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

DJ and Westphal

Expand Messages
  • Dean Oliver
    I was barely 10 yrs old when Westphal had his last good year, so forgive my memory. His numbers were quite good before he went to Seattle.... Scor. Poss.
    Message 1 of 4 , Mar 9, 2001
    • 0 Attachment
      I was barely 10 yrs old when Westphal had his last good year, so
      forgive my memory. His numbers were quite good before he went to
      Seattle....

      Scor. Poss. Floor RTG Points
      Poss. Pct. Prod.
      1978 PHX 982 1802 0.545 107.1 1930
      1979 PHX 969 1686 0.575 113.6 1914
      1980 PHX 862 1507 0.572 114.7 1729
      1981 SEA 301 568 0.531 105.1 597
      1982 NY 119 240 0.498 98.8 237
      1983 NY 458 913 0.501 100.2 915
      1984 PHX 219 414 0.529 105.4 436
      3911 7129 0.549 108.8 7758

      Def. Net Net Net Net
      Rtg. Win% W L Pts
      1978 PHX 99.7 0.764 9.6 3.0 132.6
      1979 PHX 105.3 0.776 9.6 2.8 138.8
      1980 PHX 104.6 0.822 9.9 2.1 153.2
      1981 SEA 106.1 0.461 2.2 2.6 -5.6
      1982 NY 111.1 0.127 0.3 1.8 -29.4
      1983 NY 102.0 0.430 3.6 4.8 -15.7
      1984 PHX 109.8 0.338 1.2 2.4 -18.2
      105.5 0.652 36 19 355.6

      DJ, on the other hand, doesn't look so good. This is at least
      partially due to his "Joe Dumars" factor, which is the fact that his
      defensive ability doesn't get reflected in numbers. He basically shut
      down guys by limiting their touches and their shots, not by getting
      steals and blocks. As a consequence, the numbers below may not be
      truly reflective:

      Scor. Poss. Floor RTG Points
      Poss. Pct. Prod.
      1978 SEA 537 1042 0.515 98.7 1029
      1979 SEA 647 1255 0.516 100.3 1259
      1980 SEA 779 1516 0.514 100.8 1527
      1981 PHX 736 1397 0.527 103.7 1449
      1982 PHX 783 1465 0.535 105.0 1538
      1983 PHX 584 1110 0.527 103.1 1144
      1984 BOS 556 1056 0.527 104.3 1101
      1985 BOS 693 1293 0.536 107.2 1386
      1986 BOS 652 1227 0.532 105.8 1298
      1987 BOS 623 1179 0.528 106.4 1255
      1988 BOS 583 1097 0.531 108.2 1188
      1989 BOS 438 851 0.514 102.6 874
      1990 BOS 357 660 0.541 108.4 715
      7970 15148 0.526 104.1 15762

      Def. Net Net Net Net
      Rtg. Win% W L Pts
      1978 SEA 98.4 0.510 4.7 4.5 2.6
      1979 SEA 100.2 0.502 5.7 5.6 0.7
      1980 SEA 100.6 0.507 6.6 6.5 2.7
      1981 PHX 99.6 0.658 7.7 4.0 56.4
      1982 PHX 103.6 0.555 7.1 5.7 20.4
      1983 PHX 100.9 0.588 6.1 4.3 24.4
      1984 BOS 105.4 0.457 4.6 5.5 -11.6
      1985 BOS 107.8 0.477 5.7 6.2 -7.9
      1986 BOS 104.4 0.558 6.2 4.9 18.1
      1987 BOS 108.7 0.414 4.7 6.6 -26.6
      1988 BOS 110.8 0.404 4.2 6.3 -28.2
      1989 BOS 110.3 0.235 2.0 6.6 -64.9
      1990 BOS 108.9 0.482 3.5 3.7 -3.2
      104.6 0.494 69 70 -17.0

      One thing I do when dealing with "Dumars-like" players is throw out
      their indiv. defensive stats and just work with the team defensive
      stats. Doing this for DJ helps him quite a bit, but Westphal still
      looks like the player who peaked higher....

