[rbforster]

I've been reconsidering my 2/1 bids over 1M in my strong club system (1M shows 5+M and 9-15 pts). For a long time I played this as GF ala the familiar 2/1 style, but recently I've been having second thoughts. What do you think about using 2/1 as "weak 2 bid" type hands - a decent 6 card suit and constructive but not particularly forward-going in terms of values (~6-10)?

In place of the strong 2/1 response, stronger hands would now either use a conventional raise (Bergen, Jacoby, etc) to show support, or bid 1NT forcing. After 1NT forcing, I found that using the cheapest new suit as an artificial GF works well for the GF hands, while invitational hands can bid naturally (2NT, raise partner, or bid their own good suit). Note that other weak hands (besides single suiters) can still bid 1NT forcing and then pass or take a preference to partner's major - this would only effect the single-suited weak hands.

I tried playing this way for a while and these "weak 2" type hands didn't seem to come up much. Maybe this was just my experience, but it doesn't make sense to have lots of cheap bids used for hands that aren't that likely. Another thing which I've been told is a drawback of this method is that it's vulnerable to bidding by the 4th seat after 1M-P-1NT*-(2Y). I'm not sure this is any worse than in standard however - especially given our light openings and that 1NT could be made on many of the usual weak hands too, I would think that 2Y would need to be more sound rather than less sound relative to the same bid in a 2/1 auction.

Any suggestions or experiences? I'd welcome pointers to other systems that make use of non-forcing 2/1's. Thanks!

[hrothgar]

There's lots of debate regarding "optimal" response structures to light opening bids...

I've seen two main schools

1. Transfer based response structures... There is a good write up in the Magic Diamond notes
2. 2/1 as Natural and non-forcing. First step relay is used for hands with game invite+ values

From my perspective, the first structure is probably better for constructive bidding. The second structure places more pressure on the opponents. The 2/1 bids deny a fit and are very dangerous to come in over...

Here's the response structure that I use over MOSCITO style 1♥ openings. (1♥ = 4+ SPADES)

4♣+ = Splinters
3N = To play
3♠ = Value raise, 4 card support 8+ losers
3♥ = fit showing, 6+ Hearts and 3 Spades
3♦ = Fit showing, 6+ Diamonds and 3 Spades
3♣ = Fit showing, 6+ Clubs and 3 Spades
2NT = Limit raise+ (4+ Spades, less than 8 losers)
2♠ = Value raise, exactly 3 spades
2♥ = 5+ Hearts, 7-11 HCP
2♦ = 5+ Diamonds, 7-11 HCP
2♣ = 5+ CLubs, 7-11 HCP
1N = Natural, non-forcing
1♠ = Relay

[pbleighton]

I notice in your profile you are aware of the ACBL's "oppressive regulations" - nice phrase.

In case you aren't completely aware, if you want a GCC system, the regulatory deck is very heavily stacked against you if you want to play NF 2/1s. I tried to figure something out for quite a while, then gave up.

Good luck.

Peter

[rbforster]

Thanks for the pointers Richard. I noticed that several places in both your Moscito notes and also in Viking Club's, the natural 2/1 NF responses both used 5+ suits rather than the 6+ that I suggested. This would probably go a long way towards making these hands more frequent. I agree with your assessment that they put a lot of pressure on the opponents, esp. relative to transfer methods like Magic Diamond. Do you have any additional standards in Moscito for these 2/1 bids? I saw your suggestions about being typically 5/4+ or 6+ in your suit(s), but wasn't sure if there were other considerations. I liked the Moscito continuations after the natural NF 2/1 (preemptive raises, 2NT a good raise, etc) and will probably make use of these.

pbleighton, on Aug 15 2006, 05:41 PM, said:

the regulatory deck is very heavily stacked against you if you want to play NF 2/1s.  I tried to figure something out for quite a while, then gave up.

What GCC difficulties did you encounter Peter? The NF 2/1 responses are natural, so they should be fine. Using 1NT forcing (and the normal 2/1 rebids by opener) with many of the inv+ hands also seems ok, and since you can still have the weak hands that will pass or correct to 2M, this didn't violate the rule that 1NT forcing can't guarentee inv+ values. After you get to opener's rebid, pretty much anything goes under the "constructive" bidding rule.

More generally, I'm hoping to include some limited relay methods after the 1NT forcing and new suit GF, although there won't be as much space available. Still, knowing opener's response to 1NT forcing helps convey a fair bit of information already. Another thing to consider is how the availability of 1NT forcing as an option by responder might change which hand types go into the 2/1 NF suit bids.

