Sorry, an error occurred while loading the content.

## Pace prediction

Expand Messages
• I finally did what we talked about a long time ago here. I looked at how well we can predict the pace of a single game based upon the two teams average
Message 1 of 22 , Apr 12, 2003
• 0 Attachment
I finally did what we talked about a long time ago here. I looked at
how well we can predict the pace of a single game based upon the two
teams' average paces. I used a formula like Stratomatic uses:

GamePace = TmA_avgpacediff + TmB_avgpacediff

So if TmA is 4 possessions per game faster than normal and TmB is 1
possession per game slower than league average, then they
are "expected" to play at a pace 3 possessions faster than average.

Over the last few years, this has an average error of -0.07 poss per
game. It actually slightly underpredicts pace, which surprised me,
but not by any large amount. The mean absolute error was 3.1, as
opposed to 3.8 using just the league average as a predictor. It
predicted about 71% of games correctly with regard to whether they
would be faster or slower than average.

This implies to me that a team's average pace has a pretty small
impact on the actual game pace. Reducing 3.8 to 3.1 is only about
20%. There is a fair amount of "other" factors that may account for
pace changes. I'm not sure how much we can identify in those other
factors, though.

Time to brainstorm: What independent factors are there that can be
used to a priori predict the pace of a game? Playoffs used to be a
significant one -- the pace used to get a lot slower in the playoffs.
I haven't looked in a while. If that's the case, competitiveness
(how close the teams are and how good they are) would be another
factor (correlated to playoffs). Not sure what else.

Note: I did correct for overtime games.

DeanO
• It seems to me there is one fundamental omission in the game pace model presented below and that is relative team strength. Why is this important? Well, in
Message 2 of 22 , Apr 14, 2003
• 0 Attachment
It seems to me there is one fundamental omission in the game
pace model presented below and that is relative team strength.
Why is this important? Well, in close games - a function of
relative team strength - there is an incentive for the team that is
behind at the end of the game to foul early in the opposing
team's shot clock (increasing both the expected probability of
victory and expected margins of defeat). This has the statistical
effect of increasing the number of possessions in these
contests. Furthermore, one should expect this effect to be
greatest in the games that are neither blow outs or nearly
deadlocked. As to the functional form that would best identify
this effect, offhand, I am guessing that it would have to be some
non-linear specification of relative team strength.

***********

--- In APBR_analysis@yahoogroups.com, "aaronkoo"
<deano@r...> wrote:
>
> I finally did what we talked about a long time ago here. I looked
at
> how well we can predict the pace of a single game based
upon the two
> teams' average paces. I used a formula like Stratomatic uses:
>
> GamePace = TmA_avgpacediff + TmB_avgpacediff
>
> So if TmA is 4 possessions per game faster than normal and
TmB is 1
> possession per game slower than league average, then they
> are "expected" to play at a pace 3 possessions faster than
average.
>
> Over the last few years, this has an average error of -0.07 poss
per
> game. It actually slightly underpredicts pace, which surprised
me,
> but not by any large amount. The mean absolute error was
3.1, as
> opposed to 3.8 using just the league average as a predictor. It
> predicted about 71% of games correctly with regard to whether
they
> would be faster or slower than average.
>
> This implies to me that a team's average pace has a pretty
small
> impact on the actual game pace. Reducing 3.8 to 3.1 is only
about
> 20%. There is a fair amount of "other" factors that may account
for
> pace changes. I'm not sure how much we can identify in those
other
> factors, though.
>
> Time to brainstorm: What independent factors are there that
can be
> used to a priori predict the pace of a game? Playoffs used to
be a
> significant one -- the pace used to get a lot slower in the
playoffs.
> I haven't looked in a while. If that's the case, competitiveness
> (how close the teams are and how good they are) would be
another
> factor (correlated to playoffs). Not sure what else.
>
> Note: I did correct for overtime games.
>
> DeanO
• ... That s what I called competitiveness below. But I considered it a slowing factor -- only because that had been what had been observed. You re right,
Message 3 of 22 , Apr 14, 2003
• 0 Attachment
--- In APBR_analysis@yahoogroups.com, "schtevie2003" <schtevie@h...>
wrote:
> It seems to me there is one fundamental omission in the game
> pace model presented below and that is relative team strength.
> Why is this important? Well, in close games - a function of
> relative team strength - there is an incentive for the team that is
> behind at the end of the game to foul early in the opposing
> team's shot clock (increasing both the expected probability of
> victory and expected margins of defeat). This has the statistical
> effect of increasing the number of possessions in these
> contests. Furthermore, one should expect this effect to be
> greatest in the games that are neither blow outs or nearly
> deadlocked. As to the functional form that would best identify
> this effect, offhand, I am guessing that it would have to be some
> non-linear specification of relative team strength.

That's what I called "competitiveness" below. But I considered it a
slowing factor -- only because that had been what had been observed.
You're right, though, too. Teams slow the game when coaches want
more control in the playoffs, when they want to maximize every
possession, when the defense doesn't take possessions off. But
fouling at the end of the game definitely increases possessions.
Need to come up with a model that a priori can handle this. Or maybe
it's just that the spread in competitive games is going to be greater
due to this uncertainty.

DeanO

>
> ***********
>
> --- In APBR_analysis@yahoogroups.com, "aaronkoo"
> <deano@r...> wrote:
> >
> > I finally did what we talked about a long time ago here. I
looked
> at
> > how well we can predict the pace of a single game based
> upon the two
> > teams' average paces. I used a formula like Stratomatic uses:
> >
> > GamePace = TmA_avgpacediff + TmB_avgpacediff
> >
> > So if TmA is 4 possessions per game faster than normal and
> TmB is 1
> > possession per game slower than league average, then they
> > are "expected" to play at a pace 3 possessions faster than
> average.
> >
> > Over the last few years, this has an average error of -0.07 poss
> per
> > game. It actually slightly underpredicts pace, which surprised
> me,
> > but not by any large amount. The mean absolute error was
> 3.1, as
> > opposed to 3.8 using just the league average as a predictor. It
> > predicted about 71% of games correctly with regard to whether
> they
> > would be faster or slower than average.
> >
> > This implies to me that a team's average pace has a pretty
> small
> > impact on the actual game pace. Reducing 3.8 to 3.1 is only
> about
> > 20%. There is a fair amount of "other" factors that may account
> for
> > pace changes. I'm not sure how much we can identify in those
> other
> > factors, though.
> >
> > Time to brainstorm: What independent factors are there that
> can be
> > used to a priori predict the pace of a game? Playoffs used to
> be a
> > significant one -- the pace used to get a lot slower in the
> playoffs.
> > I haven't looked in a while. If that's the case, competitiveness
> > (how close the teams are and how good they are) would be
> another
> > factor (correlated to playoffs). Not sure what else.
> >
> > Note: I did correct for overtime games.
> >
> > DeanO
• ... Just a guess, but maybe scheduling situations? Back to backs in different cities, 3 in 4 s, 4 in 5 s. Not sure if it would slow the pace due to tired
Message 4 of 22 , Apr 14, 2003
• 0 Attachment
aaronkoo wrote:
>
> Time to brainstorm: What independent factors are there that can be
> used to a priori predict the pace of a game?

Just a guess, but maybe scheduling situations? Back to backs in different
cities, 3 in 4's, 4 in 5's. Not sure if it would slow the pace due to
tired advancement or raise it due to tired defense (particularly
transition) but some effect might show up.

--
Gary Collard
SABR-L Moderator
collardg@...
• Taking one step back, I am just wondering what is the particular interest of trying to predict game pace (or more precisely, I suppose, the number of
Message 5 of 22 , Apr 15, 2003
• 0 Attachment
Taking one step back, I am just wondering what is the particular
interest of trying to predict game pace (or more precisely, I
suppose, the number of possessions) in the regular season?
Of course more knowledge is better than less, but is there any
question as to the fact that pace is the tail that doesn't wag the
dog? That is to say that pace itself doesn't determine
"competitiveness" - except in the sense that a "high" game pace
can be indicative of bad shot selection but not the cause in any
meaningful sense of the term.

That said, comparing game paces in the regular season and the
post season does seem like a very interesting avenue of inquiry.
Why? Because it allows possible insight into the optimal
performances of teams during the regular season. (Under the
assumption that teams are playing their best - or at least trying to
- in the playoffs.) For example, if a team were found, all else
equal, to play at a different game pace in the post-season
(presumably slower, reflecting a more intensive defensive effort)
one could surmise the counterfactual cost of not pursuing this
"better" strategy duing the post-season, in terms of potential
games lost and better positioning for the playoffs.

But to repeat the initial question. Why do we care about game
pace?

*********************

--- In APBR_analysis@yahoogroups.com, "aaronkoo"
<deano@r...> wrote:
> --- In APBR_analysis@yahoogroups.com, "schtevie2003"
<schtevie@h...>
> wrote:
> > It seems to me there is one fundamental omission in the
game
> > pace model presented below and that is relative team
strength.
> > Why is this important? Well, in close games - a function of
> > relative team strength - there is an incentive for the team that
is
> > behind at the end of the game to foul early in the opposing
> > team's shot clock (increasing both the expected probability of
> > victory and expected margins of defeat). This has the
statistical
> > effect of increasing the number of possessions in these
> > contests. Furthermore, one should expect this effect to be
> > greatest in the games that are neither blow outs or nearly
> > deadlocked. As to the functional form that would best identify
> > this effect, offhand, I am guessing that it would have to be
some
> > non-linear specification of relative team strength.
>
> That's what I called "competitiveness" below. But I considered
it a
> slowing factor -- only because that had been what had been
observed.
> You're right, though, too. Teams slow the game when coaches
want
> more control in the playoffs, when they want to maximize every
> possession, when the defense doesn't take possessions off.
But
> fouling at the end of the game definitely increases
possessions.
> Need to come up with a model that a priori can handle this. Or
maybe
> it's just that the spread in competitive games is going to be
greater
> due to this uncertainty.
>
> DeanO
>
>
> >
> > ***********
> >
> > --- In APBR_analysis@yahoogroups.com, "aaronkoo"
> > <deano@r...> wrote:
> > >
> > > I finally did what we talked about a long time ago here. I
> looked
> > at
> > > how well we can predict the pace of a single game based
> > upon the two
> > > teams' average paces. I used a formula like Stratomatic
uses:
> > >
> > > GamePace = TmA_avgpacediff + TmB_avgpacediff
> > >
> > > So if TmA is 4 possessions per game faster than normal
and
> > TmB is 1
> > > possession per game slower than league average, then
they
> > > are "expected" to play at a pace 3 possessions faster than
> > average.
> > >
> > > Over the last few years, this has an average error of -0.07
poss
> > per
> > > game. It actually slightly underpredicts pace, which
surprised
> > me,
> > > but not by any large amount. The mean absolute error was
> > 3.1, as
> > > opposed to 3.8 using just the league average as a
predictor. It
> > > predicted about 71% of games correctly with regard to
whether
> > they
> > > would be faster or slower than average.
> > >
> > > This implies to me that a team's average pace has a pretty
> > small
> > > impact on the actual game pace. Reducing 3.8 to 3.1 is
only
> > about
> > > 20%. There is a fair amount of "other" factors that may
account
> > for
> > > pace changes. I'm not sure how much we can identify in
those
> > other
> > > factors, though.
> > >
> > > Time to brainstorm: What independent factors are there
that
> > can be
> > > used to a priori predict the pace of a game? Playoffs used
to
> > be a
> > > significant one -- the pace used to get a lot slower in the
> > playoffs.
> > > I haven't looked in a while. If that's the case,
competitiveness
> > > (how close the teams are and how good they are) would be
> > another
> > > factor (correlated to playoffs). Not sure what else.
> > >
> > > Note: I did correct for overtime games.
> > >
> > > DeanO
• ... to ... This is probably true and the reason why we re doing this second round of polling now on who will win the title. We seemed to KNOW that the Lakers
Message 6 of 22 , Apr 15, 2003
• 0 Attachment
--- In APBR_analysis@yahoogroups.com, "schtevie2003" <schtevie@h...>
wrote:
> That said, comparing game paces in the regular season and the
> post season does seem like a very interesting avenue of inquiry.
> Why? Because it allows possible insight into the optimal
> performances of teams during the regular season. (Under the
> assumption that teams are playing their best - or at least trying
to
> - in the playoffs.) For example, if a team were found, all else
> equal, to play at a different game pace in the post-season
> (presumably slower, reflecting a more intensive defensive effort)
> one could surmise the counterfactual cost of not pursuing this
> "better" strategy duing the post-season, in terms of potential
> games lost and better positioning for the playoffs.

