[1eyedjack]

I have put this in the non-natural system discussion forum because it discusses a scheme of non-natural responses to 2N (uncontested). However the 2N opener itself is natural and if mods would prefer to move the thread to natural system discussion I do not have any objection.

There follows an early draft of a set of responses that I have been working on. I shall either bask in your potential congratulations or, more likely, crawl back under my rock when you blast it to pieces. Actually I doubt that there is much new in what follows. Although I have been working it out for myself I expect all y'all have dabbled in something like this, rejected it and moved on.

First of all I want to make a general observation about most of the schemes of responses that I see in play:
The 3C response is generally used to investigate a major suit fit, having regard to the possibility that opener may have a 5 card major. In the process the continuations are generally geared to ensuring that the contract is declared by opener.
A couple of consequences of this are:
1) A lot of information is given to the defence in the bidding, usually to no good purpose. It is rare for responder to have no interest in opener's possible 5 card majors. A 5-3 fit in that major is inherently unlikely. So on a high proportion of hands you start off with 3C and end in 3N having disclosed your shape along the way.
2) The 3S response to 2N is usually some sort of minor suit slam try. But it gets a bit unwieldy if responder has only one minor, and furthermore it is not uncommon (especially when in possession of only one minor) for responder still to have some interest in opener's 5 card major, so being left with an uncomfortable choice between 3C and 3S on the first response.

In the sequences that follow,
(1) I give up on looking for opener's 5 card major unless responder has a slam try, and while I try to get the contract declared by opener when there is an obvious route I do not obsess about it. Before you groan and stop reading further, there are some compensating benefits. It is a balancing act, and I ask you to get to the end before you form any conclusions.
(2) The 3C response is the first move on minor suit slams as well as looking for a major suit fit with or without a slam try
(3) The 3C responder promises either a slam try (any shape, especially if balanced), or, lacking a slam try, must have at least one 4 card major. This allows for opener's rebid to bypass 3N when holding both majors himself.
(4) If opener has opened 2N on skewed distribution he approximates it to the closest balanced alternative, the exception being that with 5-4 or 4-5 in the majors he gets to show precise shape.
(5) Some of the high-level continuations and further developments are not stated here. Exercise for the student etc.
Thus:

In all that follows, suit length depicted as "?" = 2 or 3 cards

2N-3C-4H = 5-4-2-2
2N-3C-4D = 4-5-2-2
2N-3C-4C = 4-4-?-? (then 4D/4H = transfers)
2N-3C-3N = ?-?-4-4
2N-3C-3S = 4-?-?-4
2N-3C-3H = ?-4-?-4 OR ?-?-?-4+
2N-3C-3D = Fewer than 4 Clubs and at most 1 x 4 card major

2N-3C-3H (?-4-?-4 OR ?-?-?-4+)
==========================
2N-3C-3H-3S = Either 4 Hearts or any slam try
2N-3C-3H-3S-4C+ = ?-4-?-4
2N-3C-3H-3S-3N = ?-?-?-4+

2N-3C-3D (Fewer than 4 Clubs and at most 1 x 4 card major)
================================================
2N-3C-3D-3N = 4 Spades, fewer than 4 Hearts, no slam interest. Opener passes or corrects to 4S with 4 Spade support.
2N-3C-3D-3S = 4 Hearts, fewer than 4 Spades, no slam interest. Opener converts to 3N or 4H depending on Heart support or lack.
2N-3C-3D-3H = both majors or slam try.

2N-3C-3D-3H (both majors or slam try)
============================
2N-3C-3D-3H-4D = 4-?-4-?
2N-3C-3D-3H-4C = ?-4-4-?
2N-3C-3D-3H-3N = ?-?-4+-?
2N-3C-3D-3H-3S = 4+-?-?-? or ?-4+-?-? (then 4H by responder is to play or convert to 4S, showing both majors with no slam interest)

Other matters
==========
The classical problem hand in basic systems of responses is when responder has 5 Spades and 4 Hearts, especially lacking slam try. Assuming that you bolt on red suit transfers to the rest of the system, then you have to use an immediate 3S response to show this. Kludgy, I know. Uses up a valuable low level response for an infrequent hand, and wrong-sides the contract much of the time that it happens. If responder has a slam try with this shape may well go through 3C anyway. Maybe suit quality would be a decider.

