Chance in Diplomacy: Some Small Points of Discussion

G

Guest

Guest
Archived from groups: rec.games.diplomacy (More info?)

http://www.diplom.org/Zine/S1995M/Dreier/NoTheory.html

Now for the second reason, a less obvious one. Allan Calhamer,
Diplomacy's
creator, said that there is no luck in Diplomacy after the initial
random
assignment of powers. He was mistaken. Diplomacy does have lots of luck
in it.
Let's see why.

Jamie Dreier
Brown University

from Spring 1995 Movement

================================================================================

Hi,

Within this news group, rec.games.diplomacy, there does exist an
article on
chance written sometime since the beginning of 2002 which contains
many, many
responses. Given the length of that thread, I have only had a chance
to glance
at some of the responses within it. So, what I may say may be a
repetition of
what was said before.

Above, is a quote from an article at the given web site address written
by
Jamie Dreier in 1995.

In this hopefully short article to the news group, I will touch upon
the issue
of luck in a small way as concerns the playing of the game, Diplomacy.

As pointed out in the above article at www.diplom.org, sometimes the
attacker is
faced with a situation where the probability of the success of an
attack of a
particular supply center is dependent upon what the defender does;
and, since
you don't know what the defender will do (unless you have good
intelligence, and
we will assume for this article that you never have reliable
intelligence), your
best strategy is to flip a coin.

Small point number 1. Actually, the article says that you may flip a
coin or
flip a weighted coin. Question: if you want to take any supply center
of the
enemy, and you want to take it as soon as possible, should the coin be
fair or
should it be weighted? Answer (in my opinion, welcoming other points
of view
if I am mistaken): the most efficient way to capture a supply center
of
opportunity is to flip an evenly weighted coin to determine how the
actual
orders will be written that turn (i.e., to determine which one of the
supply
centers will attempt to be taken that turn). This is true regardless
of any
strategy or any changing strategy used by the defender (I have not had
a chance
prove this very last point, but I feel that it is probably true).

Small point number 2: I think that the following can be said about the
situation when you are flipping a coin to decide how to grab one supply
center
of opportunity when there are two to choose from and the defender is
putting up
a fight. The probability of not capturing it by the first turn's
attempt
is 1/2. The probability of not capturing it by the second turn's
attempt is
1/4. The probability of not capturing it the third turn's attempt is
1/8. And
so one. However, keep in mind that these probabilities are determined
before
your first turn's attempt. Every time you make an attempt, the
probability of
success is always fifty percent. I guess my subtle point here is that
even
though things are not totally determinant for this situation (you don't
know
for certain if you will capture one of the supply centers of
opportunity), the
only question, really, is when you will capture one of the supply
centers. The
fact that this information can be determined probabilistically can most
likely
be leveraged into a decision matrix. In short, chance is involved, but
it
places you within the realm of the science of probability and you can
potentially base decisions upon information derived from this science.
The same
would be true if you redefined the movement of the military units in
the game
and decided to use dice to determine which units held and which
retreated. The
beauty of diplomacy, and this is another topic, is that in many
situations, if
you cooporate with another major power, you don't have to worry so much
about
these probabilities! That is, I believe, not having yet played
Diplomacy, one
of the true beauties of the game! And it clearly separates the game
from a
game where battles are resolved by throwing dice. [For instance, in a
dice game,
two major powers most probably will overcome an isolated power; but,
in
Diplomacy, two major powers cooperating will with certainty overcome an
isolated power (of course, these are general comments to make my point
that
a dice game always contains chance, but cooperating major powers that
reliably
trust one another can drastically reduce the play of chance for those
powers
cooperating).

Thanks
 
G

Guest

Guest
Archived from groups: rec.games.diplomacy (More info?)

Hi,

A summary (in case my writing was hurried and unclear above):

Dipomacy does involve chance. Chance is minimized (and you are
rewarded
for being able to more efficiently perform actions and predict events)
as your nation increases the number of units that represent honest,
reliable allies.
Chance is maximized, chance acts as a form of punishment, for the
fewer,
honest reliable military units which act as your honest ally.

The above is a generalization. For an example of an exception, if your
nation has 17 military units
and you have one enemy with only 1 military unit, your power alone will
overcome
chance and you would, I would suspect, probably win.

Thanks
 
G

Guest

Guest
Archived from groups: rec.games.diplomacy (More info?)

I disagree... What you're talking about here is psychology though, not
luck. Luck suggests a total randomness of outcomes, which is never the case
in a real game of Diplomacy.

There are many many scenarios in a game where all you can do is "guess" what
the enemy will do and plan your moves accordingly. However, you can still
maximize your chances by (1) studying your oponent's past patterns of
behavior, (2) consider a matrix of possible moves by each side and the
resulting outcome in order to devise a set of orders with the maximum
probability of success.


