Should Stayman be used with 4-3-3-3 ?

[Bertil]

I seem to recall having read someplace that the Stayman 2C bid should not
be used with a perfectly flat hand, 4-3-3-3. But I cannot find it stated
in common texts about Stayman. Am I just dreaming?

Stig

[patmp...@gmail.com]

This seems to be common knowledge among even intermediate players.

[Barry Margolin]

In article <c2e36e35-ec48-4a01...@googlegroups.com>,

I've seen it suggested in many bridge magazine articles. The idea is
that if opener has at most one doubleton, which is typical for NT
openers, there's only one ruffing opportunity, so there's not much
benefit to being in a suit rather than NT. And 9 tricks may be easier
than 10.

I don't personally subscribe to it unless I have extra values, at least
a combined 28 HCP. Ruffing isn't just for gaining tricks, it's also for
stopping the defenders from running suits. With more combined strength,
there's a better chance that every suit is double stopped and they can't
do this.

--
Barry Margolin
Arlington, MA

[Nick France]

One always wonders if your questions are for real or just to set
something else. Assuming for the moment its for real then

Both 5 Weeks to Winning Bridge and Common Sense Bidding both
explicitly state it. Many of the modern texts infer it and often show
it in one of there practice hands.

To give a modern reference look up Barbara Seagram & Linda Lee
"Stayman Auctions" in the Practice Your Bidding series where she
states this explicitly on pgs 5 and 7.

Not bidding stayman with a 4333 distribution is common knowledge for a
good teacher and he/she will buid it into the stayman convention when
it is taught.

Nick France

[Bertil]

Looking at A.Grant's two books about conventions, Bridge Basic and Commonly Used Conventions, I can find no mentioning of the 4-3-3-3 hand.
And yet, I think one should not rule out using Stayman with 4-3-3-3 with
AKQ or KQJ in the 4-card suit.

At what level of teaching should the Stayman bid be introduced? Kantar does
not mention the 4-3-3-3 hand in his "Bridge for Dummies".

Aside from this, how strong should a 4-card suit be to justify the Stayman bid, when holding 4-3-4-2?

Stig

Stig

[jonathan23]

I've seen Howard Schenken suggest that you should still Stayman with 4333
if your four-card major is strong (his examples seemed to be better than
QJxx) and you are weak elsewhere.

I think of foregoing Stayman when my major is relatively weak and my
values are distributed across the other suits.

>
> At what level of teaching should the Stayman bid be introduced? Kantar
> does not mention the 4-3-3-3 hand in his "Bridge for Dummies".

Early on, since it's so fundamental and simple and matches well with
Jacoby transfers, which they should learn too in most geographic locales.

Beginners have to learn what to do when partner opens 1NT sometime, and
it makes sense to me to teach them one scheme which is what
pretty much everyone else in their environment plays. In North America
that would be non-promissory Stayman (i.e. any response including 2D may
be passed) and simple Jacoby Transfers. They're mostly fairly mechanical
with a few specific exceptions and just a memory exercise in which
sequence shows which hand type.

The exceptions for certain 4333 hands and weak three-suiters short in
clubs can be added later on.

>
> Aside from this, how strong should a 4-card suit be to justify the
> Stayman bid, when holding 4-3-4-2?

5432 or better, since you'll only play in it if there's a 4-4 fit and as
long as you have a 4-4 fit you should be OK most of the time. "Minimum
biddable" four-card suits as played decades ago are pretty obsolete now
and were designed to avoid uncomfortable 4-3 fits.

Trying to teach beginners that there are exceptions is generally wrong in
my view. While experienced players may recognize exceptional
circumstances when they see them, beginners generally can't
and teaching them "exceptions to the rules" will mostly just confuse them
and cause them to see exceptional situations where they do not exist.

--
- Jon Campbell
Ottawa Canada a

[derek]

On Saturday, April 20, 2013 1:11:01 PM UTC-3, Nick France wrote:
> On Apr 20, 6:34 am, Bertil <stigfjor...@hotmail.com> wrote:
>
> One always wonders if your questions are for real or just to set
> something else. Assuming for the moment its for real then
>
> Both 5 Weeks to Winning Bridge and Common Sense Bidding both
> explicitly state it.

Say it ain't so! How can Stig not have memorized those two books?

[Bertil]

As usual, you are both full of crap. Sheinvold's book is not used by any
teacher today,partly because it's out of print long ago. Show me the page where he mentions not bidding Stayman with 4-3-3-3.

Much of his bidding is obsolete. E.g. he avaluates the hand :AJxxxxx-AKx-xx-x
as 20 pts by counting 12 HCp + 5 for suit length +1 for doubleton +2 for single.
In short, he counts points for both length and shortness. Who today would
evaluate this hand as 20 pts?

Also, he defines a 4-card suit as biddable if it has QJ or better.
To respond to Stayman the suit must be biddable. He says nothing about the
quality of responders 4-card suit.
He stresses the importance of having 2 quick tricks, which Grant never does.

Do you have any more stupid comments?

Stig

[Jennifer Murphy]

It is puzzling to me why you continue to post here, since pretty much
everyone is either "full of crap" or "stupid". Are you a masochist?

[Herb]

On 4/20/2013 6:10 PM, Bertil wrote:
> On Saturday, April 20, 2013 7:18:25 PM UTC-4, derek wrote:
>> On Saturday, April 20, 2013 1:11:01 PM UTC-3, Nick France wrote:
>>
>>> On Apr 20, 6:34 am, Bertil <stigfjor...@hotmail.com> wrote:
>>
>>>
>>
>>> One always wonders if your questions are for real or just to set
>>
>>> something else. Assuming for the moment its for real then
>>
>>>
>>
>>> Both 5 Weeks to Winning Bridge and Common Sense Bidding both
>>
>>> explicitly state it.
>>
>>
>>
>> Say it ain't so! How can Stig not have memorized those two books?
>
> As usual, you are both full of crap. Sheinvold's book is not used by any
> teacher today,partly because it's out of print long ago. Show me the page where he mentions not bidding Stayman with 4-3-3-3.
>

It's on the same shelf as Alan Truscott's book describing Asset points.

