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

The Curious Case of Michael Redd

Expand Messages
  • roland_beech
    Overall the on court versus off court comparisons seem to be in line with most of the assumptions about players and their value to a team (so Kevin Garnett is
    Message 1 of 9 , Dec 8 6:24 PM
      Overall the on court versus off court comparisons seem to be in line
      with most of the assumptions about players and their value to a team
      (so Kevin Garnett is by far the most significant player to the
      T'Wolves, the Hawks defense is far better when they have Ratliff in
      the game, etc.), but there are more than a few cases where obviously
      good/great players have stats showing the team has done better when
      they are off the court!

      With a guy like Carmelo Anthony, who currently has a negative rating
      (the Nuggets are +3 pts per 48min with him on the floor, +6.2 w/out)
      you can trot out the old rookie learning theory and yes, his rating
      has been improving, and yes his raw plus/minus shows a +44 on versus
      a +33 off. (by the way, Melo has hit 20 of 40 shots in clutch time
      with an eFG of 59% so far, and a nice 16 of 19 from the line)

      However, a number of the instances of a name player with a counter-
      intuitive on court versus off court team performance don't offer up
      such easy rationalizations.

      The case in point I would like to use is the cover boy for John's
      wonderful book: Michael Redd

      Some data:
      On court -10 / 743 minutes or -0.6 /48
      Off court +41 / 211 minutes or +9.3 /48
      ...for a net of -9.9 pts to the team per 48minutes

      Team pts per 100 possessions
      Offense: 106.2 with, 108.7 without
      Defense: 106.4 with, 98.5 without

      So basically on a raw level, the team plays much stronger defense
      without him, and even executes better offensively.

      An obvious argument is that perhaps "garbage time" has too much play
      in the "off court numbers", but surprisingly that's not the case --
      Mr. Redd has been dragged down primarily by the play of his team in
      clutch moments of the game (which we define as 4th qtr on, neither
      team ahead by more than 5 pts) where his +/- is -33 (or in other
      words he has a +23 "non-clutch" plus minus) and the team is +11
      without him in the clutch...

      His own stats in the clutch have been less than stellar: 30% FG,
      37% eFG, 2 assists to 3 turnovers

      The next objection is that the +/- numbers in raw fashion since they
      are not adjusting for the other players on the court could be
      rendered deceptive by in essence the "strength of schedule" effect --
      that Redd only plays on the court against stronger opposition
      players (or, possibly, with a weaker set of teammates). Of course
      the "clutch time" findings undermine this to some extent in that you
      can expect good players to be on the floor for both sides in games
      that hang in the balance, and while we do have ratings adjusted for
      the other players that do boost his standing somewhat, they don't
      explain away the difference in overall team performance.

      Perhaps in the case of Redd because he is such the focus of the
      offense when he is on the floor, his teammates are somewhat
      lethargic in their play compared to when he's resting, or perhaps
      the clutch time +/- is indicative of him being played too much and
      begin fatigued in the later stages, perhaps he indeed plays in some
      disadvantaged lineup situations as Porter elects to rest the
      other "good players" but keep Redd on the court, perhaps he is a
      sore point on the defensive end since as the main man on offense he
      needs to/feels entitled to catch a breather??

      Perhaps the Bucks are a team that doesn't intimidate and
      consequently teams are on cruise mode until the fourth quarter when
      they can take advantage (and Redd is more often than not playing)

      Maybe Porter's use of 5-man units is not optimal: the top three
      most used units for Milwaukee all include Redd and all have negative
      +/- stats, while Jones-Mason-Kukoc-Thomas-Gadzuric and Jones-
      Strickland-Mason-Kukoc-Gadzuric have both been positive in
      admittedly limited exposure.

      Lots of possible explanations come to mind...

