Loading ...
Sorry, an error occurred while loading the content.

Bob's examples

Expand Messages
  • Dean Oliver
    Gotta run pretty quick, but here are some of my # s for the guys Bob calculated. I ve always figured that how efficiently a player performs is a function of
    Message 1 of 6 , Oct 10, 2002
    • 0 Attachment

       

      Gotta run pretty quick, but here are some of my #’s for the guys Bob calculated.  I’ve always figured that how efficiently a player performs is a function of what percentage of the team’s possessions they use (% Tm Poss below), their age (improving until age 26 or 27 usually and declining around age 30), and the quality of their teammates (Fox plays a lot better because he has Shaq taking defenders away from him).  Makes it difficult to evaluate everything.  But look for yourself:

       

      Hersey Hawkins (shorter 3 pt line in ’95 and ’96)

      Season

      Tm

      G

      Sc. Poss

      Poss

      Floor %

      Off. Rtg.

      Pts Prod/G

      % Team Poss

      Age

      1989

      PHI

      79

      519

      1028

      0.51

      108

      14.0

      19%

      23

      1990

      PHI

      82

      657

      1248

      0.53

      114

      17.3

      22%

      24

      1991

      PHI

      80

      752

      1431

      0.53

      114

      20.4

      23%

      25

      1992

      PHI

      81

      651

      1248

      0.52

      113

      17.4

      21%

      26

      1993

      PHI

      81

      697

      1315

      0.53

      116

      18.8

      22%

      27

      1994

      CHA

      82

      511

      990

      0.52

      112

      13.5

      18%

      28

      1995

      CHA

      82

      480

      934

      0.51

      116

      13.2

      18%

      29

      1996

      SEA

      82

      511

      1013

      0.50

      114

      14.1

      18%

      30

      1997

      SEA

      82

      466

      883

      0.53

      121

      13.0

      17%

      31

      1998

      SEA

      82

      369

      719

      0.51

      118

      10.3

      15%

      32

      1999

      SEA

      50

      219

      456

      0.48

      109

      9.9

      15%

      33

      2000

      CHI

      61

      201

      449

      0.45

      102

      7.5

      15%

      34

      2001

      CHA

      59

      87

      161

      0.54

      120

      3.3

      13%

      35

      Became a role player in ’94 when he switched teams.  His rating didn’t improve, but he maintained the high ratings into his 30’s due to closer line and being a role player (and probably better teammates taking the pressure off in Seattle).

       

      Barkley

       

      Season

      Tm

      G

      Sc. Poss

      Poss

      Floor %

      Off. Rtg.

      Pts Prod/G

      % Team Poss

      Age

      1985

      PHI

      82

      572

      1006

      0.57

      114

      14.0

      20%

      21

      1986

      PHI

      80

      813

      1453

      0.56

      111

      20.3

      24%

      22

      1987

      PHI

      68

      808

      1381

      0.58

      119

      24.1

      25%

      23

      1988

      PHI

      80

      1056

      1713

      0.62

      125

      26.9

      27%

      24


      (Message over 64 KB, truncated)

    • bchaikin@aol.com
      deano - what does your Poss column represent? i m guessing player possessions, but for example the highest number i saw was for barkley (87-88 with 1713/80)
      Message 2 of 6 , Oct 11, 2002
      • 0 Attachment
        deano -

        what does your "Poss" column represent? i'm guessing player possessions, but for example the highest number i saw was for barkley (87-88 with 1713/80) which is 21-22 possessions per game or just 5-6 possessions per quarter....

        bob chaikin
        bchaikin@...
      • Dean Oliver
        ... possessions, but ... 1713/80) ... quarter.... ... A possession is different than a touch. A touch is pretty obvious. It s what you estimate. A guy
        Message 3 of 6 , Oct 11, 2002
        • 0 Attachment
          --- In APBR_analysis@y..., bchaikin@a... wrote:
          > deano -
          >
          > what does your "Poss" column represent? i'm guessing player
          possessions, but
          > for example the highest number i saw was for barkley (87-88 with
          1713/80)
          > which is 21-22 possessions per game or just 5-6 possessions per
          quarter....
          >
          > bob chaikin
          > bchaikin@b...

