Join Bridge Winners
All comments by Craig Zastera
You are ignoring the author of this comment. Click to temporarily show the comment.
Quite astonished at the amount of support for 2NT here, particularly as this hand has a very good 5 card suit and an unstopped major.

In 2/1 auctions that start with a 1M openings, many (most?) play the style where the 2M rebid is the catch-all, i.e. not promising a 6 card suit. I know there are some modern “shift” rebid alternatives to this as well as older “Bergen” style of using 2NT as the catchall, but I think the 2M “catch-all” rebid is (still) quite mainstream.

Here, after 1-2 (GF), I would think the arguments in favor of (usually) rebidding 2 with *5* would be even stronger because (as constrasted with 1M-2x-2M) the original 1 opening has not promised 5 (or even 4 in most styles) s.

So, by (usually) rebidding 2 with 5, we not only benefit from saving space (the main argument for 2M rebid as “catch-all” in 1M-2x auctions), but we are also giving partner some genuinely new information (that we hold 5+ s).

After 2 rebid, there is plenty of room to discover a 4-4 majors suit fit and/or to explore stoppers for NT (and, hopefully, get 3NT played from the better side when it matters).
And once in awhile, the knowledge that opener has 5 s (especially when good ones as in OP), might be what responder needs to know to get to a superior (game or) slam.
Nov. 25, 2018
You are ignoring the author of this comment. Click to temporarily show the comment.
I'm not sure if your summary covers the options.

For example, suppose opener were 4=4=3=2 (perhaps also 4=4=4=1) with not unusual honor dispersion.

What to rebid for those cases after 1-2 ?
For me, those would be 2NT rebids.
Nov. 25, 2018
You are ignoring the author of this comment. Click to temporarily show the comment.
Don't understand your comment as 2 was already game forcing.
Nov. 25, 2018
You are ignoring the author of this comment. Click to temporarily show the comment.
You seem to be arguing that this is not a forcing pass situation.

I've noticed that “forcing pass” does not seem to be enjoying its former popularity of late, but I think positing that FP does not apply on *this* auction is a very deep position! We have an opening bid with a splinter raise opposite a GF 2/1 response and a voluntarily bid slam.
I think most would consider that “enough” for FP to be “on” at the 6 level!

I see no reason to presume from the auction that partner is VOID in s. I mean it is *posible*, but I would say definitely not probable, much less certain.

If we pass (6), the normal meaning of that call is that we have first round control.
Suppose partner has some hand like:
Ax-x-AJxxx-KQJxx
He might well think our “pass” justified his bidding 7
(perhaps he'd place us with x-Ax-KQxxx-Axxxx).
Nov. 25, 2018
You are ignoring the author of this comment. Click to temporarily show the comment.
I beg your pardon? “Awful hand to open” ??

I think not. I have quite high standards for my opening bids, yet consider this hand a clear opener.

Why not?
It has:
* 12 HCPs
* 2 quick tricks
* a *good* 5 card suit headed by AQJ
* no particularly weak minor honors (J(x), Q(x), stiff K)
* no rebid problems.
* satisfies “rule of 20” and “rule of 22”.
Nov. 24, 2018
You are ignoring the author of this comment. Click to temporarily show the comment.
I didn't say double invites a grand.
I said double denies 1st round control.
Pass would promise 1st round control.
Partner is captain of this hand.
I've opened s and made a splinter raise in support of s (for better or worse). Now I tell whether I've got 1st round control or not. Partner does whatever he wants.

I suppose if I had 1st round control in some *Great* hand, I could take it upon myself to bid a grand myself.
Nov. 24, 2018
You are ignoring the author of this comment. Click to temporarily show the comment.
The alternative to the obvious 2 would be 2NT.

While I'm not 100% opposed to rebidding 2NT with 5 s, I do not think this hand is “special” enough to choose that somewhat non-standard action.

If you had KJx in *both* majors or even KJx-QJx, then I think 2NT would have more appeal.

