NT ranges

[foo]

The right question is something like

"which is more accurate: finding the proper strain and level

A= opening 1N= 10-12 and 1m rebid 1N with 13-17 (or 13-16 if you make your 2N rebid be 17-19), and 1m rebid 2N with 18-19 (or 17-19 using that method.)

or

B= Opening KNT and using 1C rebid 1N to show WNT and 1D rebid 1N to show SNT.

or

C= Passing most flat 10-12's and opening 1m rebid 1N with 12-14, opening 1N with 15-17, and 1m rebid 2N with 18-19."

 

"C" is provably the most accurate of those choices.

Well, by all means, prove it.

 

I'm especially curious as to how C can possibly be better than A in terms of finding the correct strain. If A is better than C in terms of finding the correct strain, then it is mathematically impossible to prove that C is better than A at finding the correct strain and level.

"A" is !NOT! better than "C" at finding the correct strain. It's worse.

 

"C" has 3 HCP ranges both for the 1N opening and the 1N rebid and a 2 HCP range for the 2N rebid.

 

"A" has at best a 5 HCP range for the 1N rebid and a 3 HCP range for the 2N rebid unless you accept the 1N rebid having a 6 HCP range.

 

In addition, the stronger the 1N opening, the weaker Responder can be and still take bids probing for the best strain and level. "A" has the weaker 1N opening so therefore Responder must be stronger to have the values for said probes.

 

"C" has

a= a tighter average bound on each step in its NT ladder.

b= a stronger 1N opening that makes it more likely for Responder to have enough values to probe for best contract.

QED "C" is provably better then "A" at finding both the correct strain and level.

 

 

The trade-off is that "A" gets in the opponents face more often. At the cost of less accurate constructive bidding. That's what the KNT is for. It is not a constructive bid. It's a preempt. The more bids you use for preempts, the less you have for constructive bidding and the more hand types that have to be packed into each constructive sequence. That makes each constructive sequence wider ranging; and =that= means your constructive accuracy must suffer. No way around it.

 

The case for the KNT is that the loss of constructive accuracy is worth it since you "generate enough action" to score well on enough boards to make up for it.

I, and many other players, disagree.

 

 

The hypothesis I'm presenting is that iff an appropriate defense is used with proper discipline, the "action" that the KNT generates is far more likely to be bad for the KNT side than for the opposition.

At that point, the fact that the rest of the direction can bid the non KNT hands more accurately than the KNT pair can starts to be a negative without enough compensation to make up for it.

 

But don't take mine, or Peter's, or Richard's, or Mikeh's, or anyone else's word for it.

Go test it for yourself. =Use= the defenses I'm showing you against the KNT. =Use= the run out system vs 1N X'd I posted. See for yourself. Independently investigate the truth.

People can make all sorts of claims about personal experience. Ultimately, such statements are variations of the logical fallacy of Argument From Authority.

 

The cards and the math don't give a d@mn who you are. If you do the right things with them, you'll tend to get good scores. If you don't, you'll tend to get bad scores.

 

That's the only thing that matters in defining who is "right".

[Guest Jlall]

I also have played a ton of NT ranges and the only one i don't like is 10-12 in conjunction with a natural system. I currently play strong club with 11-13 1/2 NV and 14-16 otherwise and am very happy with that. Playing a natural system I slightly prefer strong NTs to weak because of how hard competitive bidding can be playing 12-14 in a natural system base when you dont open 1N, but I would definitely agree to play 12-14 in a natural system NV if my partner wanted to. I am too old for vulnerable 12-14 NTs B)

[Finch]

I made no such assertion. I said that reliance on a gadget that gets most of its pluses because the opponents do not know how to do the right thing teaches one to be lazy bridge skills wise. I'll stand by that and we can agree to disagree if you wish.

That is true to some extent, but it's also true that playing the mini gives you a lot of practice at playing difficult 1NT contracts which is good for your bridge skills, in particular it's very good for your card-reading and psychological bridge skills.