- Herb

[Bertil]

> Yes, it would be better to have 5432, but you can only have 13 cards.

Stig

[Bertil]

Truscott's book was revised in 2004 because he had a large readership as
bridge Editor at N.Y.Times. Do you know how many copies were sold?

Stig

[Bertil]

You never miss an opportunity to stick up for the shitheads. Is it because you
identify with them? When you post at RGB you better have your facts straight.

Stig

[Barry Margolin]

In article <1ee2857b-3ce7-4383...@googlegroups.com>,

Bertil <stigf...@hotmail.com> wrote:

> On Saturday, April 20, 2013 5:55:11 PM UTC-4, jonathan23 wrote:
> > 5432 or better, since you'll only play in it if there's a 4-4 fit and as
> >
> > long as you have a 4-4 fit you should be OK most of the time. "Minimum
> >
> > biddable" four-card suits as played decades ago are pretty obsolete now
> >
> > and were designed to avoid uncomfortable 4-3 fits.
> >
> > Yes, it would be better to have 5432, but you can only have 13 cards.
>
> Stig

Why did you put a '>' prefix on your response (the sentence beginning
with 'Yes'), making it look like it was part of the message you were
responding to?

[Kenny McCormack]

In article <barmar-135C87....@news.eternal-september.org>,
Barry Margolin <bar...@alum.mit.edu> wrote:
...

>I don't personally subscribe to it unless I have extra values, at least
>a combined 28 HCP. Ruffing isn't just for gaining tricks, it's also for
>stopping the defenders from running suits. With more combined strength,
>there's a better chance that every suit is double stopped and they can't
>do this.

I approve this message.

--
But the Bush apologists hope that you won't remember all that. And they
also have a theory, which I've been hearing more and more - namely,
that President Obama, though not yet in office or even elected, caused the
2008 slump. You see, people were worried in advance about his future
policies, and that's what caused the economy to tank. Seriously.

(Paul Krugman - Addicted to Bush)

[Dave Flower]

On Saturday, 20 April 2013 11:34:47 UTC+1, Bertil wrote:
> I seem to recall having read someplace that the Stayman 2C bid should not be used with a perfectly flat hand, 4-3-3-3. But I cannot find it stated in common texts about Stayman. Am I just dreaming? Stig

I think the answer is for someone to do a simulation. However, there seem to be a number of variables:

- Strength of the 1NT bid
- Strength of responder's bid
- Number of HCP in responder's 4-card suit
- Matchpoints or IMPs

Finally, unsuccessful use of Stayman helps the defence.

Dave Flower

[Nick France]

Sheinwold is far from obsolete and you could learn a lot from his
first chapter on how to evaluate your hand. He even points out things
you have to consider in deciding if all your points are worth full
value. Try going over that chapter at least once a month until you
finally understand it.

As for the example you gave of AJxxxxx AKx xx x. The evaluation you
give is after the hand has been raised in spades. This hand is a
monster after a spade raise. Not bidding 4S with this hand after
1S-2S is a gross error. So tell me, if partner raises you and you
have the hand in question what are you doing? If the answer is you
are bidding game then you are basically agreeing with Sheinwold.

Nick France

[Will in New Haven]

It's puzzling to me why he continues to yammer about a game he never
actually plays. But it amazes me even more that anyone continues to
talk to him. And _Five Weeks to Winning Bridge_ is in print, contrary
to Mr. "You have to have your facts straight" and has probably outsold
every book, not newspaper column, Truscott ever wrote by at least two
to one.

--
Will in New Haven

[jonathan23]

I meant the four-card major suit could be 5432, not the distribution of
the hand.

[Jennifer Murphy]

On Sun, 21 Apr 2013 08:53:06 -0700 (PDT), Will in New Haven
<bill....@taylorandfrancis.com> wrote:

...snip...

>> It is puzzling to me why you continue to post here, since pretty much
>> everyone is either "full of crap" or "stupid". Are you a masochist?
>
>It's puzzling to me why he continues to yammer about a game he never
>actually plays. But it amazes me even more that anyone continues to
>talk to him. And _Five Weeks to Winning Bridge_ is in print, contrary
>to Mr. "You have to have your facts straight" and has probably outsold
>every book, not newspaper column, Truscott ever wrote by at least two
>to one.

Yes, you are right, of course. I'm am shutting up now. ;-)

[Nick France]

We probably disagree on what beginners should be taught. One of my
problems with the Audrey Grant series was that she simplified too
much. Over 1NT she neither had Stayman nor Transfers. I was happy
when Seagram came out with her beginner book that had both of these.
I understand one has to draw a line somewhere when teaching
beginners. For me Stayman is taught to beginners and the concept of
not using it on a 4333 had is also taught. To me this just follows
from the reason for using Stayman.

In any case, I would agree that there is a limit to how many
'exceptions' you can teach to beginners. And yes beginners (well
actually any level for that matter) tend to see exceptions where they
dont' exist.

Nick France

[jonathan23]

Bertil wrote:


<SNIP>

>
> As usual, you are both full of crap. Sheinvold's book is not used by
> any teacher today,partly because it's out of print long ago. Show me
> the page where he mentions not bidding Stayman with 4-3-3-3.
>
> Much of his bidding is obsolete. E.g. he avaluates the hand
> :AJxxxxx-AKx-xx-x as 20 pts by counting 12 HCp + 5 for suit length +1
> for doubleton +2 for single. In short, he counts points for both
> length and shortness. Who today would evaluate this hand as 20 pts?

Wrong, wrong, WRONG!

This is on p8 of my edition of the book. Sheinwold is following
classic Goren point count. It does value shortness and length, but not
all at once. It starts with one and then adds the other _after_ a
trump fit is established.

For the opening bid, Sheinwold values this hand at 15: 12 HCP + 1 for
the doubleton diamond and 2 for the singleton club.

