The CRAWFORD Rule, do we really need it?
Lawrence J. Gier
To me the Crawford rule makes no sense at all, whereas I can see real
reasons to do away with it entirely. In a 5 pt match the first person
to 4 gets a free shot at winning the match while the lesser side has
no chance with the next game.
This is really troubling because the Crawford rule makes THE ORDER in
which the points are accumulated during the match IMPORTANT.
The second really troubling aspect is that it distorts tremendously
the use of the doubling cube.
Finally, the last really troubling aspect is that, at least to me,
there is no good reason to have the rule in the first place.
Suppose you agree to play a 5 pt. match. The final score of the match
turns out to be 9-4 and you win the 5 pt. match. If you are playing
for money, then just collect on a per game basis. If you are playing
for rating points then calculate ratings as to the final score and not
the match length. This makes more sense anyway. If you're playing in a
tournment then base pairings on the match score rather than who wins a
match of fixed length.
The really big bonus to doing away with the Crawford rule is that then
the doubling cube would have meaning even in a 1 point match. Also,
note that if someone tried to use the cube to win a match based on
match considerations instead of game considerations then he or she
risks a greater defeat, and hence a greater loss if scores where used
rather then who wins the match.
I call upon the servers to do away with this rule and make backgammon
a more pure sport. In short, the cube can take care of itself.
Lawrence J. Gier
Kit Woolsey
When you are playing a match where the winner is the first player who
reaches a fixed number of points, the question is what should be done
about the doubling cube when one of the players has one point to go. If
there were no restrictions on doubling, the player who is behind would
obviously turn the cube at his first legal opportunity every game. This
would greatly lessen the value of being one point from winning the
match. On the other hand, if the cube were taken out of play, the player
who is behind would have to win a bunch of undoubled games in order to
recover. This would give being one point away from winning too much value.
The Crawford rule is an intelligent compromise. It gives some value for
getting to within one point of winning and still leaves the trailer a
chance to recover without having to win too many games.
At any rate, the Crawford rule is the accepted rule for match play. It
has been used at every tournament I have ever played in, which goes back
to the early 1970's. Since this is the case, it is clearly best to use
the same rule for FIBS.
Kit
Patti Beadles
Let me give you an example of the flip side of it.
THe score has just reached 4-3 in a 5-point match. Without the
Crawford rule, I may as well have not won that last point, because
it's exactly the same as it being 4-4.
-Patti
--
Patti Beadles | Knowing how to play well doesn't
pat...@netcom.com/pat...@gammon.com | win-- you must actually do it.
http://www.gammon.com/ |
or just yell, "Hey, Patti!" | No longer pa...@velo.com
Mel Rappaport
Lawrence J. Gier wrote:
> Suppose you agree to play a 5 pt. match. The final score of the match
> turns out to be 9-4 and you win the 5 pt. match. If you are playing
> for money, then just collect on a per game basis. If you are playing
> for rating points then calculate ratings as to the final score and not
> the match length. This makes more sense anyway. If you're playing in a
> tournment then base pairings on the match score rather than who wins a
> match of fixed length.
4/4 to 0/4
or 4/5 to 1/5
or 4/6 to 2/6
or 4/7 to 3/7
And perhaps we can reverse the winner's share of The 1960
Pirates Yankees match because the Yankees had more runs.
And the 144 winners of the U.S Open in Golf
can apportion themselves
{Average(participants) - (actual individual score)} / participants.
And I won't even attempt to imagine how very many winners can
apportion themselves in the U.S. Tennis Open.
And so forth :)
Albert Steg
In article <3330467...@news.kdsi.net>, ljg...@kdsi.net (Lawrence J.
Gier) wrote:
> The really big bonus to doing away with the Crawford rule is that then
> the doubling cube would have meaning even in a 1 point match. Also,
> note that if someone tried to use the cube to win a match based on
> match considerations instead of game considerations then he or she
> risks a greater defeat, and hence a greater loss if scores where used
> rather then who wins the match.
Your argument will have no currency with anyone who differentiates between
winning a *match* and winning by a point score. The fundamental between
"match" and "money" play is that the foremr is a simple "win or lose"
scenario while the latter rewards *margin* of victory.
Until you appreciate this distinction, the Crawford rule will make no
sense to you: once you do understand the distinction, the virtues of the
rule will become apparent.
Albert
Michael J Zehr
In article <3330467...@news.kdsi.net> ljg...@kdsi.net (Lawrence J. Gier) writes:
>To me the Crawford rule makes no sense at all, whereas I can see real
>reasons to do away with it entirely.
On the plus side, no more long debates over proper cube handling at
-2:-2. The doubling window would be from 50%-50%
*grin*
-michael j. zehr
Daniel Murphy
ljg...@kdsi.net (Lawrence J. Gier) writes:
>To me the Crawford rule makes no sense at all, whereas I can see real
>reasons to do away with it entirely. In a 5 pt match the first person
>to 4 gets a free shot at winning the match while the lesser side has
>no chance with the next game.
[see his post for the rest]
Welcome to backgammon, Lawrence. Perhaps when you're a little more
experienced you'll come to see the value of the Crawford rule in match
play. I might add that on this point some people do agree with you. Your
other comments on match play, however, aren't so sensible.
Yes, the Crawford rule rewards the player who first gets to within one
point of victory. Thus, that last point the leader earned has *value* for
the leader, as, I believe, it should have. Especially if, and this is
what you fail to note, the alternative to it having value for the leader
is it having *no value at all*, which is what would happen without the
Crawford rule.
Suppose we play a match to 5. The score is tied 3-3. You win one more
point to make the score 4-3. What is that last point you earned worth?
Without the Crawford rule, it has zero value. None at all. Why ? Because
in the next game, I double immediately, and the last game of the match is
worth 2 points for me, but only one for you. The score might as well be
3-3 instead of 4-3, because your lead means absolutely nothing.
About your comments about other, non-Crawford concerns, you should note
that the final score of a 5-point match is *never* 9-4, since, obviously,
the first player to reach 5 points wins the match -- regardless of how
whichever server you are playing on reports the final score. And a
one-point match is exactly that -- a match in which, unlike money play and
longer matches, the cube and gammon considerations are irrelevant. That
they are irrelevant is simply a fact, given a match length of one. Far
from being a "big bonus," your idea of using a cube in a one-point match
suggests that you don't enjoy the parameters of a one-point match, or
don't appreciate how they require you to adjust your checker play to
increase your *match*-winning chances.
Lawrence, I suspect you simply haven't played enough backgammon to realize
that above the intermediate level of play, how you handle the cube is far
more important to winning at backgammon than how you move the checkers.
It is not terribly difficult to learn to move the pieces well -- not
perfectly, but competitively. It is far more difficult to learn to
evaluate your equity in any given backgammon position, to learn to
anticipate how any given position is likely to evolve, and, based on your
analysis and understanding, make intelligent decisions about doubling and
taking.
As important and difficult as cube handling is in money play, it becomes
even more important and difficult in match play, because in match play you
cannot rely on simple formula for cube decisions without taking into
consideration your match equity (your chances of winning the match) at any
given score -- not just how many points you and your opponent have at the
moment, but how many each of you need to win the match, *and* your
relative chances of winning the match from the score that will obtain
after doubling or taking, and winning and losing -- or dropping. That's a
lot to consider *on top of* normal, money play equity considerations.
Your complaints about use of the cube in match play suggest that such
advanced considerations are not yet part of your game plan. Let me assure
you that for good players, they are extremely important. Your idea of
rewarding the winner based not on match length but on some other "final
score," no doubt skewed by some last-minute jacking up of the cube to some
level totally irrelevant to match length, throws all such considerations
out the window, and with it a great part of what is most difficult and
most intriguing about the game of backgammon.
--
_______________________________________________________
Daniel Murphy | San Francisco | rac...@cityraccoon.com
Monthly tourneys in San Mateo: See www.gammon.com/bgbb/ for details
and some excellently annotated matches. On-line: telnet fibs.com 4321.
Morten Wang
[nyc...@NOTworldnet.att.net (NYCGuy)]
>So GG folks, be careful when accepting resignations (normal resigns when G
>or BG's are in order) and modified invitations.
The easiest way to spot use of "modified invitations" is to notice that
the invitation-window becomes quite a lot larger. I messed things up
once, accepting a hypergammon invitation, and have since payed closer
attention to this. You only have to notice that there's something
different about the window, and then take your time looking at how the
options are checked.
Morten!
--
"God does not deduct from our alloted life span
the time spent playing backgammon."
-> Morty on FIBS, Mowa on GamesGrid.
Lawrence J. Gier
Then again maybe we can introduce the doubling cube into, baseball,
golf, etc,etc,etc,etc. I was talking about backgammon.
Lawrence J. Gier
Graham Trevor Price
In <33337f41...@netnews2.worldnet.att.net>
nyc...@NOTworldnet.att.net (NYCGuy) writes:
I've noticed a phenomenon on GG whereby certain players prefer to
>play without Crawford and can negate it if one is not wary.
