Post-Crawford Strategy
Hank Youngerman
Playing a FIBS match against a player with a decent rating, I took a
4-1 lead in a match to 5. I lost the next (Crawford) game. In the
following game, my opponent did not double until about the 5th or 6th
roll. My guestimate is that his winning chances at that point were in
the 60-65% range - clearly a take of course. I made a statement that
"It always surprises me how often good players (and myself too) forget
to double right away post-Crawford." My opponent said that he
intentionally waited, in the hope that he would get a free point (i.e.
his position would be strong enough to induce a drop.)
My belief is that this cannot possibly be correct strategy. First, it
just FEELS wrong. On a more analytical level, unless you think your
opponent's judgment is very poor and that he will take a clear drop,
in effect you are saying "I want to maneuver you into a position where
you have a drop. But then, even though I would like to force you to
take such a position, I want you to drop it anyway. (Since I could
"force" you to take in such a position by making a no-cost double
earlier before my position got this strong.)"
Who's right, him or me?
Kit Woolsey
Hank Youngerman (hankyou...@mindspring.com) wrote:
: Playing a FIBS match against a player with a decent rating, I took a
It depends. If the leader in the match fully understands the situation,
then this ploy obviously can gain nothing. But if he doesn't, the gain
can be considerable.
Let's suppose the trailer waits, and the game develops into a straight
race where the trailer is 90% to win (gammons not possible). He now
doubles. Would you take or pass if you were the leader? If your answer
was that you would pass, you have fallen into his trap. Now he only has
to win one game to win the match (since NEXT game he will double on the
opening roll), while if you take he has to win both this game and the next
game.
Think this is so obvious that nobody will fall for it? Think again.
Using this ploy I have often stolen a vital point, and not just against
beginners. I have seen several quite competent players fall into this trap.
Kit
Chuck Bower
In article <34a03f6f...@news.mindspring.com>,
Hank Youngerman <hankyou...@NOSPAMmindspring.com> wrote:
>Playing a FIBS match against a player with a decent rating, I took a
>4-1 lead in a match to 5. I lost the next (Crawford) game. In the
>following game, my opponent did not double until about the 5th or 6th
>roll. My guestimate is that his winning chances at that point were in
>the 60-65% range - clearly a take of course.
Of course!! Of course??
>I made a statement that
>"It always surprises me how often good players (and myself too) forget
>to double right away post-Crawford." My opponent said that he
>intentionally waited, in the hope that he would get a free point (i.e.
>his position would be strong enough to induce a drop.)
>
>My belief is that this cannot possibly be correct strategy.
"...cannot possibly..." Pretty strong words.
>First, it just FEELS wrong.
"FEEL" is in the touch of the beholder...
>On a more analytical level, unless you think your
>opponent's judgment is very poor and that he will take a clear drop,
Please define "...judgment is very poor..."
>in effect you are saying "I want to maneuver you into a position where
>you have a drop. But then, even though I would like to force you to
>take such a position, I want you to drop it anyway. (Since I could
>"force" you to take in such a position by making a no-cost double
>earlier before my position got this strong.)"
>
>Who's right, him or me?
This strategy is not new. It's been proposed in the past (distant
and not so distant) by such people with "poor judgement" and names like
Paul Magriel and Frank Frigo. (OK, I'm guilty of authoritarianism
once again.)
I don't think this is really a "who's right, who's wrong" question.
Under different circumstances, either of you could be right or wrong.
But there is a good principal of BG (and many other pursuits) which
goes like "always try to give your opponent a tough decision". Pretty
much everyone at the intermediate level and above knows to take a
post-Crawford cube at the 3-away score. Tough decision? I don't think
so.
You seem to indicate that the take/drop decision WHEN THE GAME
HAS PROGRESSED A WHILE is an easy one. It's certainly not obvious to
me. I'd like to see further explanation as to how one goes about
determining when to drop and when to take under these circumstances.
Is there an easy algorithm? If it stumps me, then I can think of
at least one opponent who I'd recommend using it against!
This discussion is at least partly analogous to the "double immediately"
strategy at the 2-away, 2-away score. The person making the drop/take
decision can always take and be no worse off than if the cube had been
sent immediately. But that's not necessarily taking best advantage of
the situation. Maybe dropping is right. Then you throw away equity.
And in the process of trying to figure out which decision is correct,
maybe an error occurs.
Possibly the strongest argument in favor of your opp's strategy comes from
your own words. I repeat:
>"It always surprises me how often good players (and myself too) forget
>to double right away post-Crawford."
Everybody knows that technically it is correct to double immediately
post-Crawford. Yet even "good players" forget. In other words, they
make mistakes. So if they make mistakes in this "obvious" situation,
maybe they can be induced to err in a less obvious one.
Chuck
bo...@bigbang.astro.indiana.edu
c_ray on FIBS
Gary Wong
> Playing a FIBS match against a player with a decent rating, I took a
> 4-1 lead in a match to 5. I lost the next (Crawford) game. In the
> following game, my opponent did not double until about the 5th or 6th
> roll. My guestimate is that his winning chances at that point were in
> "It always surprises me how often good players (and myself too) forget
> to double right away post-Crawford." My opponent said that he
> intentionally waited, in the hope that he would get a free point (i.e.