This is probably true and the reason why we're doing this second
round of polling now on who will win the title. We seemed to KNOW
that the Lakers would win the title the last couple years despite not
having the best record. Some of that is injury, but a lot of it is
people perceiving they save their A game for the playoffs. This goes
completely counter to the long-held belief in baseball that clutch
play does not exist, so putting together the story in a very
convincing manner is important.

>
> But to repeat the initial question. Why do we care about game
> pace?

The question was posed about whether a slow team or a fast team could
dictate tempo. I didn't think that either could and that a formula
like the one tested would do reasonably. It turns out to be accurate
but doesn't explain a ton of the variation, implying that there may
be cases where one can do better with other methods. Maybe we
believe that certain slow teams can dictate pace and certain fast
teams can dictate pace and we know it through some non-quantified
thing. If we can find that, it may help improve prediction.
Ultimately, this then gets at whether teams play better or worse at
certain paces. I've found that most teams don't play all that
different, but there are definite exceptions. One I remember because
it's in the book is the 2002 Raptors who played much better at a
faster pace. I have found several others, but don't remember them.

>
> *********************
>
> --- In APBR_analysis@yahoogroups.com, "aaronkoo"
> <deano@r...> wrote:
> > --- In APBR_analysis@yahoogroups.com, "schtevie2003"
> <schtevie@h...>
> > wrote:
> > > It seems to me there is one fundamental omission in the
> game
> > > pace model presented below and that is relative team
> strength.
> > > Why is this important? Well, in close games - a function of
> > > relative team strength - there is an incentive for the team
that
> is
> > > behind at the end of the game to foul early in the opposing
> > > team's shot clock (increasing both the expected probability of
> > > victory and expected margins of defeat). This has the
> statistical
> > > effect of increasing the number of possessions in these
> > > contests. Furthermore, one should expect this effect to be
> > > greatest in the games that are neither blow outs or nearly
> > > deadlocked. As to the functional form that would best identify
> > > this effect, offhand, I am guessing that it would have to be
> some
> > > non-linear specification of relative team strength.
> >
> > That's what I called "competitiveness" below. But I considered
> it a
> > slowing factor -- only because that had been what had been
> observed.
> > You're right, though, too. Teams slow the game when coaches
> want
> > more control in the playoffs, when they want to maximize every
> > possession, when the defense doesn't take possessions off.
> But
> > fouling at the end of the game definitely increases
> possessions.
> > Need to come up with a model that a priori can handle this. Or
> maybe
> > it's just that the spread in competitive games is going to be
> greater
> > due to this uncertainty.
> >
> > DeanO
> >
> >
> > >
> > > ***********
> > >
> > > --- In APBR_analysis@yahoogroups.com, "aaronkoo"
> > > <deano@r...> wrote:
> > > >
> > > > I finally did what we talked about a long time ago here. I
> > looked
> > > at
> > > > how well we can predict the pace of a single game based
> > > upon the two
> > > > teams' average paces. I used a formula like Stratomatic
> uses:
> > > >
> > > > GamePace = TmA_avgpacediff + TmB_avgpacediff
> > > >
> > > > So if TmA is 4 possessions per game faster than normal
> and
> > > TmB is 1
> > > > possession per game slower than league average, then
> they
> > > > are "expected" to play at a pace 3 possessions faster than
> > > average.
> > > >
> > > > Over the last few years, this has an average error of -0.07
> poss
> > > per
> > > > game. It actually slightly underpredicts pace, which
> surprised
> > > me,
> > > > but not by any large amount. The mean absolute error was
> > > 3.1, as
> > > > opposed to 3.8 using just the league average as a
> predictor. It
> > > > predicted about 71% of games correctly with regard to
> whether
> > > they
> > > > would be faster or slower than average.
> > > >
> > > > This implies to me that a team's average pace has a pretty
> > > small
> > > > impact on the actual game pace. Reducing 3.8 to 3.1 is
> only
> > > about
> > > > 20%. There is a fair amount of "other" factors that may
> account
> > > for
> > > > pace changes. I'm not sure how much we can identify in
> those
> > > other
> > > > factors, though.
> > > >
> > > > Time to brainstorm: What independent factors are there
> that
> > > can be
> > > > used to a priori predict the pace of a game? Playoffs used
> to
> > > be a
> > > > significant one -- the pace used to get a lot slower in the
> > > playoffs.
> > > > I haven't looked in a while. If that's the case,
> competitiveness
> > > > (how close the teams are and how good they are) would be
> > > another
> > > > factor (correlated to playoffs). Not sure what else.
> > > >
> > > > Note: I did correct for overtime games.
> > > >
> > > > DeanO
• ... It seems to me that the linguistic analogy for clutch play in baseball is the hot hand in basketball, and I think that people are correct to disbelieve
Message 7 of 22 , Apr 15, 2003
• 0 Attachment
--- In APBR_analysis@yahoogroups.com, "aaronkoo" <deano@r...> wrote:
> --- In APBR_analysis@yahoogroups.com, "schtevie2003" <schtevie@h...>
> wrote:
> > That said, comparing game paces in the regular season and the
> > post season does seem like a very interesting avenue of inquiry.
> > Why? Because it allows possible insight into the optimal
> > performances of teams during the regular season. (Under the
> > assumption that teams are playing their best - or at least trying
> to
> > - in the playoffs.) For example, if a team were found, all else
> > equal, to play at a different game pace in the post-season
> > (presumably slower, reflecting a more intensive defensive effort)
> > one could surmise the counterfactual cost of not pursuing this
> > "better" strategy duing the post-season, in terms of potential
> > games lost and better positioning for the playoffs.
>
> This is probably true and the reason why we're doing this second
> round of polling now on who will win the title. We seemed to KNOW
> that the Lakers would win the title the last couple years despite not
> having the best record. Some of that is injury, but a lot of it is
> people perceiving they save their A game for the playoffs. This goes
> completely counter to the long-held belief in baseball that clutch
> play does not exist, so putting together the story in a very
> convincing manner is important.

It seems to me that the linguistic analogy for "clutch play" in
baseball is the "hot hand" in basketball, and I think that people are
correct to disbelieve it. That said, there is a certain rationality
for a team that knows they are likely the best team in basketball not
to give its all in the regular season - especially if it doesn't feel
that home court advantage matters much to it.

>
> >
> > But to repeat the initial question. Why do we care about game
> > pace?
>
> The question was posed about whether a slow team or a fast team could
> dictate tempo. I didn't think that either could and that a formula
> like the one tested would do reasonably. It turns out to be accurate
> but doesn't explain a ton of the variation, implying that there may
> be cases where one can do better with other methods. Maybe we
> believe that certain slow teams can dictate pace and certain fast
> teams can dictate pace and we know it through some non-quantified
> thing. If we can find that, it may help improve prediction.
> Ultimately, this then gets at whether teams play better or worse at
> certain paces. I've found that most teams don't play all that
> different, but there are definite exceptions. One I remember because
> it's in the book is the 2002 Raptors who played much better at a
> faster pace. I have found several others, but don't remember them.
>

First, a comment on semantics only because it apparently influences
thinking here. How exactly can a team dictate tempo to another? If a
given team plays best (in the sense that doing so gives it its best
expectation of victory) by settling into its half-court offense, say,
how could anything that its opponent does compel a change its behavior?
I suppose if the opponent is a running team and the team, lacking
poise, decides to mimic it one could call this "dictation", but it
isn't, is it?

Second, out of curiosity, how was the exceptional behavior of the 2002
Raptors identified? That is how does one tell if a correlation of game
pace and productivity isn't spurious? The concern: even if one's null
hypothesis is that each team plays best at a given set offensive pace,
random variation would be expected to cough up a team which happens to
perform better when the game pace happens to be higher than average.
• ... ... are ... See my book in October. The original study missed something, intentionally or not. It missed its own evidence that hot hands
Message 8 of 22 , Apr 15, 2003
• 0 Attachment
--- In APBR_analysis@yahoogroups.com, "schtevie2003" <schtevie@h...>
wrote:
> --- In APBR_analysis@yahoogroups.com, "aaronkoo" <deano@r...> wrote:
> > --- In APBR_analysis@yahoogroups.com, "schtevie2003"
<schtevie@h...>
> > wrote:
> It seems to me that the linguistic analogy for "clutch play" in
> baseball is the "hot hand" in basketball, and I think that people
are
> correct to disbelieve it.

See my book in October. The original study missed something,
intentionally or not. It missed its own evidence that hot hands may
exist.

> That said, there is a certain rationality
> for a team that knows they are likely the best team in basketball
not
> to give its all in the regular season - especially if it doesn't
feel
> that home court advantage matters much to it.
>

Yes, a different type of clutch. Perhaps we need a better term for
it.

