question for movement mavens
David Babcock
1. 6.5 tables;
2. Club policy that everyone will always play at least 24 boards;
3. Club policy that no one ever has a 4-board sit-out;
4. A one-winner movement is acceptable (even with not-too-great
balance), but
5. Enough players have mobility limitations that a Howell would be a
last resort.
Does the following work?
Start as though playing a 7-table Mitchell with a 4-board sit out,
with EW 1 as the phantom pair, but modify as follows:
1. For the last two of the usual four sit-out boards in each of the
first six rounds, the EW seats of the phantom pair are taken by the NS
pair that will see those boards in the last round.
2. The last two boards of each four-board set in the 7th round are
not played (because NS have seen them).
It looks like the differences from the basic Mitchell are:
Round 1: NS 2 sit 1 EW for boards 3 and 4. EW 2, at table 2, have a 2-
board sitout.
Round 2: NS 4 sit 2 EW for boards 11 and 12. EW 3, now at table 4,
have a 2-board sit-out.
Round 3: NS 6 sit 3 EW for boards 19 and 20. EW 4, now at table 6,
have a 2-board sit-out.
Round 4: NS 1 sit 4 EW for boards 27 and 28. EW 5, now at table 1,
have a 2-board sit-out.
Round 5: NS 3 sit 5 EW for boards 7 and 8. EW 6, now at table 3, have
a 2-board sit-out.
Round 6 NS 5 sit 6 EW for boards 15 and 16. EW 7, now at table 5,
have a 2-board sit-out.
Round 7: The highest two boards in each set are not played because NS
have already seen them.
As the board progressions and the EW pair progressions are not
changed, there should be no issues there.
This seems straightforward enough that either it has already been
published, or for some reason that I'm missing, it doesn't work. Can
anyone clarify?
David
Frances
Isn't it considerably simpler just to do the following:
1. Play a 12 board 6.5 table Mitchell not playing the last round.
Let's say you have no pair NS table 1.
2. Everyone goes back to where they started.
3. Get the pair who started EW at table 2 (who have not yet sat out)
to move to NS table 1, so the Phanton pair are now EW 2
4. Play a 14- board 6.5 table Mitchell
I'm not claiming that this is any better (it's not), but doesn't it
have the same effect but with a much simpler movement?
Clearly the "right" answer is a 7-table 26-board Howell. If you are
prepared to make your own movement cards you can certainly reduce the
amount of movement required - if, say, 4 pairs wanted to be
stationary, 3 of them can sit throughout apart from the rounds when
they had to play each other...
David Babcock
> I'm not claiming that this is any better (it's not), but doesn't it
> have the same effect but with a much simpler movement?
If it has the same effect, "simple" will definitely win the
day. :-).
David
Michael Angelo Ravera
>
> - Show quoted text -
OK. The solution that we have used for this when a Howell is
undesirable is what is known as a "half-round bump"
In order for it to work, you put 32 boards in play and play 14 two-
board rounds. The odd pair sits out the first and last rounds. (This
is handy for when the 13th pair arrives late) When the bump pair isn't
involved, opponents face each other for two rounds.
In round 2 (the second half of the first round) , pair 13 bumps out
the pair at 2E-W. After that, they play first at E-W then at N-S at
each of the even tables, then each of the odd tables and then back to
table 2N-S at the 13th round.
Tim DeLaney
ACBLscore has a movement (under "external" when setting up the game)
called H7ROVER. I haven't tried to work out if it's similar to what
you describe. N-S plays 24 boards, E-W play 22, and there is no 4-
board sitout. It's a 1-winner movement, and of course it's fairest if
the N-S and E-W fields are balanced.
Perhaps simpler in concept is to employ what might be called an
artificial stationary. Let's say there are two pairs that are not
very mobile. One of them gets the stationary (Pair 14 at table 4) and
the other -- let's say pair 3 -- gets an "artificial stationary". At
the beginning of the game, all players are instructed that when they
are scheduled to play pair 3, they take the table mat and the boards
to the table where pair 3 will sit the entire session.