Then _after_ the 1S opening bid has been raised to 2S i.e. a trump fit
has been established he adds points for the long spades. Goren's
formula for this was to add 1 for the fifth spade (playing four-card
majors) and 2 for each additional card in the suit. If partner had
responded anything else he would not add for spade length.

Root in "Commonsense Bidding" does it the other way around: he adds for
long cards first, then for short ones _after_ a trump fit is
established.

You've completely missed the "point" of this anyway, which is: Every
book author wants the reader to look at this hand, open it one spade,
and then if partner raises to two spades, jump to four.

Goren aggressively valued distribution and capped his opener's re-raise
to three at 19 revalued points because he also allowed responders to
bid on shaded hands with about 3-4 HCP sometimes. So, he needed a
"safety valve" to avoid overbidding. If responder had a sound suitable
6 HCP he would go onto game over 3S and only pass with something shaded
or unsuitable. Goren had to make this hand revalue to 20 after a spade
raise so that 4S would be the "system bid".

Modern standard authors will jump to game after a single raise with a
revalued 18-19 since their responders virtually never have less than
5-6 HCP and the game bar has been lowered from 26 to 25. They can
afford to stay quiet over partner's opening one-bid with less since
they don't have to worry about monster opening one-bids that aren't
just right for a Goren strong two or 22-24 all-suits-stopped 2NT. Most
of those hands are now opened a 20-22 2NT or 2C (which isn't always
game forcing in Standard American methods).

>
> Also, he defines a 4-card suit as biddable if it has QJ or better.
> To respond to Stayman the suit must be biddable. He says nothing
> about the quality of responders 4-card suit.

This was in the context of old-fashioned four-card major standard,
which is obsolete and irrelevant to a discussion of what people should
do today. In those days, you didn't open 1NT with any doubleton weaker
than Qx, either, which almost nobody does anymore.

> He stresses the importance of having 2 quick tricks, which Grant
> never does.
>

That wasn't an original idea then, it's even older than Culbertson.
But it's not a universal one and never was. For other alternative
evaluation methods used by successful players that ignore quick tricks
completely see "The Four Horsemen's One-Over-One" by Burnstine (1932),
or "The Four Aces System of Contract Bridge" by Burnstine, Schenken,
Jacoby and Gottlieb (1935).

> Do you have any more stupid comments?
>

You've made enough for everyone.


--
- Jon Campbell
Ottawa CANADA

[jonathan23]

Nick France wrote:

> On Apr 20, 5:55 pm, jonathan23 <campb...@yahoo.ca> wrote:

<SNIP>

> >
> > Trying to teach beginners that there are exceptions is generally
> > wrong in my view.  While experienced players may recognize
> > exceptional circumstances when they see them, beginners generally
> > can't and teaching them "exceptions to the rules" will mostly just
> > confuse them and cause them to see exceptional situations where
> > they do not exist.
> >
>

> We probably disagree on what beginners should be taught. One of my
> problems with the Audrey Grant series was that she simplified too
> much. Over 1NT she neither had Stayman nor Transfers. I was happy
> when Seagram came out with her beginner book that had both of these.
> I understand one has to draw a line somewhere when teaching
> beginners. For me Stayman is taught to beginners and the concept of
> not using it on a 4333 had is also taught. To me this just follows
> from the reason for using Stayman.

I don't think we disagree much, I agree with you that Stayman and
transfers should be in on the ground floor (in areas where pretty much
everyone plays them). I just don't think that exceptions like the
decision whether to use Stayman or not with 4333 are all that simple
for beginners and can wait until the learners have assimilated the
basic methods well enough that they are applying them consistently (and
can also grok the problems that the exceptions are designed to handle).

--
- Jon Campbell
Ottawa CANADA

[vsp...@hotmail.com]

Think this is a key point. Weak 4-4 fits seen
to require more HCP for both the major suit and
notrump game to make. In other words with 25-6
HCP and a weak 4-4 fit, both games fail frequently.

[Jeff Miller]

On Monday, April 22, 2013 2:40:39 AM UTC+12, Dave Flower wrote:

> I think the answer is for someone to do a simulation.

I did some. They are reported in a brief article in the May 2005 issue of The Bridge World.

>However, there seem to be a number of variables:
> - Strength of the 1NT bid

In my simulations, opener had 13-15 HCP.

> - Strength of responder's bid

12-15

Every deal was played twice, once after the auction 1NT-3NT and a second time after a Stayman auction. For the latter, the sequences were constructed to mimic what would happen at the table (e.g., opener replied 2D, 2H or 2S, thereby giving away some info).

Following the bidding, the declarer play and defense were single-dummy (GIB).

> - Number of HCP in responder's 4-card suit

I didn't look at that, sorry.

> - Matchpoints or IMPs
I looked at it both ways. At matchpoints, the Stayman auctions scored 39% and 1NT-3NT scored 61%. At IMPs, the Stayman auctions lost an average of 2.2 IMPs per deal (not vul).

Simulations with other responder distributions showed that 3NT generally also outscores 4 of a major _at matchpoints_ when there is a 5-3 or 6-2 major-suit fit.

[Dave Flower]

On Monday, 22 April 2013 01:46:03 UTC+1, Jeff Miller wrote:
> On Monday, April 22, 2013 2:40:39 AM UTC+12, Dave Flower wrote: > I think the answer is for someone to do a simulation. I did some. They are reported in a brief article in the May 2005 issue of The Bridge World. >However, there seem to be a number of variables: > - Strength of the 1NT bid In my simulations, opener had 13-15 HCP. > - Strength of responder's bid 12-15 Every deal was played twice, once after the auction 1NT-3NT and a second time after a Stayman auction. For the latter, the sequences were constructed to mimic what would happen at the table (e.g., opener replied 2D, 2H or 2S, thereby giving away some info). Following the bidding, the declarer play and defense were single-dummy (GIB). > - Number of HCP in responder's 4-card suit I didn't look at that, sorry. > - Matchpoints or IMPs I looked at it both ways. At matchpoints, the Stayman auctions scored 39% and 1NT-3NT scored 61%. At IMPs, the Stayman auctions lost an average of 2.2 IMPs per deal (not vul). Simulations with other responder distributions showed that 3NT generally also outscores 4 of a major _at matchpoints_ when there is a 5-3 or 6-2 major-suit fit.