But here, you have good s and not even a real stopper in s, hence no reason to deviate from “normal” 2 rebid.
Nov. 24, 2018
You are ignoring the author of this comment. Click to temporarily show the comment.
If I double to deny first round control, that call in no way bars partner from bidding 7 or 7.

It just says that I do not have 1st round control.
Period. Not a penalty double.

In fact, having opened s and splintered in support of s, partner might be encouraged to learn I don't have the (possibly wasted) A.
Nov. 24, 2018
You are ignoring the author of this comment. Click to temporarily show the comment.
Rafael,
Yes, if we are just discussing the apriori (i.e. no special inferences available from bidding and/or play to change the odds), then the probabilities for *ANY TWO CARDS* whatsoever being in the same opponent's hand vs. one in each opponent's hand are the same. It doesn't matter whether the two cards of interest are in the same suit or different suits.
If in the same suit, it doesn't matter how many cards our side holds in that suit.

The calculation is simple (it can be done in other, more complicated ways, but the below is correct and easy):

1. Conceptually “place” one of the two cards
(say K in your example) in one opponent's hand–
let's say your LHO's hand.

2. Now, consider the relative liklihood of the *other*
card of interest (say the Q in your example) being
in LHO's hand (remember, we're assuming he already
has the K) vs. in your RHO's hand.

Since LHO has the K, he has only 12 empty “slots”
available which might receive the Q.

But RHO, with no cards already placed, has *13* empty
slots available which might receive the Q.

Therefore, the odds are 13 to 12 (or 52 to 48) that
RHO will have the other card (Q) rather than LHO
(who already, by assumption) has the K.

As you can see, the above analysis is very general and would apply equally to any two cards of interest.

BTW, this type of reasoning (known as “vacant spaces” analysis) can be generalized to handle situations where something is already known about the opponents' hands.

For example, suppose in your situation s were trump and your side has 9 of them (s).
Suppose you have already drawn trump and found that your LHO started with three s while your RHO had only one.

Now, what are the relative probabilities of the missing K and Q being (a) both with LHO vs. (b) both with RHO vs. © split between LHO and RHO ?

Well, because LHO started with three S, he has only *10* “vacant spaces” available for other cards (including the K and Q).
RHO, on the other hand, having started with only one has *12* “vacant spaces” remaining.

So now, we place the first card of interest, say the K.
The chances LHO has it are 10 out of 22 because there are 22 “vacant spaces” left total between LHO and RHO and LHO has only 10 of them.

Given LHO *has* the K, now he has only *9* vacant spaces left out of 21 remaining (RHO has the other 12 still).
So the probability that LHO also has the Q (given that he has the K) is only 9 out of 21.

Therefore, the overcall probability that LHO (with three s) has both the K and Q is
(10 / 22) * (9 / 21) = 90 / 462 ~ 19.5%

Doing a similar calculation for *RHO* who starts with 12 “vacant spaces” since he had only one , we compute the probability that *he* has both the K and Q as:
(12 / 22) * (11 / 21) = 132 / 462 ~ 28.6%

Notice how LHO's having more s than RHO has shifted the probability of either player having both honors from equal (24% each) to unequal (19.5% vs. 28.6%) with the opponent with fewer s becoming more likely to have both s.

To compute the probability (with s 1=3) of the honors being split, we could just subtract the sum of the above from 100% giving:
split honor probability =
1 - ((132 + 90) / 462) = (462 - 222)/462 = 240 / 462
=~ 51.9%

Notice this hasn't changed much from the apriori 52% when we have no knowledge of the distribution (it is not exactly the same, though).
Nov. 24, 2018
You are ignoring the author of this comment. Click to temporarily show the comment.
“Everything that is not compulsory is forbidden.” ??
Nov. 24, 2018
Craig Zastera edited this comment Nov. 24, 2018
You are ignoring the author of this comment. Click to temporarily show the comment.
I play Soloway jump shifts regularly.

My main comment (echoing one above) is that opener should NOT rebid 3 in reply to the SJS.

Soloway jump shifts show one of several very explicit hand types (the most important being “fit jump” types).

In order to allow responder the room he needs to clarify his hand type, opener should usually make the *cheapest rebid*, here 2NT. This rebid is basically a relay and not to be construed as saying much if anything about opener's hand.