      Def. Offensive Net
      Rtg. Win% Wins Losses Pts
      1978 SEA 95.8 0.621 6.0 3.7 30.2
      1979 SEA 98.1 0.592 7.0 4.8 28.0
      1980 SEA 99.1 0.567 8.2 6.3 24.9
      1981 PHX 97.6 0.732 9.6 3.5 85.7
      1982 PHX 100.8 0.662 9.5 4.9 61.3
      1983 PHX 98.3 0.686 7.3 3.4 52.8
      1984 BOS 102.3 0.579 5.9 4.3 21.0
      1985 BOS 104.4 0.608 7.6 4.9 36.3
      1986 BOS 100.9 0.689 8.1 3.7 61.0
      1987 BOS 105.0 0.555 6.5 5.2 16.7
      1988 BOS 107.4 0.533 5.8 5.1 9.4
      1989 BOS 107.5 0.316 2.7 5.8 -41.8
      1990 BOS 106.1 0.588 3.9 2.7 15.2
      0.602 88 58 400.7

      In this scenario, taking DJ to be an average defender on all his clubs
      (which, for complicated reasons, is not unreasonable for Dumars-like
      players), gives him a win% closer to that of Westphal and a career
      value of net points that is a little higher than Westphal's partial
      career numbers.

      Generally, it looks like Westphal was generally a better contributor.
      But stats (even now) don't adequately represent Dumars-like players
      like DJ. And, with the exception of the very best players, the need
      of a team -- the fit -- helps determine how valuable a player is.

      Dean Oliver
      Journal of Basketball Studies
    • harlanzo@yahoo.com
      It is tough to evaluate DJ because he apparently has the dreaded intangibles label where we know he played with winners and we are pretty sure that good
      Message 2 of 4 , Mar 9, 2001
      • 0 Attachment
        It is tough to evaluate DJ because he apparently has the
        dreaded "intangibles" label where we know he played with winners and
        we are pretty sure that good karma should be worth something but the
        numbers just don't bear it out.

        Evaluating DJ next to prolific scorer like Westphal invariably
        results in the scorer winning the statistical evaluation. However,
        DJ has some impressive accomplishments on his resume. The best
        player on a championship team (granted it was a marginal team and he
        may not actually be better than Sikma) and as a solid contributor on
        the revered 80s Celtics. These accomplishments may have a value that
        is not wholly reflected in statistics.

        The question of course is whether the formulas should be readjusted
        to reflect these strengths or whether these strengths are
        overstated. Being the best player on the champion is impressive but
        the state of the late 70s NBA could also indicate that the sonics
        were kind of a fluke team and that though the DJ-sonics were good
        that team would have never even sniffed a championship during the
        Bird-Magic or other storied NBA eras.

        As for the DJ's role on the Celtics, I remember as a kid being
        favorably impressed with DJ's defense and passing on the celtics. It
        was clear that they did not need him to score 25 a game. Indeed, had
        you placed Westphal (at or near peak value) on those celts it
        probably would not have been as efficient to run the bulk of the
        scoring thru him rather than Bird and Herman Munster Mchale. So it
        is possible to argue that DJ's ability to fill a niche had a value to
        the celtics in excess of what westphal could have given them.

        So how then to reflect this value? I have toyed with the idea of
        creating a system called "karma points" that add extra value to
        players on winning teams to adjust formulas. (This really just an
        extended version of what others of you have done when you might award
        points to a player making an all-defensive team for example). You
        can award points for significant contributions of complementary
        players on playoff bound and championship caliber teams. This is of
        course potentially arbitrary in application. Indeed, the formula
        might bear out that aaron mckie is too close in value to stackhouse
        or that avery johnson or eric snow have a value over marbury.
        However, if the points more carefully scrutinized (I'm still working
        out the details) it might bridge the gap that stats won't always
        reflect.







