[mr1303]

Hi. My partner and I are trying to come up with a new set of responses to 1NT

2C must be symmetric relay. (2D = 4+ hearts, 2H = 4+ spades, not 4 hearts, 2S = any 4333, 2NT = 5 diamonds, 3C-S = 5 clubs). Responder can also check range after opener's shape has been defined
2D = puppet to 2H for sign off in 2M/3m
2H/S = invite, Keri style, 4+
2NT = bal invite, no 4 card major
3C/D = invite, Keri style, 6+ cards

I'm wondering what I can now do with 3M bids direct after 1NT, and also after the 2D puppet followed by 2NT, 3NT, and 3M

Any ideas?

Mark

[jtfanclub]

How much do you enjoy slam auctions?

1NT-2D-2H-2NT could be ace-asking (number or specific), 1NT-2D-2H-3NT could be equivalent to 4NT quantitative. 1NT-2D-3H/3S could be a slam invite in the major. This would make 2D weak/strong (GF with slam aspirations or signoff).

Depending upon the strength of the 1NT, it may be worth having a direct bid of 3M be pre-emptive.

[whereagles]

1NT-3M as splinter is a gadget I've grown to like

Also, beware of the symmetric relay after a 1NT opening. Sometimes you only get to know pard has KQx opposite your singleton too late! Two opponents of mine once had a relay auction to 6NT. Upon inquiry, the relayer said "pard has this and that and we can't make a thing!" He was right - 2 down

[EricK]

It is not always right for responder to do the asking on GF hands. If you have eg a 1444 hand, you might want to know whether partner has your singleton well stopped. Finding out his shape won't necessarily do that. Also, a 2NT natural invite is possibly a waste of a bid.

So you could immediate 3♥, 3♠ and 2♦ followed by 3♥, 3♠ as splinters (♥, ♠, ♣, ♦ respectively)

2NT could be weak with both minors asking partner to give preference.

2♦ followed by 2NT could be invitational with both minors.

I suppose 2♦ followed by 3NT could be some GF or slam type hand with both minors. Are there any which aren't best handled by relays?

Eric

[luke warm]

EricK, on Sep 28 2004, 07:55 AM, said:

~~snip~~~
Are there any which aren't best handled by relays?

Eric

not that i know of

[Free]

I'm used to play a full relay structure after stayman, an ambiguous 2♦ (which can also be invitational or GF with or without ♥s) and I use 4 suit transfers (2♦ exception), so I actually have the entire 3-level free. I also use 2NT for 5-3 distributions in the Majors, so we don't miss 5-3 fits, and sometimes 3♣ is puppet stayman, to be able to investigate quick 5-3 M-fits without telling too much to opps when we don't have such fit.

For the 3-level I like 2 approaches: show a 4441 or an immediate RKC in the specified suit. They are both not very frequent, but both can be helpfull in some situations. When I play 3♣ as puppet stayman, then I use 3NT for conventional stuff as well, so with GF hands go through 2♣ or 2♦.

It works really well!

so:
2♣ = stayman (garbage stayman included), prepared to play at least 3♠ when opener has a 5 card ♠
2♦ = 5+♥ OR inv+ hands with/without ♥s, no interest in a 3-level ♠ contract when opener has a 5 card ♠
2♥ = trf ♠
2♠ = trf ♣
2NT = trf ♦ OR exactly 5-3 M with GF values (opener responds as it's trf ♦)
3♣ = Puppet stayman, no slam ambitions
3♦/♥/♠ = 4441 with stiff in the next suit and slam try opposite a good fitting hand
3NT = 4-4-1-4 (singleton ♦) with slam try opposite a good fitting hand

Negatives:
- stayman doesn't let relayer play in case of a fit (natural responses)
- because we don't use transfer bids over stayman, we need to play at least at 3-level when opener has a 5-card ♠. That's why we need some ambiguity in our 2♦ response.

Positives:
- full relay structure
- garbage stayman
- opener stays quite unknown when responder has only game-values or game-invitational values
- 5-3 fits will be found and usually rightsided (smolen principle after the 2NT relay)

[Dean]

If

2♦= Signoff (puppet to ♥)
2♥/♠/3♣/3♦= Invitational Keri style

Wouldn't you be better playing
2♦ = signoff in ♥/♣/♦ or 4=♥ and 4+♠ (to be shown by rebidding 2♠)
2♠ = signoff in spades
2♥= inv 4+♥ keri style.

This fixes the problem of both majors which Klinger obviously did lots of work on. You also have a 2NT followup after 2♦. Maybe you could use that as a puppet to clubs followed by new suit = RKCB in that suit.

[1eyedjack]

luke warm, on Sep 28 2004, 05:42 AM, said:

EricK, on Sep 28 2004, 07:55 AM, said:

~~snip~~~
Are there any which aren't best handled by relays?

