our 2C structure

[Zelandakh]

You can have stop asks over major suit transfers.

 

After transferring, if pard does not accept the transfer, he'll often rebid 2NT or 3♣. At that point, 3♦ becomes the stop ask.

 

If acceptance of the xfer, 2NT becomes stop asking, with emphasis on major suit fit first.

The problem with that approach is that you lose out on invitational hands with a 4 card major (with the 2NT bid). You also have a problem after a super-accept in that the usual solution with hearts is to use 3D as a re-transfer to cater for both weak and GF hands. With spades you obviously have 3H for that so 3D usually either shows hearts or, if you bid that hand through 2D, is natural with 5+ diamonds. It is arguable which is better but you are always giving something up here.

[keylime]

Keep in mind, that our 2♣ denies a four card major; coupled with transfers it allows much more flexibility in the bidding.

[Zelandakh]

Keep in mind, that our 2♣ denies a four card major; coupled with transfers it allows much more flexibility in the bidding.

That sounds like Unassuming Club style but even there 3D is used as a re-transfer to hearts. I guess my question here is that if 2C - 2D - 2N - 3H shows a weak hand then how do you make an invite? or show a GF hand with hearts? There just do not seem to be enough bids left over. And if a 3H rebid does not show the weak hand then how do you show it instead?

[keylime]

Invite hands run through 2NT. 2S is our diamond hand. 2D/H are xfers into the corresponding majors.

[straube]

Invite hands run through 2NT. 2S is our diamond hand. 2D/H are xfers into the corresponding majors.

What do you do with GF hands that don't have a suit? I'm thinking that 2D needs to be an asking bid of some sort. We've already established that opener has 6 clubs. Why reverse gears over that in trying to show responder's pattern? Just finish showing opener's pattern? Unless, of course, responder is very distributional.

[keylime]

GF hands without a suit would simply bid game. Since our 2♣ denies a major, xfer acceptance shows either a doubleton or tripleton.

 

FWIW, I have played prior the sequence 2C-2M forcing, and found that as long as it promised six clubs itself, it was a reasonable system. I'm not a fan of 2C being 5C, 4m admittedly (rather open that 1♦ or Precision 2♦/♥.

[straube]

Anyone feel that hiding the GF 5M/5 and the GF 6M hands in the 2D relay is unplayable?

 

For this we can...

 

show the GI 5M/5 hands and stop at the 3-level and show GI 6M with a fit and GI 6M w/out a club fit.

 

We can bid our major when GF with a major and club fit or GF and desire to play our major or 3N.

 

We can keep 2N as a constructive raise (as opposed to multi-meaning) and can use a direct 3C raise as a blocking (weak) bid.

 

I know that Meckwell uses 2N as the weak raise OR GF 5M/5 hands and they use 3C as a puppet to 3D which they can then remove to 3M to show other hands.

 

But, Meckwell has no constructive club raise. They have to use 2D which is GI and doesn't separate constructive from weak very well. Actually, they don't try (they differentiate opener's strength into 3). Also doesn't prepare partner for a competitive auction.

[awm]

Sam and I actually play that 2M...3X is forcing and that a direct 3M is a GF one-suiter. This gives us no problem with the GF hands, but obviously creates some issues with the 5/5 invites (I don't see fit jump as being so hugely necessary, but I suppose we could miss out on those too). In practice I haven't found that 5/5 exactly invitational is that common a hand type, and if those do come up I usually just overbid to game. :)

 

In terms of relaying, my feeling is that the GF one-suiter is probably okay to relay. The GF two-suiter can be annoying when you don't have a good fit, because you really need partner to have the fourth suit locked up and may not have the methods to ask for that particular stopper. Of course, there is some tradeoff between having issues on the GF two-suiter opposite no obvious fit, versus having issues on the INV two-suiter opposite no obvious fit; my view is that the GF variety is more common and also more critical to get right.

[straube]

Sam and I actually play that 2M...3X is forcing and that a direct 3M is a GF one-suiter. This gives us no problem with the GF hands, but obviously creates some issues with the 5/5 invites (I don't see fit jump as being so hugely necessary, but I suppose we could miss out on those too). In practice I haven't found that 5/5 exactly invitational is that common a hand type, and if those do come up I usually just overbid to game. :)

 

In terms of relaying, my feeling is that the GF one-suiter is probably okay to relay. The GF two-suiter can be annoying when you don't have a good fit, because you really need partner to have the fourth suit locked up and may not have the methods to ask for that particular stopper. Of course, there is some tradeoff between having issues on the GF two-suiter opposite no obvious fit, versus having issues on the INV two-suiter opposite no obvious fit; my view is that the GF variety is more common and also more critical to get right.

