Join Bridge Winners
All comments by Craig Zastera
1 2 3 4 ... 48 49 50 51 52 53 54 55 56 ... 66 67 68 69
You are ignoring the author of this comment. Click to temporarily show the comment.
I consider my pass here “not close.”
Sept. 11, 2017
You are ignoring the author of this comment. Click to temporarily show the comment.
Jordan,
After 2NT-3-3, responder with four hearts knows that no heart fit (8+ cards) can exist.
If that responder has four spades also (i.e. 4=4 majors), there is still the possibility of a spade fit, so he continues with 3 (the “spade length inquiry”). Opener replies 3 with exactly three spades, 3NT with exactly two spades, and makes a 4 level cue-bid with four spades.
Over that responder can continue according to his strength
He would also make the 3 “spade length inquiry” with 5=4 majors and now would want to play spades when opener showed three as well as four. If a spade fit is discover, responder with slam aspirations bids on (cue-bids, RKCB, even splinter jumps are available).

After 2N-3-3, if responder has four hearts but fewer than four spades, he knows that no major suit fit exsits.
Therefore, he bids 3–a relay to 3NT. Over opener's 3NT, if responder has slam interest, he can raise to 4NT or bid 4 of a minor (natural and forcing), e.g. with 2=4=5=2 or some such.
Sept. 10, 2017
You are ignoring the author of this comment. Click to temporarily show the comment.
Barry,
I'd be happy to run any simulation you think would be enlightening, but I'm not sure what it is you are asking for that is different from what I did in the simulation referenced above. That one used the actual OP hand from this discussion opposite randomly generated 15-17 HCP balanced hands containing exactly *4 spades*. Since I did not specify number of hearts, this simulation will include openers that are 4=4 in the majors too.
The simulation then compared 3NT vs. 4 (always in a 4=4 fit) and found 3NT to fare considerably better both at matchpoints and IMPs.
If you are looking for something different, please clarify and I will attempt to implement what you desire.

It is true that I omitted opening 1NT hands that contain *5 spades*. I did this for several reasons. Primarily, because partnership requirements as to when a 15-17 1NT opening with 5 spades is permitted vary widely. I could probably adjust the simulation to include any well-specified 5 spade 1NT openers.
In my case, I very rarely will open 1NT with 5 spades. Almost never. This is because playing 2/1, such hands are generally better described by starting with 1. If partner makes a 2/1, a jump to 3NT gives a good description.
If partner starts with 1NT (forcing) and preferences to 2 over opener's 2m rebid, then a 2NT continuation shows 16-17 balanced.
With a boring 5332 15, after 1-1N-2m-2, I would generally just pass.
So that leaves mostly only some 15 HCP hands with 5 *weak* spades (usually not better than JTxxx) as good candidates (in my view) for a 1NT opener with a 5 card spade suit.

But I know that others will open virtually any 5=332 hand with 15-17 with 1NT. No doubt including such a policy would shift the simulation somewhat towards invoking Stayman with the given responding hand, and how much could certainly be quantified with an appropriate simulation. My guess is that it would not be nearly enough to change the qualitative conclusion already reached–that this responding hand is better served by a direct jump to 3NT.
Sept. 10, 2017
You are ignoring the author of this comment. Click to temporarily show the comment.
Jordan,
You are correct 2N-3-3-3N shows 4=5 majors but is non-forcing. If responder has a stronger hand with 4=5 majors, he can bid 4NT (slam invitational, still not forcing) or 5NT (slam forcing).
Similarly, after 2N-3-3-3N (showing four spades), 3NT is not forcing, so if responder has a stronger hand with four spades he can bid 4NT or 5NT instead.

In a complete system, there should also be agreements for
2N-3-3-<4/4/4>.
I like to use 4 to show 5=5 majors, 4 to show 4=6 and 4 to show 6=4 (but an alternative for the 4m rebids would
be to show 4=5=1=3 and 4=5=3=1 slam tries so 4N would be a 4=5=2=2 slam invite).
If, after 2N-3-3, responder can determine that no major suit fit exists, but he now wants to show a 5+ card minor, he starts by bidding 3 (relay to 3NT), then pulls to 4m, natural and forcing.

