Singles Tennis Viewed as a PCT Interaction

[From David Goldstein (2011.09.01.1248 EDT)]

Some additional thoughts.

Tennis from a PCT View

· Get to ball so that you can hit it on the fly or before it bounces twice.

· Hit a forehand or backhand over the net within the bounds of the other person’s court.

· Place the ball as far from where the other person is as possible.

· Hit the ball flat, slice or topspin.

· Anticipate where the other person is likely to hit the ball and start to move in that direction.

· Repeat all of the above until the point is won.

···

From: Bob Hintz

Sent: Wednesday, August 31, 2011 8:37 PM

To: CSGNET@LISTSERV.ILLINOIS.EDU

Subject: Re: Singles Tennis Viewed as a PCT Interaction

bob hintz - 2011.08.31

I believe that each hit is organized not only to get the ball over the net and in bounds, but also to make it as difficult as possible for the other player to return the ball over the net and in bounds. If one has no awareness of where the other player is on the court, it becomes more difficult to achieve this second part of the reference goal. Each player is simultaneously taking the other player’s observable behavior into account as each continuously organizes his own. I have no idea how this could be included in a spreadsheet.

bob

On , Richard Marken rsmarken@gmail.com wrote:

[From Rick Marken (2011.08.31.1615)

David Goldstein (2011.08.31.1614 EDT)

DG: The idea of making it sequential makes it more realistic.

Yes, of course. I guess I’ll give it a try. It will require a major

rewrite, I think. But maybe it’s worth it. Why don’t you try it, too?

Also, I was thinking that the biggest disturbance is where the other

person hits the ball, how far it is from you, whether you are fast enough to

reach the ball, what kind of spin the person puts on the ball (flat, topspin,

slice). Each person is a disturbance for the other person.

Yes, this is true in all interpersonal conflicts (and, of course, it’s

true in my little simulation). The other person’s output joins with

any other prevailing environmental disturbances (like the wind) to

disturb the perception controlled by the other person. The other

person’s output in a conflict is, by definition, a disturbance because

it is an effect on a controlled variable that is independent of the

actions of the controller. But whereas environmental disturbances are

random and unsystematic; the disturbance created by an opponent is

systematic – it is actively working against your efforts to work

against it (and prevent it from affecting the state of the variable

you are controlling).

This is not taking away from the fact that we have an interpersonal conflict

going with one person winning and the other person losing a point.

It’s not only not taking away from the fact that there is a conflict;

it is the essence of the conflict.

Best

Rick

David

-----Original Message----- From: Richard Marken

Sent: Wednesday, August 31, 2011 2:20 PM

To: CSGNET@LISTSERV.ILLINOIS.EDU

Subject: Re: Singles Tennis Viewed as a PCT Interaction

[From Rick Marken (2011.08.31.1120)]

2011/8/31 Boris Hartman boris.hartman@masicom.net>:

BH: I think that you are doing fine in simplifying the understanding of

tennis

game, although this form is not yet enough to really understand the

essence

of it.

It seems to me I’m capturing the essence but not the “frills” (the

higher level goals like strategy and reasons for playing at all).

BH: And there is one thing about your explanation that I didn’t

understand quite well. You know my American …:))

BH: Is everything what you wrote above happening in the same moment ? Is

conflict something that is happening in the same moment ?

Yes, everything is happening at the same time, in the sense that both

players have opposing references for the state of the ball and are

acting at the same time to bring the ball’s perceived position to the

intended state. What is incorrect about the simulation is that both

players affect the ball’s position simultaneously. In real tennis, of

course, one person hits the ball and then the other does. I could make

the simulation act this way; maybe I should since it may be confusing

people the way it is. But even when there is alternation in the shots,

both players are controlling for ball position at the same time, they

just can’t have an effect on the ball at the same time (as in the

present simulation).

I’m beginning to think that maybe a simulation of a more realistic

tennis dual may be in order. This would mean putting physics into the

simulation, which will be a pain for a soft-headed psychologist like

me. But it may be worth the effort.