          A possession is different than a touch. A touch is pretty obvious.
          It's what you estimate. A guy touches the ball a lot more than he
          actually does something that contributes to the end of a team
          possession. A lot of passes are just moving the ball on to the next
          guy - neutral things that I can't afford to really care about.
          Individual possessions track only when a player contributes to the
          end of a team possession, a shot that isn't rebounded by the offense
          or a turnover mainly. I've devised individual possessions so that
          the sum of them add up pretty darn close to the actual number of team
          possessions (minus 5-s calls, 8-s calls, and 24-s calls). That means
          that a team possession where an assist is credited on a made basket --
          part of the possession goes to the assistant and part to the guy who
          made the shot, so that it adds to 1. Guys who make better passes and
          help their teammates to higher shooting percentages get more credit
          for their assists than guys who make safe passes. It's not a huge
          adjustment but it can matter.

          Anyway, your touch count should not add up at all to my individual
          possession count. I don't want every touch to be factored into an
          evaluation of how efficient a player is with the ball. If I counted
          passes that did nothing, that would be a very different stat. Not
          sure what it would mean. I basically try to capture how efficient
          players are WHEN THEY TRY TO DO SOMETHING WITH THE BALL. Then I also
          track how often they try to do something witht he ball. Stackhouse
          and Iverson often try to do something with the ball. They aren't
          very efficient with it. Jordan often tried to do something with the
          ball. And he was damned efficient with it. Brent Barry doesn't
          often try to do something with the ball, but he's very efficient when
          he tries. Dickie Simpkins doesn't often try to do something with the
          ball and he ain't efficient. He's got to be in the league for
          another reason besides offense.

          DeanO
        • bchaikin@aol.com
          I ve devised individual possessions so that the sum of them add up pretty darn close to the actual number of team possessions. ok.......gotcha.......makes
          Message 4 of 6 , Oct 11, 2002
          • 0 Attachment
            I've devised individual possessions so that the sum of them add up pretty darn close to the actual number of team possessions.

            ok.......gotcha.......makes perfect sense.....

            on another note...

            Guys who make better passes and help their teammates to higher shooting percentages get more credit for their assists than guys who make safe passes.

            do you believe, or have data, or have seen data to show that some players do indeed actually increase the FG%s of their teammates thru their passing? i had always assumed this was true but did a small study for the Nets almost a decade ago showing that while good passers do indeed increase the opportunities for their teammates to score, there was no evidence to show that they did in fact increase their FG%s (i.e. did not get them better shot selection thru their passing skills)...

            what i did was to look at all of the high assist men (PGs with high assist totals) that had been traded during their career over a 15-16 year period (late 70s to early 90s), and looked at the FG%s of their teammates before their arrival and after, and could not find a single PG that consistently increased FG% of teammates. there were numerous cases where PGs did just that at one stop but not the next, didn't at their 1st stop but did their 2nd, etc, but no data to show that any single player or players always or most of the time increased their teammates FG%s....

            what i had found (but i can't find the data now) was - as i remember it - that some players (PGs) could increase their teamates scoring avgs by getting them more opportunities to score, which makes sense if you think about it. great passers like magic and stockton shoot so infrequently and pass so often in relation to how often they handle the ball, that since they handle the ball so often in the 1st place their teammates get more chances to score. i couldn't use magic and stockton in the study because they played their entire careers with one team, but i did look at players traded to the teams of the great PGs (and possibly traded away) and looked at their FG%s before and after. again no consistent pattern as to increases or decreases in FG%s....

            haven't done a similar study to this yet for the 1990s, but i'd look at players like cassell, marbury, stoudamire, etc - PGs traded once or twice that were starters wherever they went...

            bob chaikin
            bchaikin@...










