[akhare]

After pattern resolution (using ~symmetric relays), we play:

3N: To play
4♦: Canonical terminator puppet
S1: QP ask
S2: Always RKC in ♣
S3: Always RKC in ♦
S4: Always RKC in ♥
S5: Always RKC in ♠

For example, after 3♦, showing say a 5=4=3=1 pattern, 3♠ will always be RKC in ♣.

To me, it seems that this treatment is suboptimal, if not downright silly.

IMO, it's much better to play Richard's Moscito treatment, i.e. RKC based on suit length, with ties broken in suit order. In other words, in the above case, 3♠ is RKC ♠, 4♣ is RKC ♥ and so forth.

Comments?

[Free]

If you use fixed suits, it's better to use the order ♥-♠-♣-♦ because for the respective game contracts you get similar amounts of bidding space. For example, after 3♦ showing a 5=4=3=1 pattern, 4♣ as RKC ♠ gives us the opportunity to play 4♠ unless partner has 2+Q. In your system, 4♠ is RKC ♠ which forces us to the 5-level.

When using variable RKC (like I used when playing MOSCITO), I believe the following is optimal:
- the order is: suits longest to shortest
- when suits have equal length, prioritize as follows: ♥ > ♠ > ♣ > ♦

[straube]

We just got burnt not being able to RKC cheaply for a 6-cd long heart suit. OTOH, it's been nice to RKC for the minors on occasion since we run out of room so quickly. Perhaps some rules melding longest suits with minors would work. Like ties go to the minor.

[akhare]

 straube, on 2013-July-19, 14:47, said:

OTOH, it's been nice to RKC for the minors on occasion since we run out of room so quickly. Perhaps some rules melding longest suits with minors would work. Like ties go to the minor.

Richard had the following comment in his document and I think it makes sense. The minor suit ask will still be cheap assuming that it's the longest or second longest suit.

Quote

Any other bid by the relay asker is 1430 Roman Key Card Blackwood, simultaneously setting the trump suit and asking for Aces. If the relay asker makes the lowest unreserved bid, he is bidding RKCB in partner's longest suit. If the relay asker makes the second
unreserved bid, he is bidding RKCB in partner's second longest suit. Ties are broken in order ♥> ♠> ♣> ♦.

There are very logical reasons to adopt this ordering structure when breaking ties. We initially base the asking order on RR's suit length because we are most likely to want to be able to set RR's long suit as trump. The order in which we resolve ties is designed to maximize the amount of bidding space available before passing the safety level for a given hand. There is less room to explore for game versus slam when hearts are trump than when spades are trump. As such, we ask about hearts before asking about spades.

[benlessard]

our rules are the following. Like Free its longest to shortest but never in a singleton or void, and its HSCD priority when equal. We dont stop in 4Nt or 5Nt. 4NT is a GST in clubs when 5C would be to play and we didnt have enough space to do C RKC. Extra space could be void showing but we are never using them.

If we end in 4D we don't keycard we ask for min max instead.

If we end in 4C or 3Nt and both M are possible then 4D is PES (Polish end signal)

EX
6043 ending at 4C

only 1M is possible so 4D PES doesnt apply.

4D = S KC
4H = D RKC
4S to play
4Nt C slam try

6421 ending at 3Nt.
Both M are possible so 4D PES apply

4C=S RKC
4D= PES
4H=H RKC
4S=D RKC

[yunling]

I think you don't really have much problem when your shape relay ends at 3♥ or lower.

After 3♥, it sounds very good to use
3♠ QP ask
3N To play
4♣ RKC in ♣
4♦ terminator puppet
4♥ RKC in ♦
4♠ RKC in ♥
4NT RKC in ♠

So, after 3♦, you can just keep most of these "kickback" RKCs and use something like
3♥ QP ask
3♠ RKC in ♠
3N To play
4♣ RKC in ♣
4♦ terminator puppet
4♥ RKC in ♦
4♠ RKC in ♥

It is when your shape relay ends at 3♠ or higher that this method sucks.
e.g.
3♠-5♣(RKC of ♠)-5♥-?
Now often you cannot check trump Q cheaply.

Meckwell plays 4♥/4♠/4N as RKC in low/middle/high when 3 suits are in focus.The problem still exists but at least you are no higher than standard bidders.

Though more frequently suit-length based RKCs are lower, you might have a big problem when p is 5431 and you want to play in p's 3 card minor. Kickbacks are smoother.