> > be cases where one can do better with other methods. Maybe we
> > believe that certain slow teams can dictate pace and certain fast
> > teams can dictate pace and we know it through some non-quantified
> > thing. If we can find that, it may help improve prediction.
> > Ultimately, this then gets at whether teams play better or worse
at
> > certain paces. I've found that most teams don't play all that
> > different, but there are definite exceptions. One I remember
because
> > it's in the book is the 2002 Raptors who played much better at a
> > faster pace. I have found several others, but don't remember
them.
> >
>
> First, a comment on semantics only because it apparently influences
> thinking here. How exactly can a team dictate tempo to another?
If a
> given team plays best (in the sense that doing so gives it its best
> expectation of victory) by settling into its half-court offense,
say,
> how could anything that its opponent does compel a change its
behavior?
> I suppose if the opponent is a running team and the team, lacking
> poise, decides to mimic it one could call this "dictation", but it
> isn't, is it?

Changing pace isn't hard to force on an opponent. A team that likes
to go slow and is facing a fast breaking team can go less hard to the
offensive boards, wait longer to take shots, pass up iffy fast break
opportunities, cut off passes to start breaks, and play zone (though
type of zone can matter). A team that likes to go uptempo and get in
the open court can strictly send 3 guys to the defensive boards and
strictly send guards to fast break locations (not send them to the
defensive glass). Or it can send 2 guys to the defensive boards with
3 releasing. It can play high pressure defense in a half or full
court.

>
> Second, out of curiosity, how was the exceptional behavior of the
2002
> Raptors identified? That is how does one tell if a correlation of
game
> pace and productivity isn't spurious? The concern: even if one's
null
> hypothesis is that each team plays best at a given set offensive
pace,
> random variation would be expected to cough up a team which happens
to
> perform better when the game pace happens to be higher than average.

This is a good point and one I haven't addressed in full yet in the
research. I believe the significance of what I found for the Raptors
was greater than 99%, though. So it is unlikely to be spurious but
possible. Ideally, we do more than use a statistical argument to say
this. There should be a mechanical explanation as well and I believe
there is one for the Raptors. Carter and Williams definitely prefer
an open court game. I'm not so sure about Peterson or Davis. I saw
strong correlation between forcing turnovers and pace with these
guys, so they probably should put pressure on opponents to force pace
and turnovers.

Frankly, though, I need to rework the entire methodology behind the
strategy analysis I've done. I've been doing it for a few years now
and, though I know how to use the results, they are too ugly to be
presented (too much of an art, not enough science), so they didn't
make it into the book. I've figured out a better way -- I think. I
need to test it when I have time.

DeanO
• ... Seems to me this is a much bigger factor in college, but a drop of water in the ocean in the NBA. Pro teams almost never foul intentionally until the final
Message 9 of 22 , Apr 15, 2003
• 0 Attachment
>there is an incentive for the team that is
> behind at the end of the game to foul early in the opposing
> team's shot clock (increasing both the expected probability of
> victory and expected margins of defeat)

Seems to me this is a much bigger factor in college, but a drop of
water in the ocean in the NBA. Pro teams almost never foul
intentionally until the final minute, and only do it then when the
math on playing it straight absolutely won't work. And whereas
college teams will hack even if they're down 15, NBA guys are jaded
enough that they need a reasonable chance of winning before they'll
whack a guy.
• ... of ... jaded ... they ll ... ****** Maybe so, but my guess is that it matters quite a bit. And my guess also is that teams that foul in the final minute
Message 10 of 22 , Apr 16, 2003
• 0 Attachment
--- In APBR_analysis@yahoogroups.com, "John Hollinger"
<alleyoop2@y...> wrote:
> >there is an incentive for the team that is
> > behind at the end of the game to foul early in the opposing
> > team's shot clock (increasing both the expected probability of
> > victory and expected margins of defeat)
>
>
> Seems to me this is a much bigger factor in college, but a drop
of
> water in the ocean in the NBA. Pro teams almost never foul
> intentionally until the final minute, and only do it then when the
> math on playing it straight absolutely won't work. And whereas
> college teams will hack even if they're down 15, NBA guys are
jaded
> enough that they need a reasonable chance of winning before
they'll
> whack a guy.

******

Maybe so, but my guess is that it matters quite a bit. And my
guess also is that teams that foul in the final minute are only
needing one extra possession or perhaps two for a chance at
victory. But if you are down by more, the rationality of fouling
kicks in well before two minutes. In any case, even if leads to
only two or three extra possessions, if must be a significant
amount of the unexplained variation.
• ... ... people ... something, ... may ... Divulging the type of evidence you refer to will in no way lessen my eagerness to buy the book come
Message 11 of 22 , Apr 16, 2003
• 0 Attachment
--- In APBR_analysis@yahoogroups.com, "aaronkoo"
<deano@r...> wrote:
> --- In APBR_analysis@yahoogroups.com, "schtevie2003"
<schtevie@h...>
> wrote:
> > --- In APBR_analysis@yahoogroups.com, "aaronkoo"
<deano@r...> wrote:
> > > --- In APBR_analysis@yahoogroups.com, "schtevie2003"
> <schtevie@h...>
> > > wrote:
> > It seems to me that the linguistic analogy for "clutch play" in
> > baseball is the "hot hand" in basketball, and I think that
people
> are
> > correct to disbelieve it.
>
> See my book in October. The original study missed
something,
> intentionally or not. It missed its own evidence that hot hands
may
> exist.

Divulging the type of evidence you refer to will in no way lessen
my eagerness to buy the book come October. (Or has it already
been spoken of in a previous posting?)

*******************

> > > be cases where one can do better with other methods.
Maybe we
> > > believe that certain slow teams can dictate pace and
certain fast
> > > teams can dictate pace and we know it through some
non-quantified
> > > thing. If we can find that, it may help improve prediction.
> > > Ultimately, this then gets at whether teams play better or
worse
> at
> > > certain paces. I've found that most teams don't play all that
> > > different, but there are definite exceptions. One I remember
> because
> > > it's in the book is the 2002 Raptors who played much better
at a
> > > faster pace. I have found several others, but don't
remember
> them.
> > >
> >
> > First, a comment on semantics only because it apparently
influences
> > thinking here. How exactly can a team dictate tempo to
another?
> If a
> > given team plays best (in the sense that doing so gives it its
best
> > expectation of victory) by settling into its half-court offense,
> say,
> > how could anything that its opponent does compel a change
its
> behavior?
> > I suppose if the opponent is a running team and the team,
lacking
> > poise, decides to mimic it one could call this "dictation", but it
> > isn't, is it?
>
> Changing pace isn't hard to force on an opponent. A team that
likes
> to go slow and is facing a fast breaking team can go less hard
to the
> offensive boards, wait longer to take shots, pass up iffy fast
break
> opportunities, cut off passes to start breaks, and play zone
(though
> type of zone can matter). A team that likes to go uptempo and
get in
> the open court can strictly send 3 guys to the defensive boards
and
> strictly send guards to fast break locations (not send them to
the
> defensive glass). Or it can send 2 guys to the defensive
boards with
> 3 releasing. It can play high pressure defense in a half or full
> court.
>

Pardon me if I am belaboring what might be construed as a
trivial semantic point, but as I read your examples, they do not
describe ways in which one team forces a team to play at a
certain pace. Rather they describe both optimal and sub-optimal
strategy decisions taken by teams facing given opponents, but in
no case would I refer to these factors as explaining how one
team "dictates" pace to another. And the semantic distinction is
important, I would argue, as imprecise language is the easiest
way to being led astray analytically.

Taking your examples, one by one:

Slow Team Pace Control Variables:

1) Going less hard to the offensive boards. Comment: it may be
optimal for a "slow" team to do this if it is slow in getting back
defensively, but if it is optimal, it is not so for the purpose of
controlling the pace of the game, rather for not losing the game.

2) Waiting longer to take shots. Comment: This is by definition a
sub-optimal strategy (why would one deny oneself maximum
time on the shot clock to come up with the best possible) and
though one may be dictating pace in this instance, one is also
dictating to one's own team a higher probability of losing.

3) Cut of iffy fast breaks and passes that start fast breaks.
Comment: Again, to deny one's team the potential for an easy
basket is by definition a sub-optimal strategy. Again, a team
dictates pace at a high price.

4) Playing a zone. Comment: Here again, pace isn't the issue, it
is choosing an optimal defense.

Fast Team Pace Control Variables:

I guess I won't list them, but they all describe strategy choices
that relate to controlling the probability of victory, not pace.

So to repeat the bottom line, teams making conscious optimal
choices try to maximize the probability of victory not the pace of
the game. Teams choosing to influence the pace of the game
independently do so at their peril - unless of course I am
missing something.
• ... ... people ... something, ... may ... Divulging the type of evidence you refer to will in no way lessen my eagerness to buy the book come
Message 12 of 22 , Apr 16, 2003
• 0 Attachment
--- In APBR_analysis@yahoogroups.com, "aaronkoo"
<deano@r...> wrote:
> --- In APBR_analysis@yahoogroups.com, "schtevie2003"
<schtevie@h...>
> wrote:
> > --- In APBR_analysis@yahoogroups.com, "aaronkoo"
<deano@r...> wrote:
> > > --- In APBR_analysis@yahoogroups.com, "schtevie2003"
> <schtevie@h...>
> > > wrote:
> > It seems to me that the linguistic analogy for "clutch play" in
> > baseball is the "hot hand" in basketball, and I think that
people
> are
> > correct to disbelieve it.
>
> See my book in October. The original study missed
something,
> intentionally or not. It missed its own evidence that hot hands
may
> exist.

Divulging the type of evidence you refer to will in no way lessen
my eagerness to buy the book come October. (Or has it already
been spoken of in a previous posting?)