Of course, pair 3 will have to move when they are scheduled to play
pair 14, so it's not a perfect solution. Just locate the artificial
stationary table as close to table 4 to minimize length of the move.
This works just fine if there are only 2 immobile pairs; we use it
frequently at our club. With more than two pairs, it gets a bit
hairy.
Tim
Gordon Rainsford
I think a 7-table Double Hesitation Mitchell would fulfil all your
requirements.
David Stevenson
>ACBLscore has a movement (under "external" when setting up the game)
>called H7ROVER. I haven't tried to work out if it's similar to what
>you describe. N-S plays 24 boards, E-W play 22, and there is no 4-
>board sitout. It's a 1-winner movement, and of course it's fairest if
>the N-S and E-W fields are balanced.
>
>Perhaps simpler in concept is to employ what might be called an
>artificial stationary. Let's say there are two pairs that are not
>very mobile. One of them gets the stationary (Pair 14 at table 4) and
>the other -- let's say pair 3 -- gets an "artificial stationary". At
>the beginning of the game, all players are instructed that when they
>are scheduled to play pair 3, they take the table mat and the boards
>to the table where pair 3 will sit the entire session.
>
>Of course, pair 3 will have to move when they are scheduled to play
>pair 14, so it's not a perfect solution. Just locate the artificial
>stationary table as close to table 4 to minimize length of the move.
>
>This works just fine if there are only 2 immobile pairs; we use it
>frequently at our club. With more than two pairs, it gets a bit
>hairy.
Yes, but, the OP said
>> 2. �Club policy that everyone will always play at least 24 boards;
If we do not follow that requirement it is easy: we just play a
Hesitation Mitchell, 8x3 rounds, 6 stationary pairs, easy peasy.
--
David Stevenson Bridge RTFLB Cats Railways
Liverpool, England, UK Fax: +44 870 055 7697 ICQ: 20039682
<webj...@googlemail.com> bluejak on OKB
Bridgepage: http://blakjak.org/brg_menu.htm
Sid
"David Babcock" <dp...@fastmail.fm> wrote in message
news:10b342d3-2ee0-40f8...@k19g2000yqc.googlegroups.com...
> Assume the following conditions:
>
> 1. 6.5 tables;
> 2. Club policy that everyone will always play at least 24 boards;
> 3. Club policy that no one ever has a 4-board sit-out;
> 4. A one-winner movement is acceptable (even with not-too-great
> balance), but
> 5. Enough players have mobility limitations that a Howell would be a
> last resort.
>
Here's your solution: Double-hesitation Howell, 27 boards, from the EBU
movement book. 5 Pairs are stationary which will solve your mobility
problem.
http://www.elsid.co.za/7table27bds.pdf
Sid
Sid
"Gordon Rainsford" <r...@gordonrainsford.com> wrote in message
news:aaaa16ba-7258-45a1...@j4g2000yqe.googlegroups.com...
I think a 7-table Double Hesitation Mitchell would fulfil all your
requirements.
---
Yep, it does. :) Posted a movement.
Sid
David Babcock
> ACBLscore has a movement (under "external" when setting up the game)
> called H7ROVER. I haven't tried to work out if it's similar to what
> you describe. N-S plays 24 boards, E-W play 22, and there is no 4-
> board sitout. It's a 1-winner movement, and of course it's fairest if
> the N-S and E-W fields are balanced.
The reason for my OP is that in H7ROVER, which we have used, only one
pair actually plays 24 boards; the rest play 22. (H7ROVER is a 6-
table Mitchell with a roving pair that makes 2-board roves both
ways.) I would love to satisfy club policy with something already in
ACBLScore, but there isn't anything AFAIK for 6.5 that does that.
David
Gordon Rainsford
>
> The reason for my OP is that in H7ROVER, which we have used, only one
> pair actually plays 24 boards; the rest play 22. (H7ROVER is a 6-
> table Mitchell with a roving pair that makes 2-board roves both
> ways.) I would love to satisfy club policy with something already in
> ACBLScore, but there isn't anything AFAIK for 6.5 that does that.