Over to the firing squad.

[Fluffy]

puppet stayman sucks, and this is an improvement IMO, but same as improving cappeletti, all you need is 5 minutes with a pencil+paper to improve it, so improving means nothing. You will have some random success with minor slams with your imply/deny club structure.

You need to work on continuations on what bids show what, for example after 2N-3C-3D-3H-4C = ?-4-4-? you need to show diamond slam try probably with 4♦, and heart slam try probably with 4♠, but what is 4NT? could be natural or club slam try I guess.


Also just a typo correction, hands where you show 4-4 can actually be 4-5 obviously. And for you it seems that 5-4 is also possible.

[1eyedjack]

Fluffy, on 2012-March-17, 02:22, said:

Also just a typo correction, hands where you show 4-4 can actually be 4-5 obviously. And for you it seems that 5-4 is also possible.

Yes indeed I may well open any 5422 in range with 2N. Others may think it a bad idea. I like to keep the option open even if not compulsory. The only time that the 5422 shape is expressly distinguished from 4-4 hands is when opener has both majors.

With most 5422 hands you would approximate to the corresponding 4-4. But I did say in the OP that skewed distributions are approximated to the nearest balanced equivalent, so I would not call it a typo.

And yes I also said that there was scope for further development of the high level sequences.
For example there are some interesting possibilities after
2N-3C-3D-3H-3S.
Opener here has promised a major, and there must be a fit in that major if responder has anything short of a slam try. So how would you play 3N now by responder?
It may not make much sense to have it as a non-forcing bid. It is not possible to have mirror distribution,. And then there is also 4m to define.

Thanks for the qualified support. Could have been more painful

[1eyedjack]

On reflection it may be better to switch a couple of bids around, thus:

2N-3C-3D-3H (both majors or slam try)
============================
2N-3C-3D-3H-4D = ?-4+-?-?
2N-3C-3D-3H-4C = 4+-?-?-? (then 4H = tfr)
2N-3C-3D-3H-3N = ?-?-4+-?
2N-3C-3D-3H-3S = 4-?-4-? or ?-4-4-? (then 3N asks for major)

The main reason for this suggested change is that responder with a Diamond fit and slam try has more room to express the fit.

It probably is not really important.

The suggested order of 4C and 4D above is a bit non-intuitive and is purely to ensure that Spades are played the right way up. That also is probably not important and the bids can be swapped. Give up on transfer to Spades and use 4H as slam try in Spades then.

[1eyedjack]

Have been rethinking (again) the continuations after

2N-3C-3D (<4 Clubs and at most one Major)
.....................-3H (both majors or slam try)

We are rather spoilt for choice, with something of a luxury of bidding space available to dinstingulsh a relatively small population of hand types; hence the opportunity for several schemes of reasonable followups.

The methods described earlier in this thread contained one common flaw, in that there was a respectable chance of playing in Spades the wrong way up (ie by responder) when there is a 4-4 Spade fit. There are ways round that in some cases by way of followup transfers, but that starts to make it complicated and simplicity has its virtues. Also if a bid is required as a transfer then it rather limits its potential for use as a slam try.

Two possible alternatives are as below (there may be yet others that are reasonable).
As before, "?" denotes 2 or 3 cards. "4" denotes 4 or 5 cards

Method A
--------------
2N-3C-3D-3H-3S = 4-?-?-? or 4-?-4-?
2N-3C-3D-3H-3N = ?-?-4-? (nonforcing)
2N-3C-3D-3H-4C = ?-4-4-? (then 4D agrees Diamonds?)
2N-3C-3D-3H-4D = ?-4-?-?