Thank you - I have looked up the article, and I think your contribution is worth more than any number of quotes from experts.

It is strong evidence in favour of bidding 3NT with a 4333 hand.

However, does anyone out there have any evidence of how results vary with the quality of the 4-card suit ?

Dave Flower

[Lorne]

On 20/04/2013 11:34, Bertil wrote:
> I seem to recall having read someplace that the Stayman 2C bid should not
> be used with a perfectly flat hand, 4-3-3-3. But I cannot find it stated
> in common texts about Stayman. Am I just dreaming?
>
> Stig
>

4-4 fits create an extra trick more than 50% of the time so in pairs you
should look for the fit. In teams it is more complex since you lose a
game swing when both contracts make exactly 9 tricks but gain only 1 IMP
for an extra trick so you should play 3N more often than in pairs.

However the formula changes when you have about 30 points between you -
now the odds are 11 tricks in either contract so a reasonable rule is
stayman if you point count is up to about 28 or 29 and 3N above than
unless you have good controls and might bid a 30/32 point slam with a fit.

[dake50]

AS Dave Flower, but my rule is with honors in TWO suits use Stayman.
If 3 suits have an honor, just blast 3NT.

[Nick France]

Is this estimate based on past simulations you have done? Thanks in
advance

Nick France

[bhmwe...@gmail.com]

Den lördagen den 20:e april 2013 kl. 12:34:47 UTC+2 skrev Bertil:
> I seem to recall having read someplace that the Stayman 2C bid should not
>
> be used with a perfectly flat hand, 4-3-3-3. But I cannot find it stated
>
> in common texts about Stayman. Am I just dreaming?

The reason not to use Stayman with 4333 is that it gives opponetns some info about opener's hand. And also the dummy (might help the lead). When you have 4333 you will seldom have ruffing potential even if you find a 4-card fit with opener, so it's unlikely that a major suit game will play better.

There is simple very little to gain and much to lose.

Of course, nothing prevents you from using Stayman with 4333, you can always try it out and see what happens.

I believe I read the advice about not asking in a book by Kelsey, BTW.

[Lorne]

Yes - here is a quick summary from 1,000 deals. South has 12-14 points
4S,3H,2-4 in each minor, North has 11-18 points 4333 shape so we
guarantee a 4-4 spade fit and a NT shape for South.

Total Tricks:
<=7 8 9 10 11 12+
S 44 176 323 289 131 37
NT 134 300 263 178 101 24

so spadecontracts clearly make more tricks but game is 10 tricks for
spades vs 9 for NT = 457 vs 566 so in teams the cost of game failing
more often outweighs the vaule of many overtricks and 3N is a clear
winner. In MP's the overtricks look like they swing the balance the
other way.

If I then look at the 160 deals where North is 16+
<=7 8 9 10 11 12+
S 0 1 12 65 57 25
NT 0 3 19 60 60 18

so the MP position is close to break even between 4S and 3N.

[Lorne]

On 22/04/2013 15:40, bhmwe...@gmail.com wrote:
>
>When you have 4333 you will seldom have ruffing potential even if you find a 4-card fit with opener, so it's unlikely that a major suit game will play better.
>

Do not forget the NT opener will have 4432 shape more often than 4333,
and also that even a 3-3 side suit fit may be exposed to 4 or 5 losers
in NT but only 3 in a suit contract.

I agree that giving info away is a problem however and you do that every
time you use stayman and play in NT.

[france...@googlemail.com]

On Monday, April 22, 2013 4:28:28 PM UTC+1, Lorne wrote:
> On 22/04/2013 15:40, bhmwe...@gmail.com wrote: > >When you have 4333 you will seldom have ruffing potential even if you find a 4-card fit with opener, so it's unlikely that a major suit game will play better. > Do not forget the NT opener will have 4432 shape more often than 4333, and also that even a 3-3 side suit fit may be exposed to 4 or 5 losers in NT but only 3 in a suit contract. I agree that giving info away is a problem however and you do that every time you use stayman and play in NT.

The problem with all of these simulations is that different partnerships have different rules on when they open 1NT.

Playing strong NT (just subtract a king somewhere for weak NT), think of the following hands:

Qx
KQxx
AQxxx
Kx

(surely everyone would open that 1NT)

KQxx
xx
AQxxx
Kx

(I'm fairly certain Gitelman & Moss would open that 1NT - they open all hands in range 1NT)

AKxx
K
KQxx
J10xx

bet you 1NT would be a popular choice in some circles

Overall, I'm agnostic on the benefits of always ignoring Stayman. I quite like the approach of (i) looking at your hand and considering its relative suitability for suit play and (ii) playing a method where you can find a fit and then offer partner a choice. I think this gets me to the right contract more often than insisting on either but the downside of information leakage is quite hard to measure.

[paul]

On Apr 20, 6:34 am, Bertil <stigfjor...@hotmail.com> wrote:
> I seem to recall having read someplace that the Stayman 2C bid should not
> be used with a perfectly flat hand, 4-3-3-3. But I cannot find it stated
> in common texts about Stayman. Am I just dreaming?
>
> Stig

I did some extensive analysis using a spreadsheet on this question.
There are several reasons why using Stayman may be wrong: the defense
may double the 2C bid for a lead; the defense may benefit from the
extra information; the 4-4 fit may break badly; and opener may be 4333
as well. The first three apply to all Stayman auctions, and do not
outweigh the expected benefit. The last is common enough to tilt the
balance against using Stayman. However, my analysis assumed classic
shape for the notrump bidder (no more than one doubleton, no five-card
major.) These days, it's common to open 1NT with a five-card major or
4252 shape, which makes it a closer decision. Your club holding may be
significant -- with KQJ, the risk of a lead-directing double is
minimal.