>When you invite someone to a match on GG, your default setting are
sent.
I've played a few matches allowing no crawford and using beavers etc.
I find it actually is in my favor because usually the person
who plays with these rules often has little knowledge of proper cube
play and strategy anyway and so if I fall behind I get the benefit
of free re-doubling.(down 0-3, when does leader double?)
A funny request I often get is play a 3 point match with
beavers.
This doesnt make much sense because double- redouble gets you
to the same spot so the beaver is actually redundant. Just
goes to demonstrate that many of these players are not at all
familiar with cube strategies,leverage or any strategy.
They use it because they think a big cube makes the game more fun
I guess (that's what one person told me when we played anyway).
Graham.
Mark Damish
Lawrence J. Gier (ljg...@kdsi.net) wrote:
: Then again maybe we can introduce the doubling cube into, baseball,
: golf, etc,etc,etc,etc. I was talking about backgammon.
I always go to a baseball game with a doubling cube. Given equal
strength, is it a take when you are down by 2 runs in the bottom
of the 7th when you are the home team?
--
...Mark Damish mda...@bbn.com 617.873.3735 6/498
Stephen Turner
Mark Damish wrote:
>
> I always go to a baseball game with a doubling cube. Given equal
> strength, is it a take when you are down by 2 runs in the bottom
> of the 7th when you are the home team?
>
Hmmm, an interesting thought. England could have doubled France out in the
rugby three weeks ago. Then again, if we'd doubled too early, France might
have taken, and won twice as much when they eventually turned it round.
--
Stephen Turner sr...@cam.ac.uk http://www.statslab.cam.ac.uk/~sret1/
Statistical Laboratory, 16 Mill Lane, Cambridge, CB2 1SB, England
"This store will remain open during modernisation. We apologise
for any inconvenience this may cause" Topshop, Cambridge
Graham Trevor Price
In <3331650e...@news.kdsi.net> ljg...@kdsi.net (Lawrence J. Gier)
writes:
>
>Then again maybe we can introduce the doubling cube into, baseball,
>golf, etc,etc,etc,etc. I was talking about backgammon.
win much like in backgammon where if you are -2:-1 and your score is
the -2 you might try for gammon to win the match.
In baseball they often use pinch hitters to try to generate offense
at the cost of defence if losing in the late innings,
in golf you can always make a risky shot to try to hit the green
under regulation over a water hazard or trees,
in hockey you pull the goalie.....
You give up a lot if it doesn't work but you probably give yourself
a little more chance for your risk.
And for someone betting on the outcomes of games how many times
has the other person lost maybe the first two bets and then wants
to make the 3rd bet "double or nothing". Now that's catching up
using the cube (He loses 2 bets but if he wins this doubled bet
he is even). In backgammon look up "steaming".
Graham.
Lawrence J. Gier
This is precisely why the final score should have meaning!!!!!!!!!!!
Let me make one thing clear. I have won more matches using it
than I have lost (something like 6-1). Everytime I win one I feel
like I am stealing it, and it still doesn't make any sense!!
On 20 Mar 1997 02:05:27 -0800, rac...@best.com (Daniel Murphy) wrote:
>ljg...@kdsi.net (Lawrence J. Gier) writes:
>
Lawrence J. Gier
Graham Trevor Price
In <333a9f4...@news.kdsi.net> ljg...@kdsi.net (Lawrence J. Gier)
writes:
>
>This is precisely why the final score should have meaning!!!!!!!!!!!
>
>Let me make one thing clear. I have won more matches using it
>than I have lost (something like 6-1). Everytime I win one I feel
>like I am stealing it, and it still doesn't make any sense!!
Lawrence, I would recommend that you re-read Daniel's suggestion which
should give you cause to think rather than reiterate your myopic
view of the backgammon world. BTW most strong players could not
count the number of Crawford games they have won or lost so we
are looking at a vast experience deficiency on your part.
>>Lawrence, I suspect you simply haven't played enough backgammon to
realize
>>that above the intermediate level of play, how you handle the cube is
far
>>more important to winning at backgammon than how you move the
checkers.
___________________________________________________
>>Daniel Murphy | San Francisco | rac...@cityraccoon.com
>Lawrence J. Gier
>
Then to test your theory get out there and play tough competition
and see how successful you are under the current conditions. If
you demolish everyone then your ideas may be given some credence.
A good idea might be to play say 20 matches against jf_level_five
to 7 points. At the least you will see pretty good cube handling
and find that if you take (or give) a cube that is a little too risky
you are out of the match in a rush. Best of all you will see where
you stand in relation to a strong player. A more serious approach
would be to actually study the jf- mloner matches, lots of things to
question and learn there believe me, but it requires some time
commitment.
Good Luck
Graham.
Lawrence J. Gier
It is the top of the third the visiting team has just scored two runs
and leads 2-0 and it appears their pitcher has his stuff today, this
the first game of the world series. Flash........ The team gets out of
the dug out and rolls a six foot by six foot doubling cube over to the
home team's dug out. The announcer says over the loud speaker system
"TAKES" and then the visiting team goes on to win the game by 2-0.
That is a doubled game, gammoned, end of the series, away team wins
4-0.
You could hear some fans say, "Hey, I had tickets to tomorrow's game!"
I was trying to make a serious point but Graham Trevor Price wants to
divert this discussion to other games or sports that do not use a
doubling cube. The doubling cube and fixed length matches mix like
water and oil. The Crawford rule is a very weak attempt at confronting
this problem.
On 21 Mar 1997 17:41:33 GMT, gpr...@netcom.ca(Graham Trevor Price)
wrote:
>In <3331650e...@news.kdsi.net> ljg...@kdsi.net (Lawrence J. Gier)
>writes:
>>
Lawrence J. Gier
Lawrence J. Gier
To accept the cube one needs 33% chance or better to win, but
how does this number change in an n game match (n is an integer match
length)? Obviously, early on in the match, one is more ready to double
if he has the superior postion because the first one to (n-1) gets a
free shot at winning the match via the Crawford rule.
I have seen all kinds on numerical analysis on this newsgroup, let us
see some on this absurd distortion caused by the Crawford rule. How,
come the good players are not defending this rule???? The 3-1 ratio to
accept and double change according to the match score in a complex
way, who needs it??
Also, for those who only care about who wins a match you can simply
consider the sign of the result. For instance in a 5 pt. match where
the final score was 9-3 then the winner has a +6 result while the
loser has a -6 result. The one with the plus score wins while the one
with the minus score is the loser. For some who think the score is
important in determining rating points, etc.etc. The the ENTIRE result
can be used, ie the magnatude of the victory or loss. The 6 part !
Do you see how rediculous the Crawford rule is now? The essence of
everything that has been said is this. The doubling cube and fixed
length matches do not go together well and the Crawford rule was used
to force them together. The real solution is that fixed length matches
must go to preserve the cube. You can still use the length of the
match to signify when to stop playing, ie first person to reach this
point total or greater wins the match, but ratings and money etc. must
be based on the final score. The reward is the straight forward use of
the cube that will add enjoyment to the game FOR ALL.
David Montgomery
In article <333c013...@news.kdsi.net> ljg...@kdsi.net (Lawrence J. Gier) writes:
>Do you see how rediculous the Crawford rule is now? The essence of
>everything that has been said is this. The doubling cube and fixed
>length matches do not go together well and the Crawford rule was used
>to force them together. The real solution is that fixed length matches
>must go to preserve the cube. You can still use the length of the
>match to signify when to stop playing, ie first person to reach this
>point total or greater wins the match, but ratings and money etc. must
>be based on the final score. The reward is the straight forward use of
>the cube that will add enjoyment to the game FOR ALL.
>Lawrence J. Gier
Okay Larry, I will grant you that:
The Crawford Rule IS arbitrary.
The Crawford Rule DOES distort (change) doubling -- that is, the
value of each point in the match, and so the take points, gammon
prices, doubling points, too good points, and so forth.
The Crawford Rule IS NOT necessary.
There ARE other (perhaps more elegant) rules that are possible
for playing matches to a set number of points.
HOWEVER,
THE CRAWFORD RULE IS THE RULE. It is arbitrary, but so are the
following rules:
- the fact that you have to play the small die before the large die
when you can't play both. This is arbitrary.
- the convention for the rolling the dice to start the game -- in
some countries the player who rolls higher can or must reroll, so
that opening doubles can occur. Another rule is that the player
who won (or lost) the last game goes first, and so can roll doubles.
All of this is arbitrary.
- the ability to pick and pass. In some regions, picking and passing
is not allowed. I lost in the finals of the first tournament I played
in because my opponent insisted I couldn't pick and pass, and the
tournament director didn't know enough to not believe him. Anyway,
whether or not picking and passing is allowed is arbitrary.
