Saturday, January 30, 2010

Is this a record for GIB futility?

Just playing a few hands with the GIBs on BBO.  I like this form of bridge because when my partner screws up I can yell quite happily without fear of offense.  Also, when I need to take a break, the GIBs just wait patiently.

Here's a hand where my partner GIB played well and my RHO GIB was a total moron:
Dealer North. None Vulnerable.
♠ J 
A Q 9 8 
A T 2
♣ T 8 7 4
♠ A K 8 5
6 4 3 
Q J 6
♣ Q J 3
Bridge deal ♠ Q 6 4 2 
T 7 5 
9 8 7 5
♣ A 2
♠ T 9 7 3
K J 2
K 3
♣ K 9 6 5

There were two passes to me and I had a hand that some people might possibly open in third seat. In fact, looking back over the other 15 results, one actually did open in third seat: 2♠!  This was not a huge success and ended up going quietly down 1 for the only negative score in the N/S direction and a loss of 5 imps.

But for me, if I open a substandard hand in third seat, at least the suit must be good (i.e. I want it led).  In this case there was no suit I particularly wanted led and the bad 4432 texture was not auspicious.

After three passes, LHO opened 1♣ and thereafter, everyone had something to say.  Here is the complete auction:

pp3♠ p

Given that I had earlier doubled 3 for the lead (i.e. a penalty double), and also I had passed over 3♠, my final double of 4♠ was incontrovertibly a penalty double.  According to GIB's post-analysis, they should only be down 4.  But GIB E followed up his terrible bidding with some pretty terrible play, not crediting me with four decent spades for my final double.  The result was down 5 for 1100 and a 14 imp gain on a nothing hand.

Gotta love those GIBs!

Friday, January 29, 2010


This is another topic that I've been thinking about for some time.  Len and I are trying a slightly different scheme for signaling which works as follows:

if either leader or leader's partner has strongly bid a suit, then follower skips attitude and goes directly to count; otherwise, follower starts with attitude and then goes to count. In all cases, after count, suit preference would come next.

A little explanation of "strongly bid" is called for. A "strongly bid" suit is one which either:
  • overcalled;
  • was bid twice;
  • made a preempt.
It's possible of course that your suit quality isn't all that good, especially if you made a weak jump overcall.  But it's too late to worry about that now that partner is leading your suit!  He's expecting you to have good attitude so your carding, if it means anything at all, is count.

Leader of course makes the normal lead from an honor sequence or 3rd/lowest or whatever your lead conventions are.

The point of this convention is that when one player has strongly bid a suit, there won't be many tricks standing up in the suit.  But knowing exactly how many will stand up is often crucially important.

Tuesday, January 26, 2010

A couple of difficult competitive situations

Sometimes, it seems, there just is no way to compete successfully.  You just have to give up and hope for the best.

Here's a case in point.  White on red.  You deal yourself ♠83 QJ86 KQJ3 ♣J75 and open 1NT (12-14).  LHO bids 2♠ (natural) and partner bids 2NT (Lebensohl).  RHO bids 3♠.  What now?  In my opinion, double stands out, saying please bid your suit (or 3NT if that's where you're heading).  Admittedly, we don't necessarily have the balance of power.  But we are at favorable vulnerability and we have good support for the other suits.  There's a danger that partner will pass and they'll make exactly 3♠, but if he has a good long suit of his own then he should bid it now.

Now, what if you pass and partner reopens with double?  He seems to be suggesting that he can support any suit, even though he's suggested a single-suited hand with the Lebensohl bid.  Or could it be that partner was planning on bidding 3NT (going via Lebensohl to show a stopper)?  In that case, he's suggesting that you pass.

Unfortunately, 3♠ would have been our top spot (-200) for an excellent score.  3♠X was almost a bottom (-1130).  It was one of those hands where although we (just) had the balance of power (21hcp), the hand categorically belongs to the opponents.  Our offensive par was only 70 (1).  Our defensive par (and the absolute par) was -650.  You may be wondering how anyone could score worse than -1130.  The answer is that the result at another table was 4♠X making 6 (1190).

Here's the second difficult hand: ♠A7 Q87 AKQ832 ♣AK.  Second to speak, you're debating whether to open 2♣ or 2NT (surely not 1?)  when RHO start proceedings with 1.  Now what?  You elect to double this and LHO passes and partner bids 2♣.  Since you have a GOSH (good one-suited hand), you have to bid 3 (the call that you would have made directly in the old days before weak jump overcalls took over) in order to show your strength (by agreement, 2 would suggest an off-shape non-minimum double with spades and diamonds but no enthusiasm for clubs).  Partner bids 3.  What now?  Can you be sure partner has half a stopper in hearts?  Or can you be sure that RHO has both the ace and king?

Unfortunately, I bid 3♠ to suggest a spade stop but only a tentative heart stop.  Whether that meaning is right, I'm not sure.  Surely if I wasn't interested in notrump at all, I would simply bid 4.

Anyway, partner with JT4 of hearts decided to bid 4.  This was just the holding that allowed 3NT to make and consign 4 to the trash bin.

Any comments?

Staying with the program

One of the best and most memorable bits of bridge advice I've ever received came from my own dear wife, Kim.  At the time, I thought I was the one that was supposed to be doling out the advice, but that's another story.

"Stay with the program," was what she said.  I had wimped out in the run up to a slam (or possibly I passed a forcing bid, the details have been blissfully erased from my memory).  But I have never forgotten the advice.

Here's an example of what I mean.  Names have been omitted but my partner understands that he didn't stay with the program on this occasion!

His hand was ♠AK86 Q8 A98532 ♣7, red on white at matchpoints and LHO had dealt and passed.  I opened 1NT (12-14) and partner (with hand shown) bid 2 (game-forcing Stayman).  I rebid 2NT (no four-card major and no five-card minor).  Partner bid 3.  I now bid 3.  This bid implicitly accepts diamonds as the trump suit (unless of course we later elect to play in NT) and shows a heart control, and by extension some interest in slam.  It was at this point that partner dropped the ball and bid 3NT.  Either of 3♠ or 4 ("minorwood") would have got us to the lay-down 6.  Instead, we languished in a contract that was destined to go down 1.  However, I was so discombobulated at missing the slam that I forgot I wasn't even in a contract at all and so "cross-ruffed" my way to down 3.

The moral of the story is this.  If you make a bid that invites a slam (or game) and then partner makes a forward-going bid, even if his hand is limited, it's your duty to cooperate.  You cannot wimp out.  In other words...

Stay with the program!

Saturday, January 23, 2010

Every bid tells a story

This is a topic I've been thinking about writing on for some time now.  EBTAS is essentially a group of sub-principles of bridge bidding:
  • the principle of no undos;
  • the principle of telling your