 

As for run outs, I'll even give you the best one I know:

...

If you think you know a better run out system, I'd be happy to see it.

 

Yes. Here is the run-out system I play at IMPs, rubber and aggregate after a second-seat double:

 

Pass = to play in 1NTx

Redouble = to play in 1NTxx (forcing pass)

2C, 2D, 2H, 2S = to play

2NT = strong 2-suiter

3C, 3D, 3H, 3S, 3NT, 4C, 4D, 4H, 4S = to play

 

Easy, isn't it?

 

The only conventional call is that opener after 1NT - x - P - P can redouble with a 5/6-card suit to run to, which responder can pass or pull as appropriate.

 

I know that this doesn't allow responder to show all the various two-suiters that most run-out systems do. However, I'm going to explain why, IMO, this is by far the best run-out system to play. Resons in descending order of importance:

 

i) There's little point playing a mini for its pre-emptive value if you don't allow opener to be mildly off-shape (5-card major, 6-card minor). It's much more useful to be able to get out into opener's 6-card club suit than for responder to show 4-4 in two other suits.

 

ii) If neither opener nor responder has a 5+ card suit 1NTx is frequently the opening side's best spot, particularly at favourable (and no-one vaguely sane plays the mini vulnerable). Run-outs rely on either finding a 4-4 fit or the opponents getting bored of doubling, and even if you have a 4-4 fit it needs to play 2 tricks better than 1NT.

 

iii) Artificial methods give the opponents two bites at the cherry. For example, after 1NT x xx (forcing 2C from opener), 4th hand has space to distinguish between a very weak distributional hand and a hand with some values and a good hand. After 1NT x 2S (to play in 2S) a whole level of bidding has been removed. I have played with and against the mini (and the weak NT) for years, and I much prefer it when the opponents play an artificial escape mechanism. Any evaluation of methods that doesn't consider all four hands at the table is flawed.

 

As it happens, we do play an escape mechanism at matchpoints, but this is really predicated on the benefits from the opponents getting confused during the auction outweighing the downsides above. We take teams matches more seriously! (or rather, either we are playing a weak team and it doesn't matter, or we are playing a strong team and that advantage isn't there)

[Gerben42]

Pass = to play in 1NTx

Redouble = to play in 1NTxx (forcing pass)

2C, 2D, 2H, 2S = to play

2NT = strong 2-suiter

3C, 3D, 3H, 3S, 3NT, 4C, 4D, 4H, 4S = to play

 

Brilliant. I like it B)

[foo]

Frances,

 

As usual, a thought provoking post. Thanks. There's going to be some discussions amongst some experts I know due to your post.

 

One thing I'll note is that if possible usually the KNT pair wants the 1N opening to be Dummy and Responder to be Declarer since Responder is the unlimited hand.

 

In that regard, your suggestion is certainly better than most run outs I've seen.

[pbleighton]

and no-one vaguely sane plays the mini vulnerable

 

Nothing vague about me....

 

:)

 

Peter

[MickyB]

Okay, here goes.

 

Playing 5cM and better minor, somewhere around 15-17 is best. This is because, having opened a suit, there exist many competitive auctions where it is uncomfortable for these hands to take a second call, while passing could easily miss game. 1st NV there is a case for 10-13/11-14/14-16 due to the preemptive effect, but in 2nd seat this is much less significant (only one opp to preempt, they won't have 13 opposite 13 and miss 3NT). 10-12 or 9-12 is taking it too far, you are doing serious damage to your constructive bidding - 10-13 is bad enough.

 

In systems with a way to show a strong balanced hand (min 17/18 points) with a 1NT rebid, there is a good case for 14-16 NT, as seen in Swedish Club and many short club/xfer responses systems. In Siege, there are many auctions where opening 1♣ and then doubling shows 17-19 bal - for reasons I won't get into, we don't need a pure takeout double here.

