continuations after pattern completion

[straube]

First topic

 

In the context of a strong club relay auction where a positive shows 6 QPs, if pattern is complete at 3H and we have 3S as QP ask and 4D as terminator puppet with the option to super-accept with 9....

 

3S-

.....3N-6

.....4C-7

.....4D-8

.....4H-9

.....4S-10

 

4D-

.....4H-6-8

.....4S-9

 

we're actually ahead by using the QP ask rather than the terminator puppet except for when we want to play 4H when partner has specifically 8 QPs.

 

If instead we allow a super-accept with 8 then

 

3S-

.....3N-6

.....4C-7

.....4D-8

.....4H-9

.....4S-10

 

4D

.....4H-6-7

.....4S-8

 

Now we're just way ahead using step one and not the terminator puppet. It also frees up the terminator puppet for something else, possibly a RKC ask.

 

Second topic....

 

When relaying a semipositive (say Moscito 3-5 QPs) and shape is shown at 3S or 3N, is it better for...

 

4C-QP ask

4D-terminator puppet

4H-RKC longest/highest

4S-RKC next

etc

 

or

 

4C-QP ask

4D-terminator puppet

4M-optional RKC ask

5m-slam invite

 

I'm interested in 4M here as invitational because there is less likelihood for slam than after a positive and hence no super-accept allowed (typically) after the 4D terminator puppet and you don't have

room then for responder to show extra.

 

Third topic

 

After relaying a semipositive (say Moscito 3-5 QPs) and you are using RKC asks, has anyone tried

 

S1-0 key cards

.....may ask for Q

S2-1 key card, no Q

S3-1 key card and Q

S4-2 key cards

[straube]

I'll stick this in here, too. This is use of 4C after a TOSR relay auction to announce a void by asker.

 

"4c exclusion relay:

When responder has finished showing his shape with a bid of 3h or lower, a 4c continuation by opener shows a void and asks responder to bid 4d, after which opener will bid 4h, 4s, 4nt or 5c to show a void in responder’s longest, second, third or shortest suits (ties broken in the usual way, highest first) and ask for controls outside of the excluded suit. The responses are: 0 or 1, 2, 3, and so on, after which opener can relay for DCB in the usual way, but always excluding that suit. (So if responder has a short suit somewhere else, the DCB only covers two suits!)

 

There are a number of important notes about this kind of ask, so let’s jump right in.

 

This is basically forcing to the five-level at least, so opener must have a very good hand to do it: not just any old hand with a void and a fit. Also, opener will usually have another suit with neither of the top honors; otherwise he could just relay normally and expect to figure out where responder’s cards are. In short, as opener be extremely careful about using this toy. Even if it doesn’t lead to immediate disaster it may be inferior to the regular controls+DCB route."

[Kungsgeten]

Couldn't you put a two point interval in the first step? We did this when playing DCB (but we used controls and not QP). So something like this:

 

3H-3S;

3N = 6 or 7 QP

...4C = Relay

......4D = 6 QP, 4H now initiates DCB

......4H+ = Zooms into DCB with 7 QP

4C = 8 QP

4D = 9 QP

etc

 

This gets you higher with 6-7 QP, but opener can still sign off even though responder has 8 QP.

 

I like the idea of using a modified RKC vs a semi-positive.

[straube]

Couldn't you put a two point interval in the first step? We did this when playing DCB (but we used controls and not QP). So something like this:

 

3H-3S;

3N = 6 or 7 QP

...4C = Relay

......4D = 6 QP, 4H now initiates DCB

......4H+ = Zooms into DCB with 7 QP

4C = 8 QP

4D = 9 QP

etc

 

This gets you higher with 6-7 QP, but opener can still sign off even though responder has 8 QP.

 

I kind of like separating these. With 8 opposite 10+ and with a fit (captain can bid 3N without) our chances at the 5-level or even 4N seem pretty good. Maybe if I get burned too many times...

 

I like the idea of using a modified RKC vs a semi-positive.

 

Thanks

[rbforster]

The super accept with 9+ QPs works a lot better over 3N than it does over 4D as you observe. As a consequence, you basically always ask for QPs when the shape resolves below 3S and you plan on playing in a suit. Then the question is what should you use 4D for? Maybe the void ask or some other gadget, at least if there are a couple steps still below it available to ask for QPs.

[straube]

The super accept with 9+ QPs works a lot better over 3N than it does over 4D as you observe. As a consequence, you basically always ask for QPs when the shape resolves below 3S and you plan on playing in a suit. Then the question is what should you use 4D for? Maybe the void ask or some other gadget, at least if there are a couple steps still below it available to ask for QPs.