      So I am curious to hear the opinions of this sage group in terms of
      what to make of this effect (and again it's common enough: Paul
      Pierce has a -18.9 rating, Pippen -12.9, Walker -10.6, Dampier -
      14.4...)
    • Dean Oliver
      I ve been thinking about this situation for a while actually. So far my thinking is progressing a bit like this: 1. You re measuring effectively wins and
      Message 2 of 9 , Dec 8 10:14 PM
        I've been thinking about this situation for a while actually. So far
        my thinking is progressing a bit like this:

        1. You're measuring effectively wins and losses.
        2. We know from pitcher records in baseball that wins and losses can
        be pretty erratic for even good pitchers (though great pitchers have
        less variability).
        3. Can we evaluate the statistical significance of such breakdowns?
        Yes. Can I? Not from here. Can Roland? Uh-huh.
        4. If Roland finds statistical significance for someone like Michael
        Redd, at what level do we start to believe it?
        5. Hmm, I need to calculate my individual stats for Michael Redd this
        year. That makes 2371 things on my to do list.
        6. Man, I'm tired.

        I really don't think that either of these breakdowns are statistically
        significant. The chance of a team that allows 54% scores to opponents
        (roughly half of 1.08) giving up roughly 49% scores on roughly 420
        possessions (roughly how many there'd be in 211 minutes) is pretty
        darn large. It sure ain't 5%, which is a typical significance figure.

        OK, so I guess I can evaluate statistical significance from here.
        That allows me to skip to step 6. I'm going to bed.

        DeanO

        ps -- given that I pretty much start off at step 6 these days, I could
        be wrong. I did this in about 2 minutes, so it definitely could be
        wrong... But I guess I can say it again -- this is one of the reasons
        why I think the whole Winston/Sagarin method of evaluating players
        needs a bit of refinement. Sheesh, they (like Roland) had Pippen
        ranked really high last year and poor this year -- statistical
        variation. Winston/Sagarin didn't do a very good job accounting for
        natural variation. But MarkC doesn't like me saying that.

        --- In APBR_analysis@yahoogroups.com, "roland_beech" <roland@t...> wrote:
        > Overall the on court versus off court comparisons seem to be in line
        > with most of the assumptions about players and their value to a team
        > (so Kevin Garnett is by far the most significant player to the
        > T'Wolves, the Hawks defense is far better when they have Ratliff in
        > the game, etc.), but there are more than a few cases where obviously
        > good/great players have stats showing the team has done better when
        > they are off the court!
        >
        > With a guy like Carmelo Anthony, who currently has a negative rating
        > (the Nuggets are +3 pts per 48min with him on the floor, +6.2 w/out)
        > you can trot out the old rookie learning theory and yes, his rating
        > has been improving, and yes his raw plus/minus shows a +44 on versus
        > a +33 off. (by the way, Melo has hit 20 of 40 shots in clutch time
        > with an eFG of 59% so far, and a nice 16 of 19 from the line)
        >
        > However, a number of the instances of a name player with a counter-
        > intuitive on court versus off court team performance don't offer up
        > such easy rationalizations.
        >
        > The case in point I would like to use is the cover boy for John's
        > wonderful book: Michael Redd
        >
        > Some data:
        > On court -10 / 743 minutes or -0.6 /48
        > Off court +41 / 211 minutes or +9.3 /48
        > ...for a net of -9.9 pts to the team per 48minutes
        >
        > Team pts per 100 possessions
        > Offense: 106.2 with, 108.7 without
        > Defense: 106.4 with, 98.5 without
        >
        > So basically on a raw level, the team plays much stronger defense
        > without him, and even executes better offensively.
        >
        > An obvious argument is that perhaps "garbage time" has too much play
        > in the "off court numbers", but surprisingly that's not the case --
        > Mr. Redd has been dragged down primarily by the play of his team in
        > clutch moments of the game (which we define as 4th qtr on, neither
        > team ahead by more than 5 pts) where his +/- is -33 (or in other
        > words he has a +23 "non-clutch" plus minus) and the team is +11
        > without him in the clutch...
        >
        > His own stats in the clutch have been less than stellar: 30% FG,
        > 37% eFG, 2 assists to 3 turnovers
        >
        > The next objection is that the +/- numbers in raw fashion since they
        > are not adjusting for the other players on the court could be
        > rendered deceptive by in essence the "strength of schedule" effect --
        > that Redd only plays on the court against stronger opposition
        > players (or, possibly, with a weaker set of teammates). Of course
        > the "clutch time" findings undermine this to some extent in that you
        > can expect good players to be on the floor for both sides in games
        > that hang in the balance, and while we do have ratings adjusted for
        > the other players that do boost his standing somewhat, they don't
        > explain away the difference in overall team performance.
        >
        > Perhaps in the case of Redd because he is such the focus of the
        > offense when he is on the floor, his teammates are somewhat
        > lethargic in their play compared to when he's resting, or perhaps
        > the clutch time +/- is indicative of him being played too much and
        > begin fatigued in the later stages, perhaps he indeed plays in some
        > disadvantaged lineup situations as Porter elects to rest the
        > other "good players" but keep Redd on the court, perhaps he is a
        > sore point on the defensive end since as the main man on offense he
        > needs to/feels entitled to catch a breather??
        >
        > Perhaps the Bucks are a team that doesn't intimidate and
        > consequently teams are on cruise mode until the fourth quarter when
        > they can take advantage (and Redd is more often than not playing)
        >
        > Maybe Porter's use of 5-man units is not optimal: the top three
        > most used units for Milwaukee all include Redd and all have negative
        > +/- stats, while Jones-Mason-Kukoc-Thomas-Gadzuric and Jones-
        > Strickland-Mason-Kukoc-Gadzuric have both been positive in
        > admittedly limited exposure.
        >
        > Lots of possible explanations come to mind...
        >
        > So I am curious to hear the opinions of this sage group in terms of
        > what to make of this effect (and again it's common enough: Paul
        > Pierce has a -18.9 rating, Pippen -12.9, Walker -10.6, Dampier -
        > 14.4...)
      • Dean Oliver
        so much for sleep. That defensive difference was significant at about 2%. At least as a first cut. Too tired to do it right. I still think significance is,
        Message 3 of 9 , Dec 8 10:45 PM
          so much for sleep.