*******************

> > > be cases where one can do better with other methods.
Maybe we
> > > believe that certain slow teams can dictate pace and
certain fast
> > > teams can dictate pace and we know it through some
non-quantified
> > > thing. If we can find that, it may help improve prediction.
> > > Ultimately, this then gets at whether teams play better or
worse
> at
> > > certain paces. I've found that most teams don't play all that
> > > different, but there are definite exceptions. One I remember
> because
> > > it's in the book is the 2002 Raptors who played much better
at a
> > > faster pace. I have found several others, but don't
remember
> them.
> > >
> >
> > First, a comment on semantics only because it apparently
influences
> > thinking here. How exactly can a team dictate tempo to
another?
> If a
> > given team plays best (in the sense that doing so gives it its
best
> > expectation of victory) by settling into its half-court offense,
> say,
> > how could anything that its opponent does compel a change
its
> behavior?
> > I suppose if the opponent is a running team and the team,
lacking
> > poise, decides to mimic it one could call this "dictation", but it
> > isn't, is it?
>
> Changing pace isn't hard to force on an opponent. A team that
likes
> to go slow and is facing a fast breaking team can go less hard
to the
> offensive boards, wait longer to take shots, pass up iffy fast
break
> opportunities, cut off passes to start breaks, and play zone
(though
> type of zone can matter). A team that likes to go uptempo and
get in
> the open court can strictly send 3 guys to the defensive boards
and
> strictly send guards to fast break locations (not send them to
the
> defensive glass). Or it can send 2 guys to the defensive
boards with
> 3 releasing. It can play high pressure defense in a half or full
> court.
>

Pardon me if I am belaboring what might be construed as a
trivial semantic point, but as I read your examples, they do not
describe ways in which one team forces a team to play at a
certain pace. Rather they describe both optimal and sub-optimal
strategy decisions taken by teams facing given opponents, but in
no case would I refer to these factors as explaining how one
team "dictates" pace to another. And the semantic distinction is
important, I would argue, as imprecise language is the easiest
way to being led astray analytically.

Taking your examples, one by one:

Slow Team Pace Control Variables:

1) Going less hard to the offensive boards. Comment: it may be
optimal for a "slow" team to do this if it is slow in getting back
defensively, but if it is optimal, it is not so for the purpose of
controlling the pace of the game, rather for not losing the game.

2) Waiting longer to take shots. Comment: This is by definition a
sub-optimal strategy (why would one deny oneself maximum
time on the shot clock to come up with the best possible) and
though one may be dictating pace in this instance, one is also
dictating to one's own team a higher probability of losing.

3) Cut of iffy fast breaks and passes that start fast breaks.
Comment: Again, to deny one's team the potential for an easy
basket is by definition a sub-optimal strategy. Again, a team
dictates pace at a high price.

4) Playing a zone. Comment: Here again, pace isn't the issue, it
is choosing an optimal defense.

Fast Team Pace Control Variables:

I guess I won't list them, but they all describe strategy choices
that relate to controlling the probability of victory, not pace.

So to repeat the bottom line, teams making conscious optimal
choices try to maximize the probability of victory not the pace of
the game. Teams choosing to influence the pace of the game
independently do so at their peril - unless of course I am
missing something.
• ... Basically, observers and shooters were somehow able to predict whether the next shot would go in better than would be expected by pure chance. It was
Message 13 of 22 , Apr 16, 2003
• 0 Attachment
--- In APBR_analysis@yahoogroups.com, "schtevie2003" <schtevie@h...>
wrote:
> --- In APBR_analysis@yahoogroups.com, "aaronkoo"
> <deano@r...> wrote:
> > --- In APBR_analysis@yahoogroups.com, "schtevie2003"
> <schtevie@h...>
> > wrote:
> > > --- In APBR_analysis@yahoogroups.com, "aaronkoo"
> <deano@r...> wrote:
> > > > --- In APBR_analysis@yahoogroups.com, "schtevie2003"
> > <schtevie@h...>
> > > > wrote:
> > > It seems to me that the linguistic analogy for "clutch play" in
> > > baseball is the "hot hand" in basketball, and I think that
> people
> > are
> > > correct to disbelieve it.
> >
> > See my book in October. The original study missed
> something,
> > intentionally or not. It missed its own evidence that hot hands
> may
> > exist.
>
> Divulging the type of evidence you refer to will in no way lessen
> my eagerness to buy the book come October. (Or has it already
> been spoken of in a previous posting?)

Basically, observers and shooters were somehow able to predict
whether the next shot would go in better than would be expected by
pure chance. It was significant at greater than 95%, but not framed
that way in Tversky's paper. He didn't do that test, oddly.

>
> Pardon me if I am belaboring what might be construed as a
> trivial semantic point, but as I read your examples, they do not
> describe ways in which one team forces a team to play at a
> certain pace. Rather they describe both optimal and sub-optimal
> strategy decisions taken by teams facing given opponents, but in
> no case would I refer to these factors as explaining how one
> team "dictates" pace to another. And the semantic distinction is
> important, I would argue, as imprecise language is the easiest
> way to being led astray analytically.
>
> Taking your examples, one by one:
>
> Slow Team Pace Control Variables:
>
> 1) Going less hard to the offensive boards. Comment: it may be
> optimal for a "slow" team to do this if it is slow in getting back
> defensively, but if it is optimal, it is not so for the purpose of
> controlling the pace of the game, rather for not losing the game.
>

You're assuming that there is a single "optimal" strategy for each
team. I'd say that optimal strategy varies depending upon opponent,
depending on your status as an underdog/favorite, and other things.
Against some opponents, those strategies _that lead to pace slowing_
may be more valuable than against others, regardless of your own
personnel. Slowing the pace also makes sense if you are a general
underdog but grab an early lead. That's pretty standard coaching
actually...

> 2) Waiting longer to take shots. Comment: This is by definition a
> sub-optimal strategy (why would one deny oneself maximum
> time on the shot clock to come up with the best possible) and
> though one may be dictating pace in this instance, one is also
> dictating to one's own team a higher probability of losing.
>

If you have a lead against a tougher opponent, you want to run that
clock as much as possible (shorten the game). It adds as much as
about 10% to your odds of winning. There are also theories (tougher
to prove) that defenses get less effective the longer in the clock
they have to play. I've disproved this when there are <3 s on the
clock, but it's difficult to say before that time.

> 3) Cut of iffy fast breaks and passes that start fast breaks.
> Comment: Again, to deny one's team the potential for an easy
> basket is by definition a sub-optimal strategy. Again, a team
> dictates pace at a high price.
>
> 4) Playing a zone. Comment: Here again, pace isn't the issue, it
> is choosing an optimal defense.

You are right that there is some cause-effect issue between strategy
and pace. If a zone is better against a certain team, it can slow
the pace. But there are some inherent benefits of slow pace if you
are an underdog EVEN IF you are slightly worse at working at a slow
pace.

See http://www.rawbw.com/~deano/articles/aa030597.htm for an
interactive example.

>
> Fast Team Pace Control Variables:
>
> I guess I won't list them, but they all describe strategy choices
> that relate to controlling the probability of victory, not pace.
>
> So to repeat the bottom line, teams making conscious optimal
> choices try to maximize the probability of victory not the pace of
> the game. Teams choosing to influence the pace of the game
> independently do so at their peril - unless of course I am
> missing something.

The Pistons of the late '80's were highly noted for beating the
Lakers by slowing the pace. If they could play at their tempo,
they'd win. Yes, some of that implied that they just could implement
their strategies successfully, which _caused_ the pace to slow. That
was part of the point of the discussion -- what could they do to slow
that pace successfully?

Actually, it would be interesting to look back at their series and
see whether pace was important in deciding winner-loser...

DeanO
• ... Thanks for the teaser. I look forward to October! ... Responding to the various points in your above remarks...Actually, I am not assuming that there is a
Message 14 of 22 , Apr 16, 2003
• 0 Attachment
--- In APBR_analysis@yahoogroups.com, "aaronkoo" <deano@r...> wrote:
> --- In APBR_analysis@yahoogroups.com, "schtevie2003" <schtevie@h...>
> wrote:
> > --- In APBR_analysis@yahoogroups.com, "aaronkoo"
> > <deano@r...> wrote:
> > > --- In APBR_analysis@yahoogroups.com, "schtevie2003"
> > <schtevie@h...>
> > > wrote:
> > > > --- In APBR_analysis@yahoogroups.com, "aaronkoo"
> > <deano@r...> wrote:
> > > > > --- In APBR_analysis@yahoogroups.com, "schtevie2003"
> > > <schtevie@h...>
> > > > > wrote:
> > > > It seems to me that the linguistic analogy for "clutch play" in
> > > > baseball is the "hot hand" in basketball, and I think that
> > people
> > > are
> > > > correct to disbelieve it.
> > >
> > > See my book in October. The original study missed
> > something,
> > > intentionally or not. It missed its own evidence that hot hands
> > may
> > > exist.
> >
> > Divulging the type of evidence you refer to will in no way lessen
> > my eagerness to buy the book come October. (Or has it already
> > been spoken of in a previous posting?)
>
> Basically, observers and shooters were somehow able to predict
> whether the next shot would go in better than would be expected by
> pure chance. It was significant at greater than 95%, but not framed
> that way in Tversky's paper. He didn't do that test, oddly.

Thanks for the teaser. I look forward to October!

************************

> > Pardon me if I am belaboring what might be construed as a
> > trivial semantic point, but as I read your examples, they do not
> > describe ways in which one team forces a team to play at a
> > certain pace. Rather they describe both optimal and sub-optimal
> > strategy decisions taken by teams facing given opponents, but in
> > no case would I refer to these factors as explaining how one
> > team "dictates" pace to another. And the semantic distinction is
> > important, I would argue, as imprecise language is the easiest
> > way to being led astray analytically.
> >
> > Taking your examples, one by one:
> >
> > Slow Team Pace Control Variables:
> >
> > 1) Going less hard to the offensive boards. Comment: it may be
> > optimal for a "slow" team to do this if it is slow in getting back
> > defensively, but if it is optimal, it is not so for the purpose of
> > controlling the pace of the game, rather for not losing the game.
> >
>
> You're assuming that there is a single "optimal" strategy for each
> team. I'd say that optimal strategy varies depending upon opponent,
> depending on your status as an underdog/favorite, and other things.
> Against some opponents, those strategies _that lead to pace slowing_
> may be more valuable than against others, regardless of your own
> personnel. Slowing the pace also makes sense if you are a general
> underdog but grab an early lead. That's pretty standard coaching
> actually...

Responding to the various points in your above remarks...Actually, I am
not assuming that there is a single "optimal" strategy for each team.
I am assuming that in expectation, for a given team pairing, there is
an optimal strategy for both sides (reflecting mutual expectations of
plays that are to be run.) And as such, the point about pace stands,
in reality, strategy choices are made for the purpose of winning, not
on independently influencing the pace of the game for the purpose of
winning. As for optimal strategy depending on underdog versus favorite
status, I don't see how this is the case. As an underdog, you expect
to lose; you just do the best such that if luck is on your side, you
get to "steal" a victory. As far as choosing a slower pace if you are
an underdog but early game luck sees you in the lead, of course that is
true at a certain point in the game. How late in the game one should
resort to a conscious slowing, where by definition one is forgoing
one's own offensive productivity for the added "value" (in terms of
expectation of victory) of time running off the clock, is a very
interesting research question, but I would be willing to bet money on
short odds that it is something which should occur very late in the
game (that is the time threshold is not very sensitive to the realized
lead, especially if one is an underdog).