In auctions that start 2N-3-3 (four hearts), if responder continues with 4 or 4, those are natural (5+ card suit) and forcing. As stated earlier bidding 3 over 3 is a relay to 3NT. If responder then pulls to 4m, 4, or 4 (Kickback), he is making a slam try in hearts.
Sept. 10, 2017
You are ignoring the author of this comment. Click to temporarily show the comment.
I thought I was quite clear (apparently not) that I have no desire to limit players' “judgement” in upgrading (or downgrading) hands to a value other than their objective 4-3-2-1 pointcount.
What I said was that players who do this frequently ought to be required to disclose this and not to simply state that their 1NT requirements are “15-17” (or any other range) when in fact they are not. Otherwise, they have what amounts to a private understanding not availble to the opponents.
Sept. 10, 2017
You are ignoring the author of this comment. Click to temporarily show the comment.
A related issue is when opponents open 1NT with a point count that is outside of the range stated on their card. I think this ought to be treated as at least as serious an infraction as opening 1NT with a singleton (but I think it is not).

I am in no way trying to limit an opponent's right to “upgrade” a hand (however unjustified I might think said upgrade is), but if they are prone to doing so, that ought to be clearly stated on their card. To me, if the card says “15-17” that should be what it means. If the partnership upgrades 14s regularly, then the card should say something like (14)15-17. And if questioned as to what their criteria is for upgrading a below range hand, they ought to be able to describe those requirements clearly.
Sept. 9, 2017
You are ignoring the author of this comment. Click to temporarily show the comment.
Michael,
Although I still think the North hand is too weak for any bid over opener's 2 rebid (which I do not believe implies any extra values–just 6+ spades), I do like your suggestion that *if* North were going to bid again over 2, then 2NT (natural, invitational) would perhaps be a better choice than 3 to cater to hands like KQxxxx-x-Axx-Axx opposite (where 3NT is good while 4 is not).
Sept. 9, 2017
You are ignoring the author of this comment. Click to temporarily show the comment.
Around the Seattle area where many people play this, it is referred to as “MSI” which might stand for “Major Suit Inquiry” or maybe “Modified Stayman Inquiry.”
I left out some details, e.g.:
* over a 3 (5 spades) reply, responder's 4 agrees
spades and shows slam interest

* over a 3 (4 hearts) reply, if responder has slam
interest in hearts, he starts with 3 (relay to 3NT),
then pulls (to 4m cue-bid or to 4 or to 4 Kickback)
(over 3, if responder continues with 4m directly, that
is natural and forcing, not interested in playing hearts)

* over a 3NT (5 hearts) reply, 4 is natural and forcing
while 4 shows diamonds (5+) and is forcing
(because 4 is used to agree hearts)

You may have noticed that in this scheme, opener always plays the hand in whatever major is selected.
Sept. 9, 2017
Craig Zastera edited this comment Sept. 9, 2017
You are ignoring the author of this comment. Click to temporarily show the comment.
Well, the numbers are what they are.
Remember that 15 HCP hands constitute about 43% of all 15-17 HCP balanced hands opposite a 9 count (and 17 HCP hands are about 23%), so it doesn't seem that surprising that the average goes up only 5% or so.
Also, the first simulation allowed *all* balanced shapes (I was investigating whether this responding hand should bid game or just invite), whereas this latter simulation only included opposite hands with four spades (i.e. the 3NT make percentage for 15 HCP hands with four spades might be different than for all 15 HCP balanced hands).
Sept. 9, 2017
You are ignoring the author of this comment. Click to temporarily show the comment.
Here are the results of a 5000 deal simulation which attempts to answer the question of whether responder should Stayman with this hand.
The opposite hand was constrained to be 15-17 HCP balanced hand with exactly four spades.

