[phoenix214]

I was checking out some 1NT response structures, but i don`t know which is the better one to use(Any complexity level is ok). I do not know anything much about the older ones, like 2-way Stayman and stuff. I have tried playing a 4way transfer over NT with a full relay for finding openers distribution, but that didn't feel best. For the last two years or so, i did play a version where responder shows his shape to opener.
Checking out more modern structures i have noticed that most of them use 2♣ as a puppet to 2♦(Keri, Heeman). Is hiding the openers distribution that much of a gain?
Does anyone have any opinions on what NT structure is the better one.

[TylerE]

"Best" on what criteria?

I would posit that the best thing for your local club game is not what you'd want to play in the Bermuda Bowl. Will also depend heavily on your range, wether you open 1N with 5M never, sometimes, frequently, or always, etc.

[phoenix214]

 TylerE, on 2013-July-29, 12:59, said:

"Best" on what criteria?

I would posit that the best thing for your local club game is not what you'd want to play in the Bermuda Bowl. Will also depend heavily on your range, wether you open 1N with 5M never, sometimes, frequently, or always, etc.

Well, lets put it like this. If you could learn and play any NT structure, which one would you play?
P.S. - I am thinking more on the range of strong NT if it matters, and it can be off-shape as well if it matters(I don`t think a full relay is the best idea anyway)

[akwoo]

If you are playing 25 board matchpoint games in a mediocre club field, the best NT structure for you to play is whatever everyone else plays.

Think of it this way: on hands where you are in the same contract as everyone else, you'll score 60% just by virtue of playing better. I venture no system will get you to a contract that does better than the field 1.5 times as frequently as it gets you to a contract that does worse than the field.

[awm]

Sam and I spent quite a bit of time on this, and have played a number of systems (we used Keri variants for many years) before settling on what we play now. It's basically four-way transfers with a lot of additions/modifications. Our responses:

2♣ = stayman, includes invite with 5+♠ and "minor suit stayman" type hands as well as traditional stayman hands
2♦ = transfer to hearts, won't have exactly 4♠, if 5/5 majors then exactly invitational, possibly GF with 4♥ and 6+ minor
2♥ = transfer to spades, won't have exactly 4♥, always either signoff or GF, possibly GF with 4♠ and 6+ minor
2♠ = balanced invite, or one-suited minor invite, or checkback for xx opposite xx in a major, or a pure quantitative slam try
2NT = transfer to clubs, signoff or GF, no 4M
3♣ = transfer to diamonds, signoff or GF, no 4M
3♦ = 4+♦ and 0-1♣, 3-4 in each major, GF
3♥ = 4♥ and 0-1♠, 3+ in each minor, GF
3♠ = 4♠ and 0-1♥, 3+ in each minor, GF
3NT = to play
4♣ = transfer to hearts, wants pd to declare
4♦ = transfer to spades, wants pd to declare
4M = to play

After stayman:

2♥ by responder forces 2♠; this is either a weak 5♠/4♥ hand, or various inv+ hands with both majors
.... 2NT = 4♠ 5♥ invite
.... 3m = GF longer minor with 5/4 majors either way
.... 3M = invite 6/4
2♠ by responder is invitational with 5♠ (possibly also 4♥ if opener denied hearts) and is NF
2N by responder guarantees a major
3♣ by responder is a GF relay (usually looking for 5-3 major fit or 4-4 minor fit)
3♦ by responder shows short diamonds with 4+♣ and no known fit (i.e. 3451 hand, 4441 if opener rebid 2♦)
3M by responder is smolen but guarantees 2-2 in the minors

After 2♦ transfer to hearts is accepted:

2♠ = invite with 5+♥ and less than 4♠ (opener bids 2N with min, 3♣ with max without fit)
2N = GF hearts and clubs
3♣ = GF 5+♥ and 4♦
3♦ = GF 5+♦ (maybe only 4♥)
3♥ = invitational 5/5 majors
3♠ = slam try single-suited in hearts
4m = splinter single-suited in hearts

After 2♥ transfer to spades is accepted, 2N+ similar to above (but 3♥ is GF 5/5 majors)
After 2♠, opener bids the lowest type he would reject (i.e. 2N=min, 3m=max but xx in this minor, 3M=max with xx in this major)
After 2N/3♣ transfer is accepted, 3M = 3-card major with 0-1 in the other major

There are quite a few more details (in particular, follow-ups after responder's second transfer, and continuations after 3-level splinters).