[awm]

I have a bunch of thoughts on this that seem diametrically opposed to the moscito crowd. Playing "standard" openings, the range is 11-20 or so but the vast majority of hands are in the minimum range. The reasons for this are that the more you increase high card points (above the average of 10) the less frequent the hand becomes, and also that intermediate-range balanced hands (5332 is one of the most common patterns) are often opened 1NT. With this in mind the responding ranges are roughly:

0-5: Pass
6-9: Bid something, but it's not a real invite.
10-12: Invite
13+: Game

Passing on 0-5 is relatively safe because opener is usually in the 11-14 range and you don't even have half the values. Surely you can have a game if opener is max and you have a big side suit fit, but frequency-wise this is very rare.

On the other hand, opposite a range of 8-15 or so the majority of hands tend to be around 11 hcp (close to average). Because of this, passing with hands in the 8-9 point range is uncomfortable as you will often have half the values and would like to find a reasonable partscore. Game is actually not that unlikely as the odds of opener having 15 in an 8-15 range are substantially better than the odds opener has 20 in an 11-20 range. So even though the openings are roughly 3-4 points lighter on both the top and bottom ends, you don't want to routinely pass with most 8-9 counts. This leaves you something like:

0-6: Pass
7-10: Bid something, but it's not a real invite.
11-15: Invite
16+: Game

Notice that the "pass" range doesn't change that much, since the average opener is only about a point lighter and you need to try to scramble to the best partial when you have roughly half the points. On the other hand, the "invite" range is substantially wider and the "game" range is much less frequent.

Let me now give some general rules about bidding:

(1) Shape and fit are very important. Methods which allow you to find fits early are generally more effective than methods that focus only on "points" without exploring for fits.

(2) "Multiplexing" a lot of hand types into a particular bid usually means you don't do all that well when that bid comes up. Even in uncontested auctions it can be hard to sort out exactly who has what. If the opponents happen to interfere, this will become even more difficult.

Most people are aware of these two rules. Even true 2/1 GF devotees will admit that the wide-ranging forcing 1NT response is not the best part of their system, and that they can occasionally obtain poor results when it comes up. The wins from 2/1 GF come when you can bid a game forcing 2/1, and this can help substantially in some followup auctions. People who like 2/1 will argue that the invitational range where there are sometimes "problem" hands is relatively narrow and that a game force is more likely than an invite in standard methods, all of which is true. But once you switch to the lighter openings with a wider invite range and a higher minimum for the game force, it's not true anymore...

The methods you've suggested, where 2/1s are weak hands and all invites and better in addition to balanced weak hands go through 1NT is potentially even worse. Now the 1NT bid is totally unclear about both shape and strength. I'm not saying people will bid over it much more often than a normal forcing notrump; the problem is that when they do bid it will be incredibly hard to sort things out. Even in an uncontested auction, what do you do after 1♠-1NT-2♥? Two spades has a very useful weak meaning and all the other calls have natural invitational meanings. In any case sorting this out will not be without substantial negatives.

Playing the minimum call as "inv+ relay" suffers from a lot of issues. For one, you're multiplexing a lot of hands and vulnerable to interference (hard to sort out who has what, or even whether passes are forcing). For another, if your subsequent bids focus on strength first (i.e. the lowest step shows a min, others jump into GF relays) then you can easily miss a huge side suit fit when opener shows a min and responder passes. Finally you often lose a call with a less-than-invite balanced hand type (in Moscito you can use 1NT for this but if 1♠ was natural and 1NT is relay then you have no call).

In my partnerships where we use 8-15 range openings, I generally play a 1NT bid with a similar range to in "normal" 2/1, and two-over-ones forcing one round, 12+ points but not game force. This gives us two easy invitational ranges (2/1 and make NF call is the good invite, 1NT and followup is the weaker invite). It does create some awkwardness when opener is minimum and responder has a moose (16+) and has to use fourth suit to force game, but frequency-wise this is rare.

Another reasonable option would be to play 2♣ as game force relay and let the other 2/1 bids be natural and invitational. This gives you a lot of natural options with invite hands while retaining relays for the game force and not multiplexing too much into 1NT.

[hrothgar]

Rob F, on Aug 16 2006, 03:59 AM, said:

Thanks for the pointers Richard. I noticed that several places in both your Moscito notes and also in Viking Club's, the natural 2/1 NF responses both used 5+ suits rather than the 6+ that I suggested. This would probably go a long way towards making these hands more frequent. I agree with your assessment that they put a lot of pressure on the opponents, esp. relative to transfer methods like Magic Diamond. Do you have any additional standards in Moscito for these 2/1 bids? I saw your suggestions about being typically 5/4+ or 6+ in your suit(s), but wasn't sure if there were other considerations.