*******************

> > 2) Waiting longer to take shots. Comment: This is by definition a
> > sub-optimal strategy (why would one deny oneself maximum
> > time on the shot clock to come up with the best possible) and
> > though one may be dictating pace in this instance, one is also
> > dictating to one's own team a higher probability of losing.
> >
>
> If you have a lead against a tougher opponent, you want to run that
> clock as much as possible (shorten the game). It adds as much as
> about 10% to your odds of winning. There are also theories (tougher
> to prove) that defenses get less effective the longer in the clock
> they have to play. I've disproved this when there are <3 s on the
> clock, but it's difficult to say before that time.

Where did this 10% statistic come from, and what does it depend on? I
am very curious. As far as defenses getting less effective the longer
in the clock they have to play, that may be the case, but it is besides
the point I make above. Restating: There is an expected optimal
offense which takes an expected amount of time, on average, off the
shot clock. Deciding independently to change the average time consumed
on the shot clock, will necessarily lower the expected points per
possession, unless this derivative is zero - in which case, and in only
which case, one can control pace.

> > 3) Cut of iffy fast breaks and passes that start fast breaks.
> > Comment: Again, to deny one's team the potential for an easy
> > basket is by definition a sub-optimal strategy. Again, a team
> > dictates pace at a high price.
> >
> > 4) Playing a zone. Comment: Here again, pace isn't the issue, it
> > is choosing an optimal defense.
>
> You are right that there is some cause-effect issue between strategy
> and pace. If a zone is better against a certain team, it can slow
> the pace. But there are some inherent benefits of slow pace if you
> are an underdog EVEN IF you are slightly worse at working at a slow
> pace.
>
> See http://www.rawbw.com/~deano/articles/aa030597.htm for an
> interactive example.

Yes, I remember years ago reading this article and being heartened,
perhaps for the first time, that that there was a fellow traveler in
this world. And, in fact, I was going to add some words on this in the
preceding note, but didn't because it was my impression that the "fewer
unfair coin flips if your team sucks is a good thing" effect was
ultimately more of a theoretic curiosity than significant possible
competitive advantage. A question: how confident are you that the 8%
gain shown in the Cav-Bulls example is robust in terms of behavioral
assumptions? In terms of what I wrote above, it implies that the
derivative referred to above (expected points per possession as a
function of elapsed time on shot clock) is rather flat. Because one
would never rationally turn down a fast break layup for the purpose of
slowing down the pace of the game. Now, it may have been true for the
Cavs if their baseline offense was to waste shot clock time anyway, but
I have my suspicions. Shot clock time is valuable!!!

> > Fast Team Pace Control Variables:
> >
> > I guess I won't list them, but they all describe strategy choices
> > that relate to controlling the probability of victory, not pace.
> >
> > So to repeat the bottom line, teams making conscious optimal
> > choices try to maximize the probability of victory not the pace of
> > the game. Teams choosing to influence the pace of the game
> > independently do so at their peril - unless of course I am
> > missing something.
>
> The Pistons of the late '80's were highly noted for beating the
> Lakers by slowing the pace. If they could play at their tempo,
> they'd win. Yes, some of that implied that they just could implement
> their strategies successfully, which _caused_ the pace to slow. That
> was part of the point of the discussion -- what could they do to slow
> that pace successfully?
>
> Actually, it would be interesting to look back at their series and
> see whether pace was important in deciding winner-loser...
>
> DeanO

This is the kind of impressionistic observation that I am most leery
of. Were these games slower than expected? Of course, if the Lakers
were a fast break team and there was a slow game pace, chances are
there would have been fewer realized fast breaks, hence, all else
equal, one might expect that in these games the Pistons would have been
the victors.

All this said, besides the end of game effect of intentionally slowing
the game to protect an "insurmountable" lead (but then again, in terms
of detecting this effect empirically, remember that the apparent pace
of the game in the final box score would not necessarily reflect this
eventuality for the appearance of extra foul generated possessions by
the team attempting to catch up) and the possibility of underdogs whose
half-court offensive productivities aren't dependent on time (those
with but one shooter, who can get his equivalent, open shot whenever?)
is there any other variable where a team, by choosing to influence
pace, is not sacrificing competitiveness?
• One quick response but gotta work. You are basically assuming that offensive performance declines with time left on the shot clock. What I ve found is that it
Message 15 of 22 , Apr 16, 2003
• 0 Attachment
One quick response but gotta work.

You are basically assuming that offensive performance declines with
time left on the shot clock. What I've found is that it declines
with <3 s on the clock but is not significantly impacted before
then. That implies that teams have pretty constant offensive
performance up to that point and that the assumptions in that web
page are pretty accurate up to the point of <3 s on the clock.

So, yes, I do think 8% is within the realm of possibility. 8% ain't
a lot to gain and can be offset if your team doesn't play well at a
slow pace (they defy the more general study above, which is
definitely possible). And, yes, throwing away fast breaks doesn't
make sense.

But I'd take money on that bet. I have seen many times in college
especially where slowing the game was used successfully. The most
prominent time was when I was at UNC at Florida St. got an early lead
against the eventual National Champs, let Sam Cassell dribble until
there was about 10 s on the clock, then attacked. They did that from
midway through the first half. As long as they could hold that lead,
they stuck with it. It also made sense because they weren't deep
(they could rest a lot) so there were secondary factors, but the
strategy was go slow (and shoot 3s and stick in a zone). The longer
the shot clock, the more valuable slowing can be early on. In the
NBA, it is tougher with the short clock.

DeanO

--- In APBR_analysis@yahoogroups.com, "schtevie2003" <schtevie@h...>
wrote:
> --- In APBR_analysis@yahoogroups.com, "aaronkoo" <deano@r...> wrote:
> > --- In APBR_analysis@yahoogroups.com, "schtevie2003"
<schtevie@h...>
> > wrote:
> > > --- In APBR_analysis@yahoogroups.com, "aaronkoo"
> > > <deano@r...> wrote:
> > > > --- In APBR_analysis@yahoogroups.com, "schtevie2003"
> > > <schtevie@h...>
> > > > wrote:
> > > > > --- In APBR_analysis@yahoogroups.com, "aaronkoo"
> > > <deano@r...> wrote:
> > > > > > --- In APBR_analysis@yahoogroups.com, "schtevie2003"
> > > > <schtevie@h...>
> > > > > > wrote:
> > > > > It seems to me that the linguistic analogy for "clutch
play" in
> > > > > baseball is the "hot hand" in basketball, and I think that
> > > people
> > > > are
> > > > > correct to disbelieve it.
> > > >
> > > > See my book in October. The original study missed
> > > something,
> > > > intentionally or not. It missed its own evidence that hot
hands
> > > may
> > > > exist.
> > >
> > > Divulging the type of evidence you refer to will in no way
lessen
> > > my eagerness to buy the book come October. (Or has it already
> > > been spoken of in a previous posting?)
> >
> > Basically, observers and shooters were somehow able to predict
> > whether the next shot would go in better than would be expected
by
> > pure chance. It was significant at greater than 95%, but not
framed
> > that way in Tversky's paper. He didn't do that test, oddly.
>
> Thanks for the teaser. I look forward to October!
>
> ************************
>
> > > Pardon me if I am belaboring what might be construed as a
> > > trivial semantic point, but as I read your examples, they do
not
> > > describe ways in which one team forces a team to play at a
> > > certain pace. Rather they describe both optimal and sub-
optimal
> > > strategy decisions taken by teams facing given opponents, but
in
> > > no case would I refer to these factors as explaining how one
> > > team "dictates" pace to another. And the semantic distinction
is
> > > important, I would argue, as imprecise language is the easiest
> > > way to being led astray analytically.
> > >
> > > Taking your examples, one by one:
> > >
> > > Slow Team Pace Control Variables:
> > >
> > > 1) Going less hard to the offensive boards. Comment: it may be
> > > optimal for a "slow" team to do this if it is slow in getting
back
> > > defensively, but if it is optimal, it is not so for the purpose
of
> > > controlling the pace of the game, rather for not losing the
game.
> > >
> >
> > You're assuming that there is a single "optimal" strategy for
each
> > team. I'd say that optimal strategy varies depending upon
opponent,
> > depending on your status as an underdog/favorite, and other
things.
> > Against some opponents, those strategies _that lead to pace
slowing_
> > may be more valuable than against others, regardless of your own
> > personnel. Slowing the pace also makes sense if you are a
general
> > underdog but grab an early lead. That's pretty standard coaching
> > actually...
>
> Responding to the various points in your above remarks...Actually,
I am
> not assuming that there is a single "optimal" strategy for each
team.
> I am assuming that in expectation, for a given team pairing, there
is
> an optimal strategy for both sides (reflecting mutual expectations
of
> plays that are to be run.) And as such, the point about pace
stands,
> in reality, strategy choices are made for the purpose of winning,
not
> on independently influencing the pace of the game for the purpose
of
> winning. As for optimal strategy depending on underdog versus
favorite
> status, I don't see how this is the case. As an underdog, you
expect
> to lose; you just do the best such that if luck is on your side,
you
> get to "steal" a victory. As far as choosing a slower pace if you
are
> an underdog but early game luck sees you in the lead, of course
that is
> true at a certain point in the game. How late in the game one
should
> resort to a conscious slowing, where by definition one is forgoing
> one's own offensive productivity for the added "value" (in terms of
> expectation of victory) of time running off the clock, is a very
> interesting research question, but I would be willing to bet money
on
> short odds that it is something which should occur very late in the
> game (that is the time threshold is not very sensitive to the
realized
> lead, especially if one is an underdog).
>
> *******************
>
> > > 2) Waiting longer to take shots. Comment: This is by
definition a
> > > sub-optimal strategy (why would one deny oneself maximum
> > > time on the shot clock to come up with the best possible) and
> > > though one may be dictating pace in this instance, one is also
> > > dictating to one's own team a higher probability of losing.
> > >
> >
> > If you have a lead against a tougher opponent, you want to run
that
> > clock as much as possible (shorten the game). It adds as much as
> > about 10% to your odds of winning. There are also theories
(tougher
> > to prove) that defenses get less effective the longer in the
clock
> > they have to play. I've disproved this when there are <3 s on
the
> > clock, but it's difficult to say before that time.
>
> Where did this 10% statistic come from, and what does it depend
on? I
> am very curious. As far as defenses getting less effective the
longer
> in the clock they have to play, that may be the case, but it is
besides
> the point I make above. Restating: There is an expected optimal
> offense which takes an expected amount of time, on average, off the
> shot clock. Deciding independently to change the average time
consumed
> on the shot clock, will necessarily lower the expected points per
> possession, unless this derivative is zero - in which case, and in
only
> which case, one can control pace.
>
> > > 3) Cut of iffy fast breaks and passes that start fast breaks.
> > > Comment: Again, to deny one's team the potential for an easy
> > > basket is by definition a sub-optimal strategy. Again, a team
> > > dictates pace at a high price.
> > >
> > > 4) Playing a zone. Comment: Here again, pace isn't the issue,
it
> > > is choosing an optimal defense.
> >
> > You are right that there is some cause-effect issue between
strategy
> > and pace. If a zone is better against a certain team, it can
slow
> > the pace. But there are some inherent benefits of slow pace if
you
> > are an underdog EVEN IF you are slightly worse at working at a
slow
> > pace.
> >
> > See http://www.rawbw.com/~deano/articles/aa030597.htm for an
> > interactive example.
>
> Yes, I remember years ago reading this article and being heartened,
> perhaps for the first time, that that there was a fellow traveler
in
> this world. And, in fact, I was going to add some words on this in
the
> preceding note, but didn't because it was my impression that
the "fewer
> unfair coin flips if your team sucks is a good thing" effect was
> ultimately more of a theoretic curiosity than significant possible
> competitive advantage. A question: how confident are you that the
8%
> gain shown in the Cav-Bulls example is robust in terms of
behavioral
> assumptions? In terms of what I wrote above, it implies that the
> derivative referred to above (expected points per possession as a
> function of elapsed time on shot clock) is rather flat. Because
one
> would never rationally turn down a fast break layup for the purpose
of
> slowing down the pace of the game. Now, it may have been true for
the
> Cavs if their baseline offense was to waste shot clock time anyway,
but
> I have my suspicions. Shot clock time is valuable!!!
>
>
> > > Fast Team Pace Control Variables:
> > >
> > > I guess I won't list them, but they all describe strategy
choices
> > > that relate to controlling the probability of victory, not pace.
> > >
> > > So to repeat the bottom line, teams making conscious optimal
> > > choices try to maximize the probability of victory not the pace
of
> > > the game. Teams choosing to influence the pace of the game
> > > independently do so at their peril - unless of course I am
> > > missing something.
> >
> > The Pistons of the late '80's were highly noted for beating the
> > Lakers by slowing the pace. If they could play at their tempo,
> > they'd win. Yes, some of that implied that they just could
implement
> > their strategies successfully, which _caused_ the pace to slow.
That
> > was part of the point of the discussion -- what could they do to
slow
> > that pace successfully?
> >
> > Actually, it would be interesting to look back at their series
and
> > see whether pace was important in deciding winner-loser...
> >
> > DeanO
>
> This is the kind of impressionistic observation that I am most
leery
> of. Were these games slower than expected? Of course, if the
Lakers
> were a fast break team and there was a slow game pace, chances are
> there would have been fewer realized fast breaks, hence, all else
> equal, one might expect that in these games the Pistons would have
been
> the victors.
>
> All this said, besides the end of game effect of intentionally
slowing
> the game to protect an "insurmountable" lead (but then again, in
terms
> of detecting this effect empirically, remember that the apparent
pace
> of the game in the final box score would not necessarily reflect
this
> eventuality for the appearance of extra foul generated possessions
by
> the team attempting to catch up) and the possibility of underdogs
whose
> half-court offensive productivities aren't dependent on time (those
> with but one shooter, who can get his equivalent, open shot
whenever?)
> is there any other variable where a team, by choosing to influence
> pace, is not sacrificing competitiveness?
• ... You are correct that I am effectively assuming that offensive performance declines with time left on the shot clock. But the real assumption is that teams
Message 16 of 22 , Apr 17, 2003
• 0 Attachment
--- In APBR_analysis@yahoogroups.com, "aaronkoo" <deano@r...> wrote:
>
> One quick response but gotta work.
>
> You are basically assuming that offensive performance declines with
> time left on the shot clock. What I've found is that it declines
> with <3 s on the clock but is not significantly impacted before
> then. That implies that teams have pretty constant offensive
> performance up to that point and that the assumptions in that web
> page are pretty accurate up to the point of <3 s on the clock.