 

So you're using 9? We've been using 8 lately. I guess 8 is a little optional (we're obviously not forced to super-accept) but it seems like especially with a fit and a minimum of 18 QPs that the 5-level would be safe. Aside from the risk of getting too high, we lose a step here for PCB. OTOH, it takes a little pressure off of opener to ask QPs using 4C. Any thoughts? We had been missing some slams with 9 as our super-accept. I think TOSR requires 10 so I think they or we must be wrong here.

[straube]

Thinking for the semipositives to RKC ask in the order H/S/C/D regardless of shape. Then

 

pattern at 3H

 

.....3S-QP ask

.....3N-to play

.....4C-RKC hearts

..........4D-0

..........4H-1

..........4S-2

..........4N-1 and Q, no side K

..........5C-1 and Q, side K

.....4D-terminator

.....4H-RKC spades

..........4S-0

..........4N-1

..........etc

.....4S-RKC clubs

.....4N-RKC diamonds

 

It gets better for pattern ending at 3C or 3D.

 

 

Also thinking to use 9 for super-accept of positives and 13 for super-accept of strong club which are +3 whether captain has bid 3N or 4D terminator (which applies at pattern end of 3S or 3N only) and use the same

4M and 5m optional RKC ask/slam invite that we do for the semipositives already. The target of 4M would be primarily 8 QP hands. What ought 5m ask for though? There is room to look for 8 QPs there so it seems like it should ask for something specific...like "We're missing 2 key cards and I don't have space to find that out". Any suggestions or thoughts on this?

[kenrexford]

The idea of a modified RKC ask after a semi-positive is at the heart of my Variable Key Card Blackwood. I believe this to be the innovation of the future. That said, I think you could inprove on your structure somewhat if goong in that direction.

 

VKCB employs a concept of wrapping around. If your response is the highest possible, for example 2 in your structure, dedicating one bid to show that, and then requiring one bid to ask the next question, before any answers wastes the two most productive steps. Better to have the person answering go on to the next question on his own.

 

Thus, suppose the 2 answer only leaves room for a Jack. +4 might be 2 with the lowest Jack, +5 as 2 with the second lowest Jack, etc. Skip to slam with no Jack, perhaps.

 

Often, +4 is low enough that you can bump ip one. Then, +5 actually starts, in the example, the specific Jacks. You can then make room to ask for more lower. For example, VKCB allows for discovery of next-most-interesting side cards. If no trump Queen, for example, what about the King or Queen in your longest non-trump suit. Stuff like that.

[straube]

Ken, I'm sure you're right.

 

Different topic but does anyone use an abbreviated cue bid scheme if QPs are revealed too late? Say pattern ends at 3S showing a 6430 and after a terminator 4D, RR breaks the puppet and shows 9 QPs at 4S.

We're pretty well placed just knowing pattern and QPs, but there isn't much room for locating cards.

 

Should we continue to use PCB here (or DCB), especially as there might still be room for a grand? Should we just ask # aces? Focus on the longest suit? Does anyone use two or more schemes depending at what point QPs are shown?

[rbforster]

I think the original TOSR notes switched to straight ace asking with the first step (rather than QP ask) if the level of shape resolution was too high. Maybe it was if the ask was 4C or 4H, don't remember.

 

I do think there's some improvement to be made with QP showing schemes, specifically when you want to play in a suit, since there's a lot more space. That said, I'm not sure what the best way is to go about that and there are lots of variations - ignore the terminator and have direct game bids be optional sign offs (where opener answers QPs or maybe RKC in that suit with a very good hand, else pass), weak vs strong relay asks for QPs using the first two free steps (so they start at different levels for how much the first step "min" response showed, maybe for the strong one it goes 5/6/7/8... But for the weak one it's 5-8/9/10...), always asking for QPs so you free up cheaper RKC bids in all the suits if opener would rather have that option than a QP ask, having higher RKC asks only for better hands so the extra bids below 3N besides the first step QP ask can be stopper asks, etc, etc.

 

At some point, I think having any comprehensive, memorable relay asking system for honors and/or key cards gets 90% of the improvement over standard methods.

[straube]

PCB takes up to two whole levels if you get a series of even responses. Practically I think I'd want to find QPs by 4C or 4D to mostly comfortably use it.

 

I've been noticing how often (I'd say greater than half the time) there is a slam if responder has 8 QPs to opener's minimum strong club, so I think I don't mind responder removing from 3N with 8. OTOH, if opener uses a 4D terminator, 9 would be shown at 4S (if we super-accept starting 9 in that case) so paradoxically, with a fit (opener uses 4D with a fit) we're actually on a worse start than if we had no fit or a minor suit fit and tried to sign off in 3N.