1. A 2/1 bid denies 3+ card support for partner's suit

2. A 2/1 typically denies a balanced hand. (Balanced hands will almost always prefer to bid 1NT. The only exception would be a 5332 with concentrated values)

[glen]

In ETM Tops, with 8-16 1♥/♠ openings, we used 2/1 responses with about 12+, forcing. However opener could not rebid above 2M without a standard opening bid, and responder's rebids of 2M, 2NT and 3 of responder's suit were all non-forcing, albeit with values. Responder could also pass a 2M rebid by opener.

1NT went up to 11 or poor 12 on most hands, but, also with a singleton/void in M and no six card or longer suit then 1NT handled the misfits of 13-15. The idea was if opener passed 1NT (with 8-11, not very distributional) then even it turned out the partnership had the HCP for game, a fair bit of the time it would go down since the misfit in M would make the hand play poorly.

[jtfanclub]

Rob F, on Aug 15 2006, 07:59 PM, said:

What GCC difficulties did you encounter Peter? The NF 2/1 responses are natural, so they should be fine. Using 1NT forcing (and the normal 2/1 rebids by opener) with many of the inv+ hands also seems ok, and since you can still have the weak hands that will pass or correct to 2M, this didn't violate the rule that 1NT forcing can't guarentee inv+ values.

I'm sure that's what he was talking about. We've spoken to the ACBL about this, and as long as you do it with 9 counts it seems to be OK. One long ago former partner did it with 10+ and it was ruled legal. It's pushing it, though. In one of the few times I think the ACBL doesn't make people alert enough, describing the NT as 'forcing' seems crazy to me. Forcing NT is expected to be 6-9 usually with exactly 2 card support, not 9+ with any number of trumps in support.

I've been using a system like this for some time in ACBL-land, in which 2/1 is 8-11 or so and natural (rarely 5, almost always 6+), and does not absolutely deny 3 card support (playing in a 6-2 minor suit fit isn't the worst thing in the world even when you also have a 5-3 major suit fit. Better than playing in their 8 or 9 card fit). We play 2♦ opening as 4+ in each major, which has a nice side effect: the artificial GF is is the other major, which can't be natural.

I dunno- I like it, at least.

[hrothgar]

awm, on Aug 16 2006, 04:06 AM, said:

(2) "Multiplexing" a lot of hand types into a particular bid usually means you don't do all that well when that bid comes up. Even in uncontested auctions it can be hard to sort out exactly who has what. If the opponents happen to interfere, this will become even more difficult.

I don't disagree with what you're saying, but I don't think its quite as bad as you suggestion.

MOSCITO has done a lot of work to try to minimize resulting from interference after our game invite+ relays.

First and foremost, we've removed many game invitational hands with a fit for partner's suit from the relay sequences. For example, after a 1♥ opening, if responder holds 4+ card Spade support and a limit raise or better he will (typically) prefer to bid 2NT and immediate clarify his trump support and his values. In a similar, if responder has a game invite with a 6+ card minor and 3 card trump support he will prefer to make an immediate fit showing jump. By removing many hand types from the relay sequence, we're much better positioned to show our hand types if the opponents intervene.

Equally significant, if we don't have a fit, they (probably) don't have a fit. Its a lot more dangerous for the opponents to intervene in a auction where we've shown values but don't necessarily have a fit. (In fact, our defensive methods when the opponents intervene in our relay auctions are penalty oriented) If they are going to screw with our bidding we're gonna make them pay...

[rbforster]

One of my motivations for 2/1 NF methods is to show common hands may not otherwise get a chance to bid under the usual 2/1 GF methods. For example 5-5 hands lacking a fit with opener may be forced to pass if they can't tolerate the worst response to 1NT forcing. Single suiters might have similar issues. Granted sometimes these hands can get out if partner bids a lower ranking suit in response to 1NT forcing, but without some tolerance for the major they can't always afford to bid in the first place. I suppose in a "standard" system where 1NT is NF many of these hands can bid 1NT, although having 6-4 or 5-5 hands bidding 1NT seems less than optimal. Since these are relatively weak hands (say 6-10 opposite a light opener), it stands to reason that higher bids are too dangerous for these hands to chance. Better hands can afford to spend more time bidding and can do so safely at a higher level, so on that front these methods seem to make sense.

