[mikeh]

The recent thread on masterpoints contained an extensive digression on what inferences one might be able to take from conventions listed on an opp's cc.

Fred, in his well-reasoned posts commented that he had a few conventions that he really felt were unsound, and one of them happens to be a convention that I like to play in my partnerships: two way drury.

Now, I recognize that a battle of bridge theory opinion between me and Fred would be a bit like the Polish cavalry charges against German tanks in 1939, and it doesn't take a genius to figure out who is riding the wrong horse, so to speak.

But, I would be interested in hearing why 2WD is a bad convention.

Let me start the ball rolling.

All conventions have costs and benefits and these costs and benefits depend upon the context in which the convention is used: in particular, upon other methods that may be impacted by use or non-use of the convention.

Thus, I usually, but not always, play 2WD in the context of a fairly loose weak 2♦ opening.

I am not wanting to open a debate on the wisdom of that approach, but it seems to me that the availability of this device reduces the probability of having a natural use for a response of 2♦, by a passed hand, to a 3rd or 4th seat 1Major.

In other methods, such as where 2♦ is a multi, the cost of losing the natural 2♦ may be significantly higher.

I also usually play that a 3rd seat and, to a lesser degree, 4th seat 1Major opening can be on a 4 card suit, with 4 card hearts more common than 4 card spades. Again, if one generally requires 5 cards to open 1Major in 3rd and 4th seat, the benefit from distinguishing 3 and 4 card raises is somewhat (but not entirely) reduced.

Against that backdrop, it seems to me that 2WD offers some significant advantages, which I will attempt to address before turning to costs.

1. In my serious partnerships, we have two entirely different artificial schemes for constructive bidding after a 3 card drury than we do after a 4 card drury, based on the assumption that we will not (usually) want to play in notrump once we uncover 4 card support. Thus we can dedicate 2N (and, in some sequences) 3N to artificial uses, whereas notrump tends to be a more attractive denomination after a 3 card raise.... even when opener has a 5 card major... 5332 opposite 4333 or 4432 (with 3 card support) may well play better for 9 tricks in notrump than in the major, but this is less-likely when we have a 5-4 fit.

2. While we have, largely, got by the opps already, it is not impossible for the opps to balance, especially after a 3rd seat 1♥ opening over which opener signs off in 2♥ after drury. In that case, it may be useful for opener, in deciding whether to compete, to know that partner holds 4+ support. BTW, I am not listing factors in order of perceived importance... I rate this one very low, because I play primarily at imps, where balancing is less aggressive, and, in any event, we have already got by the opponents at least once, and half the time our suit is spades, making a successful balance at the 3-level dangerous and rare.

3. When opener has a decent hand, knowing of the degree of fit can make a difference in the decision to bid game, especially if the trump suit is weak. Axxxx opposite a 3 card raise is significantly weaker than Axxxx opposite a 4 card raise, and the same is true if we hold a 6 card suit. Qxxxxx opposite xxxx is far better than Qxxxxx opposite xxx.

4. When opener has a great hand, again, the degree of fit may warrant or argue against a slam try.

These are the benefits that occur to me as I write.. others may have differing ideas.

As to cost:

1. We lose a natural 2♦. The frequency of this cost will depend upon system. Those playing a weak 2♦ will suffer less than those who have to pass with KQ109xx of diamonds and a side card.

2. As part of that cost, we will have to decide whether to play a forcing or semi-forcing or non-forcing 1N response to a 3rd and 4th seat 1Major. I personally play semi-forcing: opener always rebids with shape or with a hand that still offers any play for game opposite a maximum 1N. This might seem to enhance one's ability to back into diamonds, if opener rebids 2♣. Honesty compells me to admit that we use 2♦ there as artificial

3. We give the opps two ways to double... they can double 2♦ as well as 2♣. I think this is a straw man argument: if they play double of drury as takeout of the major, giving them a double of 2♣ or 2♦ makes no difference, and if they use the double of 2♦ as lead-directing, they probably have the hand to bid 2♦, relatively safely, over our 2♣.

I am sure that there are other, and significant, costs to the approach, but I am not aware of them. I was hoping, by this post, to engage Fred and others in a discussion... I am not wedded to any method, and if 2WD is indeed a defective convention, I look forward to abandoning it, especially if my partners can be shown the error of their (and my) ways.