          • Dean Oliver
            ... shooting ... safe passes. ... players do ... passing? i ... almost a ... to show ... better shot ... It is a belief, a prior assumption, something that
            Message 5 of 6 , Oct 12, 2002
            • 0 Attachment
              --- In APBR_analysis@y..., bchaikin@a... wrote:
              > Guys who make better passes and help their teammates to higher
              shooting
              > percentages get more credit for their assists than guys who make
              safe passes.
              >
              > do you believe, or have data, or have seen data to show that some
              players do
              > indeed actually increase the FG%s of their teammates thru their
              passing? i
              > had always assumed this was true but did a small study for the Nets
              almost a
              > decade ago showing that while good passers do indeed increase the
              > opportunities for their teammates to score, there was no evidence
              to show
              > that they did in fact increase their FG%s (i.e. did not get them
              better shot
              > selection thru their passing skills)...
              >

              It is a belief, a prior assumption, something that does wait for
              proof. I did a very short study once that suggested that assists do
              not, as a class, increase shooting percentages. Basically, guys got
              easy shots as much on their own as off of passes. It wasn't as
              thorough as I'd like it to have been and, like your study, I have
              doubts about its results. If assists don't increase the odds of
              making a basket, then why should anyone get credit for them?

              So, my answer is that I don't have evidence or proof of it, kinda
              like there is no evidence that batting order makes a difference in
              scoring runs in baseball. But people do definitely construct batting
              orders thinking they matter.

              There are definite old wives tales in sports. Some of them are
              difficult to prove but we hold on to anyway. Some of them can be
              pretty adequately dispelled. An interesting example is the concept
              of great players making their teammates better. I've generally found
              that great players make _some_ teammates better, but not necessarily
              all. Magic, Bird, and Jordan all had definite positive effects on
              some teammates. But there were plenty of players who came through
              that didn't play any better with them.

              Your test for whether certain guys improve their teammates seems like
              a decent one. But there are other tests. Maybe some assist men make
              only occasional players better shooters. That is something to
              consider. But let me think about better tests.

              DeanO

              > what i did was to look at all of the high assist men (PGs with high
              assist
              > totals) that had been traded during their career over a 15-16 year
              period
              > (late 70s to early 90s), and looked at the FG%s of their teammates
              before
              > their arrival and after, and could not find a single PG that
              consistently
              > increased FG% of teammates. there were numerous cases where PGs did
              just that
              > at one stop but not the next, didn't at their 1st stop but did
              their 2nd,
              > etc, but no data to show that any single player or players always
              or most of
              > the time increased their teammates FG%s....
              >
              > what i had found (but i can't find the data now) was - as i
              remember it -
              > that some players (PGs) could increase their teamates scoring avgs
              by getting
              > them more opportunities to score, which makes sense if you think
              about it.
              > great passers like magic and stockton shoot so infrequently and
              pass so often
              > in relation to how often they handle the ball, that since they
              handle the
              > ball so often in the 1st place their teammates get more chances to
              score. i
              > couldn't use magic and stockton in the study because they played
              their entire
              > careers with one team, but i did look at players traded to the
              teams of the
              > great PGs (and possibly traded away) and looked at their FG%s
              before and
              > after. again no consistent pattern as to increases or decreases in
              FG%s....
              >
              > haven't done a similar study to this yet for the 1990s, but i'd
              look at
              > players like cassell, marbury, stoudamire, etc - PGs traded once or
              twice
              > that were starters wherever they went...
              >
              > bob chaikin
              > bchaikin@b...
            • Mike G
              See posts #631-646 on the Kidd Effect. I began this study entirely skeptical, but became convinced that at least one PG raises his teammates shooting pcts.
              Message 6 of 6 , Oct 12, 2002
              • 0 Attachment
                See posts #631-646 on the Kidd Effect. I began this study entirely
                skeptical, but became convinced that at least one PG raises his
                teammates' shooting pcts.