 

 

I have to admit that the FSJs don't seem to come up a lot in simulation. I do seem to come across the GI 5/5 hands and it's because of concern for not finding a fit and not having the 4th suit stopped that I kind of like having the ability to stop shy of game with these.

 

If I didn't need 3D as a sign off, I could use 3D, 3H, and 3S for the GF 5/5s. Or possibly these could break relay...

 

2C-2D,

.....2H

..........3D-two lowest

..........3H-highest and lowest

..........3S-two highest

[Zelandakh]

We've already established that opener has 6 clubs. Why reverse gears over that in trying to show responder's pattern? Just finish showing opener's pattern? Unless, of course, responder is very distributional.

I think you are missing the point about transfers. They are not about showing your shape so much as asking Opener about their hand in targeted ways. Think about a 1NT opening for a moment. Here Opener has shown 8 of their cards, with the other 5 highly restricted in their placement. Yet transfers work well. Mostly it is simply a question of asking Opener if they have 2, 3 or 4-5 cards in the relevant suit. The same principle works rather nicely over a precision 2C opening too, whether that be a modern 6+ clubs or a traditional 5C-4M type.

 

 

Anyone feel that hiding the GF 5M/5 and the GF 6M hands in the 2D relay is unplayable?

I showed you already a way to make including the 5-5 GF 2-suiters possible within your opening franework. It is a matter of how you construct your relay system. If you prefer showing shortage then yes, it is best to take this hand out since you cannot find fits so easily. Unlike awm I think including the 6M hands into 2D is more difficult. If I only had room to take one of these hand types out of 2D then it would be the 1-suiters.

 

 

I have to admit that the FSJs don't seem to come up a lot in simulation. I do seem to come across the GI 5/5 hands and it's because of concern for not finding a fit and not having the 4th suit stopped that I kind of like having the ability to stop shy of game with these.

 

If I didn't need 3D as a sign off, I could use 3D, 3H, and 3S for the GF 5/5s. Or possibly these could break relay...

I am a little confused why you are worrying so much about your 5-5 GI hands when you seem to have a perfectly good way of handling these through your forcing 2M responses. If you are wanting to use the 3M responses for something other than FSJs then natural and game-forcing with a 6 card suit springs to mind as something obvious. Then your 5-5 GF hands must either relay or go via 2NT. Given your preferred relay structure the latter is probably better. We are now missing a GF hand with diamonds of course. I think I would need to see where you have gotten to to see where we might slide that in below 3NT - it might even be that that just is not possible without giving up something more important.

[straube]

The latest...

 

2D-relay

.....2H-single-suited

..........2S-GF relay

..........2N-natural, invitational

..........3C-natural, invitational

..........3D-5/5, two lower

..........3H-5/5, upper and lower

..........3S-5/5, two higher

.....2S-4 diamonds, minimum

.....2N-4 diamonds maximum

.....3C-4 spades, minimum

.....etc-4 spades, maximum

2H-5H, f

.....2S-4 spades, minimum or medium

..........2N-12-13

..........3C-nf

..........3D-5D, GI

..........3H-6H, GI

..........3S-4S, GI

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

..........4D-splinter in support of clubs

.....2N-weak

.....3C-medium

.....3H-minimum

.....other-maximum, natural

2S-5S, f

.....2N-weak

..........P-12-13

..........3C-nf

..........3D-5D, GI

..........3H-5H, GI

..........3S-6S, GI

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

..........4D-splinter in support of clubs

..........4H-splinter in support of clubs

.....3C-medium

.....3S-minimum, fit

.....3L-maximum, natural

2N-constructive raise (fit)

3C-weak raise

3D-to play

3H-forcing with diamonds

3S-LR, forcing to 3N or 4C

3N-to play

[straube]

Zelandakh,

 

Thanks for your post.

 

I'd be interested in seeing how you use transfers (continuations and such). I'm skeptical of them and list my reasons a little later.

 

I've used your idea for 2C-3S as a good club raise. Now 2C-2N is a constructive raise and 2C-3S is better in terms of points and fit.

 

If I'm using 2D to relay, then I'd rather focus on shortness. Your relay structure separates max/min and avoids the problem that we have with 7321s in knowing whether opener has 3-card support for responder's major. OTOH it loses clarity on opener's shortness which may be useful for slams and most important it is another thing we'd have to memorize (and we are up to our neck in relays).

