foo82 Member

    Joined:
    Dec 4, 2006
    Messages:
    923
    Likes Received:
    31
    Overahaul? After one feaking loss you expect an overhaul?
    What the hell is wrong with you fans?

    Adelman and his coaching staff are ecstatic over our pf situation. I'm sure you couch coaches have a much greater understanding of basketball than our GMs and coaching staff do.
     
  4. TeamUSA

    TeamUSA Member

    Joined:
    Aug 2, 2006
    Messages:
    2,770
    Likes Received:
    5
    Player Team Min PC/40MIN
    1 Francis, STEVE hou 0 0.00000000000
     
  5. kokopuffs

    kokopuffs Member

    Joined:
    Oct 19, 2006
    Messages:
    1,637
    Likes Received:
    31
    you should standardize the FG-missed against league average eFG or something.
     
  6. Zfan

    Zfan Member

    Joined:
    Nov 7, 2007
    Messages:
    754
    Likes Received:
    8
    IMHO. This equation is focus on possession. A possession is wasted or not should be out of consideration.
    Thanks duvasa. Very cool stats... :)
     
  7. durvasa

    durvasa Contributing Member

    Joined:
    Feb 11, 2006
    Messages:
    37,999
    Likes Received:
    15,462
    What I wanted to do was isolate a particular type of contribution. That McGrady scores poorly by this metric doesn't mean he's a bad player overall. If you want to win, it's not enough to just get your team the ball. You need to efficiently use those possessions. And without McGrady and Yao, we would be horrible at generating points from the possessions.

    I just wanted to highlight some things players can do that often get overlooked in the stats. McGrady and Yao get plenty of praise already for what they do.

    At the same time, I recognize the issue with this measurement. A player can essentially stand around and do nothing, and rate well above average. Some might even argue that's what Battier does (which goes back to the limitation of box score stats in capturing what a player really does).

    I could eliminate blocked shots and missed field goal attempts from the metric and just focus on rebounding, steals, and turnovers.
     
  8. anitasri

    anitasri Member

    Joined:
    Jul 28, 2002
    Messages:
    835
    Likes Received:
    0
    Missed 3 pointers that lead to a fast break?
     
  9. kokopuffs

    kokopuffs Member

    Joined:
    Oct 19, 2006
    Messages:
    1,637
    Likes Received:
    31
    Wait, no, it'd make more sense to do it against league average defensive rebounding %.
     
  10. Xerobull

    Xerobull You son of a b!tch! I'm in!

    Joined:
    Jun 18, 2003
    Messages:
    33,409
    Likes Received:
    30,984
    This is an awesome stat. Someone forward this to Morey? Waitaminute....Isn't Morey a big stat guy?

    Durvasa=Morey?
     
  11. durvasa

    durvasa Contributing Member

    Joined:
    Feb 11, 2006
    Messages:
    37,999
    Likes Received:
    15,462
    After considering your feedback, I decided to change the formula. I want to focus on net possession creation, and not how those possessions are used to generate points. Therefore, I decided to eliminate blocked shots and missed field goals from the equation. Blocking shots impacts how the opposing team uses possessions, but not directly on creating a possession for your team. Chuck might be penalized a bit by this exclusion, actually, because a lot of what gets recorded as a "blocked shot" for him are really strip moves he does on players he's guarding in the lane. They could just as easily be recorded as a steal. Oh well.

    This is the modified formula:

    net_poss_created = 0.7*oreb + 0.3*dreb + stl - tov

    Here are the league leaders (minimum 80 minutes played):

    Code:
            [B]Player              Team      Min      PC/40Min[/B]
    1       hayes,chuck         hou       190       7.2
    2       skinner,brian       pho       90        7.1
    3       evans,reggie        phi       175       6.5
    4       bynum,andrew        lal       144       6.5
    5       marion,shawn        pho       297       6.5
    6       smith,joe           chi       118       6.3
    7       okafor,emeka        cha       235       6.1
    8       camby,marcus        den       272       5.9
    9       hollins,ryan        cha       88        5.7
    10      collison,nick       sea       217       5.7
    11      foster,jeff         ind       164       5.5
    12      haywood,brendan     was       175       5.5
    13      dudley,jared        cha       86        4.8
    14      jefferson,al        min       184       4.7
    15      ilgauskas,zydrun    cle       268       4.7
    16      garnett,kevin       bos       238       4.7
    17      smith,jason         phi       106       4.3
    18      millsap,paul        uta       161       4.2
    19      randolph,zach       nyk       146       4.2
    20      najera,eduardo      den       178       4.1
    
    And here are the stats for the Rockets:

    Code:
            [B]Player              Team      Min      PC/40min[/B]
    1       hayes,chuck         hou       190       7.2
    2       mutombo,dikembe     hou       31        4.8
    3       wells,bonzi         hou       177       3.5
    4       scola,luis          hou       137       2.7
    5       battier,shane       hou       288       2.0
    6       ming,yao            hou       302       1.4
    7       james,mike          hou       193       1.2
    8       mcgrady,tracy       hou       302       0.8
    9       alston,rafer        hou       207       0.7
    10      head,luther         hou       92        0.3
    
    Now Chuck is leading the league. Looks good to me. :D

    BTW, for those who are curious, here were the league leaders last year:

    Code:
            [B]Player              Team      Min       PC/40min[/B]
    1       foster,jeff         ind       1742      6.3
    2       mutombo,dikembe     hou       1291      6.1
    3       hayes,chuck         hou       1716      5.7
    4       wallace,ben         chi       2699      5.6
    5       evans,reggie        den       1129      5.2
    6       varejao,anderson    cle       1931      5.2
    7       balkman,renaldo     nyk       1065      5.1
    8       lee,david           nyk       1729      5.0
    9       chandler,tyson      nor       2510      4.9
    10      roberts,lawrence    mem       974       4.9
    11      diop,desagana       dal       1473      4.7
    12      okafor,emeka        cha       2330      4.7
    13      gooden,drew         cle       2239      4.7
    14      millsap,paul        uta       1473      4.7
    15      biedrins,andris     gsw       2378      4.7
    16      mcdyess,antonio     det       1727      4.7
    17      camby,marcus        den       2371      4.6
    18      marion,shawn        pho       3007      4.6
    19      mohammed,nazr       det       776       4.6
    20      thomas,kurt         pho       1213      4.5
    
     
  12. kokopuffs

    kokopuffs Member

    Joined:
    Oct 19, 2006
    Messages:
    1,637
    Likes Received:
    31
    durvasa, why do you use a 0.7/0.3 coefficient for the rebounding?
     
  13. weslinder

    weslinder Contributing Member

    Joined:
    Jun 27, 2006
    Messages:
    12,983
    Likes Received:
    291
    durvasa,
    I like what you're doing, but I have a problem with the terminology. It doesn't really measure possessions created. And traditionally, an offensive rebound is considered as extending a possession.
     
  14. kokopuffs

    kokopuffs Member

    Joined:
    Oct 19, 2006
    Messages:
    1,637
    Likes Received:
    31
    I think it makes sense in this context because an o-reb (usually) results in a new 24-second shotclock. Granted, it's a little different than a new possession after the opponent scores, but I don't see a big problem with treating o-rebs as a "new" possession.
     
  15. Zfan

    Zfan Member

    Joined:
    Nov 7, 2007
    Messages:
    754
    Likes Received:
    8
    Chuck is the man. This stats is awesome :)
     
  16. NIKEstrad

    NIKEstrad Contributing Member
    Supporting Member

    Joined:
    Mar 21, 2000
    Messages:
    10,061
    Likes Received:
    3,777
    durvasa -- Any idea who the per game leaders are (not minute adjusted)? I'd imagine Marion, Camby, and Okafor are at the top, which makes plenty of sense to me.

    To get an idea of how effective Hayes was last night when the refs weren't handcuffing him, he registered 4 steals and 0 turnovers, along with 1 off. reb and 4 def. rebs -- all in 19 minutes. That's a 5.9 npc, adjusted to 40 minutes makes it a 12.42! Not surprisingly, we were +7 with Chuck out there or +14.7/40 minutes.

    One guy who I'm surprised isn't on these lists is Kirilenko -- he misses out mainly due to turnovers, I'd imagine.
     
  17. wnes

    wnes Contributing Member

    Joined:
    Feb 19, 2003
    Messages:
    8,196
    Likes Received:
    19
    I am not concerned about overall greatness of McGrady or Yao, so let's not sidetrack the topic at hand.

    I also would not quickly dismiss the impact of missed field goal attempts on possessions. Conceptually, we consider two aspects: creation of possessions, and preservation of possessions. I think you have covered both pretty well.

    Steals and defensive rebounding are directly and positively related to creation of possessions. Offensive rebounding is positively related to preservation of possession, while turnovers and missed field goals are negatively related to preservation of possessions. Made field goals indirectly (and positively) affect preservation of possessions by offsetting negative effect of missed field goal attempts. I am not sure how positive blocked shots are to creating possessions so I would not object to leaving it out.

    One defensive rebound is equivalent to one steal as far as possessions are considered, so I would give the metric a coefficient of "1," the same as that of steal. "-1" is good for turnover. Is preservation of possession just as significant as creation of possession? If yes, then offensive rebounding should be weighed the same as defensive rebounding.

    Now onto the last term in your original formula. I think it should definitely include both missed FGs and made FGs. How to weigh them is up to debate.
     
    #37 wnes, Nov 14, 2007
    Last edited: Nov 14, 2007
  18. durvasa

    durvasa Contributing Member

    Joined:
    Feb 11, 2006
    Messages:
    37,999
    Likes Received:
    15,462
    Here's my rationale, but it's open to debate. There's a roughly 30% chance a team will grab the offensive rebound on their miss. So, on a given miss, when a player gets the defensive board, he's overcome that 30% probability to help secure the possession for his team. And I use a similar rationale for offensive rebounding (overcome the 70% probability of the opposing team getting the defensive board).
     
  19. GATER

    GATER Contributing Member

    Joined:
    Jun 25, 2000
    Messages:
    8,325
    Likes Received:
    78
    The formula tells me Chuck is an incredibly valuable player.

    My eyes tell me he barely belongs in the NBA.

    For me, seeing is believing.
     
  20. durvasa

    durvasa Contributing Member

    Joined:
    Feb 11, 2006
    Messages:
    37,999
    Likes Received:
    15,462
    Agree. I couldn't think of another way to succinctly describe it, though. Any suggestions?
     

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