          That defensive difference was significant at about 2%. At least as a
          first cut. Too tired to do it right. I still think significance is,
          uh, significant.

          --- In APBR_analysis@yahoogroups.com, "Dean Oliver" <deano@r...> wrote:
          > I've been thinking about this situation for a while actually. So far
          > my thinking is progressing a bit like this:
          >
          > 1. You're measuring effectively wins and losses.
          > 2. We know from pitcher records in baseball that wins and losses can
          > be pretty erratic for even good pitchers (though great pitchers have
          > less variability).
          > 3. Can we evaluate the statistical significance of such breakdowns?
          > Yes. Can I? Not from here. Can Roland? Uh-huh.
          > 4. If Roland finds statistical significance for someone like Michael
          > Redd, at what level do we start to believe it?
          > 5. Hmm, I need to calculate my individual stats for Michael Redd this
          > year. That makes 2371 things on my to do list.
          > 6. Man, I'm tired.
          >
          > I really don't think that either of these breakdowns are statistically
          > significant. The chance of a team that allows 54% scores to opponents
          > (roughly half of 1.08) giving up roughly 49% scores on roughly 420
          > possessions (roughly how many there'd be in 211 minutes) is pretty
          > darn large. It sure ain't 5%, which is a typical significance figure.
          >
          > OK, so I guess I can evaluate statistical significance from here.
          > That allows me to skip to step 6. I'm going to bed.
          >
          > DeanO
          >
          > ps -- given that I pretty much start off at step 6 these days, I could
          > be wrong. I did this in about 2 minutes, so it definitely could be
          > wrong... But I guess I can say it again -- this is one of the reasons
          > why I think the whole Winston/Sagarin method of evaluating players
          > needs a bit of refinement. Sheesh, they (like Roland) had Pippen
          > ranked really high last year and poor this year -- statistical
          > variation. Winston/Sagarin didn't do a very good job accounting for
          > natural variation. But MarkC doesn't like me saying that.
          >
          > --- In APBR_analysis@yahoogroups.com, "roland_beech" <roland@t...>
          wrote:
          > > Overall the on court versus off court comparisons seem to be in line
          > > with most of the assumptions about players and their value to a team
          > > (so Kevin Garnett is by far the most significant player to the
          > > T'Wolves, the Hawks defense is far better when they have Ratliff in
          > > the game, etc.), but there are more than a few cases where obviously
          > > good/great players have stats showing the team has done better when
          > > they are off the court!
          > >
          > > With a guy like Carmelo Anthony, who currently has a negative rating
          > > (the Nuggets are +3 pts per 48min with him on the floor, +6.2 w/out)
          > > you can trot out the old rookie learning theory and yes, his rating
          > > has been improving, and yes his raw plus/minus shows a +44 on versus
          > > a +33 off. (by the way, Melo has hit 20 of 40 shots in clutch time
          > > with an eFG of 59% so far, and a nice 16 of 19 from the line)
          > >
          > > However, a number of the instances of a name player with a counter-
          > > intuitive on court versus off court team performance don't offer up
          > > such easy rationalizations.
          > >
          > > The case in point I would like to use is the cover boy for John's
          > > wonderful book: Michael Redd
          > >
          > > Some data:
          > > On court -10 / 743 minutes or -0.6 /48
          > > Off court +41 / 211 minutes or +9.3 /48
          > > ...for a net of -9.9 pts to the team per 48minutes
          > >
          > > Team pts per 100 possessions
          > > Offense: 106.2 with, 108.7 without
          > > Defense: 106.4 with, 98.5 without
          > >
          > > So basically on a raw level, the team plays much stronger defense
          > > without him, and even executes better offensively.
          > >
          > > An obvious argument is that perhaps "garbage time" has too much play
          > > in the "off court numbers", but surprisingly that's not the case --
          > > Mr. Redd has been dragged down primarily by the play of his team in
          > > clutch moments of the game (which we define as 4th qtr on, neither
          > > team ahead by more than 5 pts) where his +/- is -33 (or in other
          > > words he has a +23 "non-clutch" plus minus) and the team is +11
          > > without him in the clutch...
          > >
          > > His own stats in the clutch have been less than stellar: 30% FG,
          > > 37% eFG, 2 assists to 3 turnovers
          > >
          > > The next objection is that the +/- numbers in raw fashion since they
          > > are not adjusting for the other players on the court could be
          > > rendered deceptive by in essence the "strength of schedule" effect --
          > > that Redd only plays on the court against stronger opposition
          > > players (or, possibly, with a weaker set of teammates). Of course
          > > the "clutch time" findings undermine this to some extent in that you
          > > can expect good players to be on the floor for both sides in games
          > > that hang in the balance, and while we do have ratings adjusted for
          > > the other players that do boost his standing somewhat, they don't
          > > explain away the difference in overall team performance.
          > >
          > > Perhaps in the case of Redd because he is such the focus of the
          > > offense when he is on the floor, his teammates are somewhat
          > > lethargic in their play compared to when he's resting, or perhaps
          > > the clutch time +/- is indicative of him being played too much and
          > > begin fatigued in the later stages, perhaps he indeed plays in some
          > > disadvantaged lineup situations as Porter elects to rest the
          > > other "good players" but keep Redd on the court, perhaps he is a
          > > sore point on the defensive end since as the main man on offense he
          > > needs to/feels entitled to catch a breather??
          > >
          > > Perhaps the Bucks are a team that doesn't intimidate and
          > > consequently teams are on cruise mode until the fourth quarter when
          > > they can take advantage (and Redd is more often than not playing)
          > >
          > > Maybe Porter's use of 5-man units is not optimal: the top three
          > > most used units for Milwaukee all include Redd and all have negative
          > > +/- stats, while Jones-Mason-Kukoc-Thomas-Gadzuric and Jones-
          > > Strickland-Mason-Kukoc-Gadzuric have both been positive in
          > > admittedly limited exposure.
          > >
          > > Lots of possible explanations come to mind...
          > >
          > > So I am curious to hear the opinions of this sage group in terms of
          > > what to make of this effect (and again it's common enough: Paul
          > > Pierce has a -18.9 rating, Pippen -12.9, Walker -10.6, Dampier -
          > > 14.4...)
        • Mike G
          ... FG, 37% eFG,.. ... they ... effect.. So are you using raw in/out numbers because it s time-consuming to generate strength-adjusted numbers in
          Message 4 of 9 , Dec 9 6:37 AM
            --- In APBR_analysis@yahoogroups.com, "roland_beech" <roland@t...>
            wrote:
            >..: Michael Redd
            > ...
            > Offense: 106.2 with, 108.7 without
            > Defense: 106.4 with, 98.5 without ..
            >
            >... His own stats in the clutch have been less than stellar: 30%
            FG, 37% eFG,..
            > The next objection is that the +/- numbers in raw fashion since
            they
            > are not adjusting for the other players on the court could be
            > rendered deceptive by in essence the "strength of schedule"
            effect..