[rhm]

 akwoo, on 2013-July-29, 13:35, said:

If you are playing 25 board matchpoint games in a mediocre club field, the best NT structure for you to play is whatever everyone else plays.

Think of it this way: on hands where you are in the same contract as everyone else, you'll score 60% just by virtue of playing better. I venture no system will get you to a contract that does better than the field 1.5 times as frequently as it gets you to a contract that does worse than the field.

There is a lot of fallacy in this type of argument.

1) The weaker the field the more outcomes of tournaments are decided by bidding decisions not by play. Better bidding decisions tend to be more important than better play decisions. But the better the field the less scope is there for better bidding decisions. The same holds true for play decisions, only to a lesser degree.
2) If you are playing better than the field, it does not mean you come out ahead over the short term. A better line of play may well be less successful. The results in bridge books on play, where the layout always rewards the right play are misleading. Just think of opening leads. The scenario is essentially the same as in bidding. Reaching 60% constantly in a "mediocre club" by play decisions alone either means your partnership is world class or you play in a club so weak, that there would be little point playing there at all. Even if you are very good at a typical local club your technical advantage is on average much less than 60%. (BBO inspects fraud when you are doing better than 58% on average)
3) What matters in the bidding are good judgement and good agreements on scenarios, which occur with reasonable frequency. Gadgets rarely matter. Also concentrating on constructive bidding sequences at the expense of competitive ones is a common error.
4) Even for constructive bidding many spent an inordinate amount of effort on sequences after notrump, where opener has already limited his strength and distribution well, at the expense of sequences after other opening bids.

Rainer Herrmann

[Free]

 phoenix214, on 2013-July-29, 12:57, said:

Is hiding the openers distribution that much of a gain?

Hiding opener's distribution isn't a purpose on it's own, it's a matter of efficiency. It's much better that 1 player tells everything about his hand and the other one makes the decisions, rather than both telling half a story and not having a clue about who has to make which decision. Relay systems work in a similar way.

Obviously it's possible to use a relay scheme to ask about opener's hand. The 1NT opener has already described his hand quite nicely (or so you think), so there's space for full relays. However, it's impossible to make good decisions when you've got no idea about partner's honor location. Suppose you hold a 5-4-3-1 and opener shows a 3-4-2-4: as long as you don't know what strength partner has in ♣ you don't know how much your singleton is worth. Turn things around however and have responder show his entire hand: opener will know about a ♣ singleton, so with ♣5432 he will be optimistic about slam, while with ♣KJxx he will be more pessimistic about slam. Distribution is what matters here, because the 1NT opener can evaluate the unbalanced hand much better (he knows where his strength is located).

Therefore modern systems try to let responder describe as much as possible about his hand and let opener decide. A very simple example is transfer extensions (see Justin's site for example) which stay low and let responder describe his hand very well (first your 5 card M, then a side 4 card suit, then a shortness => 5431).

[hrothgar]

I'm fond of the treatments in "Notrump Bidding the Scanian Way"

Sadly, its hard to find copies this days...

[Zelandakh]

For a weak NT, Keri and ETM are both decent, as are a few others. For a strong NT or a generic method to use over multiple NT ranges then transfers are probably a better choice. The default for that is becoming 4 suit transfers and 3♣ as Puppet Stayman and that is surely a good option. An alternative of my own with 2♣ as Puppet Stayman is:-