        --- In APBR_analysis@y..., "Dean Oliver" <deano@t...> wrote:
        >
        > I was barely 10 yrs old when Westphal had his last good year, so
        > forgive my memory. His numbers were quite good before he went to
        > Seattle....
        >
        > Scor. Poss. Floor RTG Points
        > Poss. Pct. Prod.
        > 1978 PHX 982 1802 0.545 107.1 1930
        > 1979 PHX 969 1686 0.575 113.6 1914
        > 1980 PHX 862 1507 0.572 114.7 1729
        > 1981 SEA 301 568 0.531 105.1 597
        > 1982 NY 119 240 0.498 98.8 237
        > 1983 NY 458 913 0.501 100.2 915
        > 1984 PHX 219 414 0.529 105.4 436
        > 3911 7129 0.549 108.8 7758
        >
        > Def. Net Net Net Net
        > Rtg. Win% W L Pts
        > 1978 PHX 99.7 0.764 9.6 3.0 132.6
        > 1979 PHX 105.3 0.776 9.6 2.8 138.8
        > 1980 PHX 104.6 0.822 9.9 2.1 153.2
        > 1981 SEA 106.1 0.461 2.2 2.6 -5.6
        > 1982 NY 111.1 0.127 0.3 1.8 -29.4
        > 1983 NY 102.0 0.430 3.6 4.8 -15.7
        > 1984 PHX 109.8 0.338 1.2 2.4 -18.2
        > 105.5 0.652 36 19 355.6
        >
        > DJ, on the other hand, doesn't look so good. This is at least
        > partially due to his "Joe Dumars" factor, which is the fact that
        his
        > defensive ability doesn't get reflected in numbers. He basically
        shut
        > down guys by limiting their touches and their shots, not by getting
        > steals and blocks. As a consequence, the numbers below may not be
        > truly reflective:
        >
        > Scor. Poss. Floor RTG Points
        > Poss. Pct. Prod.
        > 1978 SEA 537 1042 0.515 98.7 1029
        > 1979 SEA 647 1255 0.516 100.3 1259
        > 1980 SEA 779 1516 0.514 100.8 1527
        > 1981 PHX 736 1397 0.527 103.7 1449
        > 1982 PHX 783 1465 0.535 105.0 1538
        > 1983 PHX 584 1110 0.527 103.1 1144
        > 1984 BOS 556 1056 0.527 104.3 1101
        > 1985 BOS 693 1293 0.536 107.2 1386
        > 1986 BOS 652 1227 0.532 105.8 1298
        > 1987 BOS 623 1179 0.528 106.4 1255
        > 1988 BOS 583 1097 0.531 108.2 1188
        > 1989 BOS 438 851 0.514 102.6 874
        > 1990 BOS 357 660 0.541 108.4 715
        > 7970 15148 0.526 104.1 15762
        >
        > Def. Net Net Net Net
        > Rtg. Win% W L Pts
        > 1978 SEA 98.4 0.510 4.7 4.5 2.6
        > 1979 SEA 100.2 0.502 5.7 5.6 0.7
        > 1980 SEA 100.6 0.507 6.6 6.5 2.7
        > 1981 PHX 99.6 0.658 7.7 4.0 56.4
        > 1982 PHX 103.6 0.555 7.1 5.7 20.4
        > 1983 PHX 100.9 0.588 6.1 4.3 24.4
        > 1984 BOS 105.4 0.457 4.6 5.5 -11.6
        > 1985 BOS 107.8 0.477 5.7 6.2 -7.9
        > 1986 BOS 104.4 0.558 6.2 4.9 18.1
        > 1987 BOS 108.7 0.414 4.7 6.6 -26.6
        > 1988 BOS 110.8 0.404 4.2 6.3 -28.2
        > 1989 BOS 110.3 0.235 2.0 6.6 -64.9
        > 1990 BOS 108.9 0.482 3.5 3.7 -3.2
        > 104.6 0.494 69 70 -17.0
        >
        > One thing I do when dealing with "Dumars-like" players is throw out
        > their indiv. defensive stats and just work with the team defensive
        > stats. Doing this for DJ helps him quite a bit, but Westphal still
        > looks like the player who peaked higher....
        >
        > Def. Offensive Net
        > Rtg. Win% Wins Losses Pts
        > 1978 SEA 95.8 0.621 6.0 3.7 30.2
        > 1979 SEA 98.1 0.592 7.0 4.8 28.0
        > 1980 SEA 99.1 0.567 8.2 6.3 24.9
        > 1981 PHX 97.6 0.732 9.6 3.5 85.7
        > 1982 PHX 100.8 0.662 9.5 4.9 61.3
        > 1983 PHX 98.3 0.686 7.3 3.4 52.8
        > 1984 BOS 102.3 0.579 5.9 4.3 21.0
        > 1985 BOS 104.4 0.608 7.6 4.9 36.3
        > 1986 BOS 100.9 0.689 8.1 3.7 61.0
        > 1987 BOS 105.0 0.555 6.5 5.2 16.7
        > 1988 BOS 107.4 0.533 5.8 5.1 9.4
        > 1989 BOS 107.5 0.316 2.7 5.8 -41.8
        > 1990 BOS 106.1 0.588 3.9 2.7 15.2
        > 0.602 88 58 400.7
        >
        > In this scenario, taking DJ to be an average defender on all his
        clubs
        > (which, for complicated reasons, is not unreasonable for Dumars-
        like
        > players), gives him a win% closer to that of Westphal and a career
        > value of net points that is a little higher than Westphal's partial
        > career numbers.
        >
        > Generally, it looks like Westphal was generally a better
        contributor.
        > But stats (even now) don't adequately represent Dumars-like
        players
        > like DJ. And, with the exception of the very best players, the
        need
        > of a team -- the fit -- helps determine how valuable a player is.
        >
        > Dean Oliver
        > Journal of Basketball Studies
      • Dean Oliver
        ... The big intangibles are in the defense, as far as I m concerned. But I would also point out that the Celtic defense fell apart after 1986 (Walton left and
        Message 3 of 4 , Mar 9, 2001
        • 0 Attachment
          --- In APBR_analysis@y..., harlanzo@y... wrote:
          > It is tough to evaluate DJ because he apparently has the
          > dreaded "intangibles" label where we know he played with winners and
          > we are pretty sure that good karma should be worth something but the
          > numbers just don't bear it out.