You might also be able to make use of 4H as a bid by opener, as
2N-2C-3D-3H-4D-4H can only be to play.
I suggest using it to show also ?-4-?-? as with the 4D bid but using the two bids to distinguish slam co-operation from a hand that is disinterested in context. The reason being that without this treatment that is the one sequence in which responder cannot otherwise promise slam interest without committing beyond 4M. Or you could use the two bids to distinguish between a 4 and 5 card heart suit (but I prefer the idea of co-operation v brakes).

Again, there is little point in having 3N rebid by responder over 3S as a non-forcing bid, unless it is to express a non-forcing misfitting slam try. On anything less than a slam try there will be a 4-4 Spade fit and no duplication of shape (responder also having 4 Hearts which opener has denied). If you make it forcing then you could use it as a relay to query whether opener has 4D (4C confirms, 4D denies) which will allow you to set Diamonds as trumps by bidding 4D over 4C. You could then use 4C by responder over 4S to show a mistfitting slam try. Maybe you can still find a Club slam when responder has 5?

The above is still not entirely devoid of kludginess

Method B
------------
2N-3C-3D-3H-3S = 4-?-4-? or 4-?-?-? or ?-?-4-? (ie any possible consistent combination that does not have 4 Hearts)
2N-3C-3D-3H-3N = ?-4-4-? (then 4C agrees Hearts, 4D agrees Diamonds - slamming, 4H being to play)
2N-3C-3D-3H-4C = ?-4-?-? (then 4D is slamming agreeing H, and 4H is to play)

Then
2N-3C-3D-3H-3S-3N = 4-4 in majors, no slam try. Opener usually converts to 4S with a fit
2N-3C-3D-3H-3S-4C = slam try relay
2N-3C-3D-3H-3S-4C-4D = 4-?-4-? (then 4H agrees Diamonds, 4S agrees Spades non-forcing but with slam interest)
2N-3C-3D-3H-3S-4C-4H = 4-?-?-? (then 4S agrees Spades, non-forcing but with slam interest)
2N-3C-3D-3H-3S-4C-4S+ = ?-?-4-?

There might be a case for swapping round the order of 4D/4H/4S a bit over the 4C relay.

Method B is more complicated to remember than Method A.
I think we can ignore (as a disadvantage of method B) the risk of opponents doubling the 3S bid. Doubler is likely to be on lead, and opener may have 4 Spades.
Method B may have some technical superiority but at present I am leaning in favour of the simpler method A.

[awm]

It seems like:

(1) You are giving the opponents a lot of information on game-only hands that could help them on lead/defense.
(2) You lack a way to find opener's five-card suit (major or minor) which is often the key to a good slam contract.
(3) You'll wrong-side some contracts, including (it seems) some 4M contracts on game-only hands.
(4) You seem to have some problems with the "Smolen" hand types in that none of them can be put through 3♣.
(5) It may become difficult to bid hands where responder has a single 5+ minor and slam interest, because 3♣ has some slam-killing rebids and 3♠ is something else.

I'd guess that for slam bidding, hands where responder has some shape are a lot more important than responder's flat hands. When responder is balanced you can more or less determine slam prospects on points, and you have a great degree of safety in 4NT (or even 5NT) if you have to explore at that level. When responder has some distribution, you can often make slam on a lot less values but at the same time you need to stop lower if there's a misfit on those hands. It seems like you are sacrificing some of responder's ability to describe his hand (i.e. taking away direct 3♠, having opener rebid things like 3NT and 4m sometimes over 3♣) in exchange for getting opener's 4-card minor at a low-level. I don't find this an especially good trade.

[lalldonn]

awm, on 2012-March-31, 11:33, said:

(1) You are giving the opponents a lot of information on game-only hands that could help them on lead/defense.

THIS! I wish my opponents would tell me their minor suit shapes every time they look for a major suit fit on the way to 3NT.

[1eyedjack]

Thanks for the feedback.

Re. Item 1 - I have some difficulty working out the trade-off on giving away info, but at present I think that you overstate the problem there. Lacking slam values, responder would only bid 3C when in possession of a 4 card major. Contrasted with the puppeteers, a lot of auctions will go 1N-3N where others hunt for a 5 card major and, lacking said magic 5-3 fit (which is the majority of the time) settle in 3N. They will have given information to opponents but on different hands.

