[MickyB]

Given that I haven't (yet) joined Glen and DavidC by starting a blog on bidding theory, I'm going to post some thoughts here.

- I believe that, as far as possible, balanced hands should show their strength with their opening bid (see DavidC's bridge blog). It seems many top players agree with this, as a fair few are using 2m openings to show 18-19 balanced in the Europeans.

- The ranges used in 3rd+4th should depend on those used in 1st+2nd. Basically, your lowest range should rule out bidding to game unless you have a fit, which IMO should mean that 24 points between the two hands is possible, but 25 is not. In practice, this means that, in either 1st+2nd or 3rd+4th, your lowest range will be 11-13, and your next range will be 14-16.

These two points conflict greatly. While I don't too much mind opening a natural 1m on 18-19 balanced, doing so on a balanced 17-count is undesirable, to say the least - 17-counts are a fair bit more frequent than 19-counts, and more dangerous to take another call on. Likewise, opening 2m with 17-18 or 17-19 seems to be taking a reasonable idea too far.

So, to solve this, we need a "tri-bal" system - one which allows us to show three balanced ranges immediately at the one-level.

The most obvious solution is strong club - 1♣ shows 17+, 1♦ includes 11-13 balanced, with a 14-16 1NT opening. The problem with these systems is that they give you problems on the club hands - especially when holding 4M5♣ and 14-16 or so (the 11-13 4M5♣ hands can get away with opening 1M IMO, but the 14-16s cannot - see more from DavidC). Likewise, strong diamond gives you the same problem on diamond hands.

I've posted one attempt at solving these problems already, and here is another one that I haven't really evaluated yet -

1C 17+, not 4M5+m
1D 11-13 bal/14+unbal 4M5+m
1M can be 4 if 10-13
1N 14-16
2m 11-16 no 4M

It might be better to put the 14-16 minor single-suiters into 1D, and make the 2m openings 9-13 - preemptive and constructive like Fantunes, but without the worry of opener having a four-card major on the side.

These structures bring me to a third point. It's advantageous to separate balanced minimums from unbalanced minimums, as in Polish Club, Swedish Club and the structures above. This is because you can then treat the opening as a weak no-trump; negative freebids become much easier to handle and useful when partner has a stronger hand too.

And, just to eliminate one possibility, something like 1♣ any 14-16 and 1♦ any 17+, with an 11-13 NT isn't an option - firstly because I don't think a weak NT is playable in 3rd seat, and secondly because it wouldn't be legal in the EBU - an artificial 1m opening cannot include unbalanced hands with 5+M and fewer than 16 points (unless it is a transfer opening, of course).

Hopefully I'll find something that satisfies all my aims

[kenrexford]

You started great, IMO, about the 14-16 1NT bids in 3rd/4th seat. That's what I do.

Where you lost me was in coming up with some contorted nonsense to avoid rebidding 2NT with a 17-count. I don't get the problem.

I mean, if the 17-count is a P.O.S., then open 1NT because it is really a 16-count. Otherwise, WTP?

[glen]

If you want 11-19 balanced to open 1x, and you can't use 1M to handle a range, and you don't like 1NT with 11-13, your splits of 11-13, 14-16, 17-19 will be:

S 11-13/14-16/17-19
c1) 1♣ / 1♦ / 1NT
c2) 1♣ / 1NT / 1♦
c3) 1♣ / 1♦ / 1♦
c4) 1♣ / 1♣ / 1NT
c5) 1♣ / 1NT / 1♣

d1) 1♦ / 1♣ / 1NT
d2) 1♦ / 1NT / 1♣
d3) 1♦ / 1♣ / 1♣
d4) 1♦ / 1♦ / 1NT
d5) 1♦ / 1NT / 1♦

I believe that d2 (as seen in Meckwell etc.) is the best:
- 1NT is frequent
- 1♦ 11-13 gets to finding a major fit as first priority
- 1♣ 17-19 gives room for best game/slam investigation
- no shared used, such as the Welland 1♣ in a c5 scheme

I've wondered if it is better to flip the unbal minor suits:

1♣: ♦s Unbal or 17-19 Bal
1♦: 14+ ♣s Unbal or 11-13 Bal
1NT: 14-16
2♣: 10-13 ♣s Unbal

[MickyB]

kenrexford, on Jun 27 2008, 03:56 AM, said:

Where you lost me was in coming up with some contorted nonsense to avoid rebidding 2NT with a 17-count. I don't get the problem.