 

In a strong club system, you have the choice between having one or two balanced ranges below your strong club opening. The former uses a 11/12-14/15 1NT opening, the latter uses 14-16 or so except 9-12 is probably better 1st NV.

 

The best way to play a weak NT is to use 1♣ as a very sound short club. Millennium Club/Nightmare play 1♣ as 15+ nat or bal (or any GF) with a natural 2♣ opening. Fantunes uses it as 14+nat or 15+bal (or something like that). 1st NV, Siege uses a 10-13 NT with 1♣ as 13+ with clubs or 14-19 bal (we have to pass with 11-12 points and clubs).

 

If you like to open 4 card majors in 3rd+4th seats, you should set your NT range according to the soundness of your openings in 1st+2nd. If you open flat 12s but not flat 11s in 1st+2nd, then you want your range in 3rd+4th to be 14-16/17 so after P:1M, 1N:P you have a maximum of 13 opposite 11 - I would prefer to be in 3NT with that, but not by much, and it is easily worth the other gains of the method.

 

Btw, I agree with everything Frances said, except I like to be able to run to 2♣ intending to rescue if doubled.

[jtfanclub]

"A" is !NOT! better than "C" at finding the correct strain. It's worse.

 

"C" has 3 HCP ranges both for the 1N opening and the 1N rebid and a 2 HCP range for the 2N rebid.

 

"A" has at best a 5 HCP range for the 1N rebid and a 3 HCP range for the 2N rebid unless you accept the 1N rebid having a 6 HCP range.

 

In addition, the stronger the 1N opening, the weaker Responder can be and still take bids probing for the best strain and level. "A" has the weaker 1N opening so therefore Responder must be stronger to have the values for said probes.

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

Wow. This is so incoherent that I can barely answer it.

 

OK, let's start with the 1NT rebid. It makes no difference what the range it is in terms of STRAIN. If you can't find a good place to play after a 'natural' one club opening, play something else. Does a large 1NT rebid range make it more difficult to find the right LEVEL, ie. whether we have game or not? Sure. Does it affect STRAIN? Nope. I can't say I've had any trouble finding out fits, even if the opponents interfere at the 1 level or 2m level.

 

The more hands I open at 1m, the more likely I am to find my fits. For example, if I have 16 and my partner has 6, if I open at 1NT we will miss our 4-4 major suit fit. If I open 1 of a minor we will be sure to hit it.

 

KNT NT opens 1 of a minor for ranges 12-19 for balanced hands (assuming that strong 12s open 1 of a minor). Strong NT opens 1 of a minor for ranges 12-14 and 18-19 for balanced hands. Therefore, KNT is better at finding strain for this range.

 

For the other range, 10 to a bad 12 (or 9, or 8), opening vs. not opening is the issue. I believe that opening 1NT is better at finding the strain, particularly with a good runout.

 

If partner has a 'real' opening count (good 12 or better)

If I open 1NT, partner will now look for strain.

If I pass, partner will open.

 

This can be a wash, or better for the pass, but...

 

If partner doesn't have a 'real' opening count:

If I open 1NT, partner has a raft of options. He can bid a 5 card suit, he can bid Stayman, or we can runout to the correct strain.

If I pass, partner has to pass. Oops. No chance to find the correct strain.

 

Then there's the last case.

If I pass with KNT, and partner opens, he knows I have 0-9 hcp.

If I pass with strong NT, and partner opens, he knows I have 0-11/12 hcp.

 

I can explore for a suit much more strongly if I had a KNT opening I didn't use. I don't have to worry that partner's going to misread me for a balanced 11 count.

 

Therefore, A is better than C at finding the correct strain at all ranges.

 

Now, let's hear an argument from you that doesn't mention level at all. On how you find a correct strain more often by bidding 1NT instead of bidding 1 of a minor. Or how you find it better by passing instead of opening 1NT. Or by having a narrower range for your pass.

 

Because I asked you to prove strain, and you gave me an argument about level.