 

I haven't seen how your transfer proposal would work, but I'm thinking that our 2C opening should be handled quite different from NT openings. Even NT openings use 2C (usually) as an asking bid and allowing 3 responses. It's only the next two bids that are transfers. I think it's very likely that our S1 should be a relay bid and not a transfer. Transfers over NT work well because opener is known to have tolerance for the suit and responder may big again to describe his shape to opener (who having a balanced hand is best placed to judge how the hands are meshing). Transfers over 2C have no assurance of a fit. Transfers also restrict sequence variety because opener will frequently have to just accept the transfer. I.e. transfers and puppets are costly in terms of sequence utilization. Finally, transfers describe responder's hand to opener; if opener is balanced well and good, but if opener is unbalanced, then it would have been much more efficient to have let opener complete his pattern.

 

I'm concerned about the 5/5 GI hands because having looked at some hands, they do seem to come up. If opener has a fit, they can be worth a lot. If opener doesn't, then even at the 3-level we may be too high. Frequently these 5/5s lack a stopper in the 4th suit so 3N may be unplayable. I think it's worth keeping the ability to stop at the 3-level with these.

[Zelandakh]

Zelandakh,

 

Thanks for your post.

 

I'd be interested in seeing how you use transfers (continuations and such). I'm skeptical of them and list my reasons a little later.

 

I've used your idea for 2C-3S as a good club raise. Now 2C-2N is a constructive raise and 2C-3S is better in terms of points and fit.

 

If I'm using 2D to relay, then I'd rather focus on shortness. Your relay structure separates max/min and avoids the problem that we have with 7321s in knowing whether opener has 3-card support for responder's major. OTOH it loses clarity on opener's shortness which may be useful for slams and most important it is another thing we'd have to memorize (and we are up to our neck in relays).

 

I haven't seen how your transfer proposal would work, but I'm thinking that our 2C opening should be handled quite different from NT openings. Even NT openings use 2C (usually) as an asking bid and allowing 3 responses. It's only the next two bids that are transfers. I think it's very likely that our S1 should be a relay bid and not a transfer. Transfers over NT work well because opener is known to have tolerance for the suit and responder may big again to describe his shape to opener (who having a balanced hand is best placed to judge how the hands are meshing). Transfers over 2C have no assurance of a fit. Transfers also restrict sequence variety because opener will frequently have to just accept the transfer. I.e. transfers and puppets are costly in terms of sequence utilization. Finally, transfers describe responder's hand to opener; if opener is balanced well and good, but if opener is unbalanced, then it would have been much more efficient to have let opener complete his pattern.

 

I'm concerned about the 5/5 GI hands because having looked at some hands, they do seem to come up. If opener has a fit, they can be worth a lot. If opener doesn't, then even at the 3-level we may be too high. Frequently these 5/5s lack a stopper in the 4th suit so 3N may be unplayable. I think it's worth keeping the ability to stop at the 3-level with these.

 

I gave the immediate responses further up but here's a more detailed run-through of what I use.

 

Over 2C (traditional precision shapes but 10-14 strength)

2D = 4+ hearts

2H = 4+ spades

2S = range ask

2N = 5 spades, 4 hearts, INV

3C = 3+ clubs, PRE

3D = sign-off

3H = 6+ diamonds, SI

3S = 3+ clubs, SI

3N = sign-off

4C = 4+ clubs, PRE

4D = puppet to 4H

4H = puppet to 4S

 

 

Over 2C - 2D

2H = 0-2 hearts (see below)

2S = 3 hearts, 4 spades (NB: therefore 4=3=1=5 or 4=3=0=6)

.....2N = 4 hearts, 0-3 spades, INV

.....3C = 4 hearts, 0-3 spades, 3+ clubs, INV

.....3D = puppet to 3H, 5+ hearts, weak or GF

.....3H = 5+ hearts, INV

.....3S = 4+ spades, INV

.....3N, 4S = sign-off

.....4C = 3+ clubs, SI

..........4D = decline slam try

...............4H = RKCB

..........4H = accept slam try, 1 or 4 key cards

..........4S = accept slam try, 0 or 3 key cards

..........4N = accept slam try, 2 or 5 key cards without CQ

..........5C = accept slam try, 2 or 5 key cards with CQ

.....4D = 6+ diamonds, SI

.....4H = slam try agreeing spades

..........4S = decline slam try

...............4N = RKCB

..........4N = accept slam try, diamond void

...............5C = RKCB

..........5C = accept slam try, 4=3=1=5, 1 or 4 key cards

..........5D = accept slam try, 4=3=1=5, 0 or 3 key cards

..........5H = accept slam try, 4=3=1=5, 2 or 5 key cards without SQ

..........5S = accept slam try, 4=3=1=5, 2 or 5 key cards with SQ

.....4N = INV

2N = 3 hearts, 0-3 spades, min

.....3C = 4 hearts, 2+ clubs, SO

.....3D = puppet to 3H, 5+ hearts, weak or GF [after 3H: denial cues, frivolous 3NT]