I mean, if the 17-count is a P.O.S., then open 1NT because it is really a 16-count. Otherwise, WTP?

Sorry, I wasn't clear - I was looking at was avoiding problems *in competition* on balanced 17s. After 1♣ (3♠), you are just guessing with 17-19 balanced.

I do think that rebidding 2NT on 17 is a problem - sometimes it's too high already, sometimes it preempts you out of room to investigate - but when I play 14-16 I arrange for a 17-19 1NT rebid, so that's not my main concern here.

[kenrexford]

OK, but by the structure that you have, it seems that you are simply changing the problem. Balanced hands are more easily handled but unbalanced hands are less easily handled. When the opponents do intervene, it seems that you will tend to be unbalanced more than balanced and will tend to want to act when unbalanced more than when balanced, but you have provided a better structure for the balanced hands at the cost of the unbalanced hands.

I don't get it. But, I may have a bias to pattern. I played canape for years, and I have almost always tweaked 1♣ and 1♦ to enhance pattern messages to the 1♦ opening (usually 4+ and unbalanced).

[glen]

First on blogs and bridge theory, note Ken's excellent work at: cutebiddingatbridge (okay, its cue) and Ulf's great (when he's not too too busy playing top level bridge): viewsfromthebridgetablebehindscreens

In ACBLand, the new mid-chart 15+ change will result in some interesting tri-structures - I'm posting this here as only those reading this far in this thread are likely to be interested in these options.

For example, we can have:
1♣: 11-13 bal OR various unbal
1♠: 17-19 bal or semi-bal
1NT: 14-16

In a structure with two-point ranges for 11-16 with a four card major:
1♣/♦: two-under transfer, balanced only if 11-12
1♥: 15-16 Bal or 15+ unbal with no four card major
1♠: 17-19 bal or semi-bal
1NT: 12-14, not 12 if a 4/5 major

Switching to 12-14/15-17/18-20 splits:
1♣: 18-20 Bal or natural unbal
1♥: 15-17 Bal or 15+ 5+♥s
1NT: 12-14

and 12-14/15-16/17-19 splits:

1♥: 15-16 Bal or 15+ 5+♥s
1♠: 17-19 Bal or 15+ 5+♠s
1NT: 12-14

If you want to comment/suggest/flame without posting, feel free to PM me.

[kenrexford]

As a complete aside, maybe.

If you are going to tweak the handling of balanced hands, why not consider tweaking them on a basis that reflects the unmanageable portion of "balanced" rather than the manageable portion.

As an example, suppose that you have a call that shows a range of "11-14." Like, opening a minor. 11-14 is a fairly wide range, admittedly. So, let's suppose we decide to break these 11-14 openings between 1♣ and 1♦. What options are there?

1. Traditional. The traditional distinction is that we open the "better minor," maybe with some wildly debated minor issue like what to open with 4432 shape. This helps in inferential analysis of minor suit lengths. Whoopdie Doo!

2. HCP range. Another option is to distinguish tight HCP ranges. Say, 11-12 or 13-14. That has some merit and is the idea here.

3. Control Count range. An 11-count with AKA has five controls. A 14-count with AKAK has six controls (and is probably too rich). A 12count with QJQJQJQJ has 0 controls. A 14-count with QJQJQJAJ has two controls. So, a 14-count can have 2 controls but an 11-count 5. The net control count range is about 0-5. That's a wild difference, and much more substantial than the pure HCP range of 11-14. Controls come in handy when seeking slam or when contemplating "grab 9 and go" 3NT contracts. Controls come in handy when doubling. So, have you considered he merits of something like 1♣ for 0-2 controls and 1♦ 3-5 controls, both balanced 11-14. Or, 0-3 and 4-5? That would seem to be a more useful distinction.