                --- In APBR_analysis@y..., "Dean Oliver" <deano@r...> wrote:
                > --- In APBR_analysis@y..., bchaikin@a... wrote:
                > > Guys who make better passes and help their teammates to higher
                > shooting
                > > percentages get more credit for their assists than guys who make
                > safe passes.
                > >
                > > do you believe, or have data, or have seen data to show that some
                > players do
                > > indeed actually increase the FG%s of their teammates thru their
                > passing? i
                > > had always assumed this was true but did a small study for the
                Nets
                > almost a
                > > decade ago showing that while good passers do indeed increase the
                > > opportunities for their teammates to score, there was no evidence
                > to show
                > > that they did in fact increase their FG%s (i.e. did not get them
                > better shot
                > > selection thru their passing skills)...
                > >
                >
                > It is a belief, a prior assumption, something that does wait for
                > proof. I did a very short study once that suggested that assists
                do
                > not, as a class, increase shooting percentages. Basically, guys
                got
                > easy shots as much on their own as off of passes. It wasn't as
                > thorough as I'd like it to have been and, like your study, I have
                > doubts about its results. If assists don't increase the odds of
                > making a basket, then why should anyone get credit for them?
                >
                > So, my answer is that I don't have evidence or proof of it, kinda
                > like there is no evidence that batting order makes a difference in
                > scoring runs in baseball. But people do definitely construct
                batting
                > orders thinking they matter.
                >
                > There are definite old wives tales in sports. Some of them are
                > difficult to prove but we hold on to anyway. Some of them can be
                > pretty adequately dispelled. An interesting example is the concept
                > of great players making their teammates better. I've generally
                found
                > that great players make _some_ teammates better, but not
                necessarily
                > all. Magic, Bird, and Jordan all had definite positive effects on
                > some teammates. But there were plenty of players who came through
                > that didn't play any better with them.
                >
                > Your test for whether certain guys improve their teammates seems
                like
                > a decent one. But there are other tests. Maybe some assist men
                make
                > only occasional players better shooters. That is something to
                > consider. But let me think about better tests.
                >
                > DeanO
                >
                > > what i did was to look at all of the high assist men (PGs with
                high
                > assist
                > > totals) that had been traded during their career over a 15-16
                year
                > period
                > > (late 70s to early 90s), and looked at the FG%s of their
                teammates
                > before
                > > their arrival and after, and could not find a single PG that
                > consistently
                > > increased FG% of teammates. there were numerous cases where PGs
                did
                > just that
                > > at one stop but not the next, didn't at their 1st stop but did
                > their 2nd,
                > > etc, but no data to show that any single player or players always
                > or most of
                > > the time increased their teammates FG%s....
                > >
                > > what i had found (but i can't find the data now) was - as i
                > remember it -
                > > that some players (PGs) could increase their teamates scoring
                avgs
                > by getting
                > > them more opportunities to score, which makes sense if you think
                > about it.
                > > great passers like magic and stockton shoot so infrequently and
                > pass so often
                > > in relation to how often they handle the ball, that since they
                > handle the
                > > ball so often in the 1st place their teammates get more chances
                to
                > score. i
                > > couldn't use magic and stockton in the study because they played
                > their entire
                > > careers with one team, but i did look at players traded to the
                > teams of the
                > > great PGs (and possibly traded away) and looked at their FG%s
                > before and
                > > after. again no consistent pattern as to increases or decreases
                in
                > FG%s....
                > >
                > > haven't done a similar study to this yet for the 1990s, but i'd
                > look at
                > > players like cassell, marbury, stoudamire, etc - PGs traded once
                or
                > twice
                > > that were starters wherever they went...
                > >
                > > bob chaikin
                > > bchaikin@b...
              Your message has been successfully submitted and would be delivered to recipients shortly.