To make things a little more concrete, here's a proposal for the bids over 1M:

1NT forcing 1 round, containing
    any constructive hand willing to pass a second suit rebid or correct to 2M (with a doubleton)
    any invitational hand with 3 card or fewer support (to bid 2NT, 3M, raise opener's second suit, or bid 3 of a new suit with a single suiter)
    GF hands without 4 card support (to bid 3NT or the cheapest bid after the 1NT response)
2♣ 5+ (usually 6) constructive NF (over 1♥)
2♣ 5+ (usually 6) inv NF (over 1♠)
2♦ 5+ (usually 6) constructive NF
2♥ 5+ (usually 6) constructive NF (over 1♠)
higher bids show most hands with 3+ support and single suited strong jump shifts

awm, on Aug 15 2006, 08:06 PM, said:

opposite a range of 8-15... this leaves you [responder] something like:

0-6: Pass
7-10: Bid something, but it's not a real invite.
11-15: Invite
16+: Game

Notice that the "pass" range doesn't change that much, since the average opener is only about a point lighter and you need to try to scramble to the best partial when you have roughly half the points. On the other hand, the "invite" range is substantially wider and the "game" range is much less frequent.

I'm not so sure about the wider invite range. If we open only the "better" 9 counts, then effectively a 10-15 opener means an invite is 11-13 (opener accepts with top half of his range, 13-15) and 14+ is enough to force to game. Sure there will be a few misfits on margin values that might not want to be in game, but if responder starts with 1NT forcing he can listen to opener's response and gauge his hand and values and decide to only rebid 2NT instead of 3NT.

Quote

Even in an uncontested auction, what do you do after 1♠-1NT-2♥? Two spades has a very useful weak meaning and all the other calls have natural invitational meanings.

You bring up a good point about the auctions 1♠-1NT-2♥ or 2♠. Over these, pass or 2♠ show minimum hands taking a preference. 3♣ is the artificial GF, and all the other hands can be naturally and invitationally as before. The invitational 3♣ hand bids 2/1 2♣ directly, either replacing or combining with the 2/1 NF constructive club hand. This issue doesn't arise after 1♥ since the 2♠ reverse shows enough extra values that the invitational club hand is now game forcing.

Quote

Another reasonable option would be to play 2♣ as game force relay and let the other 2/1 bids be natural and invitational. This gives you a lot of natural options with invite hands while retaining relays for the game force and not multiplexing too much into 1NT.

A good suggestion, I'll look into this.

[jtfanclub]

Rob F, on Aug 16 2006, 04:37 PM, said:

To make things a little more concrete, here's a proposal for the bids over 1M:

1NT forcing 1 round, containing
    any constructive hand willing to pass a second suit rebid or correct to 2M (with a doubleton)
    any invitational hand with 3 card or fewer support (to bid 2NT, 3M, raise opener's second suit, or bid 3 of a new suit with a single suiter)
    GF hands without 4 card support (to bid 3NT or the cheapest bid after the 1NT response)

You can even use it to show 4 card support....use a direct bid of 2NT to be 'which slam do you want to be in', and a rebid of 2NT (eg. 1♥-1NT-2♦-2NT) to be mild slam interest with 4 card. Gives you more information, without dumping a useful bid, and allows you to just jump to game if partner shows a shape you don't like.

[rbforster]

jtfanclub, on Aug 16 2006, 05:35 PM, said:

You can even use it to show 4 card support....use a direct bid of 2NT to be 'which slam do you want to be in', and a rebid of 2NT (eg. 1♥-1NT-2♦-2NT) to be mild slam interest with 4 card. 

I was actually thinking that 2NT directly over 1M was something like Jacoby, whereas 1NT...2NT was a natural invitation. Obviously there's space for lots of variety in terms of how to treat good hands with big fits since there are lots of higher bids both before and after the 1NT. Personally it seems right to make a fairly high bid immediately to show a good fit (4+ bergen, etc) since this can help shut out the opponents from finding their likely fit.

[pbleighton]

"What GCC difficulties did you encounter Peter? The NF 2/1 responses are natural, so they should be fine. Using 1NT forcing (and the normal 2/1 rebids by opener) with many of the inv+ hands also seems ok, and since you can still have the weak hands that will pass or correct to 2M, this didn't violate the rule that 1NT forcing can't guarentee inv+ values."

I found that losing the 2M response to weak hands overloaded the 1NT forcing response.

Peter

[the hog]

I agree with Richard, (Hrothgar), in that 2/1 should be limited and terminal unless opener has a good fit. From experience this puts a great deal of pressure on the opponents if they know that they have to act NOW or the bidding will probably die.