4. Pattern divergence -- major fragments. Consider the merits of one call promising at least one three-card major and the other denying a three-card major (could have 4 or 0-2, but not 3). For example, if 1♣ promised a three-card major (if balanced), then Montreal Relay responses might make sense. Conversely, if 1♦ denied a three-card major, this helps with assessment of what to do after a 1NT rebid. Both are enhanced in checkback sequences.

There are other problem-solving distinctions that make more sense to me than obsessing on range of HCP, IMO.

[OleBerg]

I have (partially) solved it this way:

1nt = 12-14
1♣ = Natural or 15-19 Bal.

Over 1♣ I play transfer responses (Walsh principles), so:

1♣-1♦/1♥
1♠ = 15-19 bal.

Now responder checks back:

1nt = To play, assuming 15-17.
2♣ = Gameforce. Opener bids 2♦ with 18-19, anything else is 15-17.
2♦ = Invitational, assuming 15-17.

When responder bids 1nt, opener can bid a four card suit at the two-level with 18-19. Thus you will always get to two clubs, and often higher, when opener has 18-19, but you have the chance to play at the 2-level.

If you dont like the weak nt, I know others have exchanged ranges, so that:

1nt = 15-17
1♣= Natural or 12-14 bal or 18-19 bal.

This scheme also frees the 1nt-rebid, making it possible to construct a system, where yoy always respond with 4 hcp and a four-card major. (I use transfer by openers first rebid, but there are many options.)

[MickyB]

Quote

When the opponents do intervene, it seems that you will tend to be unbalanced more than balanced

I seem to remember seeing some statistics suggesting this factor to be negligible.

Quote

[you] will tend to want to act when unbalanced more than when balanced

Precisely.

Unbalanced hands with extras are usually happy to take two bids, they certainly won't describe themselves well without doing so. Balanced hands are one bid hands, you want to show your approximate shape+strength then get out of the auction. This is what I'm trying to achieve.

[MickyB]

glen, on Jun 27 2008, 04:53 AM, said:

If you want 11-19 balanced to open 1x, and you can't use 1M to handle a range, and you don't like 1NT with 11-13, your splits of 11-13, 14-16, 17-19 will be:

[snip]

Thanks Glen, didn't get the chance to respond to this before I went away for the weekend but it was useful for my thought processes.

It made me realise that there are only two sensible ways of doing what I've set out to do without using artificial 1M openings -

Play 1m as 17+ or so, in which case the other minor will have to include 14-16, 4M5+m.
Use a 17-19 NT

The former is probably preferable.

So, playing strong club, the 1D opening will include 11-13 bal and 14-16 4M5+m. It's just a case of what else to stick in out of -

4M5+m 11-13
4M5+m 17+
5+m, no 4M, 14-16

[blahonga]

MickyB, on Jun 30 2008, 12:57 AM, said:

So, playing strong club, the 1D opening will include 11-13 bal and 14-16 4M5+m. It's just a case of what else to stick in out of -

4M5+m 11-13
4M5+m 17+
5+m, no 4M, 14-16

I think you should take a look at what Bertheau and Nyström play, maybe without the variable no trump opening. Something like

1♣ 16+
1♦ a: 11-13 bal b: 11-15 4M5+m, c: 12-16 4441
1NT 14-16
2m 11-15 5+, no 4M
2NT 12-15 55+m

[OleBerg]

A simple solution in strong Club systems is:

1nt = 12-15

If it is to wide, include 15 bal. in the 1♣ opener. Then 1♣ - 1nt isn't gameforce, but thats easy to live with.

Edit: And let 4-4-1-4 hands open 1♥, then 1♦ will show an unbalanced hand with 4+ diamonds.

[MickyB]

OleBerg, on Jun 30 2008, 09:53 AM, said:

A simple solution in strong Club systems is:

1nt = 12-15

If it is to wide, include 15 bal. in the 1♣ opener.

This isn't really a solution, as such, but it is an alternative philosophy to have for a strong club system.

I believe that there are conditions where it is superior to play a 15/16+ club, with descriptive, limited openings that will frequently allow responder to place the contract immediately at the expense of our constructive bidding. I have my own pet structure that satisfies these aims.