.....3H = 5+ hearts, INV

.....3S = 5+ spades, GF

3C = 3 hearts, 0-3 spades, max

.....3D = puppet to 3H, 5+ hearts, weak or GF [after 3H: denial cues, frivolous 3NT]

.....3H = 5+ hearts, INV

.....3S = 5+ spades, GF

.....3N/4S = sign-off

.....4C = 3+ clubs, SI

..........4D = decline slam try

...............4H = RKCB

..........4H = accept slam try, 1 or 4 key cards

..........4S = accept slam try, 0 or 3 key cards

..........4N = accept slam try, 2 or 5 key cards without CQ

..........5C = accept slam try, 2 or 5 key cards with CQ

.....4D = 5+ hearts, 0-1 diamonds

.....4N = INV

3D = 4 hearts, min

.....3H = sign-off

.....3S, 4m = denial cue

.....3N = slam try

.....4S = RKCB

3H = 4 hearts, max

.....3S, 4m = denial cue

.....3N = slam try

.....4S = RKCB

 

 

Over 2C - 2D / 2H

2S = 4+ spades, F1

.....2N = 0-3 spades, min

..........3C = sign-off

..........3D = checkback

...............3H = 2 hearts

...............3S = 3 spades

...............3N = diamond stop

..........3H = 6+ hearts, GF

..........3S = 5+ spades, GF

..........3N = to play

..........4C = 3+ clubs, SI

.....3C = 0-3 spades, max

..........3D = checkback

...............3H = 2 hearts

...............3S = 3 spades

...............3N = diamond stop

..........3H = 6+ hearts, GF

..........3S = 5+ spades, GF

..........3N = to play

..........4C = 3+ clubs, SI

.....3D, 3H = 4 spades, splinter

.....3S = 4 spades, min

.....3N = 4 spades, max

2N = 0-3 spades, INV

3C = 3+ clubs, INV

3D = 5+ diamonds, GF

3H = 6+ hearts, INV

3S = 6+ hearts, 0-1 spades, GF

4C = 3+ clubs, SI+

4D = 6+ hearts, 0-1 diamonds, GF

4H = mild slam try

4S = RKCB [unnecessary because of 2C - 4D - 4H - 4S]

4N = INV

 

 

Over 2C - 2H

2S = 0-2 spades

.....2N = INV

.....3C = 3+ clubs, INV

.....3D = 5+ diamonds, GF

.....3H = 5+ hearts, INV

.....3S = 6+ spades, INV

.....3N = sign-off

.....4C = 3+ clubs, SI+

.....4D = 6+ spades, 0-1 diamonds, GF

.....4H = 6+ spades, 0-1 hearts, GF

.....4S = mild slam try

.....4N = INV

2N = 3 spades, min

.....3C = 4 spades, 3+ clubs, SO

.....3D = 5+ diamonds, SI

.....3H = puppet to 3S, 5+ spades, weak or GF [after 3S: denial cues, frivolous 3NT]

.....3S = 5+ spades, INV

.....3N = to play

.....4C = 3+ clubs, SI

..........4D = decline slam try

...............4H = RKCB

..........4H = accept slam try, 1 or 4 key cards

..........4S = accept slam try, 0 or 3 key cards

..........4N = accept slam try, 2 or 5 key cards without CQ

..........5C = accept slam try, 2 or 5 key cards with CQ

.....4D = 6+ diamonds, SI

.....4S = mild slam try

3C = 3 spades, max

.....3D = 5+ diamonds, SI

.....3H = puppet to 3S, 5+ spades, weak or GF [after 3S: denial cues, frivolous 3NT]

.....3S = 5+ spades, INV

.....3N = to play

.....4C = 3+ clubs, SI

..........4D = decline slam try

...............4H = RKCB

..........4H = accept slam try, 1 or 4 key cards

..........4S = accept slam try, 0 or 3 key cards

..........4N = accept slam try, 2 or 5 key cards without CQ

..........5C = accept slam try, 2 or 5 key cards with CQ

.....4D = 6+ diamonds, SI

.....4S = mild slam try

3D = 4 spades, min

.....3H = puppet to 3S, 5+ spades, weak or GF [after 3S: denial cues, frivolous 3NT]

