[kenrexford]

Weird question, but I am wondering...

From your experience, what is the #1 (and #2 if you have one) problem response to RKCB with spades as trumps (to avoid the 1430 problem), and what is the one piece of information most likely to answer your problem?

For example, a possible answer might be, "When partner shows two without the Queen, I often am stuck wondering whether he has the Jack" or perhaps "when partner shows one key card, and sometimes two key cards, I wonder whether his trump key card is the Ace or King" or maybe "When partner has 2 or 3 key cards without the Queen, I often wonder whether he has the King in the secondary fit."

In other words, what response seems most frequently to cause a problem, and what one piece of information at that point would be nice to be able to ask about without bypassing 5♠?

I have in mind a solution for a problem, or even two problems, but I am not sure what problem(s) are most useful to solve.

[Zelandakh]

A couple I can think of off the top of my head. After a 5♥ response it is sometimes nice to know if partner holds lower trump honours which would make picking the queen up good odds. After getting a response that we are missing one key card (whichever that is) it can be usefuö, especially at Pairs, to know if partner holds a specific side suit honour which would make 6NT better than 6 of the suit.

[Vampyr]

Zelandakh, on 2012-September-19, 06:53, said:

After getting a response that we are missing one key card (whichever that is) it can be usefuö, especially at Pairs, to know if partner holds a specific side suit honour which would make 6NT better than 6 of the suit.

In the "full version" of RKCB, as published by Eddie Kantar, you can often find this out (depending on just what honour you are looking for). I don't remember the whole asking bid structure, having read the book about 20 years ago.

[Zelandakh]

Vampyr, on 2012-September-19, 08:17, said:

In the "full version" of RKCB, as published by Eddie Kantar, you can often find this out (depending on just what honour you are looking for). I don't remember the whole asking bid structure, having read the book about 20 years ago.

You can find it out if you hold all of the key cards but the space is regarded as too precious to devote bids to this if we are missing a key card.

[lalldonn]

Sometimes if we are off 1 + the queen then slam is a clear favorite if we are off KQ of trumps but an underdog if we are off an ace and queen (say fifth) of trumps. So if you are trying to solve something then I say go for finding out whether or not one of partner's keycards is the king of trumps when he doesn't hold the queen.

[ArtK78]

Vampyr, on 2012-September-19, 08:17, said:

In the "full version" of RKCB, as published by Eddie Kantar, you can often find this out (depending on just what honour you are looking for). I don't remember the whole asking bid structure, having read the book about 20 years ago.

Unless I am mistaken, any further ask is a grand slam try and guarantees possession of all of the key cards plus the trump Q or the equivalent.

[benlessard]

Quote

Sometimes if we are off 1 + the queen then slam is a clear favorite if we are off KQ of trumps but an underdog if we are off an ace and queen (say fifth) of trumps. So if you are trying to solve something then I say go for finding out whether or not one of partner's keycards is the king of trumps when he doesn't hold the queen.

Agree 100%

Difference between the A or K of trumps is by far the biggest one for me especially in a relay system. Sometimes im missing a crucial card in a side suit. AQTxx vs xxx or Jxx in my side suit. Knowing that partner has Axxx rather than Kxxx make all the difference for not missing those 75% slams.

2- extra lenght. Sometimes extra lenght is like having the Q sometimes its not.

The rest is mostly if your able to keycard low.

3- general extras. Sometimes you need a certain amount of keycards + extras. These hands are frequent and it doesnt mean that keycarding was wrong.

4- Secondary honnors

5- Stiff A, If partner is 5521,5431 I often vizualize hands that are 68% or cold if partner doesnt have the A of clubs that are pretty poor if hes has the stiff A of C.

[ggwhiz]

My big problem occured when a really good player gave me a game as an up and coming rookie. I bid 4nt kc on hearts, he showed 2 with the queen and we were missing a couple of aces.

Ever since I cue bid and quantitate instead of 4nt when appropriate. Well at least I got my eye on that ball and with a high degree of success. I consider knee jerk 4nt bids to be the real problem, not the responses Same with knee jerk splinters.

[nigel_k]

lalldonn, on 2012-September-19, 10:43, said:

Sometimes if we are off 1 + the queen then slam is a clear favorite if we are off KQ of trumps but an underdog if we are off an ace and queen (say fifth) of trumps. So if you are trying to solve something then I say go for finding out whether or not one of partner's keycards is the king of trumps when he doesn't hold the queen.

The 4NT bidder could have JT9x but very often they will need partner to have some supporting cards as well as the Ace in this situation. How many trump suits for the 4NT bidder will be significantly above 50% to have one loser opposite A432 but significantly below 50% opposite K432?

I think extra length is most useful. It won't often be the deciding factor on its own, but if there are issues in the other suits, an extra trump can easily make the difference between a decent slam and a poor one.