You are correct that I am effectively assuming that offensive
performance declines with time left on the shot clock. But the real
assumption is that teams aspire to optimal play - that is maximizing
their offensive productivity - given this assumption, the theory is
dead solid that offensive performance is expected to decline as a
function of elapsed shot clock time.

> So, yes, I do think 8% is within the realm of possibility. 8% ain't
> a lot to gain and can be offset if your team doesn't play well at a
> slow pace (they defy the more general study above, which is
> definitely possible). And, yes, throwing away fast breaks doesn't
> make sense.

I think that 8% improvement is a very large number in the context of a
possibly costless competitive advantage - an expected 4 additional wins
per season for an average team in the league if I am understanding the
statistic correctly?

> But I'd take money on that bet. I have seen many times in college
> especially where slowing the game was used successfully. The most
> prominent time was when I was at UNC at Florida St. got an early lead
> against the eventual National Champs, let Sam Cassell dribble until
> there was about 10 s on the clock, then attacked. They did that from
> midway through the first half. As long as they could hold that lead,
> they stuck with it. It also made sense because they weren't deep
> (they could rest a lot) so there were secondary factors, but the
> strategy was go slow (and shoot 3s and stick in a zone). The longer
> the shot clock, the more valuable slowing can be early on. In the
> NBA, it is tougher with the short clock.
>
> DeanO

And I too remember some atrocious pre-shot clock college games. But
let's pose the question this way. Suppose two equal and average NBA
teams are playing, and one has a 10 point lead. Pick a time X before
the end of the game at which point you think it is optimal to alter
one's offensive strategy so that the first priority is controlling
pace.