Looking back on this post, I see that I acknowledge the possible loss of a natural 2♦ call. In my most successful partnership, with which we accompanied Fred to his second BB where we crashed and burned, we actually adopted 2WD to stop us from being able to respond with a natural 2♦ i suspect we were just unlucky, but we encountered a series of hands on which responder was dealt about a 10 count with 5 chunky diamonds, and responded 2♦, catching partner with a classic 3rd seat light opening and short diamonds... typically, if memory serves, something like an 11 count 5-3-1-4 with weak spades.. and we seemed always to be vulnerable! We'd play 2♦ when either 1N or 2♣ was better.

[glen]

Some of the advantages point to knowing if responder has just 3 trumps ("degree of fit"). However that assumes there are no follow-ups to determine trump length with not-bi Drury (or the follow-ups cost too much bidding space). For example one could play:

1M-2♣ (3+ trumps);-2♦(some game interest at least)-?

2♥: 3 trumps
2♠+: various hands with 4+ trumps

Likewise one can use Lawrence style Drury.

[fred]

Hi Mike,

Try this:

Link to old rant

My dislike of 2-way Drury has mostly to do with practical considerations whereas your like of 2-way Drury seems to be mostly about the added expressiveness that this convention offers.

But added expressiveness is only a good thing if it helps your side more than it helps the opponents

Something else I hate about 2-way Drury that I did not mention in my previous rant is that I find the 2D response to 2C to be very useful. This bid effectively allows you to make a (possibly mild) game try without getting above the 2-level.

I suppose you could do something similar with the 2H response to 2D Drury if opener's suit is spades, but I would hate to give up the equivilant of 2D when opener's suit is hearts.

Fred Gitelman
Bridge Base Inc.
www.bridgebase.com

[jdonn]

fred, on Sep 5 2008, 06:19 PM, said:

Something else I hate about 2-way Drury that I did not mention in my previous rant is that I find the 2D response to 2C to be very useful. This bid effectively allows you to make a (possibly mild) game try without getting above the 2-level.

I suppose you could do something similar with the 2H response to 2D Drury if opener's bid is spades, but I would hate to give up the equivilant of 2D when opener's suit is hearts.

I have never played this, but I would be very open to the idea of playing 2-way Drury over 1♠ and 1-way Drury over 1♥.

My experience is that I have always sort of convinced myself I don't like 2-way, but one thing strongly in its favor I have noticed a few times is that if playing 1-way and the opponents balance (or overcall over the Drury bid), we don't have enough information about trump length to know whether or not we should push to the three level. It's not hard to imagine responder sitting there on a decent hand with four trumps wanting to push to the three level over a balance by the opponents but only if opener has five, and vice versa.

In any case, I have only played 2-way in two partnerships in recent years, and both were players who like to bid Drury on extremely light hands (essentially constructive raise or better.) If someone likes to bid Drury on light hands, then I think 2-way is necessary, since otherwise it's simply too many hands to stick into one bid.

[Echognome]

In addition to Fred's point about not having a mild game try of 2♦ available if 1♥ is opened, the loss of the natural 2♦ can make for some awkward auctions. It's not just the chunky 5-card diamond suits, but it's also the wrong hand type six card suits.

♠x ♥AQx ♦xxxxxx ♣Axx

Maybe this falls under your weak 2♦ opening? Maybe it's a 1♦ opener for you?

Yes we could swap the minors and you'd have the same problem, but at least we can cut the problem in half. (Also, Phil and I use 2NT as the long club hand without a fit and play 3m as a fit jump by a passed hand. We don't have room to replace both minors. It's not that I'm thrilled we have to bid 2NT, but at least it helps.)

Also, not knowing the trump-length is a two-way sword. Maybe we can play in our 9-card fit at the 2-level because we may only have a 7-card fit!

Edit: I should also note that most pairs do not have a weak 2 in clubs available, so in clubs it's possible we have a good club suit for 2NT. Phil and I happen not to have 2♦ available either because we play a mini-multi with more intermediate 2♥ and 2♠ bids, which is, of course, an aside from the main discussion.

[glen]

jdonn, on Sep 5 2008, 07:33 PM, said:

... If someone likes to bid Drury on light hands, then I think 2-way is necessary, since otherwise it's simply too many hands to stick into one bid.

The other interesting question about bi-Drury is whether the split should be on trump length or strength, for non-disclosure reasons. For example one could play:

2♣: limit Drury, 3+ trumps
2♦: constructive Drury, 3+ trumps
2M: I have a "raise"

[fred]

mikeh, on Sep 5 2008, 10:58 PM, said:

Now, I recognize that a battle of bridge theory opinion between me and Fred would be a bit like the Polish cavalry charges against German tanks in 1939, and it doesn't take a genius to figure out who is riding the wrong horse, so to speak.