[nige1]

 akhare, on 2013-July-19, 09:57, said:

After pattern resolution (using ~symmetric relays), we play:
3N: To play
4♦: Canonical terminator puppet
S1: QP ask
S2: Always RKC in ♣
S3: Always RKC in ♦
S4: Always RKC in ♥
S5: Always RKC in ♠
For example, after 3♦, showing say a 5=4=3=1 pattern, 3♠ will always be RKC in ♣.
To me, it seems that this treatment is suboptimal, if not downright silly.
IMO, it's much better to play Richard's Moscito treatment, i.e. RKC based on suit length, with ties broken in suit order. In other words, in the above case, 3♠ is RKC ♠, 4♣ is RKC ♥ and so forth. Comments?

To increase your chance of subsiding in 4M/5m, you should consider steps:

• ♥
• ♠
• ♣
• ♦

[Free]

 yunling, on 2013-July-19, 20:39, said:

After 3♥, it sounds very good to use
3♠ QP ask
3N To play
4♣ RKC in ♣
4♦ terminator puppet
4♥ RKC in ♦
4♠ RKC in ♥
4NT RKC in ♠

I disagree that this sounds good in a relay sequence. You commit to 5-level in each strain, while you want to play as low as possible when slam isn't possible (4M or 5m). True that you're no worse off then natural systems, but why settle for something 'acceptable' when you can have something much better?

Moreover, when space allows it, you can even add an extra step in the RKC responses to differentiate between min and max. With Richard we did this when RKC was below 3NT (step 1 = min, rest is max with normal responses).

[yunling]

 Free, on 2013-July-20, 00:24, said:

I disagree that this sounds good in a relay sequence. You commit to 5-level in each strain, while you want to play as low as possible when slam isn't possible (4M or 5m). True that you're no worse off then natural systems, but why settle for something 'acceptable' when you can have something much better?

Moreover, when space allows it, you can even add an extra step in the RKC responses to differentiate between min and max. With Richard we did this when RKC was below 3NT (step 1 = min, rest is max with normal responses).

But then you will have to use 4♠ for ♣ and 4NT for ♦ which I really don't like. Playing in 5M missing 2A is bad, but lacking 2A and being forced to play 6m is horrible.

[Free]

 yunling, on 2013-July-20, 23:03, said:

But then you will have to use 4♠ for ♣ and 4NT for ♦ which I really don't like. Playing in 5M missing 2A is bad, but lacking 2A and being forced to play 6m is horrible.

Why on earth would you RKC opposite a limited opener in one of his short suits without an Ace (or with only 1 Ace)?

[yunling]

 Free, on 2013-July-20, 23:18, said:

Why on earth would you RKC opposite a limited opener in one of his short suits without an Ace (or with only 1 Ace)?

e.g. After p bid 3♥ showing 5-4-1-3, don't you want to ask for ♣ keycards with KX KQX AXX AXXXX? But what if p bids 5♦…

[Free]

 yunling, on 2013-July-21, 01:15, said:

e.g. After p bid 3♥ showing 5-4-1-3, don't you want to ask for ♣ keycards with KX KQX AXX AXXXX? But what if p bids 5♦…

No, with such hand RKC is foolish. I ask for AKQ-points and let partner show his entire hand...

[yunling]

 Free, on 2013-July-21, 04:48, said:

No, with such hand RKC is foolish. I ask for AKQ-points and let partner show his entire hand...

Suspect whether p can show his ♣ holding below 5♣, say,
distinguish
AQJXX JXXX K KXX
from
AQJXX JXXX X KQX

Edited——Sorry that I got the minors reversed.

[Free]

 yunling, on 2013-July-21, 07:30, said:

Suspect whether p can show his ♦ holding below 5♣, say,
distinguish
AQJXX JXXX KXX K
from
AQJXX JXXX KQX X

First of all, your partner showed a 5=4=1=3, not a 5=4=3=1.
Second, singleton K doesn't count for a QP ask, so we can't find out about a stiff King. Usually that's ok, but from time to time it's annoying.
Third, if we swap the minors and try to determine ♣ honors, there's a difference in QP's which makes it very easy to distinguish using denial cuebids.

Suppose we hold Kx KQx Axx Axxxx, the 5413 was shown with 3♦, we ask QP with 3♥ and base level is 6:
AQJxx-Jxxx-K-Kxx
...-3♥
3♠-4♠ (6 QP's ; 4 QP's missing in useful suits is too much, I believe 4♠ is the best contract since 3NT is out of the question and 5♣ needs a trick more)

AQJxx-Jxxx-x-KQx
...-3♥
3NT-4♣ (7 QP's ; relay, 3 QP's missing in useful suits is worth investigating, partner may miss ♣KQ though)
4♥-6♣ (1/2 top honors ♠, 0/3 top honors ♥ ; we know about AQxxx-xxxx-x-KQx with maybe some Jacks on the side)

We don't even need a full round of denial cuebidding and we know waaaaaaaaay more at 4♥ than you'd know after some silly RKC response. 4♣ RKC gets a 4NT response and you know only that partner has 2 keycards with ♣Q, you don't even know which keycard you're missing.