> his position would be strong enough to induce a drop.)
>
> just FEELS wrong. On a more analytical level, unless you think your
> opponent's judgment is very poor and that he will take a clear drop,
> you have a drop. But then, even though I would like to force you to
> take such a position, I want you to drop it anyway. (Since I could
> "force" you to take in such a position by making a no-cost double
> earlier before my position got this strong.)"
>
> Who's right, him or me?
In my opinion, you are both right. Against a reasonable player, you might
as well get the cube action out of the way and toss it in their direction
as soon as possible. However, as you say, your opponent's argument has some
merit against a poor player -- the key is that in a post-Crawford game with
an odd number of points to go, winning a single point is practically as
good as winning two. The trailing player definitely wants the leader to
drop the cube (until there are significant gammon chances). So he has
nothing to lose, and potentially an entire game towards match to win, to
wait for market losers before doubling. I wouldn't complain about his cube
handling one bit (and would be tempted to emulate it against a weak
player -- I'm not one to sneeze at the chance of getting something for
nothing :-)
Obviously this has to be contrasted with the situation an _even_ number of
points away. Here, a single point is almost worthless to the trailer
(apart from using up the leader's free drop). It is surely correct to
double as soon as possible in that case, because you never want the leader
to drop (you want to put every little bit of equity you have towards winning
a game at 2 to reduce the length of the match; you don't want to waste _any_
on your opponent's free drop).
Cheers,
Gary.
PS: I hope you won the match :-) there's nothing more annoying than watching
your hard-earned lead vanish away 2 points a time after the Crawford game -- I
recently got knocked out of a competition playing 9 point matches, after
leading 8-1 :-(
--
Gary Wong, Computer Science Department, University of Auckland, New Zealand
ga...@cs.auckland.ac.nz http://www.cs.auckland.ac.nz/~gary/
Donald Kahn
>Playing a FIBS match against a player with a decent rating, I took a
>4-1 lead in a match to 5. I lost the next (Crawford) game. In the
>following game, my opponent did not double until about the 5th or 6th
>roll. My guestimate is that his winning chances at that point were in
>the 60-65% range - clearly a take of course. I made a statement that
>"It always surprises me how often good players (and myself too) forget
>to double right away post-Crawford." My opponent said that he
>intentionally waited, in the hope that he would get a free point (i.e.
>his position would be strong enough to induce a drop.)
>
>My belief is that this cannot possibly be correct strategy. First, it
>just FEELS wrong. On a more analytical level, unless you think your
>opponent's judgment is very poor and that he will take a clear drop,
>in effect you are saying "I want to maneuver you into a position where
>you have a drop. But then, even though I would like to force you to
>take such a position, I want you to drop it anyway. (Since I could
>"force" you to take in such a position by making a no-cost double
>earlier before my position got this strong.)"
>
>Who's right, him or me?
Many of the very best players on the tournament circuit do this. Paul
Magriel (x-22) calls it "the trick". The idea is this: the trailer
benefits greatly if he can steal a point here - he is then an
even-money bet in the match.
He knows you will take if he doubles at the first opportunity. If he
waits, he is giving you a chance to make a mistake. Mainly this will
happen when *there is some gammon danger*. Now you have something to
think about, don't you?
The simple arithmetic says you should take *unless* your chances of
losing a gammon are greater than your chances of winning. Are you
sure you can judge this accurately in every case? If so, hats off to
you!!
If you take, he is no worse off than having doubled at the first
opportunity. If you pass and passing is an error, he benefits. If
you take and *that* was an error, he benefits.
Looks like a 100% proposition for him to wait.
deekay
Hank Youngerman
I've learned something. Thanks to all of you.
Tapio Palmroth
hmmmm....is there a 'everybody knows what to do' ?
i doubt it very strongly :)
there must always be situations which are to be 'never-seen-before '
at least in this funny game called backgammon :)
if somebaby tells me that sure everybody knows how to handle this
situation , i welcome him/her to play asap :)
but now to this situation :
think that you are leading and also very happy with that .
now one of the 'well-known' facts is that never give the trailer
a chance to big swing.
so if he/she doubles right away , what can you do but take .
now you know how to play that game .
but if he waits , you just don't know if he has forgot or just
wait for some place to send the monster over .
and if there is some slight chance then for a gammon , ain't that
tempting to start a new and fresh game ?
it is all in your hands .
and then , if you lose a gammon , was it in your own hands ? :)
many times it has happened that when trailing big , i have managed
to steal some points easily and then my opponent gets a bit mad
when he finds that i have gained some easy points .
then he maybe take a double and loses there a gammon .
i have gained maybe 7-8 points just finishing one game :)
but there is one big hole in this : you just have to double when
there is a gammonchance ( not too big ) , and if you have fallen
to a race where it is easy to drop , sure , you have fallen to your
own trap :) if gammonchances are too big to take , same thing again ,
your own trap :)
and when your opponent don't know what you are planning to do ,
you must be ahead in the game :)
maybe ;))
tapio
ps.and do think with your own brains and never believe everything
what i maybe say , or somebody else can say :))
are you mixed enuff ? ....gooood......let's play !