            So are you using "raw" in/out numbers because it's time-consuming to
            generate "strength-adjusted" numbers in midseason?


            > ..that Redd only plays on the court against stronger opposition
            > players (or, possibly, with a weaker set of teammates). Of
            course
            > the "clutch time" findings undermine this to some extent...

            Throw into the mix the fact that Redd has never been The Man before,
            but a strong 2nd or 3rd option. If he's trying to be Kobe -- or
            even Ray Allen -- it's quite a switch for him.


            > while we do have ratings adjusted for
            > the other players that do boost his standing somewhat, they don't
            > explain away the difference in overall team performance.

            I'm not insisting you give away any privileged information; but if
            you actually possess this rating for This Player, why do you
            speculate about it?
          • Dean Oliver
            In a very interesting note, Rick Barry out here in Oakland has been quoting Roland s points per 100 possession numbers ON THE AIR!!!! Today he was talking
            Message 5 of 9 , Dec 9 7:28 PM
              In a very interesting note, Rick Barry out here in Oakland has been
              quoting Roland's points per 100 possession numbers ON THE AIR!!!!
              Today he was talking about Erick Dampier and his negative split. And
              it reminded me of a reason that may explain Redd.

              One of the problems with Roland's stuff ("problem" may be too strong,
              but it's something to think about) is that THERE WILL ALWAYS BE GUYS
              WITH NEGATIVE VALUES ON A TEAM. This is the case with Erick Dampier.
              Damp looks bad offensively even though he is improved this year
              because his primary sub is Brian Cardinal, who is playing out of his
              mind!!! If Cardinal looks good, the guy he replaces starts looking
              bad. That is Damp.

              In Milwaukee's case, Redd is replaced most prominently by Tim Thomas
              and Erick Strickland, who must be playing better defense or playing
              against scrub offensive guys. Because Redd is a good player. He is
              handling the additional responsibility about as well as would be
              predicted. There has been a decline in his offensive efficiency, but
              to a still good level.