- the definition of gammons and backgammons. Why not give 4 points
if the opponent is still on the roof? Why not award points more
continuously based on the number of checkers left and/or how far
they have to go? I have seen rules for this kind of thing, and
known people who played this way. In fact, my understanding is
that 20 years ago backgammons were commonly only scored as gammons
in Europe. Again, all of these are arbitrary decisions.
In fact, all the rules are arbitrary. They're just rules to a game,
which people made up.
So how do we decide on what rules to use? There are a lot of factors.
One of these is what might be called the inherent quality of the rule.
The idea is that some rules are better than others because they
encourage fair play, balanced competition, are elegant, easy to understand,
consistent with other rules in the game, and so forth. It is with
respect to this factor that you criticize the Crawford Rule, and I
will agree, there is a lot to criticize.
But the inherent quality of a rule is NOT the most important factor.
Tradition and historical consistency are very important. Simple
popularity. The decisions of governing bodies. It is very difficult
to overcome all these factors simply because a rule has some deficiencies.
We play games with other people, so the first requirement of
game playing is that we agree on the rules. Deleting a rule,
introducing a new rule, or changing an existing rule usually requires
that the change be very much for the better for the change to gather
sufficient momentum to catch on.
I strongly believe that the Crawford rule is here to stay, and that
you are wasting your time by trying to persuade people to drop it.
Here is why:
1) It's not that bad a rule. It is easy to state and understand,
and it has certain desirable properties, as have been pointed out
by other posters. I am not certain that you have understood these
properties, but they are real.
2) The main defect of the rule is the distortions to doubling strategy
that the rule entails, BUT, distortions of some kind are UNAVOIDABLE
for match play, as far as I know. At least I have never seen a
proposal that avoids this. If you think that you have such a proposal,
here are some of the criteria that you must meet:
Contest Description:
A match to some pre-specified number of points, such that the final
number of points scored for both sides has a simple and clear relationship
to this number.
The only relevant result from the match is who is victorious -- the
margin of victory is irrelevant. This is a crucial point for match
play that I'm not sure you have fully understood. For tournaments
with a large number of contestants, it is important that the result of
each match be of the form: X won, O lost. Degree of victory can't
be utilized in a traditional tournament bracket. (Of course you could
simultaneously try to change both the Crawford rule and the standard
tournament procedure, but I think you would have to concede that your
chances here are then even smaller.)
Some Criteria to Avoid Distortion:
- In gammonless positions, the trailer's point of last take, for all
scores and cube sizes, is when the trailer has 25% winning chances owning
the cube.
- Gammons, for all scores and cube sizes, and for both players, have
a gammon price of 0.50. That is, you can factor gammons into the takepoint
by subtracting half the trailer's gammons and adding half the leader's
gammons.
- The double point, the point at which a player breaks even on doubling
when the game will be decided immediately, is 50%, for all scores and
cube sizes.
There are other criteria, but I think that if you just try to meet these
you will see that you can't.
So yes, the Crawford rule distorts things, but this really isn't a very
big deal, because match play by its very nature distorts things.
3) The Crawford rule isn't unpopular. You don't like it, but I have
never heard from anyone else who was particularly bothered by it. There
may be some, but they are in the minority.
4) There is no *clearly* superior alternative. Its not enough that an
alternative rule set be slightly better -- to override tradition, to
rewrite all the backgammon software, to throw out all the material written
on tournament doubling up til now, to get everybody to change the way
they play, you need to have a *clearly* superior alternative.
5) Most important of all, for the backgammon players who have been
playing tournament backgammon the most and for the longest amount of
time, and for the backgammon players who are the best at match play,
there is a TREMENDOUS disincentive to change the rules. The reason:
players who have played a long time to some degree, and players who
play well to a very high degree, have LEARNED the distortions caused by
the Crawford rule, and learned how utilize them to their advantage.
Playing tournament backgammon at the highest level can include a great
deal of memorization and learning of doubling theory, all predicated on
the existence of the Crawford rule. For this reason primarily, I would
be very much opposed to changing the rules.
----
An alternative rule is a proposal I read a long time ago in either the
Chicago Point or the Flint BackgammoNews, which suggested doing away with
the Crawford rule and simply requiring that the leader win by 2, as is
done in volleyball, and as used to be done in tennis. This seems to me
like a great rule. Simple, easy to understand, consistent with approaches
taken in other games/sports. It might well be better than the Crawford
rule.
But its not *clearly* better. And adopting it would mean that a
tremendous amount of work and knowledge by a lot of the most active
backgammon players would become worthless. For this reason, I, and I
think by far most other active and/or good backgammon players would
oppose changing the rule.
And since the backgammon community itself is opposed, your crusade to
point out the deficiencies of the rule are unlikely to produce the
result you desire.
----
Re-reading your post, it seems that a primary concern to you is that
the margin of victory should matter. As I understand it, you want
the margin of victory to matter in distributing money even at a
tournament.
You could do this. You might use a swiss system. The first round
players are randomly paired up, and they play until one player or
other reaches some predetermined number of points. The + or - score
of each player is noted. For subsequent rounds, you could match
up players according to their cumulative + or - total. At the end
of the series, the tournament director would have to do a lot of
bookkeeping, because you don't know how much each point is worth
until the end of the tournament, and every player is going to get
money back except the player who lost the most.
For example, let's say we had 8 people, each of which paid a $100
entry fee, and after some number of rounds they had the following
totals:
A B C D E F G H
+34 +12 +5 0 -5 -10 -15 -21
We have $800 to give out, so the prizes would be:
A B C D E F G H
261.90 157.14 123.81 100.0 76.19 52.38 28.57 0.00
This approach comes down to the tournament being the SAME as money
play -- its just that the tournament director determines the pairings,
and the players don't know the actual stakes, they just know their
maximum loss. Here it turns out these players have been
playing for $4.76/point.
Based on this, there is little point in modifying tournament rules
in this fashion -- it doesn't produce a new game, it just gives us
a game we already have, heads-up money play. If people want to just
play money backgammon, they already can.
-----
Larry, if your main concerns are that you want the margin of victory
to matter, and you want to avoid the distortions to cube action
in match play, there is a very simple solution: play money games.
Don't play tournament matches. There are a lot of players who only
play for money -- they don't play matches because they don't like them.
Based on your posts, it seems you may be in this category.
Tournament backgammon is a different game. Many, many people love it,
just as it is. The changes you're proposing wouldn't improve tournament
backgammon, they would turn it into money backgammon, another fine, fine
game, but one that already exists.
David Montgomery
monty on FIBS
mo...@cs.umd.edu
bob koca
ljg...@kdsi.net (Lawrence J. Gier) wrote:
>To accept the cube one needs 33% chance or better to win, but
>how does this number change in an n game match (n is an integer match
>length)? Obviously, early on in the match, one is more ready to double
>if he has the superior postion because the first one to (n-1) gets a
>free shot at winning the match via the Crawford rule.
25% not 33%.
>I have seen all kinds on numerical analysis on this newsgroup, let us
>see some on this absurd distortion caused by the Crawford rule. How,
>come the good players are not defending this rule???? The 3-1 ratio to
>accept and double change according to the match score in a complex
>way, who needs it??
3-1 ratio for accepting money cubes (assuming no gammons and taking
cube power into account) is correct. However there is no simple
ratio rule for when to double. It depends on how volatile the position
is. I suspect that doubling desicions even in money games is not
as straightforward as you think.
>Also, for those who only care about who wins a match you can simply
>consider the sign of the result. For instance in a 5 pt. match where
>the final score was 9-3 then the winner has a +6 result while the
>loser has a -6 result. The one with the plus score wins while the one
>with the minus score is the loser. For some who think the score is
>important in determining rating points, etc.etc. The the ENTIRE result
>can be used, ie the magnatude of the victory or loss. The 6 part !
>Do you see how rediculous the Crawford rule is now? The essence of
>everything that has been said is this. The doubling cube and fixed
>length matches do not go together well and the Crawford rule was used
>to force them together. The real solution is that fixed length matches
>must go to preserve the cube. You can still use the length of the
>match to signify when to stop playing, ie first person to reach this
>point total or greater wins the match, but ratings and money etc. must
>be based on the final score. The reward is the straight forward use of
>the cube that will add enjoyment to the game FOR ALL.
>Lawrence J. Gier
The complexities and distortions caused by match play give an added
depth to the game. This added challenge and variety adds enjoyment
to the game for me. Thus you must amend your "FOR ALL".
If you wish, why not stick to playing money games? This avoids the
compexities you mention. However do you have a satisfactory way of
running a tournament with out playing matches?
,Robert Koca
ko...@bobrae.bd.psu.edu
David Montgomery
In article <5hhkdr$d...@twix.cs.umd.edu> mo...@cs.umd.edu (David Montgomery) writes:
I wrote, as an example of an arbitrary rule:
> - the fact that you have to play the small die before the large die
> when you can't play both. This is arbitrary.