.....3S = INV

.....3N = side void, SI

..........4C = asks for the void

...............4D/H = void in suit bid

...............4S = club void

.....4C = 3+ clubs, SI

.....4D/H = 5+ spades, singleton in suit bid, SI

.....4S = mild slam try

3S = 4 spades, max

.....3N = slam try

.....4m, 4H = denial cue

.....4N = RKCB

 

 

Over 2C - 2S

2N = min

.....3C = SO

.....3D/H/S = stop ask

3C = max

.....3D/H/S = stop ask

 

 

Over 2C - 3H

3S = diamond fit, decline of slam try

3N = no diamond fit

4D = diamond fit, accept slam try

 

 

Over 2C - 3S

3N = decline slam try

4C = accept slam try

 

 

That is everything - probably alot more detail than you would ever want. Notice also that 5S-4H GF hands start with 2D, not 2H, to facilitate the correct information exchange. This is clearly different from Jacoby transfers and illustrates the point about Opener providing the information even though it seems to be coming from Responder. Another point about this is that 2C can be 5C - 4M which means that it can hold far more hand types than your 2C opening. That means that full shape relays would simply not be possible below 3NT.

 

Now onto your system. The reason i am asking about the worries on GI hands is that your forcing 2M responses seem to cover all of these possibilities. The only GI hands that seem to go through 2D now are invitational club raises and a natural 2NT bid. What I am wondering here is if it is not possible to make the 2D relay a full game force. The first of these hands can easily be handled through 2NT - simialr arguments can be used here as for the competitive auctions discussed in another thread. The problem is the natural balanced invite and there just is not a suitable alternative.

 

Finally a little discussion about relays based on shortness (symmetric) and based on fragments. To some extent these approaches are interchangeable - the former is better at recognising a slam early, the latter is better at finding fits quickly. The only time this matters is when the shape resolution goes above 3NT. For example, with 7321 hands you can either find out by 3NT what the 3 card suit is or what the singleton is in addition to the 7 clubs. To get the full picture requires a relay above 3NT. The question is simply which information is more important to get before you need to commit. My (strong) belief is that the fit is more important to locate. That is why I use a fragment based relay approach throughout the system. It has exactly the same information capacity as the shortness method, it simply transfers the information in a different order.

 

An example of this from a more traditional point

 

1C - 1H (4+ spades)

1S - 2D (1-suited)

2H - 2S (3 hearts, 6-7 spades) [2N is 3 clubs, 3C is 3 diamonds, 3D is 8+ spades and 0-1 hearts, 3H is 8+ spades and 0-1 diamonds, 3S is 7=2=2=2, 3N is 8+ spades and 0-1 clubs, min]

2N - 3S (7=3=1=2) [3C would show 3 diamonds, 3D is 3 clubs, 3H is 6=3=2=2]

 

The symmetric relay auction is identical except that you first show a middle shortage and then show the relative lengths of the other 3 suits.

 

Now with your 2C opening the relays have to start much higher than in standard symmetric and this is why full shape resolution is difficult. If you could make 2D GF it would be possible to start the 1-suited relays at 2NT (1 step higher than standard) which would significantly reduce the difficulties. But I do not see how to achieve this so you are probably stuck with starting at an uncomfortably high level.

 

One final note on your current structure. You have

2C - 2D - 2H - 3D = diamonds and hearts

2C - 2D - 2H - 3H = diamonds and spades

2C - 2D - 2H - 3S = hearts and spades

 

It seems to me much more efficient to switch the meanings of 3D and 3S to enable Opener to indicate a fit in either major at the 3 level.

 

I hope the bidding notes have managed to survive the forum editing intact. It was really quite difficult to do due to the characteristics of the forums. First of all no tabs, and secondly every time I tried to edit something the edit window would jump to a completely unrelated point in the message. So most of it was done completely blind. Anything that does not make sense just ask and I will correct it.

[straube]

I'm really impressed. It appears we can eat our cake and have it, too.

 

Our opening denies four hearts so 2C-2D could be the start of a GF relay or some hand with 5 hearts. That means we get 2H as a transfer, etc. I guess the obvious downsides are that 1) partner won't know we're in a GF immediately and 2) we have to learn a new and perhaps better relay structure.

 

It's getting late and tomorrow looks busy, but I'll definitely have to think this over.

 

Very elegant structure. Thanks a lot.

[Zelandakh]

Oh that is potentially really quite clever. Because of your restricted opening you never need the responses with 4 hearts over 2D and you might indeed be able to build a relay structure from that. I suspect it might get a little complex in practise but it is definitely worth looking into and is a very interesting way of using the specific shape restriction in your 2C opening. It is definitely too late/early to think about now though. I might come back in a few days if I get a chance, or if not will just look at your idea and look for anything obvious.

