[kenrexford]

An idea hit me today that I have been posting in several spots because I think it is really neat. You can find mentions of it at my blog (www.cuebiddingatbridge.blogspot.com) and at the bridgewinners site.

The idea is to use immediate RKCB answers as a means to identify a second fit that it unknown to partner.

A simple example:

1NT-2♦(transfer)
2♥-3♦
3♥-3♠(cue)
4♣(cue)-4♦(last train)
?

Opener's options:

4NT = RKCB hearts
5♣ = diamond fit also, answering RKCB (or answering 6KCB if you prefer that)
5♦ = diamond fit...
5♥ = ...
5♠ = ...

That way, 6♦ ends up in focus when Opener shows the previously-unknown fit to Responder, and conversely 6♦ is a pure asking bid (whatever form you use) if Opener DENIES the diamond fit (bids 4NT).

[han]

A bit of a side comment, but I think that after 3D opener should have a way to show a fit for both suits.

I remember that there was a meckwell auction in which they could either show or ask keycards, with different meanings. I am sure that your idea is just one of the many possible applications of such a structure.

[TWO4BRIDGE]

han, on 2012-September-13, 13:31, said:

A bit of a side comment, but I think that after 3D opener should have a way to show a fit for both suits.

I've posted this a few times in the past... from a fellow nicknamed IanD :

1NT - 2D! ( transfer )
2H - 3C ( 2nd suit )
??
3D! = ( 3-"cheapest" new suit ); 4+support for Cl; only 2 cards Hts
3H = 3 cards Hts; NO 4 cards Cl
4C! = double-fit; 4+cards Cl AND 3 cards Hts... sweet !

[kenrexford]

han, on 2012-September-13, 13:31, said:

A bit of a side comment, but I think that after 3D opener should have a way to show a fit for both suits.

I remember that there was a meckwell auction in which they could either show or ask keycards, with different meanings. I am sure that your idea is just one of the many possible applications of such a structure.

The Meckwell idea is what I have expanded.

As to Opener having a means to show a double fit, the problem is space and practicality. If you dedicate an immediate bid for that purpose, like 4♣ (just noticed the above after posting this), you take away cuebidding space and do not really tell much that is useful (you have one cuebid between 4♣ and the signoff 4♥ and yet know nothing about the black suits). If you do it later, the same problem presents itself. The practical solution I am seeing is to focus on what makes sense now (the heart fit and controls), with the second-fit option held off until later, at the five-level, where the calls probably have no meaning anyway.

Plus, there are times where there simply is no space for all of that. 1♠-2♥, 3♦-? You are too damn high to incorporate some sort of double-fit bid. But, the diamond strain might be ideal. Set spades, cue, and THEN resolve the diamonds issue.

[TWO4BRIDGE]

Ken...

1) What is your option when Opener has a fit for the MINOR only ?

1NT - 2♦ ( transfer )
2♥ - 3♦
??

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Ron ( thehog ) has posted his answer ( in other posts ) for the double-fit auction cases:
He said that a "new suit" by Opener would actually be a cue-bid showing the double-fit:
1NT - 2♦ ( transfer )
2♥ - 3♦
??
3♠ = cue-bid showing 3 cards ♥ AND 4 ( or 5 ) cards ♦, ( double-fit ),
whereas,
4♣ = cuebid for the double-fit, denying a ♠ control .

[han]

If you have only 2 calls below 3NT, it is not wise to use both of them for hands with which you want to go beyond 3NT.

[kenrexford]

When the new minor is clubs, 3D suggests a fit, keeping three of other major natural. For me, that is. If new minor diamonds, other major is diamond agreement. No double fit option.

[Zelandakh]

Sorry Ken but I think this is one of the worst ideas I have read from you. I read your objections and still think it is considerably better to show the double fit immediately over 3♦. A simple structure is

3♥ = ♥ fit, no ♦ fit
3♠ = ♦ fit, no ♥ fit, poor hand
3NT = no fit
4♣ = double fit
4♦ = ♦ fit, no ♥ fit, good hand
4♥ thru 5♣ = super-accept for diamonds, RKCB responses

If you wanted to be able to split up the range of the double fit hands then one simple solution would be for 4♣ to be a double fit with a good hand and 4♥ a double fit with a bad hand. Then 4♠ to 5♦ are the super-accepts. Of course the big advantage of 4♣ as the double fit is that it still gives you a big choice of RKCB methods if you think this is important. For example, 4♥ = RKCB for diamonds; 4♠ = 6KCB; 4♦ followed by 4♠ = RKCB for hearts, along with alternative routes for showing voids.