[foo]

jtfanclub,

 

We seem to be having trouble communicating here.

 

Richard's comment about the goal of Bridge being to get the best score is dead on.

(I happen to disagree with some of his strategies for pursuing that goal, but he's quite right about the goal.)

 

To do that, we must consistently do whatever has the highest probability of scoring well on each board.

 

To do =that=, we must consistently be in the right spot on each board.

The right spot involves both strain and level. There is no way to decompose them.

If you want them to be independent of each other, go (back to) play(ing) Whist.

 

On many boards the "right contract" is "no contract". That is to say our proper strain is no strain because We should be defending if we want to maximize our chance of a good score.

 

IOW, we want to as consistently as possible hit Absolute Par on each board. To do better than that requires that the opponents play Bad Bridge. Not something one should count on. Especially if one wants to be a good player.

 

"C" is provably better than "A" at finding the correct strain because

1= it is provably better at finding the best spot when we should be bidding and

2= it is provably better at making Us the defenders (IOW, playing the strain of "no strain") when it rates to be right; and doing it while telling the opponents as little as possible about how to play the hand.

[awm]

I'm not sure how the data you've collected really tells us "what is the best notrump range." One main point does shine through though:

 

If your general bidding is based around ZAR points (or even if you just believe ZAR points are an accurate measure of hand strength), but your 1NT opening is based on traditional "high card points" then you may want to consider subtracting a point for 4333 shape. Note that the 4333s have notably fewer ZAR distribution points (8) than 4432 (10) or 5332 (11). Most valuations will indicate that 4333 is weaker than 4432 with the same high cards. So you could play something like "10-12, but 11-13 if 4333" and that might be consistent with your general approach.

 

Let me make the unpopular point that opening 1NT as 19-21 is better than 16-18 in a strong club context. Consider the following:

 

(1) Say the auction starts (opponents passing) 1♣-1♦. One of my fairly cheap rebids is 1NT. If this is to be natural, and a 1NT open would be 16-18, then 1NT rebid is 19-21. This is quite a rare hand to hold in the context of any 16+ points and a waste of a cheap rebid.

 

(2) Say the auction starts 1♣-pass-1♦-2♠. If I hold 19-21 balanced, do I pass? Partner could easily have some balanced 5-count (not enough to make a positive response) and game could be cold. Or partner could have some balanced 0-count and if I bid I could go for a number. Total guess here. Contrast this to my holding 16-18 balanced. If partner's bidding 1♦ over my 1♣ open, we probably don't have game and I can safely pass the overcall.

 

(3) Say I have 16 balanced with a five-card spade suit. If the 1NT opening is 16-18, I have to immediately decide what to open. If I open 1NT and partner has too little to invite, it is likely that I will play 1NT when we could easily be better off in spades. We might even have a spade game opposite some hands. On the other hand, I can't open 1♠, and if I open 1♣ I'm going to be stuck rebidding spades even if my spades are lousy because 1♣...1nt would be a stronger hand. On the other hand, if 1NT is 19-21, I can open 1♣ and plan to rebid spades if partner shows weakness (finding the best partial and the only possible game) or 1nt if partner shows strength (describing the essential balanced-ness of my hand in a game-going auction). Note that a "similar problem" doesn't really come up with 19 balanced and five spades, since 1NT is almost always safe and partner will almost always find a call over 19-21 balanced in any case.

 

(4) Say I open 1♣ and partner responds 1NT. Regardless of how you re-arrange your responses, it's likely that there is some hand on which partner responds 1NT. This potentially wrong-sides a notrump contract. When is it most important to play from the stronger side? Usually when the stronger side is much stronger and fairly balanced. In other words, exactly the 19-21 balanced opening. Better to open this hand with 1NT (guaranteeing opener declares the final contract in almost all cases) and let the weaker NT hand open 1♣.

[Apollo81]

Coming from a strong club context, I prefer 14-16 for two main reasons.