**************

> --- In APBR_analysis@yahoogroups.com, "schtevie2003" <schtevie@h...>
> wrote:
> > --- In APBR_analysis@yahoogroups.com, "aaronkoo" <deano@r...> wrote:
> > > --- In APBR_analysis@yahoogroups.com, "schtevie2003"
> <schtevie@h...>
> > > wrote:
> > > > --- In APBR_analysis@yahoogroups.com, "aaronkoo"
> > > > <deano@r...> wrote:
> > > > > --- In APBR_analysis@yahoogroups.com, "schtevie2003"
> > > > <schtevie@h...>
> > > > > wrote:
> > > > > > --- In APBR_analysis@yahoogroups.com, "aaronkoo"
> > > > <deano@r...> wrote:
> > > > > > > --- In APBR_analysis@yahoogroups.com, "schtevie2003"
> > > > > <schtevie@h...>
> > > > > > > wrote:
> > > > > > It seems to me that the linguistic analogy for "clutch
> play" in
> > > > > > baseball is the "hot hand" in basketball, and I think that
> > > > people
> > > > > are
> > > > > > correct to disbelieve it.
> > > > >
> > > > > See my book in October. The original study missed
> > > > something,
> > > > > intentionally or not. It missed its own evidence that hot
> hands
> > > > may
> > > > > exist.
> > > >
> > > > Divulging the type of evidence you refer to will in no way
> lessen
> > > > my eagerness to buy the book come October. (Or has it already
> > > > been spoken of in a previous posting?)
> > >
> > > Basically, observers and shooters were somehow able to predict
> > > whether the next shot would go in better than would be expected
> by
> > > pure chance. It was significant at greater than 95%, but not
> framed
> > > that way in Tversky's paper. He didn't do that test, oddly.
> >
> > Thanks for the teaser. I look forward to October!
> >
> > ************************
> >
> > > > Pardon me if I am belaboring what might be construed as a
> > > > trivial semantic point, but as I read your examples, they do
> not
> > > > describe ways in which one team forces a team to play at a
> > > > certain pace. Rather they describe both optimal and sub-
> optimal
> > > > strategy decisions taken by teams facing given opponents, but
> in
> > > > no case would I refer to these factors as explaining how one
> > > > team "dictates" pace to another. And the semantic distinction
> is
> > > > important, I would argue, as imprecise language is the easiest
> > > > way to being led astray analytically.
> > > >
> > > > Taking your examples, one by one:
> > > >
> > > > Slow Team Pace Control Variables:
> > > >
> > > > 1) Going less hard to the offensive boards. Comment: it may be
> > > > optimal for a "slow" team to do this if it is slow in getting
> back
> > > > defensively, but if it is optimal, it is not so for the purpose
> of
> > > > controlling the pace of the game, rather for not losing the
> game.
> > > >
> > >
> > > You're assuming that there is a single "optimal" strategy for
> each
> > > team. I'd say that optimal strategy varies depending upon
> opponent,
> > > depending on your status as an underdog/favorite, and other
> things.
> > > Against some opponents, those strategies _that lead to pace
> slowing_
> > > may be more valuable than against others, regardless of your own
> > > personnel. Slowing the pace also makes sense if you are a
> general
> > > underdog but grab an early lead. That's pretty standard coaching
> > > actually...
> >
> > Responding to the various points in your above remarks...Actually,
> I am
> > not assuming that there is a single "optimal" strategy for each
> team.
> > I am assuming that in expectation, for a given team pairing, there
> is
> > an optimal strategy for both sides (reflecting mutual expectations
> of
> > plays that are to be run.) And as such, the point about pace
> stands,
> > in reality, strategy choices are made for the purpose of winning,
> not
> > on independently influencing the pace of the game for the purpose
> of
> > winning. As for optimal strategy depending on underdog versus
> favorite
> > status, I don't see how this is the case. As an underdog, you
> expect
> > to lose; you just do the best such that if luck is on your side,
> you
> > get to "steal" a victory. As far as choosing a slower pace if you
> are
> > an underdog but early game luck sees you in the lead, of course
> that is
> > true at a certain point in the game. How late in the game one
> should
> > resort to a conscious slowing, where by definition one is forgoing
> > one's own offensive productivity for the added "value" (in terms of
> > expectation of victory) of time running off the clock, is a very
> > interesting research question, but I would be willing to bet money
> on
> > short odds that it is something which should occur very late in the
> > game (that is the time threshold is not very sensitive to the
> realized
> > lead, especially if one is an underdog).
> >
> > *******************
> >
> > > > 2) Waiting longer to take shots. Comment: This is by
> definition a
> > > > sub-optimal strategy (why would one deny oneself maximum
> > > > time on the shot clock to come up with the best possible) and
> > > > though one may be dictating pace in this instance, one is also
> > > > dictating to one's own team a higher probability of losing.
> > > >
> > >
> > > If you have a lead against a tougher opponent, you want to run
> that
> > > clock as much as possible (shorten the game). It adds as much as
> > > about 10% to your odds of winning. There are also theories
> (tougher
> > > to prove) that defenses get less effective the longer in the
> clock
> > > they have to play. I've disproved this when there are <3 s on
> the
> > > clock, but it's difficult to say before that time.
> >
> > Where did this 10% statistic come from, and what does it depend
> on? I
> > am very curious. As far as defenses getting less effective the
> longer
> > in the clock they have to play, that may be the case, but it is
> besides
> > the point I make above. Restating: There is an expected optimal
> > offense which takes an expected amount of time, on average, off the
> > shot clock. Deciding independently to change the average time
> consumed
> > on the shot clock, will necessarily lower the expected points per
> > possession, unless this derivative is zero - in which case, and in
> only
> > which case, one can control pace.
> >
> > > > 3) Cut of iffy fast breaks and passes that start fast breaks.
> > > > Comment: Again, to deny one's team the potential for an easy
> > > > basket is by definition a sub-optimal strategy. Again, a team
> > > > dictates pace at a high price.
> > > >
> > > > 4) Playing a zone. Comment: Here again, pace isn't the issue,
> it
> > > > is choosing an optimal defense.
> > >
> > > You are right that there is some cause-effect issue between
> strategy
> > > and pace. If a zone is better against a certain team, it can
> slow
> > > the pace. But there are some inherent benefits of slow pace if
> you
> > > are an underdog EVEN IF you are slightly worse at working at a
> slow
> > > pace.
> > >
> > > See http://www.rawbw.com/~deano/articles/aa030597.htm for an
> > > interactive example.
> >
> > Yes, I remember years ago reading this article and being heartened,
> > perhaps for the first time, that that there was a fellow traveler
> in
> > this world. And, in fact, I was going to add some words on this in
> the
> > preceding note, but didn't because it was my impression that
> the "fewer
> > unfair coin flips if your team sucks is a good thing" effect was
> > ultimately more of a theoretic curiosity than significant possible
> > competitive advantage. A question: how confident are you that the
> 8%
> > gain shown in the Cav-Bulls example is robust in terms of
> behavioral
> > assumptions? In terms of what I wrote above, it implies that the
> > derivative referred to above (expected points per possession as a
> > function of elapsed time on shot clock) is rather flat. Because
> one
> > would never rationally turn down a fast break layup for the purpose
> of
> > slowing down the pace of the game. Now, it may have been true for
> the
> > Cavs if their baseline offense was to waste shot clock time anyway,
> but
> > I have my suspicions. Shot clock time is valuable!!!
> >
> >
> > > > Fast Team Pace Control Variables:
> > > >
> > > > I guess I won't list them, but they all describe strategy
> choices
> > > > that relate to controlling the probability of victory, not pace.
> > > >
> > > > So to repeat the bottom line, teams making conscious optimal
> > > > choices try to maximize the probability of victory not the pace
> of
> > > > the game. Teams choosing to influence the pace of the game
> > > > independently do so at their peril - unless of course I am
> > > > missing something.
> > >
> > > The Pistons of the late '80's were highly noted for beating the
> > > Lakers by slowing the pace. If they could play at their tempo,
> > > they'd win. Yes, some of that implied that they just could
> implement
> > > their strategies successfully, which _caused_ the pace to slow.
> That
> > > was part of the point of the discussion -- what could they do to
> slow
> > > that pace successfully?
> > >
> > > Actually, it would be interesting to look back at their series
> and
> > > see whether pace was important in deciding winner-loser...
> > >
> > > DeanO
> >
> > This is the kind of impressionistic observation that I am most
> leery
> > of. Were these games slower than expected? Of course, if the
> Lakers
> > were a fast break team and there was a slow game pace, chances are
> > there would have been fewer realized fast breaks, hence, all else
> > equal, one might expect that in these games the Pistons would have
> been
> > the victors.
> >
> > All this said, besides the end of game effect of intentionally
> slowing
> > the game to protect an "insurmountable" lead (but then again, in
> terms
> > of detecting this effect empirically, remember that the apparent
> pace
> > of the game in the final box score would not necessarily reflect
> this
> > eventuality for the appearance of extra foul generated possessions
> by
> > the team attempting to catch up) and the possibility of underdogs
> whose
> > half-court offensive productivities aren't dependent on time (those
> > with but one shooter, who can get his equivalent, open shot
> whenever?)
> > is there any other variable where a team, by choosing to influence
> > pace, is not sacrificing competitiveness?
• ... I m not sure that the two are analogous. Playing harder in basketball helps, notably on defense and rebounding but also hustle plays. There is no
Message 17 of 22 , Apr 17, 2003
• 0 Attachment
schtevie2003 wrote:
>
> --- In APBR_analysis@yahoogroups.com, "aaronkoo" <deano@r...> wrote:
> > --- In APBR_analysis@yahoogroups.com, "schtevie2003" <schtevie@h...>
> > wrote:
> > > That said, comparing game paces in the regular season and the
> > > post season does seem like a very interesting avenue of inquiry.
> > > Why? Because it allows possible insight into the optimal
> > > performances of teams during the regular season. (Under the
> > > assumption that teams are playing their best - or at least trying
> > to
> > > - in the playoffs.) For example, if a team were found, all else
> > > equal, to play at a different game pace in the post-season
> > > (presumably slower, reflecting a more intensive defensive effort)
> > > one could surmise the counterfactual cost of not pursuing this
> > > "better" strategy duing the post-season, in terms of potential
> > > games lost and better positioning for the playoffs.
> >
> > This is probably true and the reason why we're doing this second
> > round of polling now on who will win the title. We seemed to KNOW
> > that the Lakers would win the title the last couple years despite not
> > having the best record. Some of that is injury, but a lot of it is
> > people perceiving they save their A game for the playoffs. This goes
> > completely counter to the long-held belief in baseball that clutch
> > play does not exist, so putting together the story in a very
> > convincing manner is important.
>
> It seems to me that the linguistic analogy for "clutch play" in
> baseball is the "hot hand" in basketball, and I think that people are
> correct to disbelieve it. That said, there is a certain rationality
> for a team that knows they are likely the best team in basketball not
> to give its all in the regular season - especially if it doesn't feel
> that home court advantage matters much to it.

I'm not sure that the two are analogous. Playing harder in basketball
helps, notably on defense and rebounding but also "hustle plays." There is
no reason to believe that playing harder in baseball will have much of an
effect at all. There is so little margin for error in hitting that
half-assing it would show a huge effect in performance. Pitching, one
might gain a mile an hour but might also lose some control, any effect
would be negligible. Maybe it would help a tiny bit in fielding in the OF,
but the two sports are apples and oranges vis a vis the effect of effort.

--
Gary Collard
SABR-L Moderator
collardg@...
• ... ... and the ... inquiry. ... (Under the ... trying ... else ... defensive effort) ... pursuing this ... potential ... second ... to KNOW
Message 18 of 22 , Apr 17, 2003
• 0 Attachment
--- In APBR_analysis@yahoogroups.com, Gary Collard
<collardg@e...> wrote:
> schtevie2003 wrote:
> >
> > --- In APBR_analysis@yahoogroups.com, "aaronkoo"
<deano@r...> wrote:
> > > --- In APBR_analysis@yahoogroups.com, "schtevie2003"
<schtevie@h...>
> > > wrote:
> > > > That said, comparing game paces in the regular season
and the
> > > > post season does seem like a very interesting avenue of
inquiry.
> > > > Why? Because it allows possible insight into the optimal
> > > > performances of teams during the regular season.
(Under the
> > > > assumption that teams are playing their best - or at least
trying
> > > to
> > > > - in the playoffs.) For example, if a team were found, all
else
> > > > equal, to play at a different game pace in the post-season
> > > > (presumably slower, reflecting a more intensive
defensive effort)
> > > > one could surmise the counterfactual cost of not
pursuing this
> > > > "better" strategy duing the post-season, in terms of
potential
> > > > games lost and better positioning for the playoffs.
> > >
> > > This is probably true and the reason why we're doing this
second
> > > round of polling now on who will win the title. We seemed
to KNOW
> > > that the Lakers would win the title the last couple years
despite not
> > > having the best record. Some of that is injury, but a lot of it
is
> > > people perceiving they save their A game for the playoffs.
This goes
> > > completely counter to the long-held belief in baseball that
clutch
> > > play does not exist, so putting together the story in a very
> > > convincing manner is important.
> >
> > It seems to me that the linguistic analogy for "clutch play" in
> > baseball is the "hot hand" in basketball, and I think that
people are
> > correct to disbelieve it. That said, there is a certain rationality
> > for a team that knows they are likely the best team in
basketball not
> > to give its all in the regular season - especially if it doesn't
feel
> > that home court advantage matters much to it.
>
> I'm not sure that the two are analogous. Playing harder in
basketball
> helps, notably on defense and rebounding but also "hustle
plays." There is
> no reason to believe that playing harder in baseball will have
much of an
> effect at all. There is so little margin for error in hitting that
> half-assing it would show a huge effect in performance.
Pitching, one
> might gain a mile an hour but might also lose some control,
any effect
> would be negligible. Maybe it would help a tiny bit in fielding in
the OF,
> but the two sports are apples and oranges vis a vis the effect of
effort.
>
> --
> Gary Collard
> SABR-L Moderator
> collardg@e...