2♣ = Puppet Stayman (most strong diamond-based hands are included here)
... - 2♦ = no 5 card major
... - ... - 2♥ = asks if 4 spades are held (denies interest in a 4-4 heart fit)
... - ... - ... - 2♠ = 4 spades
... - ... - ... - ... - 2NT = nat invite
... - ... - ... - ... - 3m = nat Baron-like
... - ... - ... - ... - 3♥ = GF spade raise
... - ... - ... - ... - 3♠ = nat invite
... - ... - ... - 2NT = <4 spades, min
... - ... - ... - ... - 3m = nat Baron-like
... - ... - ... - ... - 3♥ = GF with heart shortage
... - ... - ... - ... - 3♠ = GF with 5 spades (and 3 hearts)
... - ... - ... - 3m♥ = <4 spades, max, cheapest 4 card suit (Baron-like)
... - ... - 2♠ = 4+ hearts, <4 spades
... - ... - ... - 2NT = <4 hearts, min
... - ... - ... - ... - 3m = nat, Baron-like
... - ... - ... - ... - 3♥ = GF with 5 hearts (and 3 spades)
... - ... - ... - ... - 3♠ = GF with spade shortage
... - ... - ... - 3m = <4 hearts, max, cheapest 4 card suit (Baron-like)
... - ... - ... - 3♥ = 4 hearts, min
... - ... - ... - 3♠ = 4 hearts, max (3NT now is a spade cue)
... - ... - ... - 3NT = max, 4=3=3=3
... - ... - 2NT = 4-4 majors, invite
... - ... - 3♣ = 4-4/5-4 majors, GF
... - ... - ... - 3♦ = no 4 card major (then 3M = 5 in other major)
... - ... - ... - 3M = 4 card M
... - ... - 3♦ = 5+ diamonds, GF
... - ... - 3M = 4-5/5-4 minors
... - 2♥ = 5 hearts
... - ... - 2♠ = Baron range ask
... - ... - 2NT = puppet to 3♣
... - ... - 3♣ = puppet to 3♦
... - ... - 3♦ = GF heart raise
... - ... - 3♥ = nat invite
... - 2♠ = 5 spades
... - ... - 2NT = nat invite
... - ... - 3♣ = puppet to 3♦
... - ... - 3♦ = minor suit Baron (3♥ = clubs; 3♠ = diamonds)
... - ... - 3♥ = GF raise
... - ... - 3♠ = nat invite
2♦ = 5+ hearts
... - 2♥ = any non-superaccept
... - ... - 2♠ = range ask with 5 card invite; or GF with clubs; or slammy 1-suiter
... - ... - ... - 2NT = min, <3 hearts
... - ... - ... - ... - 3♣ = 4 clubs
... - ... - ... - ... - 3♦ = 5+ clubs
... - ... - ... - ... - 3♥ = one-suiter
... - ... - ... - 3♣ = max, <3 hearts
... - ... - ... - ... - 3♦ = 4+ clubs
... - ... - ... - ... - 3♥ = one-suiter
... - ... - ... - 3♦ = max, 3+ hearts
... - ... - ... - 3♥ = min, 3+ hearts
2♥ = 5+ spades
... - 2♠ = any non-superaccept
... - ... - 2NT = nat invite
... - ... - 3♣ = GF; diamonds or slammy one-suiter
... - ... - ... - 3♦ = 2 spades, 4+ diamonds
... - ... - ... - ... - 3♠ = one-suiter
... - ... - ... - 3♥ = 2 spades, <4 diamonds
... - ... - ... - ... - 3♠ = one-suiter
... - ... - ... - 3♠ = 3+ spades, <4 diamonds
... - ... - ... - others = 3+ spades, 4+ diamonds
2♠ = range ask; (semi-)balanced or clubs
... - 2NT = min
... - ... - 3♣ = sign-off
... - ... - 3♦ = 5+ clubs, 4 hearts
... - ... - 3♥ = 5+ clubs, 4 spades
... - ... - 3♠ = 5+ clubs, 5+ diamonds
... - 3♣ = max
... - ... - 3♦ = 5+ clubs, 4 hearts
... - ... - 3♥ = 5+ clubs, 4 spades
... - ... - 3♠ = 5+ clubs, 5+ diamonds
2NT = 5 spades, 4 hearts, invite
3m = nat one-suited, slammy
3♥ = 3-suited with short diamonds (the 3M responses are somewhat unnecessary really but hard to come up with anything better than a 3-suited call of some sort)
3♠ = 3-suited with short clubs
4♣ = both majors
4red = puppet to 4M

Hopefully not too many mistakes/typos from typing this in!