[kenrexford]

OK, so here's the general thinking I have been uncomfortable for a while with some sequences leaving me in doubt even after RKCB, with the question of whether the key card is or is not the trump King being one ofthe major issues (as others have suggested -- AJ109x is substantially different than KJ109x). I also have disliked the Queen-ask where partner immediately kicks into specific Kings, because (1) sometimes I want to know about a specific Queen, and (2) sometimes the answer might get past an alternative strain. Other more subtle issues also motivate me (for example, a common issue of wanting to play in the agreed major or an alternative strain -- minor? -- depending upon whether partner has the Queen in the primary suit).

So, I thought about restructuring the basic responses to RKCB completely in a way that might allow for a better unwind. The core thought I had was something along these lines. I assume spades as trumps (because Kickback for other strains would work out the same way):

The focus is on showing the presence or lack of the trump Queen immediately more often (not just when holding two keycards). One structure would be as follows:

5NT+ = specific Kings, with 4 key cards and the trump Queen
5♠ = 1 key cards, with the Queen
5♥ = 0/3 key cards, with the Queen
5♦ = 2 key cards, with the Queen
5♣ = No trump Queen

So far, notice that when you have the Queen, you immediately show the key card holdings. This avoids the forced "answer immediate kings" scenario when asker asks for the Queen and instead allows asker to decide what question to ask.

Notice also, so far, that by kicking down the "two with the Queen" response to 5♦, you allow 5♥ as a "slam last train" call, asking partner to bid 5♠ with the trump King but something else intelligent with the trump Ace (perhaps gearing toward the 6♠/6NT question?). Alternatively, in this structure (see below for the alternative structure), the 5♥ LTTC call might invite slam if partner has "the critical side King" or something else.

5♥ can also be a functional relay if asker KNOWS that partner has the trump King (if that is the strutural choice), forcing a relay and then allowing different types of questions. In other words, if partner shows two with the Queen, you could ask normal questions. If you know that partner has the trump King, however, you can bid 5♥ to "ask him," really setting up a functional relay to change the discussion to another type of sequence. Maybe this means changing from questions to instead alternative strains, or whatever, but the net functional result is a relay.

Maybe the "with the Queen" structure might be shuffled to have 5♦ (and hence the 5♥ ask/relay call thus enabled) for a different answer, such as if that question (do you have the trump King or Ace) is more likely important when the answer is 1 key card, but that is for discussion (frequency, utility, etc.).

Obviously, as well, you notice that the Queen-denial (5♣) might be useful, because cuebidding might have already told you of a missing Ace, such that the missing Queen is enough to sign off, thus leaving the opponents in the dark as to who has precisely what top honors, and how many.

For this to work, though, there is an obvious new problem, namely how to unwind the number of key cards if the person answering does not have the trump Queen. The answer is somewhat from paradox thinking. You basically state as asker what you need to move forward as a minimum. So, for example:

...4NT-P-5♣(no Queen)-P-?

5♠ = to play (I need at least 4 key cards?)
5♥ = I need at least 3 key cards
5♦ = I need at least 1 key card (5♥ says I just have one, 5♠ says none, 5NT+ says 2+ key cards)

After this "state what you need" approach, Responder will know whether he does not have enough (signs off), whether he has just enough (usually just bids the slam), whether he has one more than needed (shows grand-oriented other cards), or whether he strangely has two more than needed (how can that happen?).

Notice also that there is an ability to force a semi-relay in this sequence, as well, if asker knows the answer. For instance, if you know that partner has at least enough for slam because he has shown one key card by cuebidding, but you want to refocus somehow, you can ask for three, knowing that partner will decline, and then shift tactics. For instance, suppose a RKCB sequence where partner holding the spade Queen means that 6♠ is the contract, but if not then 6♣. He denies the spade Queen (bids 5♣). You ask if he has three key cards (5♥), knowing that he will deny that (5♠), which allows you to NOW bid 6♣ as "to play." (You could also do this in reverse, by suggesting that you need only one key card when you and partner know that he must have at least four, but that would be rare.)

Now, if you are following and liking this so far, you probably already spotted that the "do you have the trump King?" question usually matters most when MISSING the trump Queen. So, the first major alternative structure is to show immediate answers without the trump Queen:

5NT+ = 4 keys, specific Kings, no trump Queen
5♠ = 1 key, no trump Queen
5♥ = 0/3 keys, no trump Queen
5♦ = 2 keys, no trump Queen (5♥ asking if King or not)
5♣ = I have the trump Queen

The same relay approach works, as well, again potentially geared to alternative-strain calls (hear 5♦, ask for the King that you know he has, and then bid 6♣ to play, for example).

After 5♣ (I have the Queen), you might know enough already. If not, you would use the same unwind of stating what you need, with the addition of the possible relay enabling.

Other shufflings are possible, as well.


Is this total insanity, or does this seem to have potential?