The correct rule, as Robert Johnson (rjohnson) pointed out to me, is:
- you must play the larger die if you can play either but not both.
Thanks, RJ.
Graham Trevor Price
In <333c013...@news.kdsi.net> ljg...@kdsi.net (Lawrence J. Gier)
writes:
>
>To accept the cube one needs 33% chance or better to win,
that follows?
Graham.
Kevin Whyte
I had a bad reaction to match play when I first started learning.
It is rather inelegant as it mangles cube play. One can (as
people have) argue that it adds another level of skill, but I
think that misses the point. One can add skill to the game in
many ways no one would ever accept (for example, make the players
play chess in between rolls and add the scores). It seems to
me that the point is that we want to have tournaments. If you try
to count total scores in a tournament then all games will be
doubled to 64 (or higher if allowed) since anyone playing for one
point while everyone else is playing for 64 has no chance. So
how to do it? One solution is to play without the cube, but that
takes away one of the most interesting part of the game away.
Does anyone have a better suggestion for tournament play? For
matches, it does seem to me it's better to play a fixed number of
games and split the prize money according to the final score.
If we do play matches, do we need a rule to change cube
behavior? I think it could be done either way. If we do have
one, I can't think of anything better than the Crawford rule,
even though it is ugly. Were I the one making the decision,
I'd probably not use it, but I don't feel strongly.
Just my $.02.
Kevin Whyte
kwh...@math.uchicago.edu
Julian
In article <333c013...@news.kdsi.net>, "Lawrence J. Gier"
<ljg...@kdsi.net> writes
>To accept the cube one needs 33% chance or better to win,
Care to enlighten us how you arrive at this figure?
> How come the good players are not defending this rule????
But they are...
> The 3-1 ratio to
>accept and double change according to the match score in a complex
>way, who needs it??
Let me get this right. Crawford adds complexity to cube decisions in
match play, therefore it is undesirable? If you find a simpler game more
fulfilling, maybe try Ludo.
>Also, for those who only care about who wins a match you can simply
>consider the sign of the result. For instance in a 5 pt. match where
>the final score was 9-3 then the winner has a +6 result while the
>loser has a -6 result. The one with the plus score wins while the one
>with the minus score is the loser. For some who think the score is
>important in determining rating points, etc.etc. The the ENTIRE result
>can be used, ie the magnatude of the victory or loss. The 6 part !
Rubbish. Would you propose, say, that in chess the number of pieces
remaining at the end of the game be counted to provide a 'magnitude of
victory'? In fact, can you name any game or sport played at a serious
level where the 'magnitude of victory' carries the same weight as
winning or losing? Your style of play is simply a sequence of 'money
games' with a pre-determined stopping point. It's not a 'match'.
>Do you see how rediculous the Crawford rule is now?
No, I just hear you repeating yourself without listening to anyone else.
---------------------------------------------------------------------------
Julian Hayward 'Booles' on FIBS jul...@ratbag.demon.co.uk
+44-1344-640656 http://www.ratbag.demon.co.uk/
---------------------------------------------------------------------------
"Armin Schwartz had an early success [in the RAC Rally], driving as
he was in one of the few four-wheel cars..."
- The Times
---------------------------------------------------------------------------
Lawrence J. Gier
I just received an e-mail message from ,I guess, to be one of the
better players in the world or so he stated, something about being
champion of Turkey. He was very angry at these postings saying, I was
new to backgammon, I shouldn't be shouting off my big mouth, that he
doubted I would be around long, etc. I told him to fuck off.
He said everything but put up a defense of the Crawford rule!!!
Could it be that the Crawford rule is a gambler's best friend ??
Perhaps there are people out there that instead of getting a job
use the Crawford rule to steal money from marks. Maybe that's it,
becuase this rule has no merit.
Donald Kahn
On Fri, 28 Mar 1997 18:07:25 GMT, ljg...@kdsi.net (Lawrence J. Gier)
wrote:
>To accept the cube one needs 33% chance or better to win, but
>how does this number change in an n game match (n is an integer match
>length)? Obviously, early on in the match, one is more ready to double
>if he has the superior postion because the first one to (n-1) gets a
>free shot at winning the match via the Crawford rule.
>
>I have seen all kinds on numerical analysis on this newsgroup, let us
>see some on this absurd distortion caused by the Crawford rule. How,
>come the good players are not defending this rule???? The 3-1 ratio to
>accept and double change according to the match score in a complex
>way, who needs it??
The good players won't waste their time and energy arguing this
matter.
Johnny Crawford was one of the great games players of all time: expert
at bridge, poker and backgammon and maybe more. Anyone would be well
advised to take his word for it, that his rule is a good solution to
this problem: the player who gets to 1 away is deeply disadvantaged
because the cube is dead for him and therefore does not fully benefit
from the advantage he has earned.
Nothing L. Gier has written comes even close to convincing me that I
would prefer to compete under his suggested rules.
Respectfully,
Donald Kahn
bob koca
mo...@cs.umd.edu (David Montgomery) wrote:
...
many excellent points concerning rules deleted
...
It isn't the same as money play because the results of your games
directly influence how much you win per point.
For example suppose you and your opponent are playing the very last
"match" of the tournament. Going into this match it happens that
every player has a score of 0. Suppose you win by +25 points.
Then scores would be +25, 0, 0, ...,0, -25. Your system would award
4$ per point. You would get $200 dollars (for a net gain of $100).
the even players would get their entry fee back, and the loser would
get nothing back (net loss of $100). Suppose now that you win +1
points instead. The scores are instead +1, 0, 0, ...,0, -1 but now a
point is worth 100$ so the payouts are the same. Since payout depends
only on winner you and your opponent would actually be playing a
conventional tournament match (though probably without Crawford rule).
There are a lot of other undesirable behaviors that this system
could encuourage. One example is suppose you are +8, and each of 4
opponents are at -2. You are playing the final match of the
tournament. You would do better with a -1 score in this match than a
+1 score. In the -1 case you would make 7points* 50$ = 350$ net gain.
In the +1 case you would get 9 points * 33 1/3 $ = 300$ net gain.
These two examples obviously show that the strategy in such a
tounament is not the same as straight money games.
,Bob Koca
bobk on FIBS
Greycat Sharpclaw
There is an allegation that ljg...@kdsi.net (Lawrence J. Gier) wrote:
>To accept the cube one needs 33% chance or better to win,
25%, not 33% (in the simple case w/o gammon or redoubling concerns).
This is basic. And the simpliest algebra will prove it.
>I have seen all kinds on numerical analysis on this newsgroup, let us
>see some on this absurd distortion caused by the Crawford rule.
1) Match play distorts the probabilities without the Crawford rules.
2) The Crawford rule gives a significant benefit for causing match
play do be decided by skill.
3) And how is this addition of a skill factor to the game "absurd"?
Please answer by arguements that are based on better than personal
bias.
>How,
>come the good players are not defending this rule????
Well, I know several professional astronomers, and don't see many of
them spending all their time argueing against the flat earth theory.
To many people, an arguement easily gets just too trivial to be worth
bothering with.
Besides, if this were a good arguement, I'd ask you this: what good
players are advocating the elimination of the crawford rule?
Certainly there's no great rush of support.
>The 3-1 ratio to
>accept and double change according to the match score in a complex
>way, who needs it??
The many (apparently majority from my observation) that *likes* seeing
the ratio change.
That is much the _point_ of match play, after all. If you prefer pure
point play to match play, fine, play that. But please don't insist
that match play be distorted into something else, because you don't
like it. If someone prefers soccer football to baseball, he plays
soccer football, rather than insists that a "no hands" rules be put
into baseball.
>Also, for those who only care about who wins a match you can simply
>consider the sign of the result. For instance in a 5 pt. match where
>the final score was 9-3 then the winner has a +6 result while the
>loser has a -6 result.
Why? Match play is a win-lose the match competition... the final score
is *supposed* to not matter. That's what *makes* it match play.
>The one with the plus score wins while the one
>with the minus score is the loser. For some who think the score is
>important in determining rating points, etc.etc. The the ENTIRE result
>can be used, ie the magnatude of the victory or loss. The 6 part !
Totally wrong approach. Completely violating the principle of match
play (broken record, but hey). The rating system is designed to
measure match point score, _period_. So it should. Suggestions to
improve the rating system should be based on how well it does _that_.
>Do you see how rediculous the Crawford rule is now?
Why? You have not made one signle valid point!
>The essence of
>everything that has been said is this. The doubling cube and fixed
>length matches do not go together well
Wrong!!!!!!! This point, frankly, exposes your ignorance of the
nature of match play.
>and the Crawford rule was used
>to force them together.
No... it was to cover a corner case where skill was nullified by the
certain exact counts.
>The real solution is that fixed length matches
>must go to preserve the cube.
Why? *Most* serious players *like* match play.
