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

Expand Messages
  • Dean Oliver
    ... Funny, I ve never seen the studies, but I ve reached that conclusion on my own. ... Interesting that you raise this example. It was the example I used in
    Message 1 of 36 , Apr 26, 2002
    • 0 Attachment
      --- In APBR_analysis@y..., "Michael K. Tamada" <tamada@o...> wrote:

      > Yup, that adds a whole new wrinkle -- to look at, not just the
      > on-the-floor stuff, but also the costs, revenues, and profits.
      > Economists, not surprisingly, often do look at this stuff, usually at the
      > expense of paying more attention to the on-the-floor stuff. E.g. what
      > gives a bigger boost to attendance: winning 10 more games per year, or
      > winning a championship? Not surprisingly, winning a championship provides
      > a bigger boost than all but the biggest win increases. But this can
      > produce some interesting non-convexities into the profit-maximization
      > problem. One obvious and easy-to-understand example: you typically don't
      > want to follow a strategy of trying to win as many games as possible every
      > year. Better to rebuild for a few years and make a major push for a
      > championship and the resulting profit windfall. Then rebuild again.
      >

      Funny, I've never seen the studies, but I've reached that conclusion on my own.


      >
      > Calculating total or average value for basketball players is beset by
      > similar philosophical conundrums. What is Duncan's value to the Spurs?
      > Who is responsible for how much of their success? Who's responsible for
      > say the latest drunk-driving accident -- the driver who got drunk? But
      > maybe the other driver could have driven more defensively and avoided the
      > accident. Maybe the bartender shouldn't have sold that last drink. Maybe
      > alcohol manufacturers shouldn't be selling their dangerous products
      > (that's what's going on with these lawsuits against cigarette
      > manufacturers). Maybe the automobile manufacturer should have put
      > anti-lock brakes, airbags, roll bars, leakproof fuel tanks, etc. into the
      > cars it sells. Maybe the highway dept. should have built a divided
      > highway with a barrier in the middle and installed better street lighting.
      > Etc. etc.

      Interesting that you raise this example. It was the example I used in
      proposing a PhD topic in environmental policy many years ago. How do you
      assign/credit blame for things? I came up with a probabilitistic model
      that I'll explain briefly with the 2 person example -- the drunk driver and
      the guy who got hit. On a normal night, let's say that there is a 0.01%
      chance of getting hit by a drunk driver. In that case, the drunk driver is
      99.99% to blame for the accident. Let's say it's a night of a big win by
      the home football team, so the chances of getting hit by a drunk driver are
      now, say, 5%. Then the drunk is only 95% to blame for the accident. The
      concept gets more complicated, involving trees of decision, as you have
      more parties involved (yes, the car is necessary, the bartender is
      necessary, etc.). But the thought was that such a dynamic credit/blame
      system could make for the simple optimization of events. If you wanted to
      maximize the probability of an event, this system helps get there. If you
      want to minimize, it does the same thing. I don't remember all the details
      and I frankly stopped a couple months into it when my advisor got the
      boot. I ended up doing something much less controversial, something more
      esoteric, but it got me through in a hurry. And I actually use the basics
      of the technique in working on basketball.

      The method is somewhat flawed. It doesn't get used, for instance, in
      litigation much. Allocation of blame often comes down to legal "gore
      factors", which are basically subjective factors. Not only was it a death
      by a drunk, the drunk also beats his wife, so it clearly is his
      fault. Kinda weird. More legitimately, there is matter of
      intent. Sometimes people intend to do wrong. My method assumes people
      want to win, want to accomplish something good for the team.


      > > Ahh, yes, the differences between basketball and baseball. Role
      player in
      > > basketball is so different from a role player in baseball. The only
      "role
      > > player" in baseball I can think of that is considered to be all that
      > > valuable is a relief pitcher. But utility infielder, platoon player, or
      > > pinch hitter means you're not good enough to be a starter. Role players
      > > start in the NBA. They can be very valuable. As you say, you can't just
      > > plug them in anywhere and expect a constant number of wins/losses from
      > > them. You actually have to think a lot harder about them.
      >
      > Yup, marginal value analysis becomes a lot simpler, and more valid, when
      > the factors in question are more like replaceable commodities. When they
      > are unique, such as Tim Duncan, then calculating the marginal value
      > becomes beset by the very problems identified by KevinP.
      >

      So I think you're saying that James' analysis really is a marginal
      analysis. Baseball really is a bunch of guys working at the margin. They
      can't really influence other players on their team too much. There is some
      base-stealing interplay with batters, but it's pretty small. It's not like
      the interplay in basketball. Well, that would explain why I had to abandon
      most Jamesian like approaches to basketball back in '86.