Best

Rick

Richard S. Marken PhD

rsmarken@gmail.com

www.mindreadings.com

Richard S. Marken PhD

rsmarken@gmail.com

www.mindreadings.com

[From Rick Marken (2011.09.01.1510)]

bob hintz (2011.08.31)

I believe that each hit is organized not only to get the ball over the net
and in bounds, but also to make it as difficult as possible for the other
player to return the ball over the net and in bounds. If one has no
awareness of where the other player is on the court, it becomes more
difficult to achieve this second part of the reference goal. Each player is
simultaneously taking the other player's observable behavior into account as
each continuously organizes his own. I have no idea how this could be
included in a spreadsheet.

As I said, my simulation is aimed at demonstrating the essential
aspects of the conflictual interaction that exists in a game like
tennis. Among the simplifying assumptions I make (besides the
simultaneity of both players' effects on the ball) is that the players
are playing on a one dimensional court.

Each player's awareness of the position of the other player is
implicit in the definition of the controlled variable as hitting the
ball so that it is past the other player (in one space). For example,
the controlled variable for Federer is ATAN(bh/(Sp-qi)) where bh is
the height of the ball, Sp is Sampres' court position and qi is the
actual location of the ball. Both bh and Sp are constants -- no
gravity and no movement by either player -- which is another set of
simplifying assumptions. What Federer is trying to do is hit the ball
so that the an optical angle made by the ball is the angle that would
be made if the ball hit behind Sampres. That is, Federer has a
reference for the controlled variable-- optical angle of the ball --
which, if achieved, puts the ball part Sampras -- a winning shot. The
same (in the opposite direction) is true for Sampres.

It would be possible to make this model more like real tennis but it
would require a lot of work. I think my one dimensional simulation
illustrates the conflict that exists in tennis. I thought that was the
goal of this effort. I think it would be great if you would like to
build a more realistic model of a tennis game -- one that occurs in 3
space with each player alternating hits to the ball. That would
certainly be interesting but I don't really have the time to do it
myself right now. But I think it would be a great exercise for you to
give it a shot; I don't think it would be that hard, though it would
indeed be difficult to implement it in a spreadsheet. If I were doing
it I would probably use a real programming language.

Best

Rick

···

--
Richard S. Marken PhD
rsmarken@gmail.com
www.mindreadings.com

[From Rick Marken (2011.09.01.1530)]

David Goldstein (2011.09.01.1248 EDT)]

Some additional thoughts.

Tennis from a PCT View

��������� Get to ball so that you can hit it on the fly or before it bounces
twice.

��������� Hit a forehand or backhand over the net within the bounds of the
other person�s court.

��������� Place the ball as far from where the other person is as possible.

��������� Hit the ball flat, slice or topspin.

��������� Anticipate where the other person is likely to hit the ball and
start to move in that direction.

��������� Repeat all of the above until the point is won.

These are likely perceptions that an individual tennis player might be
controlling for. But you would have to test thee suggestion of course.
I do question the suggestion that the players "Anticipate where the
other person is likely to hit the ball and start to move in that
direction", at least as something that comes out of PCT. I think of
"anticipating" as something that is done in imagination; it's not
really part of real time control. And it probably doesn't help much. I
think what is happening with skilled players is not "anticipation"
but, rather, controlling for higher order perceptions, like programs
("if the opponent moves to X move to Y else move to Z") and principles
(try to stay behind the baseline until there is an opportunity to
charge in). I think games like tennis (or my game, racquetball) go too
fast for anything like "anticipation" to be involved. I think it's
all control of different types of present time perceptions
simultaneously.

But all of these suggestions pertain to what individual players might
be controlling for. I was just trying to show that, to the extent that
the players are controlling for variables at reference levels that
imply _different_ final positions of the _same_ ball, you've got a
conflict.