[semeai]

Not crazy, but it seems that the only extra you're getting is that you have that "we're missing the Queen but maybe it's still okay" LTTC in one specific case (2 keycards no queen).

Other than optimizing what holding exactly gets to have the LTTC bid, isn't this just about equivalent to something extremely close to standard:

5S: 2 with queen
5H: 2 without queen
5D: 0/3
5C: 1/4

Over 5D, 5H is the queen ask.

Over 5C, 5D is the queen ask. Then 5H = no queen, but some useful holding anyway (or whatever) [instead of showing ♥K+♠Q], 5S = no queen and no useful holding, 5N+ = yes queen

And on top of that we still have 5H over 5C free (for "I need the queen and a useful holding" if you like, or some more normal bid that guarantees all keycards). Shuffling the bids around differently may make the 5H over 5C bid more useful. [The place you lost this extra bid is that the standard approach assumes that everyone will know based on the auction whether 3 keycards a holding that can possibly stop below slam.]

[benlessard]

Interesting but you lose 2 very important sequences

4Nt--5C (1)
5H (I dont care about the Q) but i need an extra card.

While your system its going to go
4NT--5C (no Q)
5D---5H (1)
?? here i dont know if you have 1+ xtra card or not.

or

4Nt--5C (4)
5H (I know you have 4 keycards and I dont care about the Q but i need an extra card).

VS
4NT--5C (no Q)
5S--- bidding soemthing and going down because you have 4KC but dont have the extra card.



The 2nd one is

4Nt--5C
5D---5H (I have it but no extras)
5S (not enough )

vs

4NT--5S (1+Q but you dont know if extras or not)

There is also some tiny cases where you promise a monster and 4Key+Q isnt enough and responder is going to keycards despite holding none.

4NT--5Nt (maybe going down if you promise a monster)

vs
4NT--5C (obvious 4 Keycards)
5D---5H (Q?--- Yes but I have no extras)

Its rare but IMO its OK for responder to ask for keycard holding none IMO when opener promise a monster.

I had a hand recently.
AKQxx
Ax
x
AJxxx

xxx
KQxx
Kxx
Qx


1S--1NT (S 11-14 or 18-22----relay)
2NT--3C (18-22 with S+C and short D---relay)
3H--3S (5215--S rkc)
3NT--4C (4---do you have Q?)
4H--?? (yes but no K of clubs)

Many failed to stay out of 6S after a Drury sequence.

[Siegmund]

For me the #1-and-its-not-close problem is RKC wasting space that could have been used for cuebidding to pinpoint all the second round controls before we got to 5 of a major. I hate having to go to the 6 level to get specific kings out of partner.

It's not so much RKC responses' fault -- it tries to minimize wastage-- as a fact of life that RKC after cuebids is spending about half its effort repeating information already known.

Apologies for the threadjack. I am a believer in "4NT is never asking after cuebids" (but a waiting bid / trump cuebid, depending on style.)

[benlessard]

Agree 100% with Siegmund, RKC after cuebidding or cuebidding before RKC is simply inefficient.

Quote

Notice also, so far, that by kicking down the "two with the Queen" response to 5♦, you allow 5♥ as a "slam last train" call, asking partner to bid 5♠ with the trump King but something else intelligent with the trump Ace (perhaps gearing toward the 6♠/6NT question?).

A gain but not a complete one since sometimes you can bid more than 5S with 2+Q+extra K and stop in 5S when its 2+Q but no xtra K.

I must admit however that 14--2--30without--30+Q but no xtras---and 3+Q+extras might make some sense because its safer.

[mike777]

tough to quote a book that is soo complicated as Kantar's that I think 99.999% of us dont play it.

[Zelandakh]

It is often a good idea for the first step of an asking bid or relay to show roughly half the hands and the other bids to show the rest. This is an idea I played around with too some time back but I could not find any advantage in it. So it makes sense, perhaps even has potential; but to make it worthwhile you also have to show an advantage and that is not at all easy. I am very interested to hear it if you do though!

[kenrexford]

I thought about this some more, and I think I am trying to fix inherent problems with RKCB the wrong way. I think I have a much better solution in mind, but it is somewhat complicated to explain, even if it should be easy to implement. I will get back later when I work this out completely. The core idea, however, is to somewhat combine RKCB with Culbertson's 4NT/5NT convention, but with some of the requirements in the Culbertson part potentially coming from force of the auction and knowns already established. By limiting replier's possibilities and excluding redundancy, you gain space to focus more and can distinguish between an Ace and the King of trumps, and in some sequences can even include a side critical (sort of a Byzantine Blackwood idea) into the answers.

More to follow...

[benlessard]

Im sure that having 2 way to ask for keycards can greatly improve precision. A bit like Turbo if you want.

lets say that D are established and both 4D/4H are keycards/asking for A how about 4D asking for A but I have K of trumps & 4H asking for A but denying K of trumps ?