      > > Speaking of baseball, I sat in Borders today and read James' section on
      > > win-shares. I've got gift certificates to Barnes and Noble, so I'll wait
      > > to find one of those before I buy it. But has anyone thought deeply
      about
      > > James' win-share concept?
      >
      > I haven't read his win-shares book, only his new Historical Abstract. My

      Is there a Win-shares book? I only read that part of his abstract.

      >
      > So, when comparing players' careers, the long-lasting non-spectacular
      > player will have too much of an advantage over the
      > shorter-lasting-but-better player. This is precisely where value over
      > replacement player comes in so handy.

      Actually, James handled this. He made some comparison of Dimaggio to Rusty
      Staub. The two have similar career numbers of win-shares, but Dimaggio had
      higher numbers at every age of his career, missing time due to the war and
      having a shorter career. So James did something like summing the
      win-shares over 5 consecutive years or adding the top 3 season
      values. It's a sensible but not unique way to handle it.

      >
      >
      > The bits of the formula for win shares that he reveals in the Historical
      > Abstract look rather weird, some stuff about counting marginal
      > contributions (he does use the word "marginal") but he defines marginal in
      > some weird way, something like only counting runs above half the mean or
      > something bizarre like that.

      I saw that quick explanation he did with teams. I filed it away to be
      thought of later. I slept on it and I'm pretty sure I know what he did. I
      think that counting runs above half the mean works out to be something
      close to a Taylor series approximation to his Pythagorean formula. That
      would make some sense. It would also explain why it doesn't work at all in
      basketball. You'd have to count runs above 15/16th of the mean in (NBA,
      different in college, womens, HS, etc.) basketball to get something that
      worked approximately right. Actually, I'm pretty sure that the factor of a
      half would have to be era adjusted, though he doesn't admit to it. Oh
      well. He does generally know what he's doing.

      DeanO
    • Michael Tamada
      Yeah, these are things to keep in mind as we try to zero in on the replacement level. Some so-called replacement level players are clearly better than others.
      Message 36 of 36 , Mar 24, 2004
      • 0 Attachment
        Yeah, these are things to keep in mind as we try to zero
        in on the replacement level. Some so-called replacement
        level players are clearly better than others. Are they
        thus truly "freely available"? There's only one Rod
        Strickland; once he's signed, it's not possible for any
        other team to go out and sign him. On the other hand,
        he was in a sense available to any team that wanted him.

        I suspect we'll want to look at a bunch of players
        identified as replacement level and find their average value,
        to even out the occasional Strickland who brings up the
        average and the occasional Oakley or whoever who brings down
        the average.

        In other words, something more akin to KevinP's look at the
        average performance of free agents, instead of the 290th
        man approach. Although it turns out that they yielded
        very similar values.


        --MKT

        -----Original Message-----
        From: John Hollinger [mailto:alleyoop2@...]
        Sent: Monday, March 22, 2004 8:20 AM
        To: APBR_analysis@yahoogroups.com
        Subject: [APBR_analysis] Re: - wins/Tendex replacement


        I think at this time of year there's also multiple levels of "freely
        available" -- Level 1 is the guys who get waived in March (Dion
        Glover, Rod Strickland, etc.) so they can sign with a playoff
        contender. Level 2 is the guys the bad teams sign to replace them
        (Britton Johnsen, Josh Davis, etc.).... and Level 5,831 is Charles
        Oakley.