After 1-2-2NT, the meanings of responder's rebids are:
3: 0/1 , 4-5 s, 5-6 s, strength enough to
have slam interest opposite a well fitting minimum

3: 5=2=4=2 shape, slam interest strength

3: 4-5 s, 0/1 , 5-6 s, slam interest strength

3: *solid* 6+ card suit with significant side extras
(two outside kings would be a dead minimum)

3NT: 16-17 HCPs, rounded suits stopped,
5-6 s w/ 2/3 top honors, no shortness.

4: solid 6+ card suit with usually *one* outside A/K.

4NT: like 3NT except 18-19 HCPs.

I don't think we've defined jumps to 4, 4, or 4
but it wouldn't be hard to think of something (maybe to show fit with VOID for 4 and 4).

Anyway, if a partnership wanted to allow some carefully considered exceptions to the idea that opener should make the cheapest rebid, perhaps allowing 1 step and 2 step skips (3 and 3 here) with very carefully and narrowly defined meanings might be acceptable–need to consider whether responder will still be able to show all relevent hand types after that. Probably OK.

In regards to the keycard question, 6 should ask opener to bid above 6 if he holds the K. Traditionally, this would be a grand slam, but Kit Woolsey has suggested that if replier bids just one step above the small slam in our suit (here 6NT), there might occasionally be a slight additional gain possible, particularly at matchpoints.
Nov. 24, 2018
You are ignoring the author of this comment. Click to temporarily show the comment.
I stand corrected.
If we assume opener has 16-17 HCPs with three strong s, no more than three hearts, and a weak doubleton somewhere, I estimate the odds for the location of the doubleton as:
: 59%
: 37%
: 4%
and that is without even considering that with 3=3 majors and a doubleton , if his s were strong enough he might have chosen to bid 3 instead of 3. This consideration further increases the chances of the doubleton being in s.

I should add that if he really has a weak doubleton , 3NT has almost no chance and 4 is quite poor also. That would argue that passing 3 or possibly bidding 4 would be the better choices.

When opener has 16-17 with 3 strong s (at least 2/3 top honors) and a weak doubleton ,
3NT is the best contract.

But when opener has 16-17 with 3 strong s and a weak doubleton ,
3 is the best (matchpoint) spot, with 4 not too far behind.
3NT is (almost) hopeless and 4 is only around 25% chance.

Assuming 16-17 with 3 strong s and a weak doubleton *somewhere* with relative frequencies as they occur naturally (ignoring the possiblity opener might have bid 3 with 3=3 majors and good s),
passing 3 turns out to be the percentage choice.
Under those assumptions, a simulation gave the following results:
(a) opener plays in s:
4 makes: 23%
3 makes: 63%

(b) opener plays in 3NT:
makes 32%

© opener plays in 4:
makes 54%
Nov. 24, 2018
Craig Zastera edited this comment Nov. 24, 2018
You are ignoring the author of this comment. Click to temporarily show the comment.
In general with no information about possibly relevent distributions in the other suits and no clues from bidding or defense up to critical point, the apriori odds are:
52% that the honors are split
24% that LHO holds both honors
24% that RHO holds both honors.

But in specific cases, the prior play and/or bidding is likely to have generated information that would affect the above odds.

For example, if you knew that LHO had 4 cards in the suit and RHO only 1, then obviously the odds that LHO has both honors has risen from the 24% quoted above to 60%.

But, perhaps, if LHO had both of these honors together with other values discovered during the play, he would have bid something that he did not. So, again, the odds would be changed.
Nov. 23, 2018
You are ignoring the author of this comment. Click to temporarily show the comment.
I play 4 as a splinter too, BUT I play it is a splinter in support of opener's first suit (here s) not his second suit.

I think this is a logical treatment. When I hear partner's 1 opening, if I have a good trick source (here, in s but could be any non- suit) together with 3 or even 4 card support for partner's s *and* a singleton, I *plan* an auction where I first show my good 5+ card suit and GF values, then plan to splinter next round to show my support and shortness.