[awm]

Well I'm not going to convert the 2/1 NFing faithful no matter what. But I'll note that even Paul Marston (one of the leading Moscito players) has recently switched to forcing 2/1 calls. If it was so clearcut that "NF and terminal" was better, why would he switch?

In any case, the situation is very different in Moscito than in a 5cM opening system. Playing five card majors:

(1) Responder can pass opener's suit. This will often be a good option on weakish hands. This option doesn't really exist in Moscito because of the transfer opening bids.

(2) It is rarely right to pull from partner's five card suit into a five-card suit of my own a level higher. There's no particular reason to think I've improved the contract in such a situation. In Moscito, the opening showed only four cards and it could easily be right to pull from partner's bad four-carder into my decent five-carder.

(3) The opponents will not have a difficult time dealing with the NF 2/1 in a 5-card major base. They already know two suits where they don't want to play the hand. They know who has length and whether their values are well-located. It's typically not too hard for them to make the right choices. I suspect this is often true in Moscito as well, but the possibility of opening a bad four card suit (and that opponents may want to play in that suit) muddies the waters a bit.

(4) Because of the transfer opening structure, Moscito can use the minimum call as a relay and still have 1NT available to show a weakish hand with no obvious fit. Although I still think there will be issues with the "invite-plus" relay, at least it guarantees a modicum of values. With a natural opening structure, the 1NT response to 1♠ is even more overloaded when you play NF 2/1s, as it must handle both weak hands with no fit and invitational and game forcing hands.

[hrothgar]

awm, on Aug 17 2006, 06:54 AM, said:

Well I'm not going to convert the 2/1 NFing faithful no matter what. But I'll note that even Paul Marston (one of the leading Moscito players) has recently switched to forcing 2/1 calls. If it was so clearcut that "NF and terminal" was better, why would he switch?

I've had a few fairly extensive discussions with Paul on this very subject (MOSCITO's 2/1 response structure is one of the areas where we don't see eye-to-eye and I've been trying to understand why he's changing it)

Here's what I've been able to determine

1. Paul likes to experiment. He tweaks bidding systems more than almost anyone that I know. He also prefers getting practical experiece with new methods rather than relying on simulation. (There have been LOTS of changes to MOSCITO over the years, many of which have been a lot more significant than this this)

2. Paul plays with a lot of different partners, many of whom have their own opinions about what they should be playing. As a result, MOSCITO looked very different during a period when Paul was playing with Ish Del'Monte than it does when he's playing with Hugh Grosvenor

3. Looking specifically at the 2/1 response structure, Paul was quite happy with the auction structure following natural, non-forcing responses. However, he had big concerns with their frequency. The NNF 2/1 deny 3 card support for opener's major. 1♦ - 2x denies 4 spades. The point range is also fairly narrow.

[keylime]

I used to use in KLP V12-V15 the omnibus 1NT forcing call.

In V16, it has been dropped for a variety of reasons:

1) Opponents would dive bomb a three level preempt in 4th chair, i.e. 1S-P-1NT*-3D. Now you're in a big pickle on a) balanced hands (it almost forces you to play 1NT with a five card major in your opening) b.) hands that have a side suit of a suit ranking beneath the preempt c) hands that would have bid 2M but now can't take a call for fear of overbidding their values.

2) The ability to show degrees of limit raise I think at times is overrated in a forcing club method. Instead I feel it's much better for the partnership to show the limit raise immediately. We had 1M-3M as 4 card limits with the long way to 3M being the 3 card limit. Didn't come up much at all. In V16 it's Swedish 2NT with 1M-3M being a 4 card constructive raise; this occurs with much higher frequency.

3) Setting G/F's are tricky, if the bid you use to set the g/f is the same call you'd make naturally. If it goes 1S-1NT-2D-2H, and 2H is the art g/f, you have a bugger of a time unearthing your 8 card heart fit if opener does something like jump (shape showing).

4) Not having the ability to hammer their 2 level overcall is bad; i.e. 1S-P-1NT-2H (and you're sitting on 4/5 hearts). Instead you have to wait for pard to now bid a natural G/F and lose the ability to get -500 and up because they would be in a 5-1 fit.

5) Lastly, slam explorations are labor-intensive, especially over any suit overcall. I distinctly remember three situations where I as the G/F 1NT'er had to literally guess in terms of strain of contract when they overcall 2 clubs of all things. Especially of note were the minor suit hands that had slam intent - it was nearly impossible to get to six in this scheme.

In V16, 1M-2C starts all game forces. It has freed up many bids for different causes and I like the structure.