You can do point play if you prefer... no problem. Well, FIBS is
match play for ratings only because there has to be *1* method for
ratings, or *more than 1 rating*. Suggest the latter if you like,
that at leqast will make some sense.
But most tournaments, and FIBS, and numerous other "scored" backgammon
gatherings are match play. Accept it, or go play your backgammon
elsewhere. Either option makes sense.
>be based on the final score. The reward is the straight forward use of
>the cube that will add enjoyment to the game FOR ALL.
NO! Absoulutely incorrect!
The crawford rule became popular because *most* *players* *like* *it*.
Most players would find your proposal _reduces_ enjoyment rather than
adds to it.
The use of an absolute statement is dangerous, and stressing the
absolute more so.
You have made a statement that is *proven* in this thread to be
frequently wrong, and shown by experience (including the general
acceptance of the crawford rule) to be usually wrong. Yet you claim -
and stress the claim - that it's *always* right!
This hardly encourages us to accept your logic and sensibility.
If you want to *not* play matches, fine. Fibs allows for individual
games, it also allows "unlimited" matches for the style you are
favoring. And there are other places to play BG, besides FIBS. I
have no problem with you playing _your_ favorite game style. But
don't say matches should be perverted into a form that is not match
play at all!
Greycat
Gre...@tribeca.ios.com
Does anyone have any spare tunafish??
Lawrence J. Gier
I am afraid I have stirred some feelings. I got an e-mail
message from one who told me he was one of the world's
best backgammon players. He said I was new, that I shouldn't
be voicing these concerns and that I probably wouldn't be around
long, in general a real emotional, angry reply.
He forgot to defend the Crawford rule. I have been trying to find
someone to defend the logic behind this rule. One person, wanted
to talk about other sports. One person wanted to talk about the mechanics
and how it makes a bad situation WORSE. I agree.
Well I think I figured it out. Everyone knows the rule is only good for
fleecing marks and therefore they may wants it hushed up. People are
making money off the Crawford rule, am I right?
ATTENTION MARKS----- IF SOMEONE WANTS TO PLAY FOR MONEY
USING THE CRAWFORD RULE TELL THEM TO GET A JOB!!!!!!!
My latest victory using the Crawford rule occurred today. I was playing a 5
pt.
match, losing 0-4, in the Crawford rule game. The free shot my opponent had
to win the match. I had already thanked him for the games as it appeared
that I would lose
this one to, when he left a blot. I won this game by the skin of my teeth.
The next game, I
summoned all my mental resources, I got my extra computer out and did all
the
calculations. I called MIT for there advice. Then I doubled him as fast as
I could grab
my mouse. I won that one. I doubled him in the next game as well and won
that one.
Since there is no down side to my match doubling, I essentially cheated him
out of the
match. Pure and simple. Easy too. My games were worth two points, his one
point. If
there had been a stake to the match he would have been cheated out of that
too.
Lawrence J. Gier
I am afraid I have stirred some feelings. I got an e-mail
message from one who told me he was one of the world's
best backgammon players. He said I was new, that I shouldn't
be voicing these concerns and that I probably wouldn't be around
long, in general a real emotional, angry reply.
He forgot to defend the Crawford rule with any kind of logical
argument.
I have been trying to find someone to defend the logic behind this
rule. One person, wanted to talk about other sports. One person wanted
to talk about the mechanics and how it makes a bad situation WORSE. I
agree.
Well I think I figured it out. Everyone knows the rule is only good
for fleecing marks and therefore they may want it hushed up. People
are making money off the Crawford rule, am I right?
ATTENTION MARKS----- IF SOMEONE WANTS TO PLAY FOR MONEY
USING THE CRAWFORD RULE TELL THEM TO GET A JOB!!!!!!!
My latest victory using the Crawford rule occurred today. I was
playing a 5 pt. match, losing 0-4, in the Crawford rule game. The free
shot my opponent had to win the match. I had already thanked him for
the games as it appeared that I would lose this one to, when he left a
blot. I won this game by the skin of my teeth. The next game, I
summoned all my mental resources, I got my extra computer
out and did all the calculations. I called MIT for their advice. Then
I doubled him as fast as I could grab my mouse. I won that one. I
doubled him in the next game as well and won that one. Since there is
no down side to my match doubling, I essentially cheated him
out of the match. Pure and simple. Easy too. My games were worth two
points, his one point. If there had been a stake to the match he
would have been cheated out of that too.
Lawrence J. Gier
Lawrence J. Gier
Lawrence J. Gier
I am afraid I have stirred some feelings. I got an e-mail
message from one who told me he was one of the world's
best backgammon players. He said I was new, that I shouldn't
be voicing these concerns and that I probably wouldn't be around
long, in general a real emotional, angry reply.
He forgot to defend the Crawford rule. I have been trying to find
someone to defend the logic behind this rule. One person, wanted
to talk about other sports. One person wanted to talk about the mechanics
and how it makes a bad situation WORSE. I agree.
Well I think I figured it out. Everyone knows the rule is only good for
fleecing marks and therefore they may want it hushed up. People are
making money off the Crawford rule, am I right?
ATTENTION MARKS----- IF SOMEONE WANTS TO PLAY FOR MONEY
USING THE CRAWFORD RULE TELL THEM TO GET A JOB!!!!!!!
My latest victory using the Crawford rule occurred today. I was playing a 5
pt. match, losing 0-4, in the Crawford rule game. The free shot my opponent
had
to win the match. I had already thanked him for the games as it appeared
that I would lose this one to, when he left a blot. I won this game by the
skin of my teeth.
The next game, I summoned all my mental resources, I got my extra computer
out and did all the calculations. I called MIT for there advice. Then I
doubled him as fast as
I could grab my mouse. I won that one. I doubled him in the next game as
well and won
that one. Since there is no down side to my match doubling, I essentially
cheated him
out of the match. Pure and simple. Easy too. My games were worth two
points, his one
point. If there had been a stake to the match he would have been cheated
out of that
too.
Lawrence J. Gier
ljg...@kdsi.net
http://www.kdsi.net/~ljgier/
Hugh B. McNeil
Yes, I AGREE!!! the Crawford Rule is a SCAM!!!
I have played in MANY tournaments, there have been MANY great
players at them, AND ALL OF THESE TOURNAMENTS feature THE
CRAWFORD RULE!!!
And the BEST PLAYERS are all CHEATING YOU BY USING IT!!!
AND I CAN PROVE IT!!!
First, they USE THEIR SKILL to gain a SUBSTANTIAL LEAD over
WEAKER PLAYERS.
Second, they manouver the score until they they are one point
away from VICTORY!!!
AND THEN, WHEN they can USE THE CRAWFORD RULE to their
advantage, they WIN!!!
You can test THIS EMPIRICALLY, by looking at TOURNAMENT
RESULTS!!! THE BEST PLAYERS WIN (!!!) the most TOURNAMENTS
and MATCHES!!!
YOU ARE A FISH IF YOU GO TO ANY (!!!) TOURNAMENT THAT USES
THE CRAWFORD RULE!!!
I don't know if the Las Vegas Open uses the Crawford Rule
because I've asked the President (Howard whats-his-face) and
he doesn't know because he hasn't read his rules yet.
Really.
Greycat Sharpclaw
There is an allegation that "Lawrence J. Gier" <ljg...@kdsi.net>
wrote:
>He forgot to defend the Crawford rule. I have been trying to find
>someone to defend the logic behind this rule.
May I suggest you review this thread (which you started)... it has
numerous explainations, in various details, of the Crawford rule and
the reasons for it.
>Well I think I figured it out. Everyone knows the rule is only good for
>fleecing marks and therefore they may want it hushed up.
This is crap. The rule greatly improves match play. Whether or not
money is involved. Every serious player knows if favors the better
player, yet they still _want_ it.
Must you be reduced to this puerile level? Is your ego so shallow and
arrogance so great that, having been outargued at every turn, you drop
to this level?
>People are
>making money off the Crawford rule, am I right?
The better players make money off of betting on their skill. If you
want to make skill not a factor, play roulette.
>ATTENTION MARKS----- IF SOMEONE WANTS TO PLAY FOR MONEY
>USING THE CRAWFORD RULE TELL THEM TO GET A JOB!!!!!!!
If you play a better player for money, expect to lose some. If you
aren't willing to, _don't bet on backgammon at all_.
And BTW, if you _are_ going to make such foolish statements, please
don't yell.
>My latest victory using the Crawford rule occurred today.
...
> I essentially
>cheated him
>out of the match. Pure and simple.
Nope, not at all... the match was won by the proper use of match play.
Match play is not point play, ***nor is it meant to be***.
If you don't like match play, don't play it. But let those who will
play it do so.