Instead, I want to propose a different criteria for 2-way RKCB which I have played around with a little bit, namely that Opener bids 4♠ RKCB with 2 or 3 side kings but gives RKCB answers (4NT - 5♥) with 0-1 side kings. The idea is that RKCB auctions are sometimes smoother when the hand with more side kings takes control. Of course, after a 1NT opening one might argue that more is known about Opener's hand than Responder's and that therefore Opener should always describe. In that case an alternative would be for Opener simply to give key card responses starting with 4♠. Either way, I am confident the overall structure is better.

One final note is that there is an even simpler solution to this auction - using transfers after the transfer. So with hearts and diamonds you might start 1NT - 2♦; 2♥ - 3♣. Now
3♦ = ♦ fit, no ♥ fit
3♥ = ♥ fit, no ♦ fit
3♠ = cue, double fit
3NT = no fit
4♣ = cue, double fit
4♦ = no cue available, double fit
4♥ thru 5♣ = super-accept for diamonds

You get everything that was asked for plus the ability to play 2-way RKCB or direct key card responses in the given auction. How can this not be an improvement?

[kenrexford]

First of all, the specific example given is not the point. The point is using immediate answers to resolve the two fit problem when there is no space. If there is space in your methods in that sequence, then ok. I am sure you have auctions where space is not available. For instance, 2NT-3D, 3H-4D, ?

Second, though, the transfer at second bid approach is nice for diamonds but not for clubs, then. Also, even for diamonds, Opener loses the ability to show the other major naturally. More importantly, the entire idea of Opener showing a double fit seems dumb at that stage. One hand is typically strong but otherwise unknown. The other is somewhat weaker but better defined. Setting the n
Major, allowing Responder to complete picture, then having Opener ask questions or show, as he sees fit with enhanced info, seems a lot better than spending all bids below 4M to indicate what fits exists but little or nothing about side suit controls, pattern, or even internal cards.
It might be that our bidding styles are so different that what i see as an amazingly effective tool in context is useless to you because your pre-ask style is so different. But, taking on that debate is too complicated. If your style is more like mine, you will see the intense benefit in that context. Even if not, the benefits in other sequences should hopefully be apparent.

Perhaps the easiest example:
2NT-P-3♦-P-
3♥-P-4♣-P-
?

4♦ = RKCB clubs
4♥ = to play
4♠ = RKCB hearts
4NT = to play
5♣+ = 6KCB answers, double fit established

[Zelandakh]

kenrexford, on 2012-September-14, 06:32, said:

First of all, the specific example given is not the point. The point is using immediate answers to resolve the two fit problem when there is no space. If there is space in your methods in that sequence, then ok. I am sure you have auctions where space is not available. For instance, 2NT-3D, 3H-4D, ?

I play that as both majors and not wanting to make a slam try so no space problem.

kenrexford, on 2012-September-14, 06:32, said:

Second, though, the transfer at second bid approach is nice for diamonds but not for clubs, then.

I play clubs on transfer too (1NT - 2♦; 2♥ - 2♠ = range ask; or clubs; or strong 1-suiter) which tends to avoid the problems of using 3♦ for that purpose. The same after a 2NT opening btw.

kenrexford, on 2012-September-14, 06:32, said:

Also, even for diamonds, Opener loses the ability to show the other major naturally.

This is true. You have to decide whether it is more important to use the space to show 5 spades or to split the range of diamond support. This is also true for the method where 3♦ shows diamonds of course. The difference is that in that case we are forced past 3NT with diamond support. Using 3♠ as diamond support that would decline a slam try allows us to stop in 3NT when Responder just had a 2542 slam try hand.

kenrexford, on 2012-September-14, 06:32, said:

More importantly, the entire idea of Opener showing a double fit seems dumb at that stage.

Why? Both the information that we have a double fit and the information that we do not have a double fit are often useful in the subsequent auctions. We can optimise our methods according to whether one suit or two are in focus. And, in the case of transfers, we are not even giving up on any cue bids.

kenrexford, on 2012-September-14, 06:32, said:

One hand is typically strong but otherwise unknown. The other is somewhat weaker but better defined. Setting the n
Major, allowing Responder to complete picture, then having Opener ask questions or show, as he sees fit with enhanced info, seems a lot better than spending all bids below 4M to indicate what fits exists but little or nothing about side suit controls, pattern, or even internal cards.
It might be that our bidding styles are so different that what i see as an amazingly effective tool in context is useless to you because your pre-ask style is so different. But, taking on that debate is too complicated. If your style is more like mine, you will see the intense benefit in that context. Even if not, the benefits in other sequences should hopefully be apparent.