Re. Item 2 - yes there is a degree of validity to this, but again I think that this may be overstated. The argument is strongest at the game level. There remains potential to find opener's 5 card suit if slam is on the cards. I appreciate that I have not really developed the higher level continuations in this thread. I have not come across other methods that identify opener's 5 card minor.

Re. Item 3. Everyone wrong-sides some contracts. In the slam zone all systems are more prone to this than without slam tries. There is only one occasion that I am aware of in which a contract is wrong-sided where responder lacks a slam try (in these methods), and that is when responder has 4-4 in the majors and opener has Hearts but not Spades and not Clubs (may or may not have Diamonds, which distinction, incidentally, remains concealed). And of course if Clubs is the place to be (presumably in the slam zone), 3C response always places that in responder's hand straight away. Whether that is a price worth paying is something that can only considered in the context of the whole method in the round. Rightly or wrongly I don't place quite the same importance on placing declarer. Certainly there are hands where it makes a difference, but there are a lot where it does not.

Re. Item 4. Agreed. That does not make them unbiddable hands, but it may take away more useful applications for other responses than 3C. I did acknowledge this in the OP. I have not really considered or discussed the other responses in this thread.

Re. Item 5. Have to think about this. Sort of get what you are getting at but not convinced it is insurmountable (or of its importance if it is so). I don't really get your point about the responses being slam-killer. The whole point of it was to improve slam bidding, so if that is valid then it could be a system killer.

I am toying with including some shapely slam hands in the 3S response which also includes 5-4-?-? without slam interest. Opener rebids 3N with 2-3-x-x, implying some support for a minor suit slam try. The 4 level rebids need some work, and I am resigned to wrongsiding another type in the major suit game zone, but it is looking promising, and may address concerns over items 4 and 5 in one go.


Anyway, I may try these out for a bit in situations where it doesnt matter, or in pship bidding room. Will get a better feel that way I think than theorising.

[1eyedjack]

Here is a hand I spectated on BBO a few days ago

---

Everybody and his dog was ending up in 3NT on this hand, via some sort of Stayman, puppet Stayman or similar enquiry by South in response to North's 2N opener. The Club fit never comes to light.

It is rather against the odds, but if the heart finesse fails, Clubs break 3-1, diamonds break 5-3 and you sustain a Diamond lead at trick 1, then 3NT is actually at risk where 5C makes. Just an observation that I noted a long time after the event. At the time, my observations were limited to the fact that 6C is hovering around 70% (2-2 Club break or successful heart finesse). Not rock solid, but one to be in if you can find it.

Using the method at the head of this thread, the auction might start (uncontested) 2N-3C-3S showing 4-4 in the black suits.

Should South, at this point, start to get excited? I confess that if I am honest, at the table and without the benefit of sight of all of the hands, probably I would not. Maybe I should. Maybe the methods described in this thread require a more finely honed sense of judgement than I possess in order to capitalise on the possibilities that the method affords. If i were in a knockout IMP match, going into the last 8 with a deficit and looking for an opportunity to swing, having this method available at least gives the opportunity to try for a swing based on some intelligence. Perhaps I should be more concerned about the Spade suit weakness than the Diamond Suit weakness, and angle for a Club game for that (misguided, as the cards lie) reason, and having committed beyond 3N, maybe the later continuations might provide enough info for slam.

I really dunno, but just thought it was interesting. The hands that gave rise to the method do seem to crop up in real life.

[Zelandakh]

If South evaluates as a slam try opposite 20 (not for me but seems to be the premise here) then the auction

1♣ = 15+ nat/bal or 18+ any
... - 1♦ = (almost) any non-GF
1♥ = 18+ any or 23+ bal
... - 1♠ = relay
1NT = 18-20 bal
... - 2♠ = range ask including club transfers
3♣ = max
... - 3♦ = 5+ clubs, 4 hearts, SI
4♣ = nat, accepting
... - 4♦ = RKCB
4♥ = 1 or 4 key cards
... - 4♠ = ♣Q?
5♣ = no
... - 6♣

seems to be more solid than the proposed one.