 

Adam has remarked that knowing number of QPs is often enough to decide whether to slam or not, and maybe knowing 9 at 4S is sufficient. Still, avoiding 4D terminator would be nice and if we have to use it, deciding whether PCB is the right continuation or not is something to think about. Maybe DCB is better there because at least you know something even if you can't scan every suit or scan every suit a second time. Maybe it isn't.

 

Were you thinking of low level terminators?

 

For example, say shape is shown at 3H.

 

3S-terminator puppet

.....3N-5-7

..........4C-re-ask

.....4C-8

.....etc

 

That seems to work. Try for shape shown at 3D.

 

3H-terminator puppet

.....3S-5-7

..........3N-to play?

.....3N-8

 

That's not as good.

 

3H-terminator puppet

.....3S-8

..........3N-asks

.....3N-5-7

.....4C-9

 

So we don't want to remember exceptions for every shape resolution step before 3N. Maybe 3N always shows the weak group.

[shevek]

Never liked or used 4D end signal.

Partly a memory issue. A jump to game before shape is completely resolved is natural.

Dicey if you switch to RKC just because shape is out.

 

We use 4D as a mild slam try.

When shape is out but strength unknown, 4D says "Bid 4H with any minumum, bid bad trump suits upwards with a max" (Bid 4NT with a resurrect)

Max is loosely defined as +2 SPs or +1 Kontrols.

 

Sometimes you will have AKxxx opp xxx. Takes ages to track down whether one of those xxx is the queen. 4D does this

Say partner's 3-2-6-2 positive comes out at 3S (See other thread!)

 

AKJxx, AKxx, Kx, xx

 

Qxx, xx, Axxxxx, Ax

 

would be nice, if marginal. Bid 4D over 3S to check it out.

Then 4S over 4H (or pass 4S) Responder continues with actual. (And can reach grand with D-AQxxxx as well)

Note that 4C strength ask is pretty useless.

[straube]

What's a "resurrect"?

[straube]

I'm guessing 4N would just say "bad hearts". I kind of like this idea but would like to know more about it. In a sense you're scanning suits, stopping in a suit with nothing in it. Maybe not the best for grand slam bidding? Which is a minor consideration.

 

So I suppose opener's 4M shows a minimum and then optional RKC responses over that?

 

What if instead opener bid 4N or 5m. 5m a minimum? 4N?

[shevek]

What's a "resurrect"?

 

Continue after sign-off.

If shape is out & asker bids - say - 3NT, describer resurrects with (a good) base +4 SPs, so 9 for a positive to 1C.

[shevek]

I'm guessing 4N would just say "bad hearts". I kind of like this idea but would like to know more about it. In a sense you're scanning suits, stopping in a suit with nothing in it. Maybe not the best for grand slam bidding? Which is a minor consideration.

 

So I suppose opener's 4M shows a minimum and then optional RKC responses over that?

 

What if instead opener bid 4N or 5m. 5m a minimum? 4N?

 

Describer doesn't know the trump suit. Say bids 4H and asker then bids 5D. This says "Bid on with extras including good diamonds". (Where "good" is defined as Hx, Hxx, HHxx, etc.

Think of it as natural Denial Cue Bidding.

 

after 4D

4H = min or extras with bad hearts

4S = extras, good Hs, bad Ss

4NT = 9+ SPs

5C = extras, good majors, bad clubs ....

 

Can also use it after SPs known. Then functions as a trump ask.

[straube]

Thanks a lot. That makes a lot of sense. I suppose with a good enough hand, responder could bid 4N, learn the trump suit and bid a grand. 11 QPs. Possibly 10?

 

Seems like there could be a potential for a conflict. Say I want clubs to be trump and the slave hand indicates bad spades (bids 4S) and I realize that we have two quick losers in the suit. I continue to 5C and partner having good clubs raises to 6. Is that just a negligible occurrence or do you have a way of putting on the brakes? Like opener bids 4N as a puppet....

[straube]

Throwing this in the mix...

 

4.3. Relay Termination.

 

The 3♠/NT things always happens in strong club openings. 3♠ / NT

This is triggered (A) by the relays stopping at 3♠, or (B) a bid of 3♥ if the relays have stopped lower.

With 8 − 11/15+ points you bid 3♠, with 12 − 14 you bid 3 NT. After either of these bids 4♦ is the Puppet Terminator (see below), 4♣/♥/♠/NT are RKB in partners longest/next longest/etc. suit (ties broken in order ♣/♦/♥/♠), see below for continuations. 3NT over 3♠ is to play.