Pardon the imprecision, my understanding of the term "hot hand"
as used generally here and in the literature, and the sense in
which I was using it was "predictable above average
performance, unrelated to effort". In that sense, I drew the
analogy. It is the belief in such a hot hand that I have problems
with, certainly not with the idea that individuals and teams try
harder from one game to another - very often predictably.
• ... Sorry, I was conflating two things, namely upping intensity in the playoffs vs what you were talking about, which seems to be hot hand. But a hot hand
Message 19 of 22 , Apr 17, 2003
• 0 Attachment
schtevie2003 wrote:
>
> --- In APBR_analysis@yahoogroups.com, Gary Collard
> <collardg@e...> wrote:
> > schtevie2003 wrote:
> > >
> > > --- In APBR_analysis@yahoogroups.com, "aaronkoo"
> <deano@r...> wrote:
> > > > --- In APBR_analysis@yahoogroups.com, "schtevie2003"
> <schtevie@h...>
> > > > wrote:
> > > > > That said, comparing game paces in the regular season
> and the
> > > > > post season does seem like a very interesting avenue of
> inquiry.
> > > > > Why? Because it allows possible insight into the optimal
> > > > > performances of teams during the regular season.
> (Under the
> > > > > assumption that teams are playing their best - or at least
> trying
> > > > to
> > > > > - in the playoffs.) For example, if a team were found, all
> else
> > > > > equal, to play at a different game pace in the post-season
> > > > > (presumably slower, reflecting a more intensive
> defensive effort)
> > > > > one could surmise the counterfactual cost of not
> pursuing this
> > > > > "better" strategy duing the post-season, in terms of
> potential
> > > > > games lost and better positioning for the playoffs.
> > > >
> > > > This is probably true and the reason why we're doing this
> second
> > > > round of polling now on who will win the title. We seemed
> to KNOW
> > > > that the Lakers would win the title the last couple years
> despite not
> > > > having the best record. Some of that is injury, but a lot of it
> is
> > > > people perceiving they save their A game for the playoffs.
> This goes
> > > > completely counter to the long-held belief in baseball that
> clutch
> > > > play does not exist, so putting together the story in a very
> > > > convincing manner is important.
> > >
> > > It seems to me that the linguistic analogy for "clutch play" in
> > > baseball is the "hot hand" in basketball, and I think that
> people are
> > > correct to disbelieve it. That said, there is a certain rationality
> > > for a team that knows they are likely the best team in
> basketball not
> > > to give its all in the regular season - especially if it doesn't
> feel
> > > that home court advantage matters much to it.
> >
> > I'm not sure that the two are analogous. Playing harder in
> basketball
> > helps, notably on defense and rebounding but also "hustle
> plays." There is
> > no reason to believe that playing harder in baseball will have
> much of an
> > effect at all. There is so little margin for error in hitting that
> > half-assing it would show a huge effect in performance.
> Pitching, one
> > might gain a mile an hour but might also lose some control,
> any effect
> > would be negligible. Maybe it would help a tiny bit in fielding in
> the OF,
> > but the two sports are apples and oranges vis a vis the effect of
> effort.
>
> Pardon the imprecision, my understanding of the term "hot hand"
> as used generally here and in the literature, and the sense in
> which I was using it was "predictable above average
> performance, unrelated to effort". In that sense, I drew the
> analogy. It is the belief in such a hot hand that I have problems
> with, certainly not with the idea that individuals and teams try
> harder from one game to another - very often predictably.

Sorry, I was conflating two things, namely upping intensity in the playoffs
vs what you were talking about, which seems to be hot hand. But a "hot
hand" in basketball is still not the analogue for "clutch play" in
baseball, which is (however defined) play in the late portions of close
games or performance in runners on base and/or in scoring position
situations. Performance in close and late basketball situations would be
an analogue to the former, noting in basketball really to the latter (all
possessions are created equal).

There are a number of studies of "streakiness" in baseball, i.e. how well
does a hot or cold hitter hit in the near future. This would be the
analogue to "hot hands."

--
Gary Collard
SABR-L Moderator
collardg@...
• ... with ... real ... maximizing ... The real assumption is that teams aspire to win, which is different than maximizing offensive productivity. If slowing
Message 20 of 22 , Apr 17, 2003
• 0 Attachment
--- In APBR_analysis@yahoogroups.com, "schtevie2003" <schtevie@h...>
wrote:
> > You are basically assuming that offensive performance declines
with
> > time left on the shot clock. What I've found is that it declines
> > with <3 s on the clock but is not significantly impacted before
> > then. That implies that teams have pretty constant offensive
> > performance up to that point and that the assumptions in that web
> > page are pretty accurate up to the point of <3 s on the clock.
>
> You are correct that I am effectively assuming that offensive
> performance declines with time left on the shot clock. But the
real
> assumption is that teams aspire to optimal play - that is
maximizing
> their offensive productivity - given this assumption, the theory is
> dead solid that offensive performance is expected to decline as a
> function of elapsed shot clock time.

The real assumption is that teams aspire to win, which is different
than maximizing offensive productivity. If slowing the game can give
you an extra 8% and maximizing offensive productivity gives you an
extra 1%, which do you go with? I'm not sure how steep or flat you
think the curve is for offensive productivity as a function of time
left on the clock. I'm not even convinced that productivity is high
with 21-24s on the clock when turnover rates can be high.

>
> > So, yes, I do think 8% is within the realm of possibility. 8%
ain't
> > a lot to gain and can be offset if your team doesn't play well at
a
> > slow pace (they defy the more general study above, which is
> > definitely possible). And, yes, throwing away fast breaks
doesn't
> > make sense.
>
> I think that 8% improvement is a very large number in the context
of a
> possibly costless competitive advantage - an expected 4 additional
wins
> per season for an average team in the league if I am understanding
the
> statistic correctly?

Not quite. That extra 8% came from when they were an underdog and a
big underdog. In normal games, you aren't going to get 8%. Over the
course of the season, there are a few games where this matters.
Winning 4 seems a little high but I haven't done the calc.

>
> And I too remember some atrocious pre-shot clock college games.
But
> let's pose the question this way. Suppose two equal and average
NBA
> teams are playing, and one has a 10 point lead. Pick a time X
before
> the end of the game at which point you think it is optimal to alter
> one's offensive strategy so that the first priority is controlling
> pace.
>

Obviously it depends on assumptions about how a team's offensive
strategy affects its productivity. If I assume that it doesn't, I
get one answer. If I assume that it does, I get another. Though I'd
like to get the answer either way (and think I can -- I have my old
simulator sitting around somewhere), I don't know if we have a good
way to make that assumption.

DeanO
• ... ... declines ... web ... is ... give ... high ... It is hard to keep an eye on the forest for all the trees in these strings, so let me
Message 21 of 22 , Apr 17, 2003
• 0 Attachment
--- In APBR_analysis@yahoogroups.com, "aaronkoo" <deano@r...> wrote:
> --- In APBR_analysis@yahoogroups.com, "schtevie2003"
<schtevie@h...>
> wrote:
> > > You are basically assuming that offensive performance declines
> with
> > > time left on the shot clock. What I've found is that it
declines
> > > with <3 s on the clock but is not significantly impacted before
> > > then. That implies that teams have pretty constant offensive
> > > performance up to that point and that the assumptions in that
web
> > > page are pretty accurate up to the point of <3 s on the clock.
> >
> > You are correct that I am effectively assuming that offensive
> > performance declines with time left on the shot clock. But the
> real
> > assumption is that teams aspire to optimal play - that is
> maximizing
> > their offensive productivity - given this assumption, the theory
is
> > dead solid that offensive performance is expected to decline as a
> > function of elapsed shot clock time.
>
> The real assumption is that teams aspire to win, which is different
> than maximizing offensive productivity. If slowing the game can
give
> you an extra 8% and maximizing offensive productivity gives you an
> extra 1%, which do you go with? I'm not sure how steep or flat you
> think the curve is for offensive productivity as a function of time
> left on the clock. I'm not even convinced that productivity is
high
> with 21-24s on the clock when turnover rates can be high.

It is hard to keep an eye on the forest for all the trees in these
strings, so let me restate the larger "pace" argument:

Except for milking the clock when ahead at the end of the game and
possibly the "shoot slow if you suck and hope you are lucky", I see
no way, theoretically, for a team to try and control game pace
without costing itself points and the likelihood of victory.

That said, I agree with the above statement that the real assumption
is that teams aspire to win, but disagree with the statement that it
is necessarily different than maximizing offensive productivity -
except in the two realized cases above.

As to how steep or flat I think the curve is for offensive
productivity as a function of time left on the clock, I don't know
either, as I have only a general formula with hypothetical variables
and lack true data to plug in. That said, I will shuffle through
some old paper and offer up hypothetical slope values that the good
folks here can comment on.

**************************************

> > > So, yes, I do think 8% is within the realm of possibility. 8%
> ain't
> > > a lot to gain and can be offset if your team doesn't play well
at
> a
> > > slow pace (they defy the more general study above, which is
> > > definitely possible). And, yes, throwing away fast breaks
> doesn't
> > > make sense.
> >
> > I think that 8% improvement is a very large number in the context
> of a
> > possibly costless competitive advantage - an expected 4
additional
> wins
> > per season for an average team in the league if I am
understanding
> the
> > statistic correctly?
>
> Not quite. That extra 8% came from when they were an underdog and
a
> big underdog. In normal games, you aren't going to get 8%. Over
the
> course of the season, there are a few games where this matters.
> Winning 4 seems a little high but I haven't done the calc.

Can you give a graphical argument as to why the gain is greatest when
the expected loss is greatest. If I suppose a bell shaped curve
centered around this expected losing margin, then tweak the system
with a slow down strategy and this shifts the distribution such that
a greater amount of probability mass is above zero. If this is the
argument (and I am not saying it is) then suppose a more equal but
still superior opponent. Wouldn't I expect a greater gain in the
probability of victory if a thicker part of the bell curve (assume
same moments) was being pushed past zero?

**********************

> > And I too remember some atrocious pre-shot clock college games.
> But
> > let's pose the question this way. Suppose two equal and average
> NBA
> > teams are playing, and one has a 10 point lead. Pick a time X
> before
> > the end of the game at which point you think it is optimal to
alter
> > one's offensive strategy so that the first priority is
controlling
> > pace.
> >
>
> Obviously it depends on assumptions about how a team's offensive
> strategy affects its productivity. If I assume that it doesn't, I
> get one answer. If I assume that it does, I get another. Though
I'd
> like to get the answer either way (and think I can -- I have my old
> simulator sitting around somewhere), I don't know if we have a good
> way to make that assumption.
>
> DeanO

Yeah sure it depends on assumptions, but that is the game...Just
trying to establish a hypothetical tradeoff to anchor our
expectations.
• ... Forest is about the same. I think we may have different senses of the magnitude of cost . Cost of 1% vs value of 8% would need to be weighed. In some
Message 22 of 22 , Apr 17, 2003
• 0 Attachment
--- In APBR_analysis@yahoogroups.com, "schtevie2003" <schtevie@h...>
wrote:
> It is hard to keep an eye on the forest for all the trees in these
> strings, so let me restate the larger "pace" argument:
>
> Except for milking the clock when ahead at the end of the game and
> possibly the "shoot slow if you suck and hope you are lucky", I see
> no way, theoretically, for a team to try and control game pace
> without costing itself points and the likelihood of victory.
>

Forest is about the same. I think we may have different senses of
the magnitude of "cost". Cost of 1% vs value of 8% would need to be
weighed. In some cases, the cost will be greater than the benefit,
I'm sure.

> Can you give a graphical argument as to why the gain is greatest
when
> the expected loss is greatest. If I suppose a bell shaped curve

It's in the book. But here's a try. Basically the bell curve
describes the point difference between the 2 teams. Centered at +5
for the favorite, spreading the distribution (which is all that
taking a high risk/slow pace strategy does) causes the tail of the
distribution go more across that 0 line (when the underdog wins). If
it's centered at +1 (small favorite), a fair amount of the curve is
already across the 0 line, so spreading the curve doesn't make much
difference.

Is that clear?

>
> Yeah sure it depends on assumptions, but that is the game...Just
> trying to establish a hypothetical tradeoff to anchor our
> expectations.

I can add it to the list.
Your message has been successfully submitted and would be delivered to recipients shortly.