[Zelandakh]

I've rigged something up very quick and dirty to get an idea of the amount of information space available under this idea. Hopefully I didn't miss any hand shapes! The relay structure leaves lots to be desired but that can be ironed out to match what is used elsewhere in your system. The important thing is whether the information you can fit into the space below 3NT is useful enough to take the time optimising the concept.

 

2C = 6+ clubs, 0-4 spades, 0-3 hearts, 0-4 diamonds

---------------------------------------------------

2D = 5+ hearts, weak/INV; or any GF

.....2H = 0-2 hearts; 6 clubs or any 7321

..........3C = sign-off

..........3D = puppet to 3H (weak/strong)

..........3H = 6+ hearts, INV

..........2S = relay

...............2N = 1-2 spades

....................3C = relay

.........................3D = singleton spade, 2 hearts

..............................3H = relay

...................................3S = 1=2=4=6

...................................3N = 1=2=3=7

.........................3H = 2=2=3=6

.........................3S = 2=1=4=6

.........................3N = 2=1=3=7

...............3C = 3 spades, 0-1 hearts

....................3D = relay

.........................3H = 3=0=4=6

.........................3S = 3=1=3=6

.........................3N = 3=1=2=7

...............3D = 3 spades, 2 hearts

....................3H = relay

.........................3S = 3=2=2=6

.........................3N = 3=2=1=7

...............3H = 4=0=3=6

...............3S = 4=1=2=6

...............3N = 4=2=1=6

.....2S = 0-2 hearts, 7+ clubs and not 7321

..........3C = sign-off

..........3D = puppet to 3H (weak/strong)

..........3H = 6+ hearts, INV

..........2N = relay

...............3C = 0-1 spades

....................3D = relay

.........................3H = 8+ clubs

..............................3S = relay

...................................3N = singleton spade

...................................4C = spade void

.........................3S = 1=1=4=7

.........................3N = 4 diamonds, spade void (0247 and more extreme)

...............3D = heart void

....................3H = relay

.........................3S = 3=0=3=7

.........................3N = 4 diamonds (2047 and more extreme)

.........................4C = 4 spades (4027 and more extreme)

...............3D = 8+ clubs, 0-1 hearts

....................3H = relay

.........................3S = singleton heart

.........................3N = heart void

...............3H = 4 spades, 0-1 diamonds

....................3S = relay

.........................3N = 4=1=1=7

.........................4C = diamond void (4207 and more extreme)

...............3S = 2=2=2=7

...............3N = 8+ clubs, 0-1 diamonds

....................4C = relay

.........................4D = 2218

.........................4H = 3208+

.....2N = 3 hearts, 0-2 spades, 1-3 diamonds, 6-7 clubs

..........3D = puppet to 3H (weak/strong)

..........3H = 5+ hearts, INV

..........3C = relay

...............3D = 3 diamonds

....................3H = relay

.........................3S = 1=3=3=6

.........................3N = 0=3=3=7

...............3H = 2=3=2=6

...............3S = 1=3=2=7

...............3N = 2=3=1=7

.....3C = 3 hearts, 3-4 spades, 6-7 clubs

..........3H = sign-off (no space for an invite here)

..........3D = relay

...............3H = 3=3=1=6

...............3S = 3=3=0=7

...............3N = 4=3=0=6

.....3D = 0=3=4=6, min

..........3H = sign-off

..........3S = relay

.....3H = 0=3=4=6, max

..........3S = relay

2H = 4+ spades

2S = range ask (natural NT invite or 5+ hearts and invitational)

.....2N = min

..........3C = sign-off

..........3D = red 2-suiter

..........3H = 6+ hearts, NF

.....3C = max, 0-2 hearts

..........3D = red 2-suiter

..........3H = 6+ hearts, GF

.....3H = max, 3 hearts

2N = good club raise

3C = weak raise

3D = sign-off

[straube]

Great call putting the misfitting 7C hands into 2C-2D, 2S. I'd noticed that a 2H rebid could not handle all of the hand types lacking 3 hearts.

 

This seems like the winningest structure for 2C. I wonder about the best method for organization so as to remember it. Also, we are still a bit tight for space and I would like to give up on the xxx8 hands.

 

Nice that the 7321 residues are distinguished (ours currently doesn't do this).

[awm]

It may be worth commenting that at some point Sam and I developed a transfer-based structure not too different from Zelandakh's... although it was a bit more focused on identifying stoppers for notrump rather than identifying opener's exact shape (especially on hands where opener has 8+ clubs etc).