With 15+ you bid 4♣+ over a 3N T sign off/ 4♠+ over a 4♦ puppet showing 5−, 6, 7 . . . controls after which new suits are RKB in length order again, with 5NT as an attempt to sign off in a small slam (puppets 6♣).

 

The 3♠ puppet

The major alternative to the scheme above is to go straight into RKB / sign off / make a quan-

tative slam try. This isn’t always available (relays reach 3♠/NT) but normally is. At this point 3♠ puppets 3NT after which :

4♣ is RKB for partner’s “third” longest suit,

4♦ is the puppet terminator (which triggers 4♥ bid),

4♥/4♠/4N T /5♣/5♦ are mild slam tries to be accepted by a decent 12 − 14 point hands. These set trumps and after 4♥ − N T keycards (controls 4−, 5, . . . ) are shown when acccepting. After acceptance it is possible to play in a lower ranked suit slam (to facilitate 4 − 4 minor suit slams where you otherwise wish to play 4 of a major).

At the same time 3NT/4♣/4♦ direct are RKB for partner’s longest / second longest / shortest suits. If 3♥ is available then it is used for range as specified above. If 3♦ is available it acts as RKB

for the longest suit with 4♣ via 3♠ keeping its place as RKB for the third longest suit and 4♣ direct being RKB for the shortest suit.

N.B. 3♦ here will nearly always be range RKB. Plenty of space to sort it all out though :)

If you aren’t giving range and stop early then 3♠ puppet is still on but all other are RKB in normal length order including 3♥.

[straube]

When pattern ends at 3S or 3N...

 

4C-QP ask

4D-terminator puppet

.....4S=base+3

4M-nf, invites RKC responses

.....reply=base+2

 

Apparently many use 5m as invitational, but if general strength is desired, but if base + 2 is enough why not use 4C QP ask?

 

Then we can use...

 

4N-ace asking

5m-nf, key card asking

 

The next question is what the responses should show.

 

For 4N

...............5C=0 for a semipositive, 0 or 1 for a positive, and 0, 1, or 2 for a strong club

...............5D=1 for a semipositive, 2 for a positive, and 3 for a strong club

...............5H=2 for a semipositive (not actually possible for us), 3 for a positive, and 4 for a strong club

 

For 5m

...............P=0 for a semipositive, 0 or 1 for a positive, and 0, 1, or 2 for a strong club

...............raise=1 for a semipositive, 2+ for a positive and 3+ for a strong club

 

Any opinions? Tweaks?

[Kungsgeten]

We currently do not use QP/controls after our relays, but use RKC instead. However when shape is resolved opener tend to know if responder is min/max (we only play full relays after our unbalanced 1D so there's plenty of space, and also over our 15-17 1NT where opener is very limited). We also have some other relay methods, but then the full shape is not resolved. I've been thinking of using a variation of Mulberry, especially if the bidding ends at 3S or 3NT:

 

4C = Puppet to 4D. RKC in one of the three longest suits, or a maximal slam try.

...4D

.....4H/S/NT = RKC longest, second longest, second shortest

.....5X = Natural maximum slam try. Responder usually accepts.

4D = End signal or RKC in partner's shortest suit.

...4H

......Pass/4S/5m = To play

......4NT = RKC in responder's shortest suit.

4M = Natural slam try.

4NT = Quant

5m = Natural slam try.

 

I'm not sure how we would modify it if the pattern stops below 3S (which it usually does) though. Probably we won't change the method since the relays are rather cumbersome to remember already, and not that frequent.

[straube]

Thanks Kungsgeten. Concerning Mulberry, I found this post

 

 

David Yates

In a relay auction, if 4♣ is available, I play:

 

4♣ puppets to 4♦ to start a sign-off or make a natural SI in NT

—> 4♦ then 4M/5m to play 4NT SI to 6NT

 

Responder can break the puppet with about a king or ace extra for his previous bidding by bidding 4H. Ask will then either pass, corrects, pick a slam or asks responder to pick a slam (5N). If responder has even more than that, he puppets - waiting to find out what trump is (4♦->?). Then after the ‘sign-off’ responder will then take control with a ‘step’ bid as RKC .

 

4♦ is SI in hearts (if possible, I play if responder has 3+) else it is in his longest minor

4♥ is SI in spades (if possible) else his 2nd longest minor

4♠ is RKC in 3rd longest suit

4!NT is RKC in shortest suit

 

If suits are equal length, priority goes to lowest. So if responder is 2263, 4♦ is invitational RKC in ♦, 4♥ is invitational RKC in clubs 4S is RKC in hearts and 4NT is RKC in spades.