There is also the matchpoint vs. teams consideration: at teams, you
don't care if you are +400 instead of +420, but +400 beats -50 when
both contracts take exactly nine tricks. At matchpoints +420 vs. +400
is a huge benefit, and more likely (if a 4-4 fit exists) than both
contracts taking the same number of tricks.

Some British authors such as Reese and Crowhurst have pointed out that
if a fit exists and responder is 4333, opener is most likely 4432 and
the hand will usually play better in the suit. In some old Acol texts
they advocated a "Stayman in Doubt" gadget, where a 3D bid after
Stayman says "we have a fit, but I'm 4333." Aside from tying up an
important, natural rebid, this magnifies the problems of lead-
directing doubles and free information.

Bottom line: I do not think you should use Stayman on 4333 hands at
IMPs or rubber bridge. At matchpoints, I think it's a virtual toss-up,
but I tend not to use Stayman.

[paul]

Lorne, the matchpoint odds are much closer than double-dummy analysis
suggests: using Stayman gives the defense the chance to double 2C for
a lead, a negative inference when 2C is not doubled, and free
information about suit lengths. Your simulation assumes zero cost when
no fit is found, but the lead-direction and information applies every
time. There will be no fit around two-thirds of the time.

[Bertil]

That's the only smart comment you have made. You have made no positive or
constructive comment to the subject matter, just gibberish. Good riddance of
smelly rubbish. I'll not miss you.

Stig

Stig

[vsp...@hotmail.com]

On Monday, April 22, 2013 9:52:14 AM UTC-7, paul wrote:

Some British authors such as Reese and Crowhurst have pointed out that
if a fit exists and responder is 4333, opener is most likely 4432 and
the hand will usually play better in the suit. In some old Acol texts
they advocated a "Stayman in Doubt" gadget, where a 3D bid after
Stayman says "we have a fit, but I'm 4333." Aside from tying up an
important, natural rebid, this magnifies the problems of lead-
directing doubles and free information.

...............

In imps if both games make or both fail it
doesn't matter whether you use stayman or not.

So we are only interested on boards where 4M
makes, 3NT fails and 4M fails, 3NT makes.
4M needs at least two more tricks for 4M to
make and 3NT to fail.
3NT can make with 4M failing with both contracts
making exactly 9 tricks.

4333 // 4432. 25-28 HCP. Studies show 4M makes
about 1/2 a trick more than 3NT. But 4M needs
to make a full extra trick.

[derek]

On Monday, April 22, 2013 12:46:31 PM UTC-3, france...@googlemail.com wrote:
>
> Playing strong NT (just subtract a king somewhere for weak NT), think of the following hands:
>
> Qx
> KQxx
> AQxxx
> Kx
>
> (surely everyone would open that 1NT)

No. Most players I know (so mostly just club level) wouldn't consider opening that 1NT if the spades were Kx, on account of the two doubletons. I'm not sure I would with S Qx and another doubleton.

[Douglas Newlands]

So what's their fine rebid after 1D-1S?

doug

[vsp...@hotmail.com]

On Monday, April 22, 2013 8:46:31 AM UTC-7, france...@googlemail.com wrote:
Playing strong NT (just subtract a king somewhere for weak NT), think of the following hands:

Qx
KQxx
AQxxx
Kx

(surely everyone would open that 1NT)

..............

I would open 1NT with this hand. But
aren't many players playing a NT range
which would require them to open 1D.
Also whichever range you choose for your
hand, many pairs in the field will be playing
the other NT range.

++++++++++++++++++++++++++++++++

KQxx
xx
AQxxx
Kx

4=2=5=2. No rebid problem. 1D for
me on this pattern.

[Bruce Evans]

In article <7a3588e2-5ad8-4987...@b2g2000yqe.googlegroups.com>,

paul <paul...@infi.net> wrote:
>On Apr 20, 6:34 am, Bertil <stigfjor...@hotmail.com> wrote:
>> I seem to recall having read someplace that the Stayman 2C bid should not
>> be used with a perfectly flat hand, 4-3-3-3. But I cannot find it stated
>> in common texts about Stayman. Am I just dreaming?
>

>I did some extensive analysis using a spreadsheet on this question.
>There are several reasons why using Stayman may be wrong: the defense
>may double the 2C bid for a lead; the defense may benefit from the
>extra information; the 4-4 fit may break badly; and opener may be 4333
>as well. The first three apply to all Stayman auctions, and do not
>outweigh the expected benefit.

Not all to all, since they don't all apply to puppet Stayman, or 5-card
Stayman...

>The last is common enough to tilt the
>balance against using Stayman. However, my analysis assumed classic
>shape for the notrump bidder (no more than one doubleton, no five-card
>major.) These days, it's common to open 1NT with a five-card major or
>4252 shape, which makes it a closer decision. Your club holding may be
>significant -- with KQJ, the risk of a lead-directing double is
>minimal.

I think best is 5-card Stayman at the 2 or 3 level (2NT or 3C over 1NT =
Stayman). Then if a 5-4 major fit is found, play it; if a 5-3 major fit
is found, guess (opener will have doubleton); otherwise, don't look for
4-4 fits (if you do, opener's shape isn't fully disclosed, but the defense
gets to double almost all suits at the 3 level after 1NT-3C-3D-3M).

Simulations show mixed benefits for sequences that avoid 4-3-3-3 mirror
fits. These play very badly in 4M, but the sequences need to be very
complicated and/or reveal opener's exact shape, which are bad if opener
isn't 4-3-3-3. Perhaps these are best used at imps when you have 28-30
HCP and just want to find the coldest game.

>There is also the matchpoint vs. teams consideration: at teams, you
>don't care if you are +400 instead of +420, but +400 beats -50 when
>both contracts take exactly nine tricks. At matchpoints +420 vs. +400
>is a huge benefit, and more likely (if a 4-4 fit exists) than both
>contracts taking the same number of tricks.
>
>Some British authors such as Reese and Crowhurst have pointed out that
>if a fit exists and responder is 4333, opener is most likely 4432 and
>the hand will usually play better in the suit. In some old Acol texts
>they advocated a "Stayman in Doubt" gadget, where a 3D bid after
>Stayman says "we have a fit, but I'm 4333." Aside from tying up an
>important, natural rebid, this magnifies the problems of lead-
>directing doubles and free information.