          The big intangibles are in the defense, as far as I'm concerned.
          But I would also point out that the Celtic defense fell apart
          after 1986 (Walton left and the other forwards got old).
          DJ's defense is not the kind that works without the help of
          big guys (unlike Jordan's defense, for example)..... Offensive
          intangibles that I'm trying to make tangible are primarily the fact
          that teams could not lay off him to double team -- but I think that is
          relatively minor compared to other players at his position (more
          important if you want to be a pain and say that Jerome Kersey had
          similar efficiency numbers so substitute him).

          >
          > Evaluating DJ next to prolific scorer like Westphal invariably
          > results in the scorer winning the statistical evaluation. However,

          Not true. There are plenty of prolific scorers who come out pretty
          negative by my count (Stackhouse, A. Walker, Marbury normally, Ron
          Harper in his early years, Iverson in past years, Tom Chambers several
          times). Their "intangible benefit" then becomes whether they create
          opportunities for others with all their shots (something I am starting
          to measure).

          > DJ has some impressive accomplishments on his resume. The best
          > player on a championship team (granted it was a marginal team and he
          > may not actually be better than Sikma) and as a solid contributor on
          > the revered 80s Celtics. These accomplishments may have a value
          that
          > is not wholly reflected in statistics.

          Sikma was better. That Sonic team was, by many people's take, a very
          odd champion. I think they started out the season 5-13.... Being a
          starter on a championship team, especially the 4th or 5th scorer,
          should not be over-rated either.

          >
          > As for the DJ's role on the Celtics, I remember as a kid being
          > favorably impressed with DJ's defense and passing on the celtics.
          It
          > was clear that they did not need him to score 25 a game. Indeed,
          had
          > you placed Westphal (at or near peak value) on those celts it
          > probably would not have been as efficient to run the bulk of the
          > scoring thru him rather than Bird and Herman Munster Mchale. So it
          > is possible to argue that DJ's ability to fill a niche had a value
          to
          > the celtics in excess of what westphal could have given them.
          >

          Just by saying that Bird, McHale, and Parish were the focuses of that
          team, I think it (properly) minimizes the value of DJ. What DJ was
          doing for that team, a few other guards could have done. Ainge kind
          of did it on his own, too. The Celtics won a title before he got
          there. Basically, the Celtics did not want a starting PG who shot too
          much and wasn't careful with the ball. They didn't need a great
          creator like a Stockton or Magic or Cheeks or Mark Jackson or Terry
          Porter, though any of them would have done equally well or better.
          Isiah Thomas is a little more questionable -- as good as he was --
          because he did shoot a lot and wasn't the best at it. Derek Harper
          was a very similar type of guard. So was Nate McMillan. DJ wasn't a
          spectacular player. I certainly would not consider him a Hall of
          Famer, even though the 66 ranking Mike gave is suggestive of the Hall.
          I actually worry that Mike's ranking overweights the playoffs in this
          case because DJ had the privilege of playing with Bird, McHale, and
          Parish.

          > So how then to reflect this value? I have toyed with the idea of
          > creating a system called "karma points" that add extra value to
          > players on winning teams to adjust formulas. (This really just an
          > extended version of what others of you have done when you might
          award
          > points to a player making an all-defensive team for example). You
          > can award points for significant contributions of complementary
          > players on playoff bound and championship caliber teams. This is of
          > course potentially arbitrary in application. Indeed, the formula
          > might bear out that aaron mckie is too close in value to stackhouse
          > or that avery johnson or eric snow have a value over marbury.