The method you suggest is for 3♦ to be natural, and then: 3M to be bid on any major suit support (with space for cues and serious/frivolous); 3oM to be natural without fit; 3NT to be any other hand without fit; and 4m to show diamonds (no space for cues). What I am suggesting is for 3♣ to show diamonds, and then: 3♦ to show diamond fit (full range of cues + space for serious/frivolous if desired); 3♥ to show heart fit (full range of cues + space for serious/frivolous); 3NT to show no fit; and 3♠/4♣ to show a double fit (also with space for cues). This is a no-brainer. Not only have we passed more shape information, we also have more space for cue bidding. The only thing given up is the ability to show no fit and 5 spades. In return we get not only the above advantages but can also play the simple rebid structure: 2♠ = clubs or range ask; 3♣ = diamonds rather than having to make a special case of RKCB, much more likely to be forgotten.

I hope you know by now that I love reading your ideas Ken, and often enough they inspire my own thought processes. But I think this one is really wrong, at least in this context. It may well be that in other auctions I would see a bigger benefit. What I do think is that while our styles of thinking are sometimes similar, the style for bidding is sometimes not. In this thread and some others I have the impression that you like to get a fit and then to go for the controls big time. A secondary fit is something that perhaps comes along during the control bids. I suspect I am concerned about showing more shape early and not worrying too much about controls until after the 3 level. Naturally that is a big simplification - my preferred methods with a big fit for Opener are probably more one-suit focused than yours, although there is some recovery space built in for side suits.

It would certainly be interesting for me to hear what you think are our differences in style. As I wrote earlier, I think our thought processes often run along similar lines, essentially finding patterns of bids that better optimise the available hand types than natural. My aim is usually to make such an optimisation as logical and consistent as possible across similar auctions and, ideally, as early as possible in the auction. This is probably another reason I am so opposed to this convention in this 1NT auction, since it creates unnecessary exceptions where there is an alternative solution that improves not only the 1NT structure but also that over 2NT, as well as being logical and occurring reasonably early. It would be nice to see an example of a non-1NT auction with this convention being used. I think seeing that might help me to judge it in a more overall context and that might even provide some pointers on where it could be improved or what kinds of rules would be necessary to incorporate it unambiguously.

[kenrexford]

Zelandakh, on 2012-September-14, 07:25, said:

I hope you know by now that I love reading your ideas Ken, and often enough they inspire my own thought processes. But I think this one is really wrong, at least in this context. It may well be that in other auctions I would see a bigger benefit. What I do think is that while our styles of thinking are sometimes similar, the style for bidding is sometimes not. In this thread and some others I have the impression that you like to get a fit and then to go for the controls big time. A secondary fit is something that perhaps comes along during the control bids. I suspect I am concerned about showing more shape early and not worrying too much about controls until after the 3 level. Naturally that is a big simplification - my preferred methods with a big fit for Opener are probably more one-suit focused than yours, although there is some recovery space built in for side suits.

It would certainly be interesting for me to hear what you think are our differences in style. As I wrote earlier, I think our thought processes often run along similar lines, essentially finding patterns of bids that better optimise the available hand types than natural. My aim is usually to make such an optimisation as logical and consistent as possible across similar auctions and, ideally, as early as possible in the auction. This is probably another reason I am so opposed to this convention in this 1NT auction, since it creates unnecessary exceptions where there is an alternative solution that improves not only the 1NT structure but also that over 2NT, as well as being logical and occurring reasonably early. It would be nice to see an example of a non-1NT auction with this convention being used. I think seeing that might help me to judge it in a more overall context and that might even provide some pointers on where it could be improved or what kinds of rules would be necessary to incorporate it unambiguously.

Actually, I have a lot of bids that are oriented to alternative fit analysis. A simple example: 1♠-P-2♠-P-, 3♣-P-3NT? I play that 3NT call as a super-accept of clubs, designed to help find 6♣.

You are right that I have a tendency to focus on one project first and then to delay the double-fit scenario until later. For example, with both a major and a minor fit, I will establish the major as the focus for cuebidding (because space is so precious) and delay discussion of the second fit until later. That said, it is not necessarily about controls only. For example\, in one partnership the continuation after a transfer, second minor, agree the major is pattern completion. For example: 1NT-2♦, 2♥-3♦, 3♥-? After hearts are agreed, Responder completes pattern (3♠ = low short = short club, 3NT = high short = short spade, 4♣ = low void = void club, 4♦ = high void = void spade). In that structure, there is no other means of showing stiff-stiff-void-void below 4♥ effectively. I mean, 3♠ could also handle that, but this causes a problem with a mild try (hear 3♠ from Opener and then bid 3NT as a passable "mild interest only" sequence). The use of 4♣ for the double fit clearly allows almost no pattern unwind at all.

As far as a non-NT sequence, consider 1♠-2♥, 3♦ as a start. Not bidding 3♠ as Responder because you have a double fit (spades and diamonds, maybe 3-5-4-1) is not advisable, because you first have to find a call for that and second have to realize the space consumption. This might be a great spot for this tool. 1♠-2♥, 3♦-3♠, 4♣(cooperative non-serious cue)-? That type might be the 1x-2y-3z auction type.