Thus, after 1-2, if partner rebid 2 or 2NT, “everyone” would be down with a jump rebid of 4 or 4 as being delayed splinters in support of his s.

So I feel that if partner “crosses me up” by rebidding in a red suit, I should still be able to complete my plan of splintering in support of his original suit.

If I happen to have support for his second suit, I simply raise it to the 3 level, whether I have shortness in the unbid suit or not.

Sure, such a raise is not as precise as would be a splinter jump in support of his second suit, but I feel that having the “planned delayed splinter” in support of opener's 1st always available is more important than being able to show an “accidental splinter” when opener happens to rebid in an unexpected suit for which I happen to have support.
Nov. 23, 2018
You are ignoring the author of this comment. Click to temporarily show the comment.
Is there a reason you're giving partner a weak doubleton ?

Even if we assume his 3 is based on some maximum hand with three good s and a weak doubleton, then statistically speaking that doubleton is much more likely to be in s than in either red suit.
Nov. 23, 2018
You are ignoring the author of this comment. Click to temporarily show the comment.
Simulation suggests that OP's choice of auction is correct for the given conditions.

Opposite 15 HCP balanced hands with no 4+ card major, 3NT makes about 36% of the time.

If we increase (same shapes) openers to 16 HCPs, then 3NT make percentage goes up to 56%

I am assuming that, roughly speaking, opener will decline invites with 15 but accept with 16-17.
Nov. 23, 2018
Craig Zastera edited this comment Nov. 23, 2018
You are ignoring the author of this comment. Click to temporarily show the comment.
I picked “double” and was (pleasantly?) surprised to find that the big majority choice.

I actually consider this hand rather close (between DOUBLE and 2) given the 6th , poor () suit, and limited defense. If partner leaves the double in, he will surely lead a which may not be best.

The deciding factor for me in preferring “double” is that expecting opener to balance with a “double” whenever at all rational on this auction type is a major “hot button” for me, particularly in the all-too-frequent cases where partner does not honor this expectation.

I consider “straining” to re-open with double (as opposed to some other call) as often as possible to be an essential aspect of playing “negative” doubles by responder.

Of course, sufficiently “extreme” hands for opening bidder may justify choosing a different option. Maybe a “light” 6-5 hand with a void in overcaller's suit. Or a minimal hand with a 7 card suit lacking the ace, etc.

Minimal opener's with *length* in LHO's suit may, of course, judge to PASS if they believe that it is unlikely in the extreme that responder can have a “penalty double” type of hand.

In my view, this OP hand is not quite “extreme” enough for me to want to violate partner's expectation of a re-opening double.
But judging from what my partners tend to do, I would have guessed, though, that a 2 re-opening with this hand would have proven much more popular than it appears to be.
Nov. 23, 2018
Craig Zastera edited this comment Nov. 23, 2018
You are ignoring the author of this comment. Click to temporarily show the comment.
Henry,
Your whimsical example is different.

Besides being unlikely, the key difference is that you do not have anything resembling the hand you have represented.

Thus, it would not be unreasonable to bid 4NT now. Since that is an impossible bid, partner can only infer something like what has happened.

The OP problem is in no way similar as you have (as far as you know) bid all 13 of your cards correctly.
Nov. 23, 2018
You are ignoring the author of this comment. Click to temporarily show the comment.
Irrelevent.
It is a matter of partnership trust and discipline.

When partner asks for key-cards, he is in charge of placing the contract (except in some specific cases, e.g. a king ask guarantees all the keys are held, and replier is invited to bid a grand directly if he can based on that information).
Nov. 23, 2018
You are ignoring the author of this comment. Click to temporarily show the comment.
That is certainly one point in favor of passing.

But I think it is even simpler–partner has taken over the auction with a key-card ask and then, over your reply, said that he wants to play 6 when he could have done other things (e.g. 5N) if he wanted to investigate a grand.

Partner is captain here. He asked a question, you answered, and he set the final contract. You have not been invited to venture an alternative opinion.
Nov. 23, 2018
.

Bottom Home Top