Crawford rule
Andrew Grant
Why does the Crawford rule only apply for one game? As I understand it, the
reason for the Crawford rule is to stop the trailing player doubling
immediately in order to catch up quicker if he wins. But in that case why
not forbid doubling for the whole of the remainder of the match?
Andrew Grant.
Walter Trice
would make it too hard to catch up, which would give too large a reward for
a lucky early win. E. g. in a 5 point match, one player could double and win
a gammon to go up 4-0. With a permanent Crawford, the trailer would then
have to win 5 in a row, or perhaps 3 in a row with 2 of them gammons. This
is just much too difficult.
The real reason for the Crawford rule is not some abstract sense of
'fairness.' The rule is used because without it there would be some very
strange cube action near the end of the match. For example, consider the
score 2-3 in a 5 point match with no Crawford rule. If the trailing player
doubles, the leading player can take ANYTHING AT ALL if there is no gammon
threat. The worst that could happen is he loses, goes down 4-3, doubles
immediately next game, and is 50-50 except for the small 'free drop' equity.
Of course the Crawford rule seems weird and unnatural, and it's just one
more thing that has to be explained to people who are new to match play. I
once proposed replacing it with the rule that you have to win by two points
(like in tennis, etc.) This rule would also eliminate a lot of the
'anomolous' cube action that comes up near the end. The problem was that
organizers objected that a match could go on 'forever' with the players
swapping one-point leads. I kind of like the idea anyway. But no one else
does ;-)
--Walter Trice
Andrew Grant wrote in message <861igd$719$1...@news7.svr.pol.co.uk>...
Wolf Pfeiffer
http://www.bkgm.com/glossary.html
Andrew Grant <a...@honinbo.freeserve.co.uk> schrieb in im Newsbeitrag:
861igd$719$1...@news7.svr.pol.co.uk...
wc...@my-deja.com
"Walter Trice" <w...@world.std.com> wrote:
> The real reason for the Crawford rule is not some abstract sense of
> 'fairness.' The rule is used because without it there would be some
very
> strange cube action near the end of the match. For example, consider
the
> score 2-3 in a 5 point match with no Crawford rule. If the trailing
player
> doubles, the leading player can take ANYTHING AT ALL if there is no
gammon
> threat.
The problem that I have as a new BG player with cubing is that it
doesn't make sense to me for matches that just terminate after someone
reaches a fixed score. I think a better system would be if a double
doubled the *equity* you were playing for in a particular game instead
of the nominal score.
Let me explain what I mean. Take the simple case where each game is one
point (no gammons) and we're playing to 2 points without the cube. This
is just a best of 3 match. What happens is that the first two games are
worth 0.5 equity and the last game if played is worth 1.0, the entire
equity of the match. It's equivalent to the following scoring method:
each player starts out with two chips. The first two games each player
wagers one chip on the outcome--if one player has all the chips at the
end of both games he wins the match. Otherwise what happens is that in
the third game the limits double and the players are wagering two chips
on the outcome.
All play-to-n matches can be described in this manner. Let's examine
the 3 point case because it includes the idea of an all-in. Ok, in the
play-to-3 point case, each player starts out with 8 chips. The wager is
3 chips the first two games. If the players split the limits increase
to 4 chips the next two games and 8 chips the last game. Consider what
happens someone wins the first two though--this player will have a 14:2
chip lead. Now the wager can only be 2 chips instead of 4 because it
wouldn't be right for the player who is ahead to put more chips in
jeapordy than the other player. If the player who is behind 2-0 wins
then he is again "all-in" for 4 chips and then finally 8 chips.
So, in short the way I believe the cube could work to closely resemble a
money match is that both players start out with the same amount of
chips. Now at the beginning, each game is worth x in chips, each gammon
worth 2x, and each backgammon worth 3x in chips. Now to insure that the
match terminates in a timely manner, the limits can be raised every now
and then. A double would just be a doubling in stakes for that game, if
both sides have enough chips to cover the double. However in matches we
have an all-in rule--a player cannot have more chips at risk than his
opponent. If a player has less than x chips left and each game is worth
x, neither side may double. In fact this player should be able to only
win an amount equivalent to what he has in front of him. the all-in
player does not have any extra risk due to gammons or backgammons so it
stands to reason he should not gain from them either.
Bill
Sent via Deja.com http://www.deja.com/
Before you buy.
Patti Beadles
Part of the skill in tournament backgammon is that cube decisions (and
checker play decisions too) change during the match, based on the
current scoring situation. People often come up with various schemes
to make tournaments more like money play, but I argue with the
underlying goal... I think that the intricacies of tournament play make
the game far more interesting.
Compare it to poker tournaments, something I know you understand.
It's certainly possible to have a poker tournament where you just play
a fixed number of hands (or a fixed length of time) and then divide up
the prize money based on the chips you have in front of you. But
that's not the way (post) poker tournaments are done. Tournament
poker is much different than ring game (money) poker, because chips
now have different value based on how many of them you have.
Scoring points and acquiring chips are different things conceptually.
You can lose chips, but once you score a point, it's yours forever.
-Patti
--
Patti Beadles |
pat...@netcom.com/pat...@gammon.com |
http://www.gammon.com/ |
or just yell, "Hey, Patti!" | Quisque comoedus est.
Chuck Bower
>In article <FoJAD...@world.std.com>,
This sounds interesting. First off, it has some things in common
with a "freeze out", and even more in common with the rules of "summer
freeze out" as contested as a side event at the annual Michigan Summer
Championships conducted by Carol Joy Cole (and the third highest attended
annual tournament in the US, I believe).
In a freeze-out, the first person to take a lead of N or more points
wins. So maybe you set N = 5. As soon as one player leads on the scoresheet
by 5 or more s/he is declared the match winner. One problem with the
freeze-out is the lack of a time limit. The score could reach 100-98, for
example. That would be a long match!
Bill's proposal recognizes this shortcoming and so does CJC's version
of the freeze-out. I don't have the rules in front of me, but after a
certain number of games the cube begins on 2. The next game on 4. The game
after that 8. That way a realistic time limit is enforced.
There are tournament promoters who like to spice up their events by
offering sidegames with modified rules. Bill Davis experiments during his
springtime event--The Midwest Championships. I'll forward your proposed
rules to Bill and Carol and see if it generates any interest. It would
be interesting if someone would try this out and see if there are any
anomolous/undersirable side-effects. (Or maybe you already have tried it
out??)
Chuck
bo...@bigbang.astro.indiana.edu
c_ray on FIBS