>
> David
Is it hard to put movements into ACBLscore? (genuine question)
David Babcock
> Is it hard to put movements into ACBLscore? (genuine question)
ACBLscore has no documented capability to import movements. Each of
the "External Pairs" movements is a binary file that you can edit in a
human-readable format with an ACBLscore command (EDMOV), and you can
save your changed file under a new name, but this is meant for simple
things like changing arrow switches or re-ordering boards, not for
things like adding or subtracting rounds. I've worked out most of the
file structure except for a few mystery bytes -- no big achievement
when you study a binary dump next to ACBLscore's own printout of the
movement -- but hacking binaries is nobody's idea of fun.
A better capability would certainly be welcome, since Bump Mitchells,
Hesitation Mitchells, and Pivot Mitchells do not exist in ACBLScore.
Rover Mitchells it has, at least.
Dave
Michael Angelo Ravera
I'm actually working on it!
All that you have to do to import a movement is to place it in the
appropriate directory MOV, BAM, or IND. In fact, all of those external
movements are not native to ACBLSCOR. It only knows about Several
kinds of Mitchell, Howell, Bounce, and Web. It has no native support
for most numbers of tables for BAM movements where the Bounce Movement
doesn't work, for instance or most non-prime numbers of tables for
indivivuals where neither the rainbow nor Mitchel movements work.
David Babcock
wrote:
> All that you have to do to import a movement is to place it in the
> appropriate directory MOV, BAM, or IND.
Yes. What I meant is that, AFAIK, you have to create the binary
yourself if you are not hacking an existing movement file, and the
format itself takes a bit of work to figure out (and there are still a
few gaps in my understanding of it -- the array of three-byte values
for NS pair, EW pair, and board set, I get, but I haven't worked out
everything preceding those -- I've got most of it, but not
everything) . If there is some way to create a .MOV file more easily,
my life will be easier if someone will tell me what that way
is! :-) A Hesitation Mitchell for 11 tables would be an improvement
over the 3-board rounds that we now do because of our 24-board policy,
but I haven't had the time yet even to figure out which existing file
to hack for that.
David
Michael Angelo Ravera
If you want to create a movement that isn't very much like one of the
others, you can do so in ACBLSCOR, but you have to do it as part of a
game and then save the movement file.
As you create your game and are asked "well, what kind of a movement
do you want?", if you scroll down far enough in either the Pairs, BAM,
or Indi movements, there is an option to "Create your own". After you
finish the rest of the set up, you can go edit the movement and save
it. This option isn't permitted in the Movements menu, for some
reason.
BTW: Blackpool is the definitive movement for 11 full tables when
there is no possibility that you will have 11.5.
Put boards 1-20 on tables 1-10 two per table. Put boards 21-22 on
table 11 along with boards 23-26 on a bye-stand near by table 11. For
the first 11 rounds everyone moves as normal. In round 12 (except at
tables 5,6,7, see below) it is a simple revenge round. If you want to
play a 13th round, it is inside out (all of the E-W pairs subtract
their pair number from 12 and go there).
This movement is very satisfactory with no changes whatever for 12
rounds. For 13 rounds, you end up with pairs 6 playing each other 3
times. There are about 6 ways to avoid this. Most of the easier ones
involve a share between two of the tables 5-7 at rounds 12 and 13 and
a switch of the opponents and boards that they would have had at round
13 playing in round 12.
David Babcock
wrote:
> If you want to create a movement that isn't very much like one of the
> others, you can do so in ACBLSCOR, but you have to do it as part of a
> game and then save the movement file.
>
> As you create your game and are asked "well, what kind of a movement
> do you want?", if you scroll down far enough in either the Pairs, BAM,
> or Indi movements, there is an option to "Create your own". After you
> finish the rest of the set up, you can go edit the movement and save
> it. This option isn't permitted in the Movements menu, for some
> reason.
This reassures me that I am justified in wondering sometimes whether
there is something strange in the water in Memphis. But I will
certainly look into that option.
> BTW: Blackpool is the definitive movement for 11 full tables when
> there is no possibility that you will have 11.5.