        --- In APBR_analysis@yahoogroups.com, "Michael Tamada" <tamada@o...>
        wrote:
        > -----Original Message-----
        > From: Kevin Pelton [mailto:kpelton08@h...]
        > Sent: Wednesday, March 17, 2004 11:45 AM
        >
        > >> > I'm not sure it will be easy to identify a replacement level.
        > >>
        > >> Wouldn't be easy, but I'd try various techniques: WAG (wild a**
        > >> guessing: what number seems reasonable?); looking at actual
        > >> transactions of players cut and added, especially those on 10-day
        > >> contracts; actual statistics of actual 12th men, etc. These
        > >> different techqniques would undoubtedly lead to different
        > >> definitions of what the replacement level is, but now at
        > >> least the range has been narrowed down.
        > >
        > >MikeT, I know you've mentioned this kind of study a number of
        times,
        > >and since I probably make more use of replacement level than
        anyone
        > >else here (Dan being the possible exception), I've wanted to do it
        > >for a long time and finally got around to it:
        > >
        > >http://www.hoopsworld.com/article_7557.shtml
        >
        >
        > That's really good stuff, exactly the sort of empirical study that
        was
        > needed. And I think we've got a good estimate of replacement
        level,
        > at least as measured by your efficiency statistic.
        >
        > The point about in-season vs truly "freely available" free agents
        is a
        > good one, and your stats show the importance of the distinction.
        You
        > seem to be saying, and I agree, that the higher level, out-of-
        season,
        > freely available free agent replacement level is the better one to
        use.
        > At least for making long run, multiple-season, comparisons of
        players.
        > Teams in the short run may end up with less-than-replacement-level
        > players due to contract restrictions, salary caps, or what not, but
        > those conditions will not or at least need not persist in the long
        run.
        >
        > I think your .425 or .430 estimate of the replacement level is
        probably
        > a good one, maybe it won't turn out to be 100% accurate but I'll
        bet it's
        > reasonably close to whatever the true figure is. Because you've
        used
        > multiple techniques to arrive at the same estimate: the 10th player
        > technique, and the free agents' stats technique.
        >
        > There is a subtle problem however with the 10th player technique.
        Intuitively,
        > one would think that the 12th player would be a better measure of
        the replacement
        > level player. But the stats of these 12th players might actually
        be poor
        > measures; they might be BELOW replacement value, and are only on
        the roster
        > due to guaranteed contracts or what not. Or they might be an 18-
        year old kid
        > being stashed on the roster but not expected to contribute yet. Or
        they might
        > just have turned out to have a bad year, with horrendous stats that
        plummet
        > them to 12th.
        >
        > So those are good reasons why the 12th men's stats are perhaps not
        a good
        > measure, they're likely to give us a figure that's below
        replacement level.
        > So taking the 10th man's stats, or more precisely the stats of the
        290th
        > guy in the league, might lead to better estimates.
        >
        > The subtle problem with all this is that your technique stacked the
        deck
        > against the 12th men from the beginning, because you order the
        players
        > by efficiency. So the 12th men are guaranteed to have the very
        lowest
        > efficiencies on your rosters, and thus have stats that are likely to
        > be below replacement level.
        >
        > And thus to come up with a figure that is a better estimate, you
        have to
        > "move up" the roster to the 10th position.
        >
        > A theoretically better technique (but one that is perhaps
        impossible to
        > actually execute) would be to look at the stats of players who we
        identify
        > as the 12th man *a priori*, i.e. before looking at their stats.
        Some of
        > these 12th men will be the Brian Cardinal types, with some decent
        stats.
        > Some of them will be the Ansu Sesay types, basic deadwood there to
        fill
        > the roster, literal replacement level players. Averaged together,
        their
        > efficiency will be higher than the average efficiency of the 12th
        man
        > calculated by looking at their actual stats and ordering them ("a
        posteriori"
        > stats rather than "a priori"). Indeed, if done correctly, I'd
        expect these
        > 12th men to have stats slightly above the replacement level, simply
        because
        > they were the 12th guys and not the 13th guys, the ones who truly
        will be
        > the replacements.
        >
        > The trouble with my a priori idea of course is how do we know,
        before
        > looking at the stats, who the 12th man is? For some teams, it's
        fairly
        > easy to identify, but for others, not. Probably we'd have to pick
        a couple
        > of players from each team, and estimate the stats of the 11th-12th
        men
        > collectively.
        >
        > There's another reason why the .425/.430 might be just a little
        high. It's
        > possible that the off-season pickups might not be as freely
        available as
        > we might think -- though undrafted, there might be a few teams
        willing to
        > snatch up these players, for their low cost, low risk, and
        potential
        > contributions. I.e. there may be some semi-decent free agents out
        there,
        > whose stats bring up the average, but who weren't really freely
        available
        > because other teams went after them and signed them. They were a
        bit better
        > than the truly freely available replacement players.
        >
        > However, you've got a great quote from Brian Cardinal which
        suggests that
        > this is not the case.
        >
        > Bottom line: I think this is a good estimate which likely is
        getting us into
        > the neighborhood of the replacement level. My guess is that if it
        turns out
        > not to be the correct value, it's because it's a shade high.
        >
        >
        > --MKT





        Yahoo! Groups Links
      Your message has been successfully submitted and would be delivered to recipients shortly.