          The way I have interpreted players like DJ is that they don't detract
          from a good team and they have the ego to accept that. Ron Harper
          turned out that way, though who would have suspected it with all his
          awful shooting early on in his career. Give Harper/DJ karma points
          for their championships, but they still contribute a net of about
          20-50 pts/season with his best teams (or 1-2 extra wins) and are
          pretty replaceable (not easily because they _are_ starters, but
          still...).

          Dean Oliver
          Journal of Basketball Studies
        • msg_53@hotmail.com
          I was surprised when I saw an article lamenting the omission of Dennis Johnson from the NBA alltime Top 50. I certainly did not predict any statistical
          Message 4 of 4 , Mar 24, 2001
          • 0 Attachment
            I was surprised when I saw an article lamenting the omission of
            Dennis Johnson from the NBA alltime Top 50. I certainly did not
            predict any statistical evaluation could place him near that level.
            DJ played among talented players in all 3 of his NBA locations, and
            he was instrumental in each case, as Robert says, by deferring to
            the "real stars" of those teams. Part of the reason DJ enhanced each
            team he played for was that he was the ultimate player who would do
            whatever was most needed.
            A bigger scorer with the Suns, but a bigger playmaker with the Celts,
            I seem to remember him taking, and making, an inordinate number of
            buzzer-beaters, with the shot-clock winding down. And his playoff
            numbers were generally superior to his regular seasons'.
            Why are K.C. Jones and Bill Bradley in the Hall, except that they
            were major contributors to champions? They took care of the ball,
            and did what they had to do.
            DJ played 13 productive seasons, and 16% of his minutes were in
            postseason (only Magic, Bird, Jordan, and Pippen, are higher).
            Oddly, his biggest numbers were in Phoenix, where his 3 playoff
            outings were very brief.
            I was not a fan of Dennis Johnson, until he had proven over and over
            that he could hit the big shot as regularly as any member of those
            great Celtic teams of the '80s.

            You can add playoff numbers into the regular season numbers, thus
            giving credit to players on good teams: if you are getting big
            minutes with better players around you, you must have something going
            on. You might even assign greater value to postseason minutes; I do
            this by adding the square root of the regular season totals to the
            square root of the playoff totals.
            The average player (good player, not scrub), plays 90% of his minutes
            in the reg. season, and 10% in the postseason: a 9:1 ratio. So, on
            average, only 1/10 of a player's lifetime production shows up in the
            "games that count".
            Comparing square roots, however, changes the 9:1 ratio to a 3:1
            ratio, and so 1/4 of a typical player's career might be his playoff
            record. This sounds about right to me.
            Magic, Bird, Jordan, Pippen, Russell, Shaq, West, McHale, Worthy,
            Horace Grant, Mikan, DJ, and Sam Jones, played 15-18% of their
            minutes in playoff competition; and as such, their playoff careers
            count, in my system, between 28-33% of their totals.
            Very few players in history consistently did better in postseason;
            but the ones who have, increase the worth of their playoff minutes,
            in estimating their careers. Jordan, Hakeem, Russell, Elvin Hayes,
            Rick Barry, Isiah, Reggie Miller, Cowens, Worthy, Mikan, and others
            excelled at the increased level of competition, over their careers.

            > > So how then to reflect this value? I have toyed with the idea of
            > > creating a system called "karma points" that add extra value to
            > > players on winning teams to adjust formulas. (This really just
            an
            > > extended version of what others of you have done when you might
            > award
            > > points to a player making an all-defensive team for example).
            You
            > > can award points for significant contributions of complementary
            > > players on playoff bound and championship caliber teams. This is
            of
            > > course potentially arbitrary in application. Indeed, the formula
            > > might bear out that aaron mckie is too close in value to
            stackhouse
            > > or that avery johnson or eric snow have a value over marbury.
            >
            > The way I have interpreted players like DJ is that they don't
            detract
            > from a good team and they have the ego to accept that. Ron Harper
            > turned out that way, though who would have suspected it with all
            his
            > awful shooting early on in his career. Give Harper/DJ karma points
            > for their championships, but they still contribute a net of about
            > 20-50 pts/season with his best teams (or 1-2 extra wins) and are
            > pretty replaceable (not easily because they _are_ starters, but
            > still...).
            >
            > Dean Oliver
            > Journal of Basketball Studies
          Your message has been successfully submitted and would be delivered to recipients shortly.