Best

Rick

···

--
Richard S. Marken PhD
rsmarken@gmail.com
www.mindreadings.com

[David Goldstein (2011.09.01.2009 EDT)]

Rick,
What is the purpose of creating the model of tennis that you did?
What are the benefits?
David
PS If you look at the top tennis players, you will sometimes see them take off in one
direction before the opponent hits the ball, and then be surprised when the opponent
does something different.

···

-----Original Message----- From: Richard Marken
Sent: Thursday, September 01, 2011 6:34 PM
To: CSGNET@LISTSERV.ILLINOIS.EDU
Subject: Re: Singles Tennis Viewed as a PCT Interaction

[From Rick Marken (2011.09.01.1530)]

David Goldstein (2011.09.01.1248 EDT)]

Some additional thoughts.

Tennis from a PCT View

� Get to ball so that you can hit it on the fly or before it bounces
twice.

� Hit a forehand or backhand over the net within the bounds of the
other person�s court.

� Place the ball as far from where the other person is as possible.

� Hit the ball flat, slice or topspin.

� Anticipate where the other person is likely to hit the ball and
start to move in that direction.

� Repeat all of the above until the point is won.

These are likely perceptions that an individual tennis player might be
controlling for. But you would have to test thee suggestion of course.
I do question the suggestion that the players "Anticipate where the
other person is likely to hit the ball and start to move in that
direction", at least as something that comes out of PCT. I think of
"anticipating" as something that is done in imagination; it's not
really part of real time control. And it probably doesn't help much. I
think what is happening with skilled players is not "anticipation"
but, rather, controlling for higher order perceptions, like programs
("if the opponent moves to X move to Y else move to Z") and principles
(try to stay behind the baseline until there is an opportunity to
charge in). I think games like tennis (or my game, racquetball) go too
fast for anything like "anticipation" to be involved. I think it's
all control of different types of present time perceptions
simultaneously.

But all of these suggestions pertain to what individual players might
be controlling for. I was just trying to show that, to the extent that
the players are controlling for variables at reference levels that
imply _different_ final positions of the _same_ ball, you've got a
conflict.

Best

Rick
--
Richard S. Marken PhD
rsmarken@gmail.com
www.mindreadings.com

[From Rick Marken (2011.09.01.1820)]

David Goldstein (2011.09.01.2009 EDT)--

Rick,
What is the purpose of creating the model of tennis that you did?

To show how an interaction between two control systems works in the
context of a very simplified game.

What are the benefits?

The benefit was supposed to be to allow a person to see what happens
when two separate control systems act to try to get the same variable
(perceived ball position) into two different reference states.

Actually, my main goal was to demonstrate the elements (control
systems) of what could become an actual model of a tennis game. I
wanted to show how the basic equations of control can be used to
develop a model of a realistic two person interaction, like a tennis
match. I see it as analogous to using a "toy" example to show how the
basic equations of, say, physics can be used to build a real thing,
like a bridge. I hoped to show what kinds of things you have to think
about in order to develop a working model of a tennis game.

PS If you look at the top tennis players, you will sometimes see them take
off in one direction before the opponent hits the ball, and then be surprised
when the opponent does something different.

Yes, I can even see that with the top racquetball players (like my
partner and me;-) I myself rarely try to anticipate where my partner
is going to hit the ball; but my partner apparently does sometimes try
to anticipate where I will hit it because I often see him going to
where he thought I was going to hit it (my point!). So, yes, I guess
players do try to anticipate where the ball is going to go and this
would probably have to be put into a model of a real tennis game, for
some players, anyway.

By the way, I'm not that interested in building a high fidelity model
of a tennis match because there is no data I know of to test it
against. I'm sure I could build a model of a tennis match, but I'm
more interested in building models to see if they fit data, not just
to see if they work (though building working models of complex
behavior, just to see if they can be made to work, is certainly a
worthwhile endeavor also).

Best

Rick

···

--
Richard S. Marken PhD
rsmarken@gmail.com
www.mindreadings.com