Nice of you to say Mike (I think!) but bidding theory is not something that I consider to be one of my real strengths as a player (perhaps because it doesn't really interest me that much).

I do think I am good at finding holes in things that sound good from a theory point of view, but have significant practical downsides that many strong theorists often don't consider.

If you are looking for a bidding theorist to do battle with, they don't come much tougher than our friend Tyrannosaurus Rexford

Fred Gitelman
Bridge Base Inc.
www.bridgebase.com

[kenrexford]

Never thought about one thing before, but Mike perhaps inadvertently suggested something.

If you play 2♦ as a weak two, then it seems perhaps more effective to have 2♦ as Drury and 2♣ as natural, rather than the usual way.

Hmmmm.

[glen]

kenrexford, on Sep 5 2008, 09:41 PM, said:

Never thought about one thing before, but Mike perhaps inadvertently suggested something.

If you play 2♦ as a weak two, then it seems perhaps more effective to have 2♦ as Drury and 2♣ as natural, rather than the usual way.

Hmmmm.

This was in The Bridge World quite a while back (Mike Massimilla, "Three Method Twists", March 00 BW), and Karen & I (and others with weak twos in ♦s) played it before and still do - we define Pass-1♦/♥/♠;-2♣ as a "weak two" in ♣s.

I still would like (albeit unlikely to happen here - if it doesn't I'll blog it later to cover it) bidding theory discussion on
1) is showing degree of fit important on first response
2) is showing less than a limit important via a non-2M raise

--- ---

> Jim Hudson, "On Passed-Hand Raises", August 05 BW

main value (of two-way Drury) is judging what to do when opponents compete - not necessary if responder has limit since opps have not enuff values to compete
2♣: limit 3+ trumps, 2♦=single raise 3 trumps, 2M=single raise 4+ trumps

> Gary Bernstein, "TATA Drury-Fit", December 04 BW

"tell-ask-tell-ask" Drury Fit - 2♣: 3 trumps, 2♦: 4+ trumps. Cheapest new suit bid ask, step replies involving no shortness or shortness

> Alvin P. Bluthman, "All-Purpose Passed-Hand Major Suit Raises", August 06 BW

1NT: includes 3 card limits, 2M: single raise, above 2M: various 4+ raises

[han]

Yep, that's a point awm has often brought up here. Of course you lose the 2D call after drury, a loss too large for me.

[kenrexford]

Well, I like 2♦ openings for different causes anyway. Oh well. LOL

[rbforster]

Remember that you don't need Drury at all if you only open "real" opening hands in 3rd/4th seat. Drury is primarily a psychic control so that you don't get too high when partner in 3rd seat decided to tell you what major to lead against their game by bidding 1♠ on:

AKxx
xxx
xxx
xxx

or even

KQJx
xx
Kxxx
xxx

The downside of 2-way Drury comes in terms of helping out the defenders against 2M. Leading a trump is often right against the 2M contract in a 4-3 fit when the defenders have the majority of the strength, while it's usually wrong against a 5-3 or 5-4 or 4-4 fit where the defense usually wants to make more aggressive suit leads.

An ambiguous 1-way Drury raise doesn't provide this information to the defense, leaving them much more in the dark. The 4 card raise is very effective when it comes up (telling partner how much to compete, if at all), but the 3-card raise hurts when opener was "joking" and announces a weak hand with potentially only 4 trump. Now when the opponents choose to defend 2M, they know whether or not to lead trump.

kenrexford, on Sep 5 2008, 09:41 PM, said:

If you play 2♦ as a weak two, then it seems perhaps more effective to have 2♦ as Drury and 2♣ as natural, rather than the usual way.

Certainly there's an argument for playing 2♦ as 1-way Drury so that the much more likely 6-11 hand with 5+ clubs (and no weak 2 bid) can be shown with 2♣, rather than the much more limited set of equivalent hands with diamonds than were unsuitable for a weak 2♦ opening.