              DeanO

              --- In APBR_analysis@yahoogroups.com, "Dean Oliver" <deano@r...> wrote:
              > so much for sleep.
              >
              > That defensive difference was significant at about 2%. At least as a
              > first cut. Too tired to do it right. I still think significance is,
              > uh, significant.
              >
              > --- In APBR_analysis@yahoogroups.com, "Dean Oliver" <deano@r...> wrote:
              > > I've been thinking about this situation for a while actually. So far
              > > my thinking is progressing a bit like this:
              > >
              > > 1. You're measuring effectively wins and losses.
              > > 2. We know from pitcher records in baseball that wins and losses can
              > > be pretty erratic for even good pitchers (though great pitchers have
              > > less variability).
              > > 3. Can we evaluate the statistical significance of such breakdowns?
              > > Yes. Can I? Not from here. Can Roland? Uh-huh.
              > > 4. If Roland finds statistical significance for someone like Michael
              > > Redd, at what level do we start to believe it?
              > > 5. Hmm, I need to calculate my individual stats for Michael Redd this
              > > year. That makes 2371 things on my to do list.
              > > 6. Man, I'm tired.
              > >
              > > I really don't think that either of these breakdowns are statistically
              > > significant. The chance of a team that allows 54% scores to opponents
              > > (roughly half of 1.08) giving up roughly 49% scores on roughly 420
              > > possessions (roughly how many there'd be in 211 minutes) is pretty
              > > darn large. It sure ain't 5%, which is a typical significance figure.
              > >
              > > OK, so I guess I can evaluate statistical significance from here.
              > > That allows me to skip to step 6. I'm going to bed.
              > >
              > > DeanO
              > >
              > > ps -- given that I pretty much start off at step 6 these days, I could
              > > be wrong. I did this in about 2 minutes, so it definitely could be
              > > wrong... But I guess I can say it again -- this is one of the reasons
              > > why I think the whole Winston/Sagarin method of evaluating players
              > > needs a bit of refinement. Sheesh, they (like Roland) had Pippen
              > > ranked really high last year and poor this year -- statistical
              > > variation. Winston/Sagarin didn't do a very good job accounting for
              > > natural variation. But MarkC doesn't like me saying that.
              > >
              > > --- In APBR_analysis@yahoogroups.com, "roland_beech" <roland@t...>
              > wrote:
              > > > Overall the on court versus off court comparisons seem to be in
              line
              > > > with most of the assumptions about players and their value to a
              team
              > > > (so Kevin Garnett is by far the most significant player to the
              > > > T'Wolves, the Hawks defense is far better when they have Ratliff in
              > > > the game, etc.), but there are more than a few cases where
              obviously
              > > > good/great players have stats showing the team has done better when
              > > > they are off the court!
              > > >
              > > > With a guy like Carmelo Anthony, who currently has a negative
              rating
              > > > (the Nuggets are +3 pts per 48min with him on the floor, +6.2
              w/out)
              > > > you can trot out the old rookie learning theory and yes, his rating
              > > > has been improving, and yes his raw plus/minus shows a +44 on
              versus
              > > > a +33 off. (by the way, Melo has hit 20 of 40 shots in clutch time
              > > > with an eFG of 59% so far, and a nice 16 of 19 from the line)
              > > >
              > > > However, a number of the instances of a name player with a counter-
              > > > intuitive on court versus off court team performance don't offer up
              > > > such easy rationalizations.
              > > >
              > > > The case in point I would like to use is the cover boy for John's
              > > > wonderful book: Michael Redd
