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

Value of timeouts

Expand Messages
  • HoopStudies
    I am working on a research project right now to assess the value of timeouts. Right now, the specific focus is whether a timeout helps the offense or the
    Message 1 of 8 , Aug 30, 2002
    • 0 Attachment
      I am working on a research project right now to assess the value of
      timeouts. Right now, the specific focus is whether a timeout helps
      the offense or the defense. When an offense comes out of a timeout,
      do they have a better chance of scoring because they can set up a
      play or a worse chance because the defense can change?

      One of the issues we've run into is a bias associated with when
      timeouts are taken. Many timeouts are taken when a team doesn't have
      a great chance of winning. Down by 3 with 4 seconds to go, an
      offense is going to go for a three against a defense that knows they
      will go for a three -- the odds of the offense scoring are going to
      be low. I am considering ways of accounting for this so that I don't
      lose the info. Throwing out all "hopeless" scenarios is not my
      favorite option.

      But what I'm wondering is what other issues we should consider when
      looking at timeouts. I think the value of the timeout is fundamental
      to most anything else -- does it give you a better chance to score at
      all?

      Ideas, suggestions?

      DeanO
    • Michael K. Tamada
      ... Presumably you ll throw out the timeouts called to try to ice the free throw shooter... although I suppose a slightly interesting sub-project would be to
      Message 2 of 8 , Aug 30, 2002
      • 0 Attachment
        On Fri, 30 Aug 2002, HoopStudies wrote:

        >
        > I am working on a research project right now to assess the value of
        > timeouts. Right now, the specific focus is whether a timeout helps
        > the offense or the defense. When an offense comes out of a timeout,
        > do they have a better chance of scoring because they can set up a
        > play or a worse chance because the defense can change?
        >
        > One of the issues we've run into is a bias associated with when
        > timeouts are taken. Many timeouts are taken when a team doesn't have
        > a great chance of winning. Down by 3 with 4 seconds to go, an
        > offense is going to go for a three against a defense that knows they
        > will go for a three -- the odds of the offense scoring are going to
        > be low. I am considering ways of accounting for this so that I don't
        > lose the info. Throwing out all "hopeless" scenarios is not my
        > favorite option.
        >
        > But what I'm wondering is what other issues we should consider when
        > looking at timeouts. I think the value of the timeout is fundamental
        > to most anything else -- does it give you a better chance to score at
        > all?
        >
        > Ideas, suggestions?

        Presumably you'll throw out the timeouts called to try to "ice" the free
        throw shooter... although I suppose a slightly interesting sub-project
        would be to see if FT shooters ever get rattled by those timeouts.

        If your sample is big enough, I would go ahead and throw out the
        desperation timeouts. But I don't know if your sample is big enough.


        My superficial impressions is that the Lakers from the 1980s to today seem
        to have a much better chance of scoring after they call timeout.


        --MKT
      • Chuck Frizzell
        My initial impression is offenses are more successful. However, I wouldn t be surprised to find less disciplined offenses (young, turnover prone, etc) to be
        Message 3 of 8 , Aug 30, 2002
        • 0 Attachment
          My initial impression is offenses are more successful.
          However, I wouldn't be surprised to find less
          disciplined offenses (young, turnover prone, etc) to
          be marginally successful, if not unsuccessful.

          Depending on your sample, comparing field goal
          percentage and turnover rates out of a timeout vs. the
          rest of the game would obviously be a telling stat
          (for O and D). Perhaps comparing the success of
          timeouts with personnel changes (O and D), as opposed
          to without would give some insight on coaching
          decisions.

          --- HoopStudies <deano@...> wrote:
          >
          > I am working on a research project right now to
          > assess the value of
          > timeouts. Right now, the specific focus is whether
          > a timeout helps
          > the offense or the defense. When an offense comes
          > out of a timeout,
          > do they have a better chance of scoring because they
          > can set up a
          > play or a worse chance because the defense can
          > change?
          >
          > One of the issues we've run into is a bias
          > associated with when
          > timeouts are taken. Many timeouts are taken when a
          > team doesn't have
          > a great chance of winning. Down by 3 with 4 seconds
          > to go, an
          > offense is going to go for a three against a defense
          > that knows they
          > will go for a three -- the odds of the offense
          > scoring are going to
          > be low. I am considering ways of accounting for
          > this so that I don't
          > lose the info. Throwing out all "hopeless"
          > scenarios is not my
          > favorite option.
          >
          > But what I'm wondering is what other issues we
          > should consider when
          > looking at timeouts. I think the value of the
          > timeout is fundamental
          > to most anything else -- does it give you a better
          > chance to score at
          > all?
          >
          > Ideas, suggestions?
          >
          > DeanO
          >
          >


          __________________________________________________
          Do You Yahoo!?
          Yahoo! Health - Feel better, live better
          http://health.yahoo.com
        • thedawgsareout
          ... I think it would be very nice to have a statistical idea of the value of momentum timeouts. It might be difficult to pick out a specific set of criteria to
          Message 4 of 8 , Aug 30, 2002
          • 0 Attachment
            --- In APBR_analysis@y..., "HoopStudies" <deano@r...> wrote:
            > But what I'm wondering is what other issues we should consider
            > when looking at timeouts. I think the value of the timeout is
            > fundamental to most anything else -- does it give you a better
            > chance to score at all?