I suspect this is only the right philosophy in first seat - after all, if we are aiming to have responder place the contract immediately, it will be best if he does so before one of the opponents gets the chance to speak at all. Even in first seat, I suspect that it's better to focus more on constructive bidding at unfavourable, when we have least to gain by preempting and the opponents have most to gain by preempting over our strong club.

[MickyB]

blahonga, on Jun 30 2008, 08:42 AM, said:

MickyB, on Jun 30 2008, 12:57 AM, said:

So, playing strong club, the 1D opening will include 11-13 bal and 14-16 4M5+m. It's just a case of what else to stick in out of -

4M5+m 11-13
4M5+m 17+
5+m, no 4M, 14-16

I think you should take a look at what Bertheau and Nyström play, maybe without the variable no trump opening. Something like

1♣ 16+
1♦ a: 11-13 bal b: 11-15 4M5+m, c: 12-16 4441
1NT 14-16
2m 11-15 5+, no 4M
2NT 12-15 55+m

Yup, that is one of the options, have checked out their continuations - thanks.

[MickyB]

MickyB, on Jun 30 2008, 06:57 AM, said:

So, playing strong club, the 1D opening will include 11-13 bal and 14-16 4M5+m. It's just a case of what else to stick in out of -

4M5+m 11-13
4M5+m 17+
5+m, no 4M, 14-16

So, which of these options is best?

4M5+m minima need to go somewhere. The options are 1D, 1M and 2m. I'm inclined to think 1M is best -

If you put them into 1D, you could no longer assume that partner has 11-13 balanced and bid accordingly. Negative freebids become much more problematic, to the point where it might be better to play bids as forcing instead. It's much like the difference between competing opposite a Polish Club compared with opposite a Short Club, only it is even more significant.

The disadvantages of opening 2m are obvious - you risk losing major-suit fits.

1M seems the least problematic, I don't think it indicates raising any less aggressively in competition - you'll often be happier opposite 4315 than 5323, whatever the LAW may say. On the other hand, the other two options both solve what to do with 4M6m minimum, whereas this still leaves you with a decision between 1M and 2m. Maybe you should either "upgrade" these hands to 1D, or pass and back in later implying the side-suit.

Removing 4M5+m hands from the strong club opening would help these hands a bit in competition - no risk of losing the major suit when you bid the minor. However, we're already opening 1C fairly infrequently, so this doesn't feel quite right.

Removing 5+m, no 4M from the 2m openings seems pretty useful. It makes them narrower in range and available on lighter hands, and doesn't seem to pose any significant issues to the 1D opening. Think I'll run with this idea.

So, I now have two structures I have advocated. They have a fair bit in common -

1M is limited to 16, either 5+cards or 10-13 with 4M5m.
2m is 9-13, either 6 cards or 5-4 minors.
2M and higher are available for preempts

"Revision" -
1C = 11-13 bal/14+clubs/various 17+unbal/20+bal
1D = 14+ bal or diamonds
1N = 17-19

"Tribal"
1C = 17+
1D = 11-13 bal or 14-16 unbal without a five-card major
1N = 14-16

Which is superior?

11-13 balanced does slightly better in Tribal, in that 1D:1M, 1N or 1D:P are better tactically than opening 1C (giving the oppo more room when we don't benefit much from it) and disclosing more about our hands to the oppo (1C:1M, 1N shows we have reasonable values playing Polish Club style, and 1C:1D, 1M sequences give away more info too).

14-16 balanced gets to open 1NT in Tribal, the advantage of this is difficult to quantify without empirical data but I suspect most will consider it to be reasonable.

17-19 balanced opens 1NT in Revision and a strong club in Tribal, I'd expect this to be a small gain for Revision [thoughts?]

So TriBal is a fair bit ahead on balanced hands - actually I'd expect TriBal to be a fair bit ahead of any other system on balanced hands, although some four-card majorites might argue otherwise.

14-16 with clubs is probably a wash. In TriBal, you are ahead on auction like -

1m (2m) 2H (P), because you can bid 2S to indicate 4S5+m.