John Clements
In article <333dddc...@news.kdsi.net>,
>playing a 5 pt. match, losing 0-4, in the Crawford rule game. The free
>shot my opponent had to win the match. I had already thanked him for
>the games as it appeared that I would lose this one to, when he left a
>blot. I won this game by the skin of my teeth. The next game, I
>summoned all my mental resources, I got my extra computer
>I doubled him as fast as I could grab my mouse. I won that one. I
>doubled him in the next game as well and won that one. Since there is
>no down side to my match doubling, I essentially cheated him
>out of the match. Pure and simple. Easy too. My games were worth two
>points, his one point. If there had been a stake to the match he
>would have been cheated out of that too.
>
So what you're saying is that since your opponent was one-away, there
was no reason not to double immediately, which you did. You then
say that this is essentially cheating. Note that this sort of doubling
is exactly what the Crawford rule _prevents_. In other words, if you
feel that you should not have been able to spin the cube freely in
the late stages of the game, what you're really saying is that you
_like_ the Crawford rule.
Very confusing.
john clements
Carl Tait
In article <01bc3c9e$58d22e40$2c7067ce@ljgier>,
Lawrence J. Gier <ljg...@kdsi.net> wrote:
>
>My latest victory using the Crawford rule occurred today. I was playing a 5
>Since there is no down side to my match doubling, I essentially cheated him
>out of the match. Pure and simple. Easy too. My games were worth two
This makes no sense as an argument *against* the Crawford rule. Instead,
you seem to be saying that the cube should be tossed out *entirely*
(not just for one game) when one player reaches match point so that the
underdog's games will not be "worth two points." Do you have the mistaken
impression that this is the normal match procedure without Crawford?
--
Carl Tait IBM T. J. Watson Research Center
ta...@watson.ibm.com Yorktown Heights, NY 10598
Lawrence J. Gier
Davy at least puts forth some defense of the Crawford rule. He
deserves thanks for that.
In the book by Crawford and Jacoby, they note that the doubling cube
was introduced in the 1920's. They also say that the New York Tennis
and Racquet club, I believe that is correct, set about to standardize
the rules in the 1930's. That is not a lot of tradition when
considering only the modern game, post doubling cube.
NOW, is the time to get this game right. The internet will introduce
this game to millions of new players, and I would think that one who
loves backgammon would want them to see as logical a game as possible
for then one would get more enjoyment from the game.
willingOn 28 Mar 1997 18:32:11 -0500, mo...@cs.umd.edu (David
Montgomery) wrote:
>In article <333c013...@news.kdsi.net> ljg...@kdsi.net (Lawrence J. Gier) writes:
>>Do you see how rediculous the Crawford rule is now? The essence of
>>everything that has been said is this. The doubling cube and fixed
>>length matches do not go together well and the Crawford rule was used
>>to force them together. The real solution is that fixed length matches
>>must go to preserve the cube. You can still use the length of the
>>match to signify when to stop playing, ie first person to reach this
>>point total or greater wins the match, but ratings and money etc. must
>>be based on the final score. The reward is the straight forward use of
>>the cube that will add enjoyment to the game FOR ALL.
>>Lawrence J. Gier
>
>Okay Larry, I will grant you that:
>
>The Crawford Rule IS arbitrary.
>The Crawford Rule DOES distort (change) doubling -- that is, the
> value of each point in the match, and so the take points, gammon
> prices, doubling points, too good points, and so forth.
>The Crawford Rule IS NOT necessary.
>There ARE other (perhaps more elegant) rules that are possible
> for playing matches to a set number of points.
>
>HOWEVER,
>
Because it is the rule is not a defense, it must stand or fall on its
merit.
>THE CRAWFORD RULE IS THE RULE. It is arbitrary, but so are the
>following rules:
>
>- the fact that you have to play the small die before the large die
> when you can't play both. This is arbitrary.
>- the convention for the rolling the dice to start the game -- in
> some countries the player who rolls higher can or must reroll, so
> that opening doubles can occur. Another rule is that the player
> who won (or lost) the last game goes first, and so can roll doubles.
> All of this is arbitrary.
>- the ability to pick and pass. In some regions, picking and passing
> is not allowed. I lost in the finals of the first tournament I played
> in because my opponent insisted I couldn't pick and pass, and the
> tournament director didn't know enough to not believe him. Anyway,
> whether or not picking and passing is allowed is arbitrary.
>- the definition of gammons and backgammons. Why not give 4 points
> if the opponent is still on the roof? Why not award points more
> continuously based on the number of checkers left and/or how far
> they have to go? I have seen rules for this kind of thing, and
> known people who played this way. In fact, my understanding is
> that 20 years ago backgammons were commonly only scored as gammons
> in Europe. Again, all of these are arbitrary decisions.
>
>In fact, all the rules are arbitrary. They're just rules to a game,
>which people made up.
>
It is easy to decide which rules to use. We use the rules that are
logical in execution. For instance, the Crawford rule is not logical
because it distorts play of the cube. The play of the cube is highly
logical, everybody loves it, when applied to the actual position of
the checkers on the board.
>So how do we decide on what rules to use? There are a lot of factors.
>One of these is what might be called the inherent quality of the rule.
>The idea is that some rules are better than others because they
>encourage fair play, balanced competition, are elegant, easy to understand,
>consistent with other rules in the game, and so forth. It is with
>respect to this factor that you criticize the Crawford Rule, and I
>will agree, there is a lot to criticize.
>
Davy, we disagree completely as to the following.
>But the inherent quality of a rule is NOT the most important factor.
>Tradition and historical consistency are very important. Simple
>popularity. The decisions of governing bodies. It is very difficult
>to overcome all these factors simply because a rule has some deficiencies.
>We play games with other people, so the first requirement of
>game playing is that we agree on the rules. Deleting a rule,
>introducing a new rule, or changing an existing rule usually requires
>that the change be very much for the better for the change to gather
>sufficient momentum to catch on.
>
>I strongly believe that the Crawford rule is here to stay, and that
>you are wasting your time by trying to persuade people to drop it.
>Here is why:
>
>1) It's not that bad a rule. It is easy to state and understand,
>and it has certain desirable properties, as have been pointed out
>by other posters. I am not certain that you have understood these
>properties, but they are real.
>
They are avoidable, and easily so, see my original postings.
>2) The main defect of the rule is the distortions to doubling strategy
>that the rule entails, BUT, distortions of some kind are UNAVOIDABLE
>for match play, as far as I know. At least I have never seen a
>proposal that avoids this. If you think that you have such a proposal,
>here are some of the criteria that you must meet:
>
>Contest Description:
> A match to some pre-specified number of points, such that the final
> number of points scored for both sides has a simple and clear relationship
> to this number.
>
To others, the margin of victory is relevant.
> The only relevant result from the match is who is victorious -- the
> margin of victory is irrelevant. This is a crucial point for match
> play that I'm not sure you have fully understood. For tournaments
> with a large number of contestants, it is important that the result of
> each match be of the form: X won, O lost. Degree of victory can't
> be utilized in a traditional tournament bracket. (Of course you could
> simultaneously try to change both the Crawford rule and the standard
> tournament procedure, but I think you would have to concede that your
> chances here are then even smaller.)
>
>Some Criteria to Avoid Distortion:
> - In gammonless positions, the trailer's point of last take, for all
> scores and cube sizes, is when the trailer has 25% winning chances owning
> the cube.
> - Gammons, for all scores and cube sizes, and for both players, have
> a gammon price of 0.50. That is, you can factor gammons into the takepoint
> by subtracting half the trailer's gammons and adding half the leader's
> gammons.
> - The double point, the point at which a player breaks even on doubling
> when the game will be decided immediately, is 50%, for all scores and
> cube sizes.
>
>There are other criteria, but I think that if you just try to meet these
>you will see that you can't.
>
It is when you could have a system in which no distortion of the very
popular cube idea occurs at all. Plowing old ground.
>So yes, the Crawford rule distorts things, but this really isn't a very
>big deal, because match play by its very nature distorts things.
>
>3) The Crawford rule isn't unpopular. You don't like it, but I have
>never heard from anyone else who was particularly bothered by it. There
>may be some, but they are in the minority.
I hear arguments like this, I reach for my wallet to be sure I still
have it.
>
Of course there is, see my original postings.
>4) There is no *clearly* superior alternative. Its not enough that an
>alternative rule set be slightly better -- to override tradition, to
>rewrite all the backgammon software, to throw out all the material written
>on tournament doubling up til now, to get everybody to change the way
>they play, you need to have a *clearly* superior alternative.
>
>5) Most important of all, for the backgammon players who have been
>playing tournament backgammon the most and for the longest amount of
>time, and for the backgammon players who are the best at match play,
>there is a TREMENDOUS disincentive to change the rules.
IT'S CALLED MONEY.
>The reason:
>players who have played a long time to some degree, and players who
>play well to a very high degree, have LEARNED the distortions caused by
>the Crawford rule, and learned how utilize them to their advantage.
>Playing tournament backgammon at the highest level can include a great
>deal of memorization and learning of doubling theory, all predicated on
>the existence of the Crawford rule. For this reason primarily, I would
>be very much opposed to changing the rules.