            I think it would be very nice to have a statistical idea of the
            value of momentum timeouts. It might be difficult to pick out a
            specific set of criteria to determine what timeouts are called to
            stop a run, but if you could isolate those and see what happens
            afterward compared to when teams on the same type of runs are not
            stopped by a timeout, that would be very valuable.
          • HoopStudies
            ... In our preliminary look, this is the only thing that was close to being statistically significant. Whereas our look at offensive success seemed to show
            Message 5 of 8 , Aug 30, 2002
            • 0 Attachment
              --- In APBR_analysis@y..., "thedawgsareout" <kpelton08@h...> wrote:
              > --- In APBR_analysis@y..., "HoopStudies" <deano@r...> wrote:
              > > But what I'm wondering is what other issues we should consider
              > > when looking at timeouts. I think the value of the timeout is
              > > fundamental to most anything else -- does it give you a better
              > > chance to score at all?
              >
              > I think it would be very nice to have a statistical idea of the
              > value of momentum timeouts. It might be difficult to pick out a
              > specific set of criteria to determine what timeouts are called to
              > stop a run, but if you could isolate those and see what happens
              > afterward compared to when teams on the same type of runs are not
              > stopped by a timeout, that would be very valuable.

              In our preliminary look, this is the only thing that was close to
              being statistically significant. Whereas our look at offensive
              success seemed to show almost no difference from normal (looking at
              play % -- scores divided by plays to account for FG%, FT attempts,
              and turnovers), teams that call timeouts to stop a run actually had a
              bit better chances of stopping that run than would be expected. I
              actually need to verify this.

              More reports later.

              DeanO
            • mikel_ind
              ... What do we know about runs of momentum? At the professional level, where panic is less frequent, and coaches get paid big bucks, there will still be some
              Message 6 of 8 , Aug 30, 2002
              • 0 Attachment
                --- In APBR_analysis@y..., "HoopStudies" <deano@r...> wrote:
                > --- In APBR_analysis@y..., "thedawgsareout" <kpelton08@h...> wrote:
                > > ... nice to have a statistical idea of the
                > > value of momentum timeouts. It might be difficult to pick out a
                > > specific set of criteria to determine what timeouts are called to
                > > stop a run, but if you could isolate those and see what happens
                > > afterward compared to when teams on the same type of runs are not
                > > stopped by a timeout, that would be very valuable.

                What do we know about runs of momentum? At the professional level,
                where panic is less frequent, and coaches get paid big bucks, there
                will still be some random accumulations of points favoring one team
                or the other.

                If a run is just statistical noise, then a timeout may well have a
                50% chance of coinciding with the end of the run. But so might a
                hard foul, a hustle play, or any of a number of basic strategies.


                > ...teams that call timeouts to stop a run actually had a
                > bit better chances of stopping that run than would be
                expected. ...> DeanO

                And, what is the value of an unused timeout? Being out of timeouts
                gives the opposition more options. Saving timeouts is good.

                The more polished team will not benefit from timeouts as much as the
                disorganized team will benefit. Bob Knight was the master at gaining
                the advantage in uninterrupted play. Having 4-5 seniors on the floor
                helped.
              • HoopStudies
                ... to ... not ... The accumulations do seem to be random, but the perception is real. Coaches definitely call timeouts when there seems to be a run. We re
                Message 7 of 8 , Aug 30, 2002
                • 0 Attachment
                  --- In APBR_analysis@y..., "mikel_ind" <msg_53@h...> wrote:
                  > --- In APBR_analysis@y..., "HoopStudies" <deano@r...> wrote:
                  > > --- In APBR_analysis@y..., "thedawgsareout" <kpelton08@h...>
                  wrote:
                  > > > ... nice to have a statistical idea of the
                  > > > value of momentum timeouts. It might be difficult to pick out a
                  > > > specific set of criteria to determine what timeouts are called
                  to
                  > > > stop a run, but if you could isolate those and see what happens
                  > > > afterward compared to when teams on the same type of runs are
                  not
                  > > > stopped by a timeout, that would be very valuable.
                  >
                  > What do we know about runs of momentum? At the professional level,
                  > where panic is less frequent, and coaches get paid big bucks, there
                  > will still be some random accumulations of points favoring one team
                  > or the other.
                  >

                  The accumulations do seem to be random, but the perception is real.
                  Coaches definitely call timeouts when there seems to be a run. We're
                  looking to see whether those timeouts cause the end of a run at a
                  rate that is significantly different than what you'd expect from
                  random chance.


                  > > ...teams that call timeouts to stop a run actually had a
                  > > bit better chances of stopping that run than would be
                  > expected. ...> DeanO
                  >
                  > And, what is the value of an unused timeout? Being out of timeouts
                  > gives the opposition more options. Saving timeouts is good.
                  >

                  Definitely true. You don't want to not have a timeout when you need
                  one.

                  > The more polished team will not benefit from timeouts as much as
                  the
                  > disorganized team will benefit. Bob Knight was the master at
                  gaining
                  > the advantage in uninterrupted play. Having 4-5 seniors on the
                  floor
                  > helped.

                  Well, this sounds like a future study.

                  DeanO
                • alleyoop2
                  Dean-- Great idea for a study. A couple random thoughts: -- Substitution. Teams can not only call a play or change defenses during timeouts, they can also
                  Message 8 of 8 , Sep 2, 2002
                  • 0 Attachment
                    Dean--

                    Great idea for a study. A couple random thoughts:

                    -- Substitution. Teams can not only call a play or change defenses
                    during timeouts, they can also change who is on the court. Of course,
                    trying to separate out the two effects might be a real drag.

                    -- Losing teams. As has been pointed out, teams often call timeout
                    when they are faced with a run by the opponent. This strongly
                    suggests, although I have no hard date to back this up, that losing
                    teams will use more timeouts than winning teams, and could
                    potentially bias the data if they used enough more.

                    Best of luck with it and keep us posted....

                    John
                  Your message has been successfully submitted and would be delivered to recipients shortly.