The common thread will surely be the major fit, minor fit secondary scenario, obviously.

[TWO4BRIDGE]

Zelandakh, on 2012-September-14, 02:21, said:

A simple structure is :

3♥ = ♥ fit, no ♦ fit
3♠ = ♦ fit, no ♥ fit, poor hand
3NT = no fit
4♣ = double fit
4♦ = ♦ fit, no ♥ fit, good hand
4♥ thru 5♣ = super-accept for diamonds, RKCB responses

GMTA... back in January, I proposed the same 4♣! ( next unbid suit ) to show the double-fit in a similar auction:
http://www.bridgebas...-your-thoughts/ ... ( post # 16 )

But I won't go so far as to say my thought processes are similar to KenRex.... lol ... although his posts, like yours, inspire me also.

[dake50]

Clever idea for a set of "wasted bids".
More generally, when one designs a system, what ARE those
"wasted bids" good for? Information theory would suggest it is efficient
to get some rarer cases out of the puzzle.
Maybe even as revealing as we can include those cases in an initial bid
EXACTLY because "wasted rebids" will show that case. Eg. include strong
hands in a 1H overcall because "wasted rebids" will get to a good game
decision. Eg2. "impossible negative" to 1C-force.
So freeing our partnership to infer 'partner didn't use "wasted rebid",
what is now being shown?' Must be something else.

[Fluffy]

it looks nice, it seems that options for 1NT opener on this context are to bid 4♥ or ask for keycards, how did you miss to gain one step using 4♠ as keycard for hearts and next for double fit showing ken?

[kenrexford]

4NT as double fit 6KCB is also an option, but it suffers from the same problem kickback solves, which is probably worse in 6KCB.

[Fluffy]

I you are answering me, I meant same structure you proposed, but starting in 4♠, not in 4NT

[kenrexford]

Fluffy, on 2012-September-15, 15:33, said:

I you are answering me, I meant same structure you proposed, but starting in 4♠, not in 4NT

Oh. I would, but I did not get that detailed in my example.

[Zelandakh]

The 1♠ - 2♥ - 3♦ auction is interesting because it is one of the worst in natural bidding. There are simply too many hand types for the amount of bidding space available. These kinds of auctions are one of the reasons I started looking round for alternatives (relays, limuted openers, etc). First of all, I think we should look at what Responder can do here and then we can consider what bids we can find for double fits. First of all, in nearly all auctions of this type where we are essentially guessing it is helpful to have a grope available and usually that should be the cheapest bid. Therefore I would suggest 3♥ has to be a multi-way bid, perhaps heart length and slam interest but more often just looking for the best contract. It also seems unreasonable for 3♠ to guarantee 3 card support here - 2 has to be enough; if 3 card support then slam interest since 3 card support and no slam interest would just bid 4♠. Similarly for long hearts and no slam interest (4♥). Finally, with diamond support and slma interest there is 4♦.

That leaves 4♣. What should this mean here? My default meaning for the 4th suit at the 4 level is a slam try agreeing the last bid suit; but that makes ltitle sense here since 4♦ already has that meaning. Therefore I can see 3 main possibilities:-
1. Some sort of choice of game bid. Does this seem reasonable given the lack of space? probably not.
2. A slammy hand with a major (either specify one or leave it ambiguous). This is a good idea since it takes some pressure off the 3M bids.
3. A double fit hand with slam interest.

So if option 3 was played then we do indeed have a bid available to show the double fit immediately. I cannot say if that would be good but it is certainly better than not assigning it a meaning at all. I am curious what you play here since you are generally creative in tight spots such as this. One thought that occurs to me is to try and combine options 2 and 3 into a single convention. Thus 1♠ - 2♥; 3♦ - 4♣ = slam try in one or both majors. Then 4♦ accepts the slam try and asks for the major (bid the major, or 4NT+ for double fit) and 4♥ = pass/correct, decline the slam try. (Note that I have not tested this; the idea is for illustrative purposes only)

One difference here is that this is not so easy to remember as over a NT opening. That becomes a major plus for your idea, especially if it carries across to other auctions too. On the other hand, having direct ways to show a double fit is also quite a powerful concept if built directly into the bidding system. The key to any way of showing something "extra" like this is for it to be applied consistently and logically.

Finally, to Fluffy, Opener's options are not limited to 4♥ and 4NT if playing either of the methods I suggested earlier. Playing direct key card responses, Opener would bid 4♠ with 1 or 4, 4NT with 0 or 3, etc with any hand that was willing to go beyond 4♥. Playing the version of 2-way KCB I put forward, Opener would bid 4♠ with 2-3 side kings or give a key card response (4NT = 1 or 4, 5♣ = 0 or 3, etc) with 0-1 side kings. It would be a good idea to compare Ken's suggestion with alternative ways of using these bids.