>
>----
>
>An alternative rule is a proposal I read a long time ago in either the
>Chicago Point or the Flint BackgammoNews, which suggested doing away with
>the Crawford rule and simply requiring that the leader win by 2, as is
>done in volleyball, and as used to be done in tennis. This seems to me
>like a great rule. Simple, easy to understand, consistent with approaches
>taken in other games/sports. It might well be better than the Crawford
>rule.
This too lacks merit as doubling would have to be suspended after the
Crawford rule game. The solution is easy, see my original postings.
Even so IT WOULD BE CLEARLY better.
>
>But its not *clearly* better. And adopting it would mean that a
>tremendous amount of work and knowledge by a lot of the most active
>backgammon players would become worthless. For this reason, I, and I
>think by far most other active and/or good backgammon players would
>oppose changing the rule.
BECAUSE OF MONEY?
>
>And since the backgammon community itself is opposed, your crusade to
>point out the deficiencies of the rule are unlikely to produce the
>result you desire.
Well I can try.
>
>----
>
>Re-reading your post, it seems that a primary concern to you is that
>the margin of victory should matter. As I understand it, you want
>the margin of victory to matter in distributing money even at a
>tournament.
My main care is rating points on FIBS and the other servers. Playing
for money is not a big feature of the game to me, thats why I dont
care for the Crawford rule.
There is another simple solution, see my original postings.
>Don't play tournament matches. There are a lot of players who only
>play for money -- they don't play matches because they don't like them.
>Based on your posts, it seems you may be in this category.
Visa versa I think. Some people that play for money LOVE match play!
Start out with a small stake, lose a few games, double the stake, turn
on the skill play past the Crawford game, double at will and make a
killing against some poor mark.
>
>Tournament backgammon is a different game. Many, many people love it,
>just as it is. The changes you're proposing wouldn't improve tournament
>backgammon, they would turn it into money backgammon, another fine, fine
>game, but one that already exists.
>
>David Montgomery
>monty on FIBS
>mo...@cs.umd.edu
>
Lawrence J. Gier
Daniel Murphy
ljg...@kdsi.net (Lawrence J. Gier) writes:
>Davy at least puts forth some defense of the Crawford rule. He
>deserves thanks for that.
Besides David Montgomery, whose article you quote (but do not respond to)
in its entirety, several other posters did much more than what you
uncharitably cast as "at least put forth some defense" of the Crawford
rule. Your gratitude, if that's what it is, really doesn't matter much,
so long as you fail to give any indication of having understood any of
the arguments given.
--
_______________________________________________________
Daniel Murphy | San Francisco | rac...@cityraccoon.com
Chuck Bower
In article <raccoon.860226868@shellx>, Daniel Murphy <rac...@best.com> wrote:
>(name deleted), here you simply reveal yourself to be a frigging know-nothing
>idiot.
(snip)
Daniel, Daniel, Daniel! (I'd say "Raccoon, Raccoon, Raccoon" but that's
a tongue twister.) Take it easy on the blood pressure!! Your teammates need
you in their next international match. You seem to be letting your emotions
get the best of you. Don't stoop to such name-calling. I know you're too
smart to have to resort to wrestling in the playground dirt. And with respect
to your language--I like to use this newsgroup to learn new words (like
"proffer" from Steve Mellen) but my dictionary doesn't have this "frigging"
word in it. Considering it is modifying "idiot", it must not be too endearing.
I haven't heard anyone lately being called "lovely idiot" or "brilliant idiot".
The word "idiot" usually stands alone, with the possible exceptions of the
adjectives "bloody" and "bloomin'". Of course some of our distant cousins
from other hemispheres like to attach these to just about any noun they find...
(snip)
>You know next to nothing about backgammon history,...
Thanks for the leadin to my next topic: The "Holland Rule". This beating
that poor Johnny Crawford has been taking lately just isn't fair. Why not
pass the whip around and let Tim Holland share some of the flatulations, err,
I mean "flagellations"--must be some kind of Freddian slop...
Tim Holland won the World BG Championship three consecutive years ('67-'69)
according to Robertie's "Lee Genud vs. Joe Dwek" book. Also interesting is
that Carol Crawford won it the next year! I'm pretty sure she was Johnny
Crawford's wife. (BG trivia question: who is the only other woman to win
the World Championship of Bg? Hint--I already gave it to you!) Back then,
though, I don't think tournaments were as widely advertised, and this was
before the BG boom of the 70's. Still, Holland was considered a strong
player. For example, in the Magriel interview from 1975 excerpted below,
Paul says Holland is one of the world's top players. Tim also authored
three books: "Beginning BG", "Better BG", and "BG for People Who
Hate to Lose"--a candidate for catchiest BG title. The last of
these is probably worth a read, though my recollection is that it can't
compete with the best books available today.
In Magriel's classic "Backgammon" (NOT in the running for catchy title...)
in the glossary, he defines "Holland Rule" as follows:
In a match play, after one player has reached match point, and after
the Crawford Rule game has been played, for the next two games
neither(!) player may double until two full rolls on each side have
been completed.
I have never played in a tournament where this was used. I think it is
definitely obsolete. I suspect it was never in common use at the top
levels. Anybody know? Somehow I doubt the Crawford bashers would like
this, either.
Speaking of Paul Magriel, I have a couple historical tidbits to report.
In the obscure (rare?) magazine "Backgammon Illustrated" vol. 1, #1, ONLY
ONE?!, published in either late 1975 or early 1976 (I think this was done by
Canadians from Montreal--is this last phrase politically correct?), there
is an interview with a young (29 year old) 'Backgammon Computer' named Paul
Magriel. The interviewer (Aubrey Zelman, also the Executive Editor) asked:
ZELMAN: Tell me, do you like the Crawford rule in tournament play?
MAGRIEL: Do I like it? Yes, it's a standard rule and it affects how you
play matches very much.
ZELMAN: What other rules would you like to introduce or delete from
tournament play?
MAGRIEL: I'd like to have some tournaments at a championship level where
there's less luck involved than these 15- and 17-point matches. (!!!)
...One possible way is instead of having a match you have what is
sometimes called a "freeze out". In a 15-point freeze-out instead
of the first person to reach fifteen points, you just score as if you
would normal money play and I have to get 15 points ahead of you.
(snip) This would take the luck out of it. (snip) The present
tournaments are very well structured and they give everybody some
chance, the better players have a better chance, but everybody has
some chance. But at some point I'd like to see a tournament where
there'll really be a test of skill just for the better players.
Looks like there's ammo here for both sides of the arguement... This is
a long and interesting interview. A fun read if you get the chance.
Easier to find, but just as interesting is the June 4, 1979 issue
of "Sports Illustrated". On page 64 is a column entitled "Soccer" by
Clive Gammon! (Why do I find this interesting? Because I didn't think
any Americans were interested in this sport back in 1979.) If you
continue to page 68 there is an article by Roger Dionne titled "A
Gamesman's War Against Luck, Disorder, and Surface Chaos". (This last
term sounds like a physics research project.) It's a several page
article about Magriel, with many color pictures of the now (then) 32-year
old "X-22, Human Computer". Part of that article includes the following:
Not long before (this article), Magriel and Roger Low, a 21-year-
old backgammon whiz from New York City, had represented the U.S. in a
three-day match against the best of Europe at the Mount Parnis Casino
outside Athens. Their opponents were a pair of tough, experienced
gamblers whom most devotees of the game consdiered the finest players
on the Continent, if not the world--Joe Dwek of London and Kumar
Motakhasses, a Londoner of Iranian birth. It was the opinion of
European experts that Dwek's brilliance with the doubling cube--a
crucial factor in modern backgammon--and Motakhasses' mastery of
movement of the checkers would undo the young American ex-mathematics
professor called "the Human Computer", and the unknown kid he had
brought along as a consultant. But after falling behind at the end
of the second day, the Americans spent the night in Magriel's room
at the Grand Britannia Hotel reappraising their strategy. The result
was a rally and a 63-61 victory.
I seem to recall that Magriel-Low won around $10,000 in this match. I'm
sure it is chronicled elsewhere (in Deyong's "Playboy Book of BG" possibly).
Anyway, I bet Dwek-Motakhasses wished they were playing a freeze-out!
Well, enough history. Can we summarize the points of the thread with
some rationality. (Has there been any? Just kidding.) Larry proposes
throwing out the fixed length match concept (as we know it) and replacing
it with the following.
1) Predetermined match "length" only decides when match ends--if neither
player has reached this level at the start of a game, continue. Otherwise
match is over.
2) "Score" of match is the difference between winning and losing scores.
ALL games are scored as in money play (cube X factor, where "factor" =
1, 2, or 3 depending on simple, gammon, backgammon outcome).
3) Ratings are determined (in an as yet unspecified way) based on final
score, and not on "predetermined match length" as is done in the current
systems.