Better if only responder shows 4333. Using 5-card Stayman when ordinary
Stayman is also available tends to indicate this shape for responder, and
could possibly guarantee it.

>Bottom line: I do not think you should use Stayman on 4333 hands at
>IMPs or rubber bridge. At matchpoints, I think it's a virtual toss-up,
>but I tend not to use Stayman.

This is a good rule unless you have a fancy system or highly developed
bidding judgement.

Bruce

[Adam Lea]

2H

[Barry Margolin]

In article <3YWdnSyD182nqOvM...@bt.com>,

You have pretty light requirements for a reverse. What if one of the red
queens were a jack?

--
Barry Margolin
Arlington, MA

[Adam Lea]

2D

[Nick France]

On Apr 23, 4:32 am, Barry Margolin <bar...@alum.mit.edu> wrote:
> In article <3YWdnSyD182nqOvMnZ2dnUVZ8midn...@bt.com>,

16 HCP seems ok for the reverse as the Qx is in partner's suit.
Switching the two doubletons makes the reverse questionable as you
have to give full value to Qx which cant be right.

Nick France

[Bertil]

On Sunday, April 21, 2013 11:25:09 AM UTC-4, Nick France wrote:
> On Apr 20, 9:10 pm, Bertil <stigfjor...@hotmail.com> wrote:
>
> > On Saturday, April 20, 2013 7:18:25 PM UTC-4, derek wrote:

>
> > > On Saturday, April 20, 2013 1:11:01 PM UTC-3, Nick France wrote:
>
> >
>
> > > > On Apr 20, 6:34 am, Bertil <stigfjor...@hotmail.com> wrote:
>
> >
>

> > > > One always wonders if your questions are for real or just to set
>
> >
>
> > > > something else.  Assuming for the moment its for real then
>
> >
>
> > > > Both 5 Weeks to Winning Bridge and Common Sense Bidding both
>
> >
>
> > > > explicitly state it.
>
> >
>

> > > Say it ain't so! How can Stig not have memorized those two books?

>
> >
>
> > As usual, you are both full of crap. Sheinvold's book is not used by any
>
> > teacher today,partly because it's out of print long ago. Show me the page where he mentions not bidding Stayman with 4-3-3-3.
>
> >
>
> > Much of his bidding is obsolete. E.g. he avaluates the hand :AJxxxxx-AKx-xx-x
>
> > as 20 pts by counting 12 HCp + 5 for suit length +1 for doubleton +2 for single.
>
> > In short, he counts points for both length and shortness.  Who today would
>
> > evaluate this hand as 20 pts?
>
> >
>

> > Also, he defines a 4-card suit as biddable if it has QJ or better.
>
> > To respond to Stayman the suit must be biddable. He says nothing about the
>
> > quality of responders 4-card suit.
>

> > He stresses the importance of having 2 quick tricks, which Grant never does.
>
> >
>

> > Do you have any more stupid comments?
>
> >
>

> > Stig
>
>
>
> Sheinwold is far from obsolete and you could learn a lot from his
>
> first chapter on how to evaluate your hand. He even points out things
>
> you have to consider in deciding if all your points are worth full
>
> value. Try going over that chapter at least once a month until you
>
> finally understand it.
>
>
>
> As for the example you gave of AJxxxxx AKx xx x. The evaluation you
>
> give is after the hand has been raised in spades. This hand is a
>
> monster after a spade raise. Not bidding 4S with this hand after
>
> 1S-2S is a gross error. So tell me, if partner raises you and you
>
> have the hand in question what are you doing? If the answer is you
>
> are bidding game then you are basically agreeing with Sheinwold.
>
>
>
> Nick France

Sheinwold seems to have been of two minds in 1962 when he co-wrote the K-S
book "How to play winning bridge".
They added 2 pts for a 7
-card suit and 1 for a singleton but nothing for
a doubleton. Thus the hand in question would count as 17 after a raise.
Later in 1982 Kaplan would evaluate it as 16.55 with his 4C. By my simplified count it would be 16.4.
You still have to document on what page Sheinwold discussed Stayman with 4333

You are entitled to your own opinion, however wrong it may be, but you are not
allowed to state false facts, or overlook contrary writings.

The Sheinwold book might have been reprinted in 2011 at a price of $30.
Barnes & Noble can not get it I was told. What teacher would use it today?

I'll probably have to wait till hell freezes over before you correct your
false claim.

Stig

[france...@googlemail.com]

On Monday, April 22, 2013 10:01:12 PM UTC+1, derek wrote:
> On Monday, April 22, 2013 12:46:31 PM UTC-3, france...@googlemail.com wrote: > > Playing strong NT (just subtract a king somewhere for weak NT), think of the following hands: > > Qx > KQxx > AQxxx > Kx > > (surely everyone would open that 1NT) No. Most players I know (so mostly just club level) wouldn't consider opening that 1NT if the spades were Kx, on account of the two doubletons. I'm not sure I would with S Qx and another doubleton.

OK, if you don't like to open 1NT with two doubletons, that's a perfectly valid decision (although perhaps you might look at the records of, say, the last world championship to get an idea what the top players do). But the agreement of what you open 1NT on affects whether or not you should ever use Stayman on a 4333 in response.

Never believe someone else's simulation results unless they match your system.

[Barry Margolin]

In article <80f9327d-6522-4131...@googlegroups.com>,

I think you need to be careful of recommending actions based on what
players do in world championships. They play a very different style from
more typical players -- aggressive, high pressure, and high variance (a
close match will often have lots of double-digit IMP swings in both
directions). Most people don't need to play like that, just as you don't
need to throw a 90 MPH pitch to do well in a neighborhood softball game.

[vsp...@hotmail.com]

On long knockout matches bridge needs to
stop playing total score. Leads to 'bad'
gambling bridge by the team which is behind.