...For
> the first 11 rounds everyone moves as normal. In round 12 (except at
> tables 5,6,7, see below) it is a simple revenge round.
I hope we can do better than to have one pair play 4 boards against
the best pair the other way while their scores are compared against
another pair who play 4 boards against the weakest pair. But maybe we
can't. A Hesitation Mitchell is a one-winner movement that doesn't
have the aforementioned problem, but maybe the cure is worse than the
disease because a one-winner Mitchell or near-Mitchell is horrific on
other grounds. Dunno. The revenge movement has the practical problem
that the effect of the flaw can be assessed by even the least skilled
pair..
David
David Babcock
> On Nov 14, 2:38 pm, Michael Angelo Ravera <marav...@prodigy.net>
> > This option isn't permitted in the Movements menu, for some
> > reason.
>
> This reassures me that I am justified in wondering sometimes whether
> there is something strange in the water in Memphis. But I will
> certainly look into that option.
and now, having looked into it, I must thank Michael for having
pointed me to it :-) To clarify for the next adventurer: it is in
the External area that you scroll down.
David
Michael Angelo Ravera
But with 13 rounds, you end up playing a second set against some
random other pair. Isn't it better to have some pair have to play 4
boards against 2 pairs and 2 against the 9 others and to play all of
the boards than to have some pairs play 3 boards against 9 pairs (or
even 8) and NONE against 2 or 3 others and to miss 6 or 9 boards
entirely?
Yes, ideally, you'd like to play everybody in the room an equal number
of times with equal cards and with equal shots at every dealer and
vulerability combination. The problem is that you can't do that short
of playing 336 boards (21 rounds of 16 boards)for a 11-table game. OK,
so there are two separate winners for N-S and E-W, so you could do it
in 176 boards. OK, maybe just once around with everyone as dealer and
each vulnerability, 44 boards. Even that's still too long for a normal
club game. It would be 5.5 hours at the prescribed rate of play of 7.5
minute per board. Yeah, you could make it a fast event and get it done
in 3 hours and 40 minutes at 5.5 minutes per board. But, given that
you are going to play between 24 and 28 boards and that you want
everybody to play all of the boards, a movement with a revenge round
and an inside out round are the easiest implementation (but DO
eliminate the triple play at table 6).
David Babcock
On Nov 16, 2:27 am, Michael Angelo Ravera <marav...@prodigy.net>
wrote:
> But with 13 rounds, you end up playing a second set against some
> random other pair. Isn't it better to have some pair have to play 4
> boards against 2 pairs and 2 against the 9 others and to play all of
> the boards than to have some pairs play 3 boards against 9 pairs (or
> even 8) and NONE against 2 or 3 others and to miss 6 or 9 boards
> entirely?
We are considering whether we should run our 11-table games as
Hesitation Mitchells *instead of* playing 3-board rounds (given the
club's 24-board policy).
Dave
Fred
Perhaps you should reconsider the club policy. Perhaps the policy
should be to use the best duplicate movement instead of focusing on
the number of boards played. Yeah i know you've heard me say this
before. lol. :) Nothing annoys me more than playing in some movement
where i miss 3 or more pairs and some number of boards because the
director is looking for an easy way out or the club is trying to play
as many boards as possible to make the lols smile.
Somebody knock me off my soap box please.
Fred
Frances
> I hope we can do better than to have one pair play 4 boards against
> the best pair the other way while their scores are compared against
> another pair who play 4 boards against the weakest pair. But maybe we
> can't. A Hesitation Mitchell is a one-winner movement that doesn't
> have the aforementioned problem, but maybe the cure is worse than the
> disease because a one-winner Mitchell or near-Mitchell is horrific on
> other grounds. Dunno. The revenge movement has the practical problem
> that the effect of the flaw can be assessed by even the least skilled
> pair..
>
> David
What is the problem with a one-winner Mitchell? With 11 tables a
Hestitation Mitchell is easily my preferred movement. With 12 tables I
play an arrow-switched share & relay Mitchell. I _want_ one winner.