              > > >
              > > > Some data:
              > > > On court -10 / 743 minutes or -0.6 /48
              > > > Off court +41 / 211 minutes or +9.3 /48
              > > > ...for a net of -9.9 pts to the team per 48minutes
              > > >
              > > > Team pts per 100 possessions
              > > > Offense: 106.2 with, 108.7 without
              > > > Defense: 106.4 with, 98.5 without
              > > >
              > > > So basically on a raw level, the team plays much stronger defense
              > > > without him, and even executes better offensively.
              > > >
              > > > An obvious argument is that perhaps "garbage time" has too much
              play
              > > > in the "off court numbers", but surprisingly that's not the case --
              > > > Mr. Redd has been dragged down primarily by the play of his team in
              > > > clutch moments of the game (which we define as 4th qtr on, neither
              > > > team ahead by more than 5 pts) where his +/- is -33 (or in other
              > > > words he has a +23 "non-clutch" plus minus) and the team is +11
              > > > without him in the clutch...
              > > >
              > > > His own stats in the clutch have been less than stellar: 30% FG,
              > > > 37% eFG, 2 assists to 3 turnovers
              > > >
              > > > The next objection is that the +/- numbers in raw fashion since
              they
              > > > are not adjusting for the other players on the court could be
              > > > rendered deceptive by in essence the "strength of schedule"
              effect --
              > > > that Redd only plays on the court against stronger opposition
              > > > players (or, possibly, with a weaker set of teammates). Of course
              > > > the "clutch time" findings undermine this to some extent in that
              you
              > > > can expect good players to be on the floor for both sides in games
              > > > that hang in the balance, and while we do have ratings adjusted for
              > > > the other players that do boost his standing somewhat, they don't
              > > > explain away the difference in overall team performance.
              > > >
              > > > Perhaps in the case of Redd because he is such the focus of the
              > > > offense when he is on the floor, his teammates are somewhat
              > > > lethargic in their play compared to when he's resting, or perhaps
              > > > the clutch time +/- is indicative of him being played too much and
              > > > begin fatigued in the later stages, perhaps he indeed plays in some
              > > > disadvantaged lineup situations as Porter elects to rest the
              > > > other "good players" but keep Redd on the court, perhaps he is a
              > > > sore point on the defensive end since as the main man on offense he
              > > > needs to/feels entitled to catch a breather??
              > > >
              > > > Perhaps the Bucks are a team that doesn't intimidate and
              > > > consequently teams are on cruise mode until the fourth quarter when
              > > > they can take advantage (and Redd is more often than not playing)
              > > >
              > > > Maybe Porter's use of 5-man units is not optimal: the top three
              > > > most used units for Milwaukee all include Redd and all have
              negative
              > > > +/- stats, while Jones-Mason-Kukoc-Thomas-Gadzuric and Jones-
              > > > Strickland-Mason-Kukoc-Gadzuric have both been positive in
              > > > admittedly limited exposure.
              > > >
              > > > Lots of possible explanations come to mind...
              > > >
              > > > So I am curious to hear the opinions of this sage group in terms of
              > > > what to make of this effect (and again it's common enough: Paul
              > > > Pierce has a -18.9 rating, Pippen -12.9, Walker -10.6, Dampier -
              > > > 14.4...)
            • Kevin Pelton
              ... Funny, I was about to make the same post about Desmond Mason, who is a well-regarded, if overrated, defender. Mason has been starting lately, because
              Message 6 of 9 , Dec 9 8:16 PM
                --- In APBR_analysis@yahoogroups.com, "Dean Oliver" <deano@r...>
                wrote:

                > One of the problems with Roland's stuff ("problem" may be too
                > strong, but it's something to think about) is that THERE WILL
                > ALWAYS BE GUYS WITH NEGATIVE VALUES ON A TEAM. This is the case
                > with Erick Dampier. Damp looks bad offensively even though he is
                > improved this year because his primary sub is Brian Cardinal, who
                > is playing out of his mind!!! If Cardinal looks good, the guy he
                > replaces starts looking bad. That is Damp.
                >
                > In Milwaukee's case, Redd is replaced most prominently by Tim
                > Thomas and Erick Strickland, who must be playing better defense or
                > playing against scrub offensive guys. Because Redd is a good
                > player. He is handling the additional responsibility about as
                > well as would be predicted. There has been a decline in his
                > offensive efficiency, but to a still good level.

                Funny, I was about to make the same post about Desmond Mason, who is
                a well-regarded, if overrated, defender. Mason has been starting
                lately, because Thomas has been injured, but he still represents
                about 2/3 of that "not Redd" rating. Strickland appears to have been
                backing up the two recently, and he is also a very good defender.

                In general, I would say the Bucks' bench (with Mason) is more highly
                regarded than its starters. You've got guys like Kukoc and Mason,
                who are highly regarded, and Dan Gadzuric is pretty clearly a better
                player than Daniel Santiago.

                Of course, the "clutch" point tends to reduce the concerns about the
                quality of Milwaukee's bench. Why might the Bucks be struggling in
                crunch time?

                - No established scorer
                While Redd has scored very well this season and is still reasonably
                efficient, this is his first time as "the man" in the NBA. It's not
                unreasonable to expect him to perhaps struggle with that pressure.
                I've not seen him much in that kind of situation, but as primarily a
                jump shooter, he may be struggling to create his own shot a la the
                kind of "clutch ability" we've discussed in the past.

                - T.J. Ford
                I'm just throwing this out there, but my cousin is a Bucks fan
                because of Mason, and she was regularly moaning during the early
                part of the season about how much Ford was shooting in close games
                (which, given his 32.5% overall percentage, is a rather dubious
                proposition). Having a complete bricklayer on the court in halfcourt
                situations at the ends of games may be forcing the Bucks to play 4-
                on-5, in effect, which might not be such a problem earlier in games
                when there are more transition opportunities.
              • Mike G
                ... Dampier. ... his ... Here again, I ask: Does Cardinal play mostly against scrubs, while Dampier seldom does? There s plenty of statistical evidence to
                Message 7 of 9 , Dec 10 5:56 AM
                  --- In APBR_analysis@yahoogroups.com, "Dean Oliver" <deano@r...>
                  wrote:
                  > ... THERE WILL ALWAYS BE GUYS
                  > WITH NEGATIVE VALUES ON A TEAM. This is the case with Erick
                  Dampier.
                  > Damp looks bad offensively even though he is improved this year
                  > because his primary sub is Brian Cardinal, who is playing out of
                  his
                  > mind!!! If Cardinal looks good, the guy he replaces starts looking
                  > bad. That is Damp.

                  Here again, I ask: Does Cardinal play mostly against scrubs, while
                  Dampier seldom does?

                  There's plenty of statistical "evidence" to suggest Cardinal should
                  be starting, and getting 40 minutes, and going to the allstar game.

                  Such a suggestion might provoke snickers from those who know better;
                  and well it should.

                  Since I don't know, but merely suspect, my guess is that one guy is
                  starting and playing against better competition than the guy who
                  comes off the bench in the 2nd quarter.