Be like baseball and basketball. First team
to win 4 out of 7 wins. Play 16-20 board
segments. First team to win 4 segments wins
the match.

[Nick France]

You use opening hand evaluations and then compare them to an
evaluation of a hand after it is raised by partner. In particular you
do realize that the 4C is an estimate of opening hand value before any
other bids are made. Do you understant the difference?

And Sheinwold was not of 2 different minds. The book you quote is not
meant for beginners. It shows off the KS system and is meant for
those that have already learned standard american and wanted to
improve their bidding.

It really is long past time for you to realize their is more to bridge
than opening bridge and that any opening evaluation is done in a
vacuum and is modified as the bidding proceeds.

Nick France

[Jennifer Murphy]

I think, as my mother used to say, you are trying to make a silk purse
out of a sow's ear.

[Bertil]

Let's look at what it says in the KS book on p.68-69 about rebid after a raise
"Opening bidder will try for game over the raise if he has 17 pts including
distriburion ( two pts for a 7-card suit and 1 pt for a singleton) and will
insist on game with 20 pts. Thus there is no mentioning here or anyplace of
upgrading the opening hand.
Now apply this to the hand AJxxxxx-AKx-xx-x and consider responder having
just 5 pts. The most favorable 5 pts would be KQ in spades. Opener could then
count on 7 tricks in spades plus 2 in hearts. Where would the tenth trick
come from? In the short suits? What is the likelihood of responder having
KQ of spades? If his points are in the short suits, they may be totally
wasted. So why would opener bid 4 spades rather than 3 spades?

Stig

"

[Bertil]

Maybe you should ask Piglet for help. She seems to have a wealth of knowhow
about bidding.

Stig

[Nick France]

So now you are calling yourself names.

Nick France

[Will in New Haven]

KQ of Spades would look great to responder. However, opener is not
terribly worried about a Spade loser opposite Kxx or two Spade losers
opposte xxx. That is why the seven-bagger opposite a fit is so
valuable.

The Queen of Hearts would be worth much more than the Queen of Spades,
as would a doubleton Heart. Either minor-suit Ace would be at least as
valuable as the KQ of Spades, although it would not be sufficient for
ten tricks on its own.

You will not be able to refute this because it is accurate. So you
will resort to insult.

--
Will in New Haven

[Will in New Haven]

> Nick France- Hide quoted text -

Did you _read_ his analysis saying that KQ of Spades would be the most
valuable points responder could have when opener has AJxxxxx? Comedy
at its finest. Stiganalysis.

[derek]

On Tuesday, April 23, 2013 7:25:18 AM UTC-3, Adam Lea wrote:
> On 23/04/13 09:32, Barry Margolin wrote:
>
> > In article<3YWdnSyD182nqOvM...@bt.com>,
> > Adam Lea<lea...@btinternet.com> wrote:
> >
> >> On 22/04/13 22:27, Douglas Newlands wrote:
> >>> On 23/04/13 7:01 AM, derek wrote:
> >>>> On Monday, April 22, 2013 12:46:31 PM UTC-3, france...@googlemail.com
> >>>> wrote:
> >>>>> Playing strong NT (just subtract a king somewhere for weak NT), think
> >>>>> of the following hands:
> >>>>>
> >>>>> Qx
> >>>>> KQxx
> >>>>> AQxxx
> >>>>> Kx
> >>>>>
> >>>>> (surely everyone would open that 1NT)
> >>>>
> >>>> No. Most players I know (so mostly just club level) wouldn't consider
> >>>> opening that 1NT
> >>>> if the spades were Kx, on account of the two doubletons. I'm not sure
> >>>> I would with S Qx and another doubleton.
> >>>>
> >>> So what's their fine rebid after 1D-1S?
> >>>

> >> 2H
> >
> > You have pretty light requirements for a reverse. What if one of the red
> > queens were a jack?
> >
> 2D

I don't have a problem with the reverse to 2H - it's not great but it's also not terrible. I do have a problem with 2D - being of the opinion that you don't rebid 5-card suits if you have alternatives. If we're now talking about a 15pt hand I'll bid 1NT - feeling reasonably certain we're not getting to 3N without partner having a spade stopper. My usual agreement is that my 1N rebid is 12-15 HCP even though my 1N opening is 15-17.

[derek]

On Tuesday, April 23, 2013 3:09:20 PM UTC-3, Barry Margolin wrote:
>
> I think you need to be careful of recommending actions based on what
> players do in world championships. They play a very different style from
> more typical players -- aggressive, high pressure, and high variance (a
> close match will often have lots of double-digit IMP swings in both
> directions). Most people don't need to play like that, just as you don't
> need to throw a 90 MPH pitch to do well in a neighborhood softball game.

"Rounders", Barry!

[derek]

On Tuesday, April 23, 2013 12:54:51 PM UTC-3, france...@googlemail.com wrote:
> On Monday, April 22, 2013 10:01:12 PM UTC+1, derek wrote:
>
> > On Monday, April 22, 2013 12:46:31 PM UTC-3, france...@googlemail.com wrote:
> > > Playing strong NT (just subtract a king somewhere for weak NT), think of the following hands:
> > > Qx
> > > KQxx
> > > AQxxx
> > > Kx
> > > (surely everyone would open that 1NT)

> > No. Most players I know (so mostly just club level) wouldn't consider opening that 1NT if the spades were Kx, on account of the two doubletons. I'm not sure I would with S Qx and another doubleton.

> OK, if you don't like to open 1NT with two doubletons, that's a perfectly valid decision (although perhaps you might look at the records of, say, the last world championship to get an idea what the top players do).

I certainly don't have a problem opening 1NT with two doubletons, though as I say, I don't think I would with that one - switch the doubletons, and I would (which works with what Nick said about being OK with reversing with this hand, but not with the black suits reversed). But your comment about what the top players do is exactly why I mentioned that most of the people I play against who wouldn't open that hand 1NT are exclusively club players.