On the other hand, the Polish/Revision style will sometimes have better definition in uncontested auctions, e.g. 1m:1H, 1S shows an unbal hand in both cases, but in TriBal the minor suit has not yet been determined.

14-16 with diamonds probably does better in Revision. again because the minor has already been determined. Likewise 17+ with a primary minor suit.

Feels to me like TriBal is ahead overall.

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

What about TriBal vs Swedish Club? For simplicity, I will compare it with the version of Swedish that opens 1M on 4M5C.

17+ hands are showing their strength immediately in TriBal but not in Swedish. This has obvious advantages, but disadvantages too - it encourages aggressive preemption. Still, I'd expect 17-19 balanced to fare a fair bit better, and the other hands to be slightly ahead too.
11-13 bal slightly ahead in Tribal for the reasons mentioned above.
14-16 bal equal
5+M and 4M5C slightly ahead in Tribal (because you're not opening 1M on 14-16 4M5C, which can be slightly problematic)
6+D min slightly ahead in Tribal, other diamond hands (including 5C4D) a fair bit ahead in Swedish because they can open a natural 1D
6+C ahead in Tribal (tighter range, although 5C4D is now a possibility)

Seems like TriBal is probably further ahead here.

So, have I missed something, or is this actually pretty good? Might have to give the system a proper run-out!

[awm]

Agree that Tribal is the best of these options. It also looks amusingly close to what Sam and I play.

A few things you didn't cover however:

(1) What do you bid with 4441 hands? Presumably you either open a 4cM or 1♦, but it's not clear which.

(2) There are some issues with 4M-5m hands opening 1M. For example take the auction 1♠-1N-2♣. If you don't know which suit is longer, it is not clear what responder is to do with 2-2 or even 2-3 in the blacks. There are also competitive issues, take 1♠-3♦-??? for example, with three card spade support. Opposite five spades you may well want to bid 3♠, opposite four spades your best fit could be in clubs and it's not clear you want to bid to the four-level or to play 3♠ on a 4-3.

(3) It is not really clear what you open with 4M-6m hands in the 10-13 range. If 1M then the problems mentioned in (2) become even more severe. Even moscito has given up opening with a call showing the four-card major when holding a 6cm in the latest versions.

(4) What do you open with extreme hands with both minors (say 5-5, 6-5)? Opening 2m seems too likely to be passed (it doesn't promise the other minor at all, and your offensive potential is awfully high) whereas opening 1m doesn't necessarily allow you to show both suits (presumably rebidding 2m shows a one-suiter).

What Sam and I play is close to Tribal, as mentioned before. The 2m opening is more like 11-15 and guarantees six. We will probably tend to do better when we open 2m and a bit worse when you open 2m and we don't. The 1♦ opening is then 11-13 balanced or 11-15 with a 3-suited type pattern (including 4225 as balanced and 4351/3145 as three-suited) or 11-15 with 5-5 in the minors. This lets partner pretty much assume the balanced option in competition (way more frequent, plus three-suiters are kind of like balanced hands in some ways). Finally, we play true 5-card majors with 4M-5m hands opening 1♦ (they are either 4225-ish semi-balanced hands or 4351-ish "three suited" hands).

[MickyB]

awm, on Jul 15 2008, 09:49 PM, said:

Agree that Tribal is the best of these options. It also looks amusingly close to what Sam and I play.

Yes, thought it looked a bit like that - tried looking for your notes and could only find something similar with 1C and 1D inverted.

Quote

(1) What do you bid with 4441 hands? Presumably you either open a 4cM or 1♦, but it's not clear which.

As 4M5m - so 1M if min, 1D if intermediate.

Quote

(2) There are some issues with 4M-5m hands opening 1M. For example take the auction 1♠-1N-2♣.

The 4M5m hands pass 1NT.

Quote

(3) It is not really clear what you open with 4M-6m hands in the 10-13 range. If 1M then the problems mentioned in (2) become even more severe. Even moscito has given up opening with a call showing the four-card major when holding a 6cm in the latest versions.

Yup, I mentioned this in my last post - don't think 1M is an option really.