If you only want to look for slam opposite 4 clubs then

1♣ = 15+ nat/bal or 18+ any
... - 1♦ = (almost) any GF
1♥ = 18+ any or 23+ bal
... - 1♠ = relay
1NT = 18-20 bal
... - 2♣ = 5 card major ask
2♦ = no 5 card major
... - 2♠ = 4 hearts, INV+
3♣ = no 4 hearts, 4+ clubs, max
... - 4♣ = slam try
4♥ = accepting, 1 or 4 key cards
... - 4♠ = ♣Q?
5♣ = no
... - 6♣

is also ok.

The deal is probably more interesting if you add the ♣Q to the North hand. Then South has a legitimate slam try and (almost) everyone is starting their auction at the 2 level. You might also find a few pairs taking a punt on the grand or, if MP, 6NT, at least in real life. The main point of this hand is that 2NT openers with 8 controls plus fit are slam giants. That is why some methods have specialised super-accepts for these kinds of hands. Naturally this is harder where the fit is a minor. The deal itself illustrates one problem with the OP method - information is given away on every hand whereas it is only a very small proportion of hands where it is useful. Even some hands where it could be useful cannot fully benefit from this.

[1eyedjack]

Personally I would not rate the South hand as worth a GF opposite what might be a balanced 15 count. Maybe I should., but I stopped reading the rest of the recommended sequences after seeing responder's first bid.

But in a sense that is a bit beside the point, because I would be the first to agree that a sensible strong club structure that has the luxury of an uncontested run (a significant assumption) is almost always going to outbid the best possible response structure to a natural 2N opener. The arguments for and against a strong club structure are far more complex than that comparison alone and it is not an argument that I intend to get drawn into in this thread. I think it only sensible, for the purposes of this thread, to take it as read that your methods require you to open 2N on the North hand and within that context argue for or against alternative response structures, and I am not particularly interested in an argument that the response structure is sub-optimal based on the proposal that the prescribed opening bid is sub-optimal.

Quote

If South evaluates as a slam try opposite 20 (not for me but seems to be the premise here)

But that is NOT the premise here. South does not evaluate as a slam try opposite 20 unless and until he learns of the 4 card Club support.

Quote

The deal is probably more interesting if you add the ♣Q to the North hand

I assume you mean the South hand? In that case I agree. If the original South hand was not a slam try opposite a 21 count with with a 9 card Club fit, give South the Club Queen and it certainly is. I suspect a lot of the room would still start by investigating the Heart fit over a 2N opener, and a substantial proportion will not then have a method to find the Clubs when the Hearts don't pan out. And with a combined 30 count a good number of them will still stop in 3N.

Quote

The main point of this hand is that 2NT openers with 8 controls plus fit are slam giants

And the main point of the OP method is to find that fit. Which is why this hand is put forward as an example to illustrate its benefit. I have never denied that there are other, bad hands for the method.

[awm]

Obviously there do exist hands where 5m is better than 3NT, or where you can make a minor-suit slam on very light values. It is true that "standard" methods over 2NT openings are poor at discovering these situations.

However, it is far more frequent that you have two balanced hands opposite one another, and just want to check for a major suit fit and play 3NT if none exists. The problem with the proposed method is that it offers a huge amount of information to the defense on these very common hands, in exchange for the occasional win on the (rare but existing) types of hands described above.

It is also worth noting that if we replace opener's hand on the given example with:

♠KQJx ♥Kxx ♦AJ ♣AQJx

then we have no play for 5♣ on a diamond lead (which becomes pretty likely when the defense knows opener has 4/4 in the blacks) and 3NT is cold. Also 4NT is down on (perhaps unlikely) best defense of a major lead to the ace and a diamond through. The point being that the "type" of opener's values is usually more critical on this sort of hand than opener's exact distribution anyway.

[Zelandakh]

1eyedjack, on 2012-May-23, 08:35, said:

Personally I would not rate the South hand as worth a GF opposite what might be a balanced 15 count.