                  (You know both Cardinal and Brad Miller played at Purdue; both went
                  undrafted, but persevered; and both are lighting it up in northern
                  Cali right now)
                • Dean Oliver
                  Cardinal is not just playing against scrubs. He is just playing remarkably well. I haven t had a chance to figure out whether it s sustainable. I don t
                  Message 8 of 9 , Dec 10 7:15 AM
                    Cardinal is not just playing against scrubs. He is just playing
                    remarkably well. I haven't had a chance to figure out whether it's
                    sustainable. I don't think it's quite sustainable, but it does appear
                    to be a sign of a good player -- just not the all-star level he has
                    been playing at.

                    Also, Cardinal is not the defensive player that Damp is. He plays
                    when the W's go small.

                    DeanO

                    --- In APBR_analysis@yahoogroups.com, "Mike G" <msg_53@h...> wrote:
                    > --- In APBR_analysis@yahoogroups.com, "Dean Oliver" <deano@r...>
                    > wrote:
                    > > ... THERE WILL ALWAYS BE GUYS
                    > > WITH NEGATIVE VALUES ON A TEAM. This is the case with Erick
                    > Dampier.
                    > > Damp looks bad offensively even though he is improved this year
                    > > because his primary sub is Brian Cardinal, who is playing out of
                    > his
                    > > mind!!! If Cardinal looks good, the guy he replaces starts looking
                    > > bad. That is Damp.
                    >
                    > Here again, I ask: Does Cardinal play mostly against scrubs, while
                    > Dampier seldom does?
                    >
                    > There's plenty of statistical "evidence" to suggest Cardinal should
                    > be starting, and getting 40 minutes, and going to the allstar game.
                    >
                    > Such a suggestion might provoke snickers from those who know better;
                    > and well it should.
                    >
                    > Since I don't know, but merely suspect, my guess is that one guy is
                    > starting and playing against better competition than the guy who
                    > comes off the bench in the 2nd quarter.
                    >
                    > (You know both Cardinal and Brad Miller played at Purdue; both went
                    > undrafted, but persevered; and both are lighting it up in northern
                    > Cali right now)
                  • Matthew Lloyd
                    For what its worth... Brian Cardinal was the Detroit Pistons second-round pick (44 overall) in the 2000 NBA Draft. In his three seasons prior to 2003-04 he
                    Message 9 of 9 , Dec 10 11:43 AM
                      For what its worth... Brian Cardinal was the Detroit
                      Pistons' second-round pick (44 overall) in the 2000
                      NBA Draft. In his three seasons prior to 2003-04 he
                      appeared in 28 games, averaging 1.9 ppg and 1.2 rpg.


                      --- Mike G <msg_53@...> wrote:
                      > --- In APBR_analysis@yahoogroups.com, "Dean Oliver"
                      > <deano@r...>
                      > wrote:
                      > > ... THERE WILL ALWAYS BE GUYS
                      > > WITH NEGATIVE VALUES ON A TEAM. This is the case
                      > with Erick
                      > Dampier.
                      > > Damp looks bad offensively even though he is
                      > improved this year
                      > > because his primary sub is Brian Cardinal, who is
                      > playing out of
                      > his
                      > > mind!!! If Cardinal looks good, the guy he
                      > replaces starts looking
                      > > bad. That is Damp.
                      >
                      > Here again, I ask: Does Cardinal play mostly against
                      > scrubs, while
                      > Dampier seldom does?
                      >
                      > There's plenty of statistical "evidence" to suggest
                      > Cardinal should
                      > be starting, and getting 40 minutes, and going to
                      > the allstar game.
                      >
                      > Such a suggestion might provoke snickers from those
                      > who know better;
                      > and well it should.
                      >
                      > Since I don't know, but merely suspect, my guess is
                      > that one guy is
                      > starting and playing against better competition than
                      > the guy who
                      > comes off the bench in the 2nd quarter.
                      >
                      > (You know both Cardinal and Brad Miller played at
                      > Purdue; both went
                      > undrafted, but persevered; and both are lighting it
                      > up in northern
                      > Cali right now)
                      >
                      >
                      >


                      __________________________________
                      Do you Yahoo!?
                      New Yahoo! Photos - easier uploading and sharing.
                      http://photos.yahoo.com/
                    Your message has been successfully submitted and would be delivered to recipients shortly.