 

In any case, we decided that this was better than the relay-oriented approach, but then scrapped it from our system anyway. The main issue was that it was an awful lot of stuff to remember for an auction that came up rarely and provided only a small marginal benefit. It bore no significant resemblance to anything else in our methods, so there were no "shortcuts" to remembering it better. The relay-based method is similar to symmetric relay (which we play in a lot of other auctions) so much less memory-intensive.

[straube]

It may be worth commenting that at some point Sam and I developed a transfer-based structure not too different from Zelandakh's... although it was a bit more focused on identifying stoppers for notrump rather than identifying opener's exact shape (especially on hands where opener has 8+ clubs etc).

 

In any case, we decided that this was better than the relay-oriented approach, but then scrapped it from our system anyway. The main issue was that it was an awful lot of stuff to remember for an auction that came up rarely and provided only a small marginal benefit. It bore no significant resemblance to anything else in our methods, so there were no "shortcuts" to remembering it better. The relay-based method is similar to symmetric relay (which we play in a lot of other auctions) so much less memory-intensive.

 

Yeah, I'd have trouble remembering it and I have just a touch of concern about interference. But it's very well put together. It would let us relay and correct to a major when responder has a major suit bust. I've never seen a structure that could do both of those things at the same time.

[straube]

Zelandakh and awm, how is this for memorability?

 

2C-2D,

.....2H-heart shortness

..........2S-relays

...............2N-SS

....................3D-3136

....................3H-2137

.....................3S-3127

.....................3N-3037

................3C-diamonds

.....................3H-2146

.....................3S-3046

.....................3N-2147

.....................4C-1147

................3D-4126

................3H-4036

................3S-4027

................3N-4117

 

.....2S-heart doubleton

..........2N-relays

...............3C-SS, bal

....................3D-relays

.........................3H-2236

..........................3S-3226

.........................3N-2227

...............3D-SS, unbal

....................3H-relays

.........................3S-1237

.........................3S-3217

...............3H-four diamonds

....................3S-relays

.........................3N-1246

.........................4C-0247

...............3S-4216

...............3N-4207

 

.....2N-3 hearts, higher

...............3C-relays

....................3D-1-3-3-6

....................3H-1-3-2-7

....................3S-0-3-3-7

.....3C-3 hearts, lower

..........3D-relays

....................3H-3-3-1-6

....................3S-2-3-1-7

....................3N-3-3-0-7

.....3D-0-3-4-6

.....3H-2-3-2-6 (10-13)

.....3S-4-3-0-6

[Zelandakh]

If this is the structure you prefer and is easier for you to remember then this is definitely the way to go. My relays weren't designed for ease of memory since what I would come up with would anyway not match your system and mainly I was interested in the amount of information exchange since that is the key point before optimising. Now on yours I noticed a small point that might be improved:-

2C-2D,

.....2H-heart shortness

..........2S-relays

...............2N-3 spades and/or 2-3 diamonds [3C-relays]

....................3D-3136 or 3046 <-----

.........................3H-relays

..............................3S-3136

..............................3N-3046

....................3H-2137

.....................3S-3127

.....................3N-3037

................3C-4 diamonds, <3 spades

.....................3H-2146

.....................3S-2047 <-----are these 2 shapes the correct way round?

.....................3N-1147 <-----you tend to split by balance and this seems to go against your rule

 

...has a small advantage in keeping all hands at 3NT or below. It is arguably more important to show the (2)3 spades than the 4 diamonds at this stage too.

 

The other thing I do not like is that you have gone above 3H in the initial rebid. I think this is dangerous if you are going to allow Responder to bid 2D with a bust and long hearts. Perhaps you could move the 0346 hand into the high shortage response with:-

.....2N-3 hearts, higher

...............3C-relays

....................3D-1-3-3-6/0-3-4-6 <-----

.........................3H-relays

..............................3S-1-3-3-6

..............................3N-0-3-4-6

....................3H-1-3-2-7

....................3S-0-3-3-7

.....3C-3 hearts, lower

..........3D-relays

....................3H-3-3-1-6

....................3S-2-3-1-7

....................3N-3-3-0-7

.....3D-2-3-2-6

.....3H-4-3-0-6

 

You could also make 3H 3316/3307 (going above 3NT on the follow-up relay ask but on very similar hand types) and move 4306 into the heart shortage section:-

.....3C-3 hearts, lower

..........3D-relays

....................3H-3-3-1-6/3-3-0-7

.........................3S-relays

..............................3N-3-3-1-6

..............................4C-3-3-0-7

....................3S-2-3-1-7

....................3N-4-3-0-6

.....3D-2-3-2-6, min

.....3H-2-3-2-6, max

 