Double Dummy Results:
3NT makes: 3121 (62.42%)
4S makes: 2217 (44.34%)

Matchpoint analysis:
3NT beats 4S on 2522 deals
4S beats 3N on 1537 deals
3NT ties 4S on 941 deals
matchpoint conclusion:
don't Stayman as 3NT beats 4S with 59.85% (BAM score)
and that's not even considering the extra info
that Stayman gives to the opponents for nothing
when no spade fit exists.

IMPs analysis:
3NT beat 4S by 8738 IMPs NV and 10,921 IMPs VUL
IMPs conclusion:
Don't Stayman at IMPs either.
Sept. 9, 2017
You are ignoring the author of this comment. Click to temporarily show the comment.
This is a somewhat less than maximally vexing example of a common problem type.
To illustrate, there are *2* problems in this month's Bridge World Master Solvers' Club that are similar (but harder):
Problem “C” : IMPs, VUL vs not
you hold: QT-KT754-AK-A763
(2S)- ??

Problem “F” : IMPs, VUL vs not
you hold: K96-KJ842-AKT4-Q
(2S)-P-(3S)-???

I don't have experience with the super-light openers you describe, but playing “normal” opening requirements, your problem seems like an only slightly uncomfortable 2NT overcall to me. But perhaps the probabilities shift when partner opens “many 10s and most 11s”.

Still, VUL at IMPs means to me “don't miss any games”, so I think passing this 16 count is too conservative.
Sept. 9, 2017
You are ignoring the author of this comment. Click to temporarily show the comment.
I just completed a 5000 deal simulation where opener is constrained to have exactly 15 HCPs (and fewer than 5 spades).
These are the hands with which he would presumably decline a NT game invite.

I found that (double dummy), 3NT made on 57.48% of the deals.

Single dummy, 3NT contracts typically give declarer a small
advantage vs. double dummy, so the real life “make” percentage for 3NT opposite 15 HCP balanced hands is probably more like 60%.

So clearly, this responding hand should commit to
game.

Determining whether or not responder should bother to check for a spade fit would require a different simulation, and also, some more info about under what conditions (if any)
opener is allowed to hold *5* spades in the partnership methods.
Sept. 8, 2017
You are ignoring the author of this comment. Click to temporarily show the comment.
Although I answered that I play Smolen (2 level) only over 1NT, I actually have a structure over 2NT that gives the benefits of both Puppet Stayman (can check for opener's 5 card major) *and* Smolen (responder can show 5=4 and 4=5 majors). It works as follows:
2N-3:
3 = 5 spades
3N = 5 hearts (then responder's 4D transfers to Hs)
3 = 4 hearts (may have 4 spades too)
then responder's 3 relays to 3NT while
responder's 3NT (and 4NT etc.) show 4 spades

3 = < 5 spades and < 4 hearts
then: responder's 3 asks opener's spade length
replies: 3 = 3; 3N = 2; 4 any = 4
responder's 3 relays to 3NT

If responder is 5=4 or 4=5 in the majors, he finds the fit
immediately if opener replies 3 (5 spades), 3NT (5 hearts)
or 3 (4 hearts).
If opener replies 3, then:
with 5=4: responder bids 3H to see if opener has 3 spades
with 4=5:
responder bids 3NT (or 4NT, etc. if strong enough)
which shows 4=5 by definition.
Sept. 8, 2017
You are ignoring the author of this comment. Click to temporarily show the comment.
I think more info may be needed to answer this question.
I think what kind of IMP event it is may be relevent.
In a short match (e.g. Swiss), the overtrick IMP may not matter at all (because the Victory Point result may be unaffected by an extra IMP). Thus, in a Swiss, the argument is stronger for playing to make. In a long match, the “expected value” argument perhaps has more validity.
Sept. 8, 2017
You are ignoring the author of this comment. Click to temporarily show the comment.
I like to play that 3 is weak here too (so not appropriate with this hand), but that agreement requires playing either
(a) an original 3 response is game invitational
or
(b) a 2 response followed by a 3 rebid is not GF