4) There would be no need for a Crawford Rule.
Does this system make sense? On a server like FIBS, where the
"reward/punishment" of matches is ratings, it might work. There is
both a "prize" for the winner (increased rating) and a "raspberry"
for the loser (decreased rating). As now occurs on FIBS, there would
be loopholes which could allow unscrupulous players to circumvent the
system and build unrealistic ratings.
As far as the arguments: " 'cause that's the way it's done",
" 'cause it would throw out the current match strategy (and equity
tables)" " 'cause I don't like it and I'd take my ball and go home",
well, you probably know how I feel about emotional, irrational
arguments. I think David Montgomery's question "is it better or
worse than the current system?" makes sense. Unfortunately, this
isn't easy to answer. On the surface, it's just DIFFERENT, not
obviously better or worse (IHMO) than the way things are currently
done, except...
In tournaments (on the network or otherwise) I see a flaw.
Although I tried to say this in an earlier post, let me make the
following example:
A tournament is conducted in a "modified Swiss" format, where players
are randomly paired in the first round (as opponents) and then paired
in subsequent rounds by score, where highest score plays next highest,
etc. Let's say there are eight rounds. At the end of seven (sanely
performed) rounds the following is the score:
1st David M. 91 pts.
2nd Larry L. 88 pts.
(bunch of others)
next to last Daniel M. -65 pts.
dead last Chuck B. -99 pts.
D.M. and L.L. have a close, tough, hard-fought match with Larry
coming out on top by 12 points. He's getting ready to collect the
winner's check when up comes an excited D.M., turning in the score
for his just completed match.
Apparently without any discussion (prior to or during the
match) both D.M. and C.B. independently realize that a) At most
one of them will finish in the money, and b) In order to get into
the money, a large score differntial is required IN THIS MATCH.
Within the rules, they turn the cube at EVERY AVAILABLE OPPORTUNITY.
Daniel wins the first (and only) game, a backgammon with the cube
on 2^googleplex (is that the right spelling)? So, the final score is
1st Daniel M. 3*2^g.p. - 65
2nd Larry L. 103
.
.
.
LAST Chuck B. - 3*2^g.p. -99
D.M. has "edged out" L.L. by coming from next to last to first
in the last round. C.B. receives NO PUNISHMENT (except humiliation,
which hasn't seemed to phase him over the last 20+ years of poor play)
for compiling the worst score in the history of all competions ever
recorded in any endeavor. THE EARLIER 7 ROUNDS ARE DEEMED WORTHLESS
IF THE D.M.-C.B. STRATEGY IS USED. Does this sound fair?
There may be ways around this. Magriel's suggestion of "freeze
outs" is basically an intermediate solution. Each match has the same
value, ("1 match unit") as in current practice, but determining the
winner is done differently. There would certainly be a new and
different strategy for playing this kind of match. It wouldn't be
SIMPLE MONEY PLAY strategy. It would also be a nightmare for weekend
tournament directors, since a match has NO TIME LIMIT! You might wait
several days for a couple of timid cube players. I suspect this latter
possibility is why we don't (and maybe have never?) seen this kind of
tournament.
A person who wants to see an alternate method is free to begin
his/her own server and/or tournaments. Convincing people who like
the current practices to join might be difficult. It seems like
there are other pursuits which might prove more satisfying.
Chuck
bo...@bigbang.astro.indiana.edu
c_ray on FIBS
Lawrence J. Gier
I dont mind playing with the Crawford rule in effect. I just think it
is stupid as rules go. After reading some of the responses in this
thread, I am beginning to think that backgammon causes brain cancer.
On 21 Mar 1997 17:41:33 GMT, gpr...@netcom.ca(Graham Trevor Price)
wrote:
>In <3331650e...@news.kdsi.net> ljg...@kdsi.net (Lawrence J. Gier)
>writes:
>>
>>Then again maybe we can introduce the doubling cube into, baseball,
>>golf, etc,etc,etc,etc. I was talking about backgammon.
>Ironically all these sports do have ways of increasing risk to try and
>win much like in backgammon where if you are -2:-1 and your score is
>the -2 you might try for gammon to win the match.
>In baseball they often use pinch hitters to try to generate offense
> at the cost of defence if losing in the late innings,
>in golf you can always make a risky shot to try to hit the green
> under regulation over a water hazard or trees,
>in hockey you pull the goalie.....
> You give up a lot if it doesn't work but you probably give yourself
>a little more chance for your risk.
>And for someone betting on the outcomes of games how many times
>has the other person lost maybe the first two bets and then wants
>to make the 3rd bet "double or nothing". Now that's catching up
>using the cube (He loses 2 bets but if he wins this doubled bet
>he is even). In backgammon look up "steaming".
>Graham.
>
Lawrence J. Gier
Daniel Murphy
bo...@bigbang.astro.indiana.edu (Chuck Bower) writes:
> Daniel, Daniel, Daniel! (I'd say "Raccoon, Raccoon, Raccoon" but that's
>a tongue twister.) Take it easy on the blood pressure!! Your teammates need
>you in their next international match. You seem to be letting your emotions
>get the best of you. Don't stoop to such name-calling.
Thanks, mum. The name deleted was Larry, I think Raccoon raccoon
raccoon is easier to say than Daniel daniel daniel, my blood
pressure's always within normal range, and 'frigging' ought to be in the
O.E.D. in the same volume with 'fibs,' which is what (name deleted) was
saying about the First Internet Backgammon Server. Notice that it was
to those inanities to which I was responding. The other inanities were
equally inaccurate but even less interesting.
>before the BG boom of the 70's. Still, Holland was considered a strong
>player. For example, in the Magriel interview from 1975 excerpted below,
>Paul says Holland is one of the world's top players. Tim also authored
>three books: "Beginning BG", "Better BG", and "BG for People Who
>Hate to Lose"--a candidate for catchiest BG title. The last of
>these is probably worth a read, though my recollection is that it can't
>compete with the best books available today.
I have these three books. "Beginning," I think, is an acceptable
beginner book, with the proviso (as always) that Magriel's Backgammon
makes all other beginner books irrelevant. "Better BG" consists of 64
problems and is ok but slim. And the last book, "Backgammon for People
Who Hate to Lose: the Psychology of Playing a Winning Game" analyzes
parts of 7 games, along with miscellaneous comments on play and
psychology. Also slim, but worth reading.
It also contains this passage: "In other words, sucess and skill in
backgammon are determined by the margin of victory. If by the prudent
appication of psychology you are able to win 20 points instead of 10
points, then in this given situation you are obviously a better player
than some other player whom you may consider more expert than you, but
who, under the same circumstances, wins only 10 points." But do note that
Holland is referring to money, not match play.
> Easier to find, but just as interesting is the June 4, 1979 issue
>of "Sports Illustrated". On page 64 is a column entitled "Soccer" by
>Clive Gammon! (Why do I find this interesting? Because I didn't think
>any Americans were interested in this sport back in 1979.)
YMCA league 1970-73, school team 1974-75.
Daniel Raccoon "A book entitled 'Backgammon for People Who Hate to Win'
could be assembled from some of the ideas posted here" Murphy
Kit Woolsey
I did play in a couple of experimental freeze-out tournaments a while
ago. The problem of a "match" going on forever was solved by starting
the cube at 2 after the first ten games, starting it at 4 after the next
ten games, etc. I don't remember the rest of the details, but it was an
interesting idea.
Kit
Graham Trevor Price
In <3345559a...@news.kdsi.net> ljg...@kdsi.net (Lawrence J. Gier)
writes:
>
>Visa versa I think. Some people that play for money LOVE match play!
>Start out with a small stake, lose a few games, double the stake, turn
>on the skill play past the Crawford game, double at will and make a
>killing against some poor mark.
>>
>Lawrence J. Gier
Where does this type of action exist? So if I play some strong
player a match he is going to let me get to Crawford to prove
that he is better by forcing himself to win every remaining game
in order to win money? Of course I would not double the stakes
after reaching Crawford and then what would he do? And if you
are playing for money usually beavers and raccoons are more
popular than a forced Crawford game.
I think this thread is going nowhere.
Graham.
Robert-Jan Veldhuizen
On 05-apr-97 20:20:12, Chuck Bower wrote:
[all snipped except for the important part:]
CB> Thanks for the leadin to my next topic: The "Holland Rule".
Holland rulez ! I knew it, I knew it... :)
--
Zorba/Robert-Jan
bob koca
kwoo...@netcom.com (Kit Woolsey) wrote:
>Kit
Have you gotten the match equity tables done yet Kit? :)
There would be a tremendous amount to learn. In conventional
matches if one knows a table up to 15 away 15 away it can be used
for any match up to 15 points. Under the freeze out rules a new table
would have to be used for each amount that must be won by.
If that weren't enough, if the starting value of the cube is
increased after a certain number of games, there would need to be
a chart for each number of games played so far.