Aside from that, all I can see is that you have lost some clarity on the freak hands but that is not a big deal. One final idea that might be even easier to remember would be to use spade length as your distinguisher for the heart doubleton line:-

.....2S-heart doubleton

..........2N-relays

...............3C-0-1 spades

....................3D-relays

.........................3H-1246

..........................3S-1237

.........................3N-0247

...............3D-2 spades

....................3H-relays

.........................3S-2236

.........................3N-2227

...............3H-3 spades

....................3S-relays

.........................3N-3226

.........................4C-3217

...............3S-4216

...............3N-4207

 

Again, only you can make the judgement as to what is more logical and what matches the rest of your system best. Just on a technical point in the above - it is actually better to switch the 3D and 3H bids since the difference between 2236/2227 is 'less' than between 3226/3217 and so you would prefer to have the former as the hand type that spills above 3NT. However it then becomes much more difficult to remember so (personally) I would lose the optimisation here.

 

Anyway, nice job. You might be close to something that is ready to be run through some testing. My suggestion at this stage would be to generate a lot of hands and to bid each of them using all of the different structures you have. Get a feel for which is most accurate and which gives the easiest decision-making for the way your mind works. Then do the same but adding in various levels of interference and see if your preferred method has a problem. It is only by such a practical approach that you can turn these lovely theoretical discussions into a working bidding system. Of course, once you have it fully optimised (and memorised) you have to take the biggest test of all and play it against live opps. At least that is the way I do it.

[Zelandakh]

While I was checking through my last post I also noticed another option for follow-ups with the 0-1 heart hands that looks interesting and might also be good for memory, namely splitting by heart length:-

2C-2D,

.....2H-heart shortness

..........2S-relays

...............2N-1 heart [3C-relays]

....................3D-3 spades

.........................3H-relays

..............................3S-3136

..............................3N-3127

....................3H-2146

.....................3S-2137

.....................3N-1147

................3C-heart void

.....................3H-3046

.....................3S-3037

.....................3N-2047

 

Just another option. Ideally you would have a single rule that worked for all cases - split by heart length, then by spade length. Or split by high-low shortage. Or whatever. At this stage it's probably best for you to work through the options and details with your partner on what is easiest and then move onto tests as above. Fun though it is for me to play around with this... :)

[straube]

Yeah, that 2C-2D, 3S is a weakness.

 

I'm less bothered by the spillovers past 3N because it only happens after responder has discovered heart shortness and four diamonds and gives opener permission to show his exact pattern. I.e. responder must have slam interest at that point to bother asking.

 

I think it's right to immediately separate short hearts from two because this allows responder to 1) play 6-1 heart fits at the 2-level and 2) sign off or invite with 6-2 (or even 5-2) heart fits at the 3-level.

 

Partner isn't jazzed about memorizing this though.

[straube]

I'm envisioning responder only replying with something like 9 hcps and a 6cd major when not fitting clubs. After all, this pushes us to the 3-level whenever a doubleton is found. I'm declining to introduce a 4-cd spade suit as well.

 

2C structure

 

2D-5+ hearts or GF relay

.....2H-0 or 1 heart

..........2N-four diamonds

..........3C-single-suited

..........3D etc-4126, 4036, 4027, 4117

.....2S-2 hearts

..........3C-bal

..........3D-four diamonds

..........3H-1237, 3127

..........3S etc-4216, 4207

.....2N-3 hearts, higher shortness

.....3C-3 hearts, higher shortness

.....3D-0346

.....3H-2326

.....3S-4306

 

when 3D is available as a relay break, it shows a sign off or GF in hearts, then 3H (instead of 3D) shows a GI in hearts

 

2H-5+ spades, not GF unless prepared game bid

.....2S-0-1 spades

..........2N-inv

..........3C-constructive

..........3D-GI, 5D

..........3H-GI, 5H

..........3S-GI, 6S

......2N-2 spades, strong

......3C-2 spades, weak

......3D-3 spades, strong

......3H-3 spades, medium

......3S-3 spades, weak

 

2S-invitational, artificial, not necessarily club fit

.....natural

2N-constructive raise

3C-preemptive raise

3D-to play

3H-forcing, diamonds

3S-LR clubs

[Zelandakh]

Looks fine. The main issue will be stopper asks and perhaps the red 2-suiter. You will have to find out in play-testing if these are a problem in practise. If they are then you might need to restructure your 2S response to compensate. Otherwise, I hope your partner is not too upset with my input...and good luck!