That said, the more people are opening "light" in 1st/2nd seat, the less use these 2/1 bids have by passed hands. In that regard, it's not such a big loss to use 2♦ for something, it's just that I don't think 3 vs 4 trumps is the right distinction. My personal preference (back before I started opening all 10 counts ) was to play encrypted 1-way Drury, where you showed the number of AK honors held in partner's suit (2♣=1, 2♦=0/2). Together with the reasonable requirement that partner must have at least 1 of the AK in his suit to open "light", this allows the signoffs and game tries to be encrypted in many cases to prevent the opponents from knowing what was going on. For example, in "normal" Drury, you have an auction

P-1♥
2♣/♦*-2♥ Drury; weak sub-minimum opener

When 2♣ or 2♦ shows a normal Drury raise (1-way or 2-way), the opponents now both know they can balance with the good shape since their side has about half the deck. On the other hand, in the encrypted method (which works best when the opener's side each have one of the A and K between them), it goes:

P-1♥
2♣*-2♥* 3+ limit raise with A or K♥; weak with the A or a balanced game try with the K♥

Now the next hand doesn't know whether or not it's safe to bid, and even when it gets to the opponent in balance chair, he can't tell whether responder passed since he knew opener was weak or whether he passed because he declined a balanced game try. Balancing with shape is very likely right in the former case, and very likely wrong in the latter and the encryption, when it works (A and K split between opener's side), keeps them in the dark about this. Even if opener doesn't have the weak hand, he can make encrypted game tries (short suit vs long suit) which responder can understand but the opponents cannot. Since game tries often help direct defense to lead (or not lead) the suit mentioned based on whether it was showing length or shortness, the encryption helps make the opening lead more difficult for the defense.

[Echognome]

Rob F, on Sep 5 2008, 09:06 PM, said:

Remember that you don't need Drury at all if you only open "real" opening hands in 3rd/4th seat. Drury is primarily a psychic control so that you don't get too high when partner in 3rd seat decided to tell you what major to lead against their game...

I disagree strongly for several reasons...

1. I like having drury even if I'm opening "sound" but minimum openers in 3rd or 4th. Yes 3rd is wide-ranging, but in 4th we're passing those hands and people *still* play drury. It's just a different situation when partner is a passed hand.

2. We might have no interest game opposite our minimum opener simply because partner is a passed hand. Take, e.g. QJx AQJxx Qxx Jx. We open 1♥ in 3rd and if partner bids 2♣, I'm bidding 2♥.

3. The drury bidder should press on over 2M with a suitable hand. If I hold, e.g. xx Axxxx xx AQxx and partner opens 1♥, then I'm taking us to the 3-level regardless if partner tries to sign-off in 2♥. We also may have a fit jump or some other higher call. Opening light in 3rd is not without risk. Maybe the next hand will make a jump overcall and we make a negative double.

4. I think that pairs that use it strictly as a psychic control, should be dealt with accordingly.

[EricK]

Why play "artificial" Drury at all? Why not play 2-level responses (including 1♠-2♥) as fit non-jumps? Wouldn't that give opener useful information along with the knowledge that you have a good hand with a trump fit? I suspect this is more useful than immediately knowing the degree of trump fit (as with 2 way Drury) or just knowing of the trump fit (as with 1 way Drury).

[awm]

I've always been a bit suspicious of the use of 2♣ drury where 2♦ announces a "real opening hand." This suggests that you are going to open hands which are so bad that even opposite what would be a game force across from a real opening bid, you are not willing to risk the three level. Some go so far as to say that the auction Pass-1♠-2♣-2♠ bars responder from bidding on no matter what he has. This sort of approach seems mighty close to a psychic control to me.

As to the three-card vs. four-card distinction, you generally have a lot of space to make game tries over drury. You can sort out responder's degree of support when opener needs to know while keeping opponents in the dark when opener doesn't care. So I don't think the need to know this for constructive reasons is that big a deal. Assuredly it might help you somewhat in competitive auctions, but in competitive auctions the drury bid helps opponents a lot too (i.e. they know that they should normally balance over a declined four-card raise because you have at least an eight card fit and generally nine, whereas they know that balancing over the three-card raise is more dangerous).

I like 2♦ as the only drury, leaving 2♣ free as a natural call (since I play a weak 2♦ but not weak 2♣).

[rbforster]

I agree that there's some merit to having an invitational raise below 2M so you can stop low, and that's true both in 3rd and 4th. Drury is fine for that. Too bad that's not allowed in 1st/2nd seat at most levels :/.