I made a typo there, now corrected in red, it should read "non-GF". It is a normal, negative 1♦ response.

1eyedjack, on 2012-May-23, 08:35, said:

But that is NOT the premise here. South does not evaluate as a slam try opposite 20 unless and until he learns of the 4 card Club support.

My second auction addressed this point. Here Responder learns about Opener having 4 clubs with Opener's 3♣ bid.

1eyedjack, on 2012-May-23, 08:35, said:

I assume you mean the South hand?

No I meant the North hand. If Opener has 22hcp then even strong clubbers are quite likely to be starting at 2NT as well as natural systems (after a 2♣ or 2♦ opener). More than that, the South hand is certainly interested in slam opposite 22 balanced and a known club fit.

[1eyedjack]

Just out of interest, I ran a sim that generated 500 deals in which South had the specified hand, North had 20-22 balanced with 4-4 in the blacks. Of these, almost exactly 50% (249) deals made slam at double dummy (and of those, 69 were grand). Game could not be made on 6 deals.
I am having a bit of difficulty subdividing the remainder into those where 3N (with or without overtricks) is the only making game. There are quite a few where 4H is par, but I do not expect to find those at the table in any method (and probably not the grands either).

And yes, I am aware that a slam that makes at double dummy is not necessarily a good slam, but I take what I hope is the reasonable view that these are broadly evenly counterbalanced by occasions in which slam is better than 50% at single dummy but which the double dummy par rates as game.

[1eyedjack]

If you humour me for a moment, I have just run another sim of 500 deals, with the South hand fixed with the same precise 13 known cards, but relaxed the criteria for the North hand so that North is known to hold 4 Spades, 2 or 3 Hearts, and unknown minor suit length (except that he must be balanced), again with 20-22 HCP.
The purpose of the exercise was to try go get a feel for the importance (albeit just on this hand) of South identifying the 4 card Club support as shown in the previous sim, contrasted with a normal Staymanic sequence that just identifies major suit distribution.

This run resulted in, of the 500 deals:

161 slam hands (of which 20 were grand) (compared with 249 and 69 respectively)

4 hands that could make no game (compared with 6)

To *me* the indications of this is that the example hand is indeed one which argues in favour of the OP method. Yes it is only one hand. And yes there will be downsides because of information imparted about minor suit shape on normal 3N hands.

But I am concerned that the detractors of this method are overegging the pudding, by trying to assert that this is not a good example in favour of the proposed method, in addition to the valid concerns expressed.

It is I think natural, if you dislike a method, that having so concluded to (perhaps subconsciously) artificially bolster the arguments against it by belittling the genuine points in its favour.

I don't think that I would recommend this method in a MP event. Overtricks in 3N are far too influential. Concentrating on an IMP environment I would observe that

1) It is not every hand on which you end up in 3NT in which knowledge of opener's 4 card minor is critical to the opening lead (and that knowledge becomes of decreasing relevance as the hand subsequently progresses)

2) It is not every hand on which you end up in 3NT AND the knowledge of opener's minor suit shape makes the difference between the contract making and being defeated. More frequently it will make a difference to overtricks and undertricks, which is why I acknowledge that the method is severely disadvantaged in an MP event.

3) In IMP events, the amount at stake on a particular borderline slam hand places a premium on accurate slam bidding. Capitalise on that premium and you would happily forego a few overtricks in the odd 3NT contract by reason of information imparted en route. Not every hand carries equal weight in the argument.

Anyway, I have yet to work out how to extract from these deals the hands where 3N makes but 5C fails and vice versa. Possibly more anon.

[awm]

You are too focused on this specific responder hand. The more relevant questions are:

1. In general opposite a 2n opening, how often can you make 5m but not 3n? Definitely possible but I bet rare.

2. How often can you make 6m with responder holding less than 10 hcp and no 6m? Again possible but rare.

Your method will gain when one of these holds. But you will lose on many other hands, sometimes in subtle ways, by giving opponents more information on defense. Its hard to measure this quantitatively, but "game only" hands are awfully common