"NewsGroupUser" <Google2007@mailinator.com> wrote in message
news:1105563629.538536.92110@z14g2000cwz.googlegroups.com...
> http://www.diplom.org/Zine/S1995M/Dreier/NoTheory.html
>
> Now for the second reason, a less obvious one. Allan Calhamer,
> Diplomacy's
> creator, said that there is no luck in Diplomacy after the initial
> random
> assignment of powers. He was mistaken. Diplomacy does have lots of luck
> in it.
> Let's see why.
>
> Jamie Dreier
> Brown University
>
> from Spring 1995 Movement
>
> ================================================================================
>
> Hi,
>
> Within this news group, rec.games.diplomacy, there does exist an
> article on
> chance written sometime since the beginning of 2002 which contains
> many, many
> responses. Given the length of that thread, I have only had a chance
> to glance
> at some of the responses within it. So, what I may say may be a
> repetition of
> what was said before.
>
> Above, is a quote from an article at the given web site address written
> by
> Jamie Dreier in 1995.
>
> In this hopefully short article to the news group, I will touch upon
> the issue
> of luck in a small way as concerns the playing of the game, Diplomacy.
>
> As pointed out in the above article at www.diplom.org, sometimes the
> attacker is
> faced with a situation where the probability of the success of an
> attack of a
> particular supply center is dependent upon what the defender does;
> and, since
> you don't know what the defender will do (unless you have good
> intelligence, and
> we will assume for this article that you never have reliable
> intelligence), your
> best strategy is to flip a coin.
>
> Small point number 1. Actually, the article says that you may flip a
> coin or
> flip a weighted coin. Question: if you want to take any supply center
> of the
> enemy, and you want to take it as soon as possible, should the coin be
> fair or
> should it be weighted? Answer (in my opinion, welcoming other points
> of view
> if I am mistaken): the most efficient way to capture a supply center
> of
> opportunity is to flip an evenly weighted coin to determine how the
> actual
> orders will be written that turn (i.e., to determine which one of the
> supply
> centers will attempt to be taken that turn). This is true regardless
> of any
> strategy or any changing strategy used by the defender (I have not had
> a chance
> prove this very last point, but I feel that it is probably true).
>
> Small point number 2: I think that the following can be said about the
> situation when you are flipping a coin to decide how to grab one supply
> center
> of opportunity when there are two to choose from and the defender is
> putting up
> a fight. The probability of not capturing it by the first turn's
> attempt
> is 1/2. The probability of not capturing it by the second turn's
> attempt is
> 1/4. The probability of not capturing it the third turn's attempt is
> 1/8. And
> so one. However, keep in mind that these probabilities are determined
> before
> your first turn's attempt. Every time you make an attempt, the
> probability of
> success is always fifty percent. I guess my subtle point here is that
> even
> though things are not totally determinant for this situation (you don't
> know
> for certain if you will capture one of the supply centers of
> opportunity), the
> only question, really, is when you will capture one of the supply
> centers. The
> fact that this information can be determined probabilistically can most
> likely
> be leveraged into a decision matrix. In short, chance is involved, but
> it
> places you within the realm of the science of probability and you can
> potentially base decisions upon information derived from this science.
> The same
> would be true if you redefined the movement of the military units in
> the game
> and decided to use dice to determine which units held and which
> retreated. The
> beauty of diplomacy, and this is another topic, is that in many
> situations, if
> you cooporate with another major power, you don't have to worry so much
> about
> these probabilities! That is, I believe, not having yet played
> Diplomacy, one
> of the true beauties of the game! And it clearly separates the game
> from a
> game where battles are resolved by throwing dice. [For instance, in a
> dice game,
> two major powers most probably will overcome an isolated power; but,
> in
> Diplomacy, two major powers cooperating will with certainty overcome an
> isolated power (of course, these are general comments to make my point
> that
> a dice game always contains chance, but cooperating major powers that
> reliably
> trust one another can drastically reduce the play of chance for those
> powers
> cooperating).
>
> Thanks
>
 
G

Guest

Guest
Archived from groups: rec.games.diplomacy (More info?)

In article <cs6fc0$b44$1@news.Stanford.EDU>,
adam silverman <agman1010@yahoo.com> wrote:

>There are many many scenarios in a game where all you can do is
>"guess" what the enemy will do and plan your moves accordingly.
>However, you can still maximize your chances by (1) studying
>your oponent's past patterns of behavior, (2) consider a
>matrix of possible moves by each side and the
>resulting outcome in order to devise a set of orders with the maximum
>probability of success.

If the opponent is concerned about (1), though, he may choose
to employ a randomizer. "I can't protect both Berlin and
Munich. It would be nice not to lose either one. I think my
opponent is better at predicting my moves than I am at
predicting his. Better flip a coin!"

Once he does this, you are reduced to guessing the outcome of
a coin flip, which is as close to "pure chance" as anything I
can think of.

(2) does not remove chance, it only maximizes your results in
the face of chance. That's all you can do, of course, but
I don't feel that eliminates chance.

Consider the class of positions described in some Dip Pouch
articles where the first player has a move X or a move Y. Move X
holds the line against defense A and wins against defense B.
Move Y wins against defense A and collapses into a stalemate
against defense B.

In a game with no time limit the first player is very likely
to win by playing X over and over and then suddenly
sneaking in Y. The second player can survive only by guessing
when Y will be played (to respond with B) and the first player
can make this pure chance by using a randomizer.

If I were the first player, using a randomizer, and I failed to
win this game because the second player played Y at the precisely
needed instant, I would consider that result chance, not skill.
If I weren't using a randomizer, I might have drawn due to
the opponent's skill in reading me--but I *would* use one, at
this point, because I have far more to lose by being predicted
than to gain by predicting my opponent.

Mary Kuhner mkkuhner@eskimo.com