Eric

not that i know of

Well, I thought I gave one example, at

http://bridgebase.lunarpages.com/~bridge2/...indpost&p=36457

[sceptic]

no idea what it is called but either 3 hearts or 3 spades to show

single of major bid

3 of other major

5/4 minor split

[Dean]

Wouldn't you be better playing
2♦ = signoff in ♥/♣/♦ or 4=♥ and 4+♠ (to be shown by rebidding 2♠)
2♠ = signoff in spades
2♥= inv 4+♥ keri style.


Oops. Now no way to show invitational in spades.

Sorry
Dean

[luke warm]

1eyedjack, on Sep 28 2004, 10:48 PM, said:

luke warm, on Sep 28 2004, 05:42 AM, said:

EricK, on Sep 28 2004, 07:55 AM, said:

~~snip~~~
Are there any which aren't best handled by relays?

Eric

not that i know of

Well, I thought I gave one example, at

http://bridgebase.lunarpages.com/~bridge2/...indpost&p=36457

yes, you gave an excellent example of an alternative method... i wasn't then nor am i now convinced that relays still aren't better, in the majority of cases... sure, either of us can construct hands where one or the other method is better, but the test comes about on the hands not constructed... of course, that's just my opinion

[Free]

I've noticed that slam approach with a balanced hand vs an unbalanced hand is usually best handled by the balanced hand. He can place his honour strength quite nice. If you could introduce some kind of method to invite slam when partner has nothing in a certain suit, then relays would be the best way to go... At this point there are lots of hands which will go wrong by using relays, and lots of hands will work very well using relays. I like the relay possibility included in my structure, but I also like to tell my own hand sometimes.

[1eyedjack]

luke warm, on Sep 29 2004, 06:33 AM, said:

i wasn't then nor am i now convinced that relays still aren't better, in the majority of cases...

I might be pursuaded of that, on the grounds that balanced hands are more frequent than unbalanced hands. On the other hand, unbalanced hands worth a slam try may be more frequent than balanced hands worth a slam try. If that is the case then I would have to differ. I am not a good enough statistician.

luke warm, on Sep 29 2004, 06:33 AM, said:

either of us can construct hands where one or the other method is better, but the test comes about on the hands not constructed...

There are two tests, in my opinion.
The first is in identifying a pattern, if there is one, that indicates at the outset which method is better. If you can identify that pattern, you may not have to settle for a method that simply gains on the "majority of cases", but choose to relay (if it is indeed appropriate) on the majority of cases, but to describe (when the pattern so indicates) on that (perhaps) minority of cases where that action is indicated. Whilst the example that I cited was just one example, it is I think not hard to identify why, in that example, it works better for responder to describe. Then it is only a small step to appreciate that the reasons that apply to that example are likely to apply in a general case to situations where responder is unbalanced. Here are a couple more examples:

Example 3:
West opens 1N (12-14)
Top spot 5D.
3N reckons to fail most days.

---

Example 3:
West opens 1N (12-14)
Top spot 5D.
3N reckons to fail most days.

---

Example 3:
West opens 1N (12-14)
Top spot 5D.
3N reckons to fail most days.

---

Example 3:
West opens 1N (12-14)
Top spot 5D.
3N reckons to fail most days.

---

In all of the above examples, West is much better placed to decide what to do after East has described his hand than the reverse.

I have yet to see any convincing examples that show relays as providing a marked advantage when responder is distributional, although the quality of your high level continuations after responder has described is bound to be relevant.

The second test is to work out what you have to give up in non-game-forcing hands in order to have the priviledge of choice over who describes their hand in the GF sequences. Clearly, if you have just one response to show GF hands you will have an advantage on non-GF hands over someone who employs more bids devoted to GF hands.

I guess it is time to put up or shut up. I shall post the responses that I use, in a separate thread. May take a day or so.

[1eyedjack]

There is one more illustration that I would like to put forward in support of my criticism of the unilateral use of relays by responder on all GF hands. I mention this because the point is entirely different from points that I have made earlier.

Example 7:
West opens 1N (12-14)
North passes
East responds 2D (GF relay)
South overcalls 3S
At the other table the auction is the same, except that 2D was a transfer to H

---

Example 7:
West opens 1N (12-14)
North passes
East responds 2D (GF relay)
South overcalls 3S
At the other table the auction is the same, except that 2D was a transfer to H

---

The point?
As with many systems where you make a single, low-level bid that indicates GF values without any shape clarification (ie as with Precision 1C opener) you are particularly vulnerable to preemption by the opponents. There is no guarantee of coping with a preemptive overcall, but it is my experience that shape is more important than strength in deciding what to do when it happens, and if responder has shown some feature of his distribution before the preemption, even if he has not by then guaranteed GF values, you are on balance better equipped to cope with the preemption.