Quote

(4) What do you open with extreme hands with both minors (say 5-5, 6-5)? Opening 2m seems too likely to be passed (it doesn't promise the other minor at all, and your offensive potential is awfully high) whereas opening 1m doesn't necessarily allow you to show both suits (presumably rebidding 2m shows a one-suiter).

2N as 5+/5+ minors, 9-13, is probably worthwhile.

Quote

The 1♦ opening is then 11-13 balanced or 11-15 with a 3-suited type pattern (including 4225 as balanced and 4351/3145 as three-suited) or 11-15 with 5-5 in the minors. This lets partner pretty much assume the balanced option in competition

Do you play NFBs?

So, my assessment of your stuff vs TriBal - for simplicity, assume the strong club starts at the same level.

17+ any and 14-16 NT - the same
You are ahead on 5M
You are probably very slightly ahead on 6m hands - you do better at showing shape, we do better at showing range
We are ahead on other hands - 4M5m, minors and 11-13 bal

[awm]

A version of our notes is at cs.ucla.edu/~awm/bridge/IMPrecision.pdf

I'm not convinced that opening 1M with 4M/5m puts you ahead on these hands, and I'm also pretty sure we are doing better on 11-13 balanced. The point is that the hands that open 1♦ in our style are all quite similar. There is not really that much difference between 4441 and 4342 for example. On the other hand I think 1237 or 1426 is effectively a very different hand from any balanced shape you can name (no we don't consider 2227 as balanced). Opening 1♠ and passing 1NT with 4315 when you could have a ten card club fit (for example) doesn't really appeal -- I think this is different from opening 1♠ with some flat hand like 5332 or 4324 and passing 1NT. On the 2m hands, your range is as wide as ours. You're just opening 2m on lighter hands and moving the heavier hands to 1♦. But our opening is more descriptive in terms of shape (6+ in the minor opened, won't have five in any other suit). Admittedly you might win by preempting on the 9-10 point hands that we generally pass (but some of these will qualify for 3m anyway).

We don't play negative free bids -- we currently play transfer free bids over 1♦ and interference. Opener's cheapest reject of the transfer shows a three-suiter short in the suit transferred to.

[MickyB]

awm, on Jul 16 2008, 08:12 AM, said:

I'm also pretty sure we are doing better on 11-13 balanced.

I'd be surprised. The key to the 1D opening I've suggested is that responder can bid as if opposite a 1NT opening, and if opener is unbalanced he will have extras with it, and a clear pull of the NFB, describing his hand quite precisely in the process. I can see what you are saying about 3145 and 3244 being similar, but that won't make the 5-1 heart fit play any better I may be biased here, I've played a fair bit of Polish+Swedish club over the last year or so and I really like the effect of separating the balanced and unbalanced minima.

Quote

On the 2m hands, your range is as wide as ours. You're just opening 2m on lighter hands and moving the heavier hands to 1♦. But our opening is more descriptive in terms of shape (6+ in the minor opened, won't have five in any other suit). Admittedly you might win by preempting on the 9-10 point hands that we generally pass (but some of these will qualify for 3m anyway).

Maybe 'range' wasn't quite right word. I could choose to play 2m as 11-13 if we wanted a direct comparison, but I think opening tens (and some nines) is probably worth it.

If you only have one way to show minor hands, it will show 11-15, and the 16-counts have to be opened 1C. This will have a negative impact on 17-19 balanced hands. [Of course, it isn't clear that this is a bad thing, because a 16+ club limits the 1M openings a little more as well].

[lexlogan]

MickeyB, in terms of what you're trying to accomplish, please explain what's wrong with:

1C = 17+ balanced, 16+ unbalanced
1NT = 14-16
1D includes any 11-13 balanced.

1C opener passes over interference with 17-18 balanced. Opener strives to bid or double with 19+ balanced.

Also, note to Glen -- I haven't a clue what your post means -- 1♣ 1NT 1♦ ? What on earth are you saying?

Oh, wait, I get it -- you've listed three ranges at the top, and are specifying the opening bid for each range. So d2 is the scheme above.