Frances
I don't agree with this. I think your policy should depend on
context. For a club championship, play the best movement. For a
random club evening, I am going for an enjoyable evening and don't
really care who wins. I want to play as many boards as possible,
because that's what I am paying for. I certainly don't want to to sit
out. In particular, I don't want to sit out for three boards.
On Monday (in the simultaneous pairs) we only played 22 boards as it
was a 24-board movement with a sit-out. Given that you need a very
large score to win a simultaneous pairs, this was good for us as it's
easier to get over 70% in 22 than in 24 boards. However I think we'd
still have preferred to get more bridge played.
Michael Angelo Ravera
EBU May be different, but in the ACBL there is huge masterpoint
incentive to get one winner at least up to about 14 tables. After 16
tables, you don't have to run a one-winner movement to give overalls
(but, for fairness, you'd like to).
David Babcock
> the
> director is looking for an easy way out or the club is trying to play
> as many boards as possible to make the lols smile
or just maybe the director understands that the whole argument is
academic if he doesn't get people *in the door*.
David
David Babcock
> What is the problem with a one-winner Mitchell? With 11 tables a
> Hestitation Mitchell is easily my preferred movement. With 12 tables I
> play an arrow-switched share & relay Mitchell. I _want_ one winner.
I do too, but I accept that sometimes that choice does come at an cost
in terms of fairness of comparisons. A couple of days ago, I set up a
Hesitation Mitchell in ACBLscore (arrow switch last round except at
table 11, per recommendation of a UK club -- Hertfordshire??). I am
sure the players will be happy with it, but I confess the balance
report is not a thing of beauty:
Balance frequency table:
Comparisons Frequency
0 1
1 19
2 19
3 81
4 0
5 0
6 0
7 0
8 36
9 19
10 55
11 1
But I will run this movement, and happily, because (a) it is *good
enough* that the pair playing the best bridge on the day figures to
win and (b) the players should be fine with it..
David
Derek Broughton
Indeed - we all tend to want to play the most boards possible, in the least
time, but generally without playing speedball :-) The only people happy
with a sit-out are the ones who sit out the last round (and go home).
--
derek
David Babcock
> The only people happy
> with a sit-out are the ones who sit out the last round (and go home).
Two-board sit-outs come with the territory (unless you have an on-call
player or pair), but avoiding three-board sit-outs is certainly a
worthy objective, though perhaps involving the same trade-offs we have
been discussing.
David
Barry Margolin
<a01a6ff9-f3da-462a...@k17g2000yqh.googlegroups.com>,
David Babcock <dp...@fastmail.fm> wrote:
If you have a small game, the real problem is avoiding 4-board sitouts.
That's when the movements that involve half-round sitouts become
invaluable.
--
Barry Margolin, bar...@alum.mit.edu
Arlington, MA
*** PLEASE don't copy me on replies, I'll read them in the group ***
David Babcock
> If you have a small game, the real problem is avoiding 4-board sitouts.
> That's when the movements that involve half-round sitouts become
> invaluable.
and that was the subject of my OP regarding 6.5 tables. :-) Frances
Hinden improved on my suggestion, and I have set up something very
similar to what she suggested in ACBLscore. It looks solid but I
haven't exercised it yet in practice. I could post the EDMOV printout
if anyone would care to take a look before I blow up a day's bridge at
the club. :-)
David
David Stevenson
>Perhaps you should reconsider the club policy. Perhaps the policy
>should be to use the best duplicate movement instead of focusing on
>the number of boards played. Yeah i know you've heard me say this
>before. lol. :) Nothing annoys me more than playing in some movement
>where i miss 3 or more pairs and some number of boards because the
>director is looking for an easy way out or the club is trying to play
>as many boards as possible to make the lols smile.
>
>Somebody knock me off my soap box please.
Keeping the customers happy *should* be the first principle of running
a club. Players pay so much money, and expect so many boards. If they
play too few boards they feel cheated and rightly so. I know I do!
Best is to have a movement that fits all the parameters of a club, and
a minimum number of boards played is a normal and reasonable parameter.
So is avoiding four board sitouts.