otherwise, responder has no way to show all 3 possible
ranges of club single suiters (weak, invitational, GF).
Sept. 8, 2017
You are ignoring the author of this comment. Click to temporarily show the comment.
I actually would have bid 3 (natural, invitational) initially. But presumably your methods don't include invitational 3 level jump shifts, so you have to bid strength ambiguous 3 now.
Sept. 7, 2017
You are ignoring the author of this comment. Click to temporarily show the comment.
Seems to me that I should be thrilled here that RHO's double has afforded me the opportunity to PASS rather than being forced to choose a call with a hand that really has nothing much more to say. If forced to bid (i.e. if RHO had passed), I would probably choose 2 (which should be agreed as the default with nothing useful to add), not a silly 3 on 7xx.
But with RHO's actual double, PASS seems perfect as it leaves maximum room for partner to show his intentions while simultaneously not sending any additional message which would be a distortion.
With the option of PASS available, 2 now should suggest 6, or at least a very strong 5 card suit (4/5 top honors if only 5).
Sept. 7, 2017
You are ignoring the author of this comment. Click to temporarily show the comment.
The answer to this question might depend on whether or not you play BART.
BART allows the generation of many extra sequences that allow responder to make finer strength (and shape) differentiations than he could without BART.
For example,
1-1N-2-2 vs. 1-1N-2-2(BART)-2-2, where one sequence (usually the second but could be reversed) is stronger than the other, e.g. 5-7 vs. 8-10.

So, BART aficionados may like to have opportunities to use their plaything occur as often as possible. This can be achieved by allowing opener to rebid 2 after 1-1NT
with 5=3=3=2 shape as well as 5=3=2=3 (and, of course, other shapes with 4+ clubs).

Is the slight loss of definition about opener's club length worth the gain in more frequent opportunities to use BART?
I certainly don't know.
Personally, I have not found the theoretical benefits of BART to occur often enough in practice to even play the convention in most partnerships (after all, BART does have, at least, the downside of making it impossible to play in 2).
Sept. 6, 2017
You are ignoring the author of this comment. Click to temporarily show the comment.
Steve,
While I agree with you that playing “forcing” allows responder to start with 1NT on some (balanced) hands with fully GF values (probably should be some upper limit, though), thereby adding (some) more definition to the GF 2/1 responses, I omitted listing this as a “virtue” of the forcing approach because some may find this usage unappealing. It does have the drawback of further expanding the already wide range of the 1NT response and may leave responder awkwardly placed to find a suitable (forcing) continuation.
So I confined my arguments to extolling the advantages of 1NT forcing even when the partnership opts to limit that response to a maximum of 12 HCPs.

Personally, I allow foricng 1NT on up to soft 13 or even 14
HCPs that will be content to bid 3NT or 4M next time.
But with a goodish, even 4 card, minor suit and “hard” values, I think a 2/1 (in the 4 card suit) is preferable as there might be a slam, perhaps in a 4=4 minor fit which will not be biddable otherwise.

An alternative approach with GF balanced hands after partner opens 1M is to play *2NT* response as natural and GF (13-15 or 18-19). But that gives up on using 2NT as a GF 4+ card raise, a problem for me since I use responder's 3 level jump shifts as “natural, invitational” (hence, e.g. 1S-3C not available as an alternative GF raise).
Sept. 5, 2017
You are ignoring the author of this comment. Click to temporarily show the comment.
There is nothing wrong with fit-jumps that force to game. In fact, I might even claim that those are the most common kind, e.g. 1-(2)-4 is a GF fit jump.
Sept. 5, 2017
1 2 3 4 ... 48 49 50 51 52 53 54 55 56 ... 66 67 68 69
.

Bottom Home Top