 

1. If I open 1♦ and the opps interfere, I prefer to have no extras so I can pass (unless forced) with a clear conscience.

 

2. Most opponents defend 14-16 as a strong notrump. Many do not play penalty doubles, and many play very wide ranging overcalls. Thus the 14-16 opener can occasionally talk the opponents out of a game. This seems to happen a lot less frequently playing 15-17.

 

The one other NT range I would recommend is 10-12, but only 1/2W at MPs in a strong club context.

[blackshoe]

In his book The Power System, Ron Klinger told us "all [opening] NT ranges are wrong" - and then gave us a range of 17-20 for that bid. :P

 

I agree with Justin that 10-12 probably works better with a system whose 1 level suit openings (except perhaps 1♣) are limited, and I'd say better still at favorable (or at least not vulnerable) vulnerability.

 

"Standard" Romex doesn't use the mini-NT, but 1NT is artificial and the natural one openings are limited to about 18 HCP. A 1NT rebid shows 12-16, and the system using 2 way checkback by responder to clarify that. Works well.

 

You can play Romex with mini-NT. 1NT rebid now shows 13-16, and 2 way checkback is on. If I were doing this, I'd vary according to vulnerablity and scoring.

 

The current Romex approach varies the convention card according to vulnerabiltiy. At favorable (IMPs) or not vulnerable (MPs), 1♣ is forcing, 17+, and 1NT is 10-12, so the rebid (after a natural opening) is 13-16. At his vulnerabiltiy, in 3rd seat, 1NT is 12-14, and the rebid is 15-16. 1♣-1♦-1NT is 17-20.. The relevant NT ladders are:

 

Standard Romes

• 12-16: open 1any, rebid 1NT
  
• 17-18: open 1any, rebid 2NT
  
• 19-20: open 1NT, rebid 2NT
  
• 21-22: open 2♦, rebid 2NT
  
• 23-24: open 2♣ rebid 2NT
  
• 25-26: open 2NT
  
• 27-28: open 2♦, rebid 3NT
  
• 29-30: open 2NT (or 2♣), rebid 3NT
  

and

 

Romex Foricing club

• 10-12: open 1NT (in 3rd seat, 12-14)
  
• 13-16: open 1any, rebid 1NT (in 3rd seat, this shows 15-16)
  
• 17-20: open 1♣, rebid 1NT
  
• 21-22: open 2♦, rebid 2NT
  
• 23-24: open 1♣ rebid 2NT
  
• 25-26: open 1♣, rebid 2♥ (Kokish relay, forces 2♠, then 2NT
  
• 27-28: open 2♦, rebid 3NT
  
• 29-30: open 1♣, rebid 2♥, then 3NT
  

You can use Kokish playing "standard" Romex as well, freeing 2NT for some other meaning in both cases.

[firmit]

(...)So you could play something like "10-12, but 11-13 if 4333" and that might be consistent with your general approach.(...)

 

Let me make the unpopular point that opening 1NT as 19-21 is better than 16-18 in a strong club context. Consider the following.

4333 - 1hp. I agree, and this has been implemented.

 

I like you reasoning reagarding the 19-21hp 1NT range too.

[jtfanclub]

jtfanclub,

 

We seem to be having trouble communicating here.

 

Richard's comment about the goal of Bridge being to get the best score is dead on.

(I happen to disagree with some of his strategies for pursuing that goal, but he's quite right about the goal.)

 

To do that, we must consistently do whatever has the highest probability of scoring well on each board.

 

To do =that=, we must consistently be in the right spot on each board.

Good Lord. Leaps...In...Space!

 

I don't know what you're playing where you think the way get the best score is to be 'consistently be in the right spot on each board'. But it isn't bridge. Or Whist, either. At a guess, what you're playing involves double dummy and the opponents being barred from the bidding. But that's just a guess.

 

I'll wager that you can think of several things that are more important than finding 'the right spot'.