Correct me if I'm wrong, but I have the impression that you've never played QP ask with denial cuebids, otherwise you wouldn't suggest to RKC with a balanced hand.

[benlessard]

Quote

No, with such hand RKC is foolish. I ask for AKQ-points

If ive understood correctly its

Kx KQx Axx Axxxx vs a 5413.

Here for slam to be decent you need 2 out of 3 keycard and the Q of trumps. 3 keycard with Q of trump will give you good play for 7. So how is it possible that a bid that ask for keycard is worse than a bid that ask for more general strenght ? Your point just doesnt make any sense. Fixing the trumps suit higher waste space vs making the cheapest bid, but you will need to setup trumps or make a guess anyway.

[Zelandakh]

My belief is that the calls below 3NT should be used to improve game bidding, filling some of the gaps of relaying that are easily handled by natural systems. I also switch the 4♦ terminator for direct sign-offs, which avoids any annoying lead-directing doubles. So, if the last step was 3♦:

3♥ = relay
3♠ = stopper ask in partner's fragment
3NT = to play
4♣ = puppet to 4♦ (then 4♥/4♠/4NT = RKCB for diamonds/hearts/spades)
4♦ = RKCB for clubs
4M/4NT/5m = natural

Fragment suits are arranged with priority 3 card suits before 2 card suits, ♠ > ♥ > ♦ > ♣ for equal length. The argument for having RKCB asks in the order ♥ > ♠ > ♣ > ♦ looks convincing if you only consider stopping in game but looks dubious when you consider bidding space below slam. If you often use RKCB as a slam try then sure but I think most relayers expect to finish in slam more often than not when they make this relay break so the ♣ - ♦ - ♥ - ♠ order is not such a bad idea after all.

[Free]

 benlessard, on 2013-July-21, 15:31, said:

If ive understood correctly its

Kx KQx Axx Axxxx vs a 5413.

Here for slam to be decent you need 2 out of 3 keycard and the Q of trumps. 3 keycard with Q of trump will give you good play for 7. So how is it possible that a bid that ask for keycard is worse than a bid that ask for more general strenght ? Your point just doesnt make any sense. Fixing the trumps suit higher waste space vs making the cheapest bid, but you will need to setup trumps or make a guess anyway.

Read my post right above yours...

With relay systems QP ask with denial cuebids is like 1000 times better than RKC. You indeed get general strength of the hand first, but you also get a better location of honors very quickly. Moreover, it's very easy to set trumps: bid anything other than step 1.

You claim that slam will be decent if you have 2 keycards with trump Q, but that's not true at all. When you're missing the ♣K as a keycard, you can easily have 2 trump losers. Do you really want to be in slam oposite AQxxx-Axxx-x-Qxx? RKC won't show you that you're missing ♣K, while denial cuebids will.

In my experience QP is much more accurate, while RKC is only better when you know slam is on and when you're looking for a few specific cards (keycards just to check you're not missing 1 or 2 + top honors in at most 1 side suit).

Really, based on the arguments against QP ask and the examples given, it seems like I'm having a discussion with people who don't have a clue what QP ask is about. Please be informed properly before entering a discussion, and reading all of my posts can also be helpful.

[yunling]

 Free, on 2013-July-21, 14:39, said:

First of all, your partner showed a 5=4=1=3, not a 5=4=3=1.
Second, singleton K doesn't count for a QP ask, so we can't find out about a stiff King. Usually that's ok, but from time to time it's annoying.

I count singleton K as 1QP(which is another question, though)so I have to worry whether p holds DQ or CK
For me both hands would go
3♥
3NT-4♣(7QPs)
4♦-4♥(0/2 honors in ♠)
4♠-???(0/2 honors in ♥)

[benlessard]

Every system work well when the correct answer is the cheapest.

AQxxx,Axxx,x,Qxx or
Axxxx,Axxx,x,Kxx become

ending at 3H (like it was in the original question)

3S-4K(8)
4H-5K or 5H (depending if its)

so your in 5D/5K and because trumps isnt set your options are almost over.

If you want we can compare 30-50 slammish hands of your choosing anytime. My previson is that on 5422 DCB will work better but in all other case RKC will work better. This is because the more distribution you have the more likely the RKC in the 2 long suit will dominate DCB.

[Free]

 akhare, on 2013-July-19, 09:57, said:

For example, after 3♦, showing say a 5=4=3=1 pattern

 benlessard, on 2013-July-22, 10:08, said:

ending at 3H (like it was in the original question)

Do I need to say more?