Echognome, on Sep 6 2008, 01:35 AM, said:

4. I think that pairs that use it strictly as a psychic control, should be dealt with accordingly.

If by "dealt with" you mean encouraged, that's why Drury gets it's own specifically allowed spot on the GCC so you can play even if it's mainly used to cater to partner's very light openings. Other methods to do the same thing aren't allowed, but in their wisdom the ACBL decided that this one was just fine. After all, if everyone plays it, it must be legal, right? Isn't that the ACBL's default policy on just about anything?

[cherdano]

I have never seen anyone use drury as a psychic control. I would certainly open KQJx xx Kxxx xxx in third seat white vs red, but then I always check "very light 3rd seat openings", so this is not a psych.

[Gerben42]

I would use only 2♦ as Drury, when playing a W2 in ♦. The reason why 2♣ we need TWO Drurys has always escaped me.

[fred]

If you, like most people, use a passed hand 2/1 to say "I have a 5-card suit and close to an opening bid", I believe it is not very important that you maximize your opportunities to make such a bid (particularly in a minor). I will try to explain why. If you buy what I am saying then you might want to reconsider the concept of using 2D as 1-way Drury (instead of 2C) on the theory of "I don't have a weak 2 in diamonds but I might have a weak 2 in clubs".

First thing to recognize is that it is rarely important to make a passed hand 2/1 for the purposes of "showing your points". If you are concerned about missing game, it is fine to bid 1NT - if you have a game then partner will usually bid again.

More often it will be important for a passed hand to be able to say "I have a long suit", especially if responder has an unbalanced hand and especially if you belong in a partial (otherwise, as above, bidding 1NT will usually work fine).

But it should be recognized that sometimes a passed hand 2/1 will get you to a worse partial than the alternative of 1NT. For example, if opener has a minimum or sub-minimum hand with a singleton in the 2/1 suit, he may well want to pass or even be "forced" to pass if his alternative rebid(s) are would be forcing. You will sometimes end up playing in a 5-1 fit as a result.

Kokish actually believes that, for these reasons, the passed hand 2/1 should deliver the at least a 6-card suit. This makes considerable sense to me (though I don't play it myself), but it clearly reduces the # of opportunities for making passed hand 2/1s.

Not that this is necessarily a bad thing since, as far as I can tell, there exists no entirely satisfying rebid scheme for opener after a passed hand 2/1. No matter which bids you define as forcing, opener will always be stuck without an appropriate rebid a significant % of the time.

So to me it makes sense that, if a passed hand 2/1 will often leave opener with an unsolvable rebid problem and will sometimes leave the partnership in a 5-1 fit, you should not be going out of your way to make these bids. This takes some of the luster off the concept of using 2D as 1-way Drury - if passed hand 2/1s are to be avoided then using 2D instead of 2C for Drury in order to increase the number of opportunities you have to make a 2/1 starts to sound a lot less appealing.

Fred Gitelman
Bridge Base Inc.
www.bridgebase.com

[awm]

But the important hands for "2♣ natural" are actually the hands with six clubs -- the ones where if your minors were switched you would open a weak two in diamonds.

I think it's clear that all else being equal, we'd rather have a natural club bid than a natural diamond bid over a 1M opening. The question is exactly how much is gained/lost by playing 2♦ as drury versus 2♣.

Basically the losses will come when responder has a really good hand, typically one that would force game opposite a first seat opening. On these hands playing 2♦ drury you will see the auction:

Pass - 1M - 2♦ drury - 2M signoff - 3M - Pass or 4M

Whereas playing 2♣ drury you could bid either:

Pass - 1M - 2♣ drury - 2M signoff - PASS

Pass - 1M - 2♣ drury - 2♦ "real opener" - 4M

So basically the big difference is that when opener's hand is total trash and responder has a game force opposite a real opening, you get out a level lower. How big of a win this is obviously depends on how often you open on total trash!

My tendency is to open third seat about a queen lighter than in first or second (i.e. I open a lot of 9-10 counts with good suits), and this is also the tendency that most good pairs seem to disclose if asked. If this is your style, then playing 3M when you are about a queen light of game will not be all that costly, and the fact that responder can make a descriptive game try (rather than just bidding 3M as in the first example) might actually get you to some good light games and thus compensate for playing 3M when you decline the invite.

On the other hand, if you are actually opening a lot of six-counts in third chair, then being able to stop in 2M is a huge win. Of course, this is not the style that most people disclose and it's also of questionable legality (assuming systemic openings on less than 8 hcp are disallowed, as they are in many places). This is why my comment about 2♣ drury often being used as a "psychic control."