 

Invitational RKC (IRK in our notes) has immediate step of ‘no interest’ with relay demanding KC count over an attempt to sign-off. The 3+ for M requirement is a function of the bidding style. Ask (strong club bidder) would normally show a 6CM and not relay over a positive unless very strong or a fit for responder’s suit. With normal range hands and a show, a natural auction would have set the trumps earlier and subsequent bidding would be turbo assisted cue bidding and not RKC.

 

I do not know enough about the alternative method to contrast benefits/disadvantages. I put this together maybe 25 years ago. Played it some - when I play ‘seriously’ - and never had a problem. Some of my friends & partners play it and not a problem. From a theoretical standpoint, since ASK is usually going to be declarer, the 4♣ -> 4♦ puppet for the sign-off does give the opposition a potential lead director. From a practical matter, I cannot recall this

[nullve]

My somewhat Mulberry-like structure when shape resolution is almost complete and the range shown is 2 (Bergen) points wide:

 

http://www.bridgebase.com/forums/topic/72952-2c-gf-relay-response-in-4cm/page__pid__870516#entry870516

 

(post #8)

[shevek]

Thanks a lot. That makes a lot of sense. I suppose with a good enough hand, responder could bid 4N, learn the trump suit and bid a grand. 11 QPs. Possibly 10?

 

Seems like there could be a potential for a conflict. Say I want clubs to be trump and the slave hand indicates bad spades (bids 4S) and I realize that we have two quick losers in the suit. I continue to 5C and partner having good clubs raises to 6. Is that just a negligible occurrence or do you have a way of putting on the brakes? Like opener bids 4N as a puppet....

 

 

In practice, 4D is a trump ask, so not used to find a good holding in suit x in order to play slam in y.

With that, you are stuck with SPs & DCB.

I think we play 4D - 4M - 4NT as natural invite.

[straube]

Looking at this. It's a windfall for the resurrects/super-accepts at the expense of the QP ask.

 

If shape resolved at 3H (or deduct steps for 3D or 3C)

 

3S-terminator puppet, usually no trump will have been bid by now but occasionally the slave hand will have to declare NT

3N-QP ask

4C-RKC H

4D-RKC S

4H-RKC C

4S-RKC D

 

If shape resolved at 3S/3N

 

4C-QP ask

4D-terminator puppet, partner to super-accept with base +3

4M-invites RKC with base +2

4N-ace ask, partner to bid 5C with a predetermined base, else one step for each additional

5L-invites the suit, asks a raise with more RKC cards than base

 

Any comments? Good or bad?

[straube]

Tested it a bit and so far I like it. I'll contrast what we used to do and what we're trying out now. Say 3415 is shown at 3D. Our former scheme...

 

3H-QP ask

.....3S-6

.....3N-7

.....4C-8

.....etc

3S-RKC C

3N-to play

4C-RKC H

4D-terminator

4H-RKC S

 

Present scheme...

 

3H-terminator

.....3S-6 or 7

.....3N-8

.....etc

3S-QP ask

.....3N-6

.....4C-7

.....4D-8

.....etc

3N-RKC H

4C-RKC S

4D-RKC C

4H-RKC D

 

Maybe the most obvious downside is that for the QP ask we have lost a step.

 

What I've found though is that we used to ask QPs (when space allowed) as a courtesy ask. Maybe we have a minimum hand, but partner has a good hand? Save space then, especially if our alternative action is the 4D terminator puppet. In practice, when pattern is resolved at low levels, the terminator puppet is wasting a bid because it isn't used.

 

Now we very/most frequently use the step 1 terminator with those same minimums but in these instances partner's super-accept saves us a step. In the example above, 8 is shown at 4C the old way but 3N the new way.

 

With the old way, if we asked with 3H and partner had base (6), he would respond 3S and now 3N was to play and 4C asked cue bidding. So we break even on our ask when partner has 6 because our ask here is 3S and partner bids 3N to show 6.

 

When now we use our QP ask, it obviously shows more interest in slam, some combination of good fit or extra values. On average then, the slave hand will tend to have a minimum hand (fewer QPs likely available). If he bids 3N or 4m here we will still have a pretty effective cue bidding auction.

 

Having 3N as a RKC ask and combining it with the suit order H/S/C/D means that we usually have space for meaningful RKC auctions before the trump suit has been passed.

 

We haven't tried signing off in 4m yet, but I think this will be a very rare option after the terminator. It's a small side benefit.