> But the agreement of what you open 1NT on affects whether or not you should ever use Stayman on a 4333 in response.

I'm sure.

> Never believe someone else's simulation results unless they match your system.

Fine advice.

[vsp...@hotmail.com]

> > Never believe someone else's simulation results unless they match your system.
>
>
>
> Fine advice.

Never believe someone else's simulation results unless they have shown that the appropriate side conditions have been placed on the test.

[Bertil]

I'm actually surprised that the KS authors did not point out that there is
about 2/3 chance that the partnership has at least 20 HPC together if opener has 12. Thus partner is likely to have at least 8 HCP and an aggressive opener would bid game based on that.
The Jogs formula has an uncertainty of at least 1 trick.

Stig

[france...@googlemail.com]

"I'm actually surprised that the KS authors did not point out that there is about 2/3 chance that the partnership has at least 20 HPC together if opener has 12. Thus partner is likely to have at least 8 HCP and an aggressive opener would bid game based on that. The Jogs formula has an uncertainty of at least 1 trick. Stig "

This is warped logic.
Before you know anything about partner's hand, it's true that if you have 12, then partner is 67% to have at least 8. However, on the hand being discussed, partner has made a single raise. For the sake of argument, let's say that shows 5-9 high card points (although some people would play it as weaker than that). Partner is under 50% to have 8 or 9.

The reason you would bid game looking at AJxxxxx AKx xx x opposite a single raise is that you know it is odds on to make opposite the right 4 count (a doubleton heart and either minor suit ace) but not opposite indeed the wrong 11 count (KQx xxx QJx QJxx on a heart lead) and that partner won't necessarily know what a good hand is opposite a game try. For example, he might think that trump honours are good (when in fact they aren't necessarily); a short suit try in clubs might discourage him with Q10x Qxx xxx KJxx (when game is on a finesse) and a long suit try in hearts will discourage him with xxx xxxx KQx Kxx (when game is playable)

[Will in New Haven]

Note, not a hint of acknowledgment that KQ of trump is _not_ the best
five HCP that responder can have, not even close. Oh, and he thinks 3S
on the auction is a game try. Well, it can be but how many use it that
way?

[vsp...@hotmail.com]

An equation for partnership which uses both combined
trump length and combined HCP has a std dev of
1 to 1.3 tricks/brd for estimated tricks.
An equation for partnership which uses only combined
HCP ignoring trump length has a std dev of 1.3 to 1.7
tricks/brd.
Any evaluation method which uses only 13 cards from
one hand has a std dev of over 3 tricks/brd.

[Bertil]

On Friday, April 26, 2013 10:23:10 AM UTC-4, france...@googlemail.com wrote:
> "I'm actually surprised that the KS authors did not point out that there is about 2/3 chance that the partnership has at least 20 HPC together if opener has 12. Thus partner is likely to have at least 8 HCP and an aggressive opener would bid game based on that. The Jogs formula has an uncertainty of at least 1 trick. Stig "
>
>
>
> This is warped logic.
>
> Before you know anything about partner's hand, it's true that if you have 12, then partner is 67% to have at least 8. However, on the hand being discussed, partner has made a single raise. For the sake of argument, let's say that shows 5-9 high card points (although some people would play it as weaker than that). Partner is under 50% to have 8 or 9.
>
>

Where did you get your stats? You seem to select a few specific hands out of
maybe thousands if not millions that responder could hold. Only a proper
simulation could resolve this situation.

Stig

[france...@googlemail.com]

Before you know anything about partner's hand, it's true that if you have 12, then partner is 67% to have at least 8. However, on the hand being discussed, partner has made a single raise. For the sake of argument, let's say that shows 5-9 high card points (although some people would play it as weaker than that). Partner is under 50% to have 8 or 9.

> > Where did you get your stats?

I calculated them. All the combined probabilities (if I have x HCP, what is the chance that partner has y) are not that hard to work out from first principles if you have, say, a spreadsheet and some basic knowledge of mathematics. If you have 12 HCP and partner has 5-9, he is about 47% to have 8 or 9.

>> You seem to select a few specific hands out of maybe thousands if not millions that responder could hold. Only a proper simulation could resolve this situation. Stig

A selected a few specific hands to demonstrate why making one particular game try instead of another is not always going to get you to the right contract.

A simulation might 'resolve' the situation for you, but make sure you know exactly what you think a single raise shows because different pairs have different styles. My simulation would tell you that in my partnerships it's percentage just to bid game but you may think a raise to 2S shows something else.

[france...@googlemail.com]

Oh, and he thinks 3S on the auction is a game try. Well, it can be but how many use it that way? -- Will in New Haven

Actually, I do. I was persuaded by my partner fairly recently to change from the (I admit more common) pre-emptive approach. His arguments were that (i) it's useful as a random game try with 6 trumps, (ii) why bid 3 if you might be able to play in 2, (iii) if they protect and you then compete to the 3-level you know more, not less, about the hand when you come to play it while RHO bidding a suit does not necessarily help LHO with his lead, (iv) it's extremely rare that they will correctly compete to the 4-level after having each passed on the first round so the downside is very low.

We play 1H-2H-3H as a game try as well, although he admits the logic for this is slightly shakier.

[Will in New Haven]

I don't think the preemptive re-raise is all that great but the game-
try reraise is just a lot less common. We play a relay game try that
says nothing about opener's hand and let responder describe but I
prefer to try for game by bidding it.

[David Babcock]

On Monday, April 22, 2013 12:44:19 PM UTC-4, paul wrote:
> In some old Acol texts
>
> they advocated a "Stayman in Doubt" gadget, where a 3D bid after
>
> Stayman says "we have a fit, but I'm 4333." Aside from tying up an
>
> important, natural rebid, this magnifies the problems of lead-
>
> directing doubles and free information.

An immediate 3D response saying "I'm 4-3-3-3 with a 4-card major, show me your major unless you are also 4-3-3-3" has fewer problems of that kind. Whether it's enough better to gain a full trick when it finds a 4-4 fit is another question.

DavidB