[gatorlaw]

 rhm, on 2013-July-30, 02:44, said:

... What matters in the bidding are good judgement and good agreements on scenarios, which occur with reasonable frequency. Gadgets rarely matter.

... Also concentrating on constructive bidding sequences at the expense of competitive ones is a common error.

... Even for constructive bidding many spent an inordinate amount of effort on sequences after notrump, where opener has already limited his strength and distribution well, at the expense of sequences after other opening bids.

Rainer Herrmann

Very good advice.

[gatorlaw]

 Free, on 2013-July-30, 06:39, said:

... it's impossible to make good decisions when you've got no idea about partner's honor location. Suppose you hold a 5-4-3-1 and opener shows a 3-4-2-4: as long as you don't know what strength partner has in ♣ you don't know how much your singleton is worth. Turn things around however and have responder show his entire hand: opener will know about a ♣ singleton, so with ♣5432 he will be optimistic about slam, while with ♣KJxx he will be more pessimistic about slam. Distribution is what matters here, because the 1NT opener can evaluate the unbalanced hand much better (he knows where his strength is located).

Therefore modern systems try to let responder describe as much as possible about his hand and let opener decide....

NT auctions are a good illustration of "the Balanced Hand Principle" and "the Concealed Hand Principle." The "Balanced Hand Principle" is that the unbalanced hand should describe itself to the balanced hand because only the balanced hand will know if there are duplication of values in the short suit.

The "Concealed Hand Principle" is that, given a choice of methods, the hand that is going to be dummy should describe itself to the concealed hand. Giving away information about the concealed hand to the opponents results in better defense often enough so that it eliminates any gain to the declaring side.

The practical effect of these ideas for NT bidding is that it's better for the responding structure to use transfers and describe responder's shortness to the balanced hand than it is to use two-way checkback and methods to have opener describe his pattern to responder.

[gatorlaw]

Some practical advice:

(1) The "best" method to play is the one that both you AND YOUR PARTNER know and understand perfectly. A bad method completely understood and played perfectly is better than a perfect method poorly understood and played badly.

(2) It doesn't help to play a great system if you don't have any partners to play it with. For example, "NT bidding the Scanian Way" is a fine method with lots of merit to it. You won't find anyone in America to play it with. So don't go there.

(3) Don't reinvent the wheel. There are lots of complete methods already out there. Don't invent a homegrown one. Play something widespread enough that you'll find lots of partners to play it with. You'll get to play it more often and know it better.

(4) Keep it as simple as possible. Complicated agreements about unusual situations rarely come up. When they do, they usually won't give you an advantage, and even when in theory they should result in a net gain, in practice either you or your partner will forget them often enough to result in a net loss.

(5) Use similar bidding methods for similar situations. Let Meckstroth and Rodwell have a hundred specific different agreements to cover situations that seem to be similar. You're not Meckstroth and Rodwell, and that's not why they win anyway.

(6) Complicated bidding methods use up mental energy you need for judgment about bidding, opening leads, defense, and declarer play. At a lecture, I heard Rodwell say he was really impressed with the accuracy of Helgamo and Hellness' relay system. When they gave it up, he asked them why. They said they had to put so much energy into the bidding they couldn't take all their tricks.

[mgoetze]

 phoenix214, on 2013-July-29, 13:16, said:

 TylerE, on 2013-July-29, 12:59, said:

"Best" on what criteria?

I would posit that the best thing for your local club game is not what you'd want to play in the Bermuda Bowl. Will also depend heavily on your range, wether you open 1N with 5M never, sometimes, frequently, or always, etc.

Well, lets put it like this. If you could learn and play any NT structure, which one would you play?

Well, it depends on the level of the opposition, my 1NT range, how frequently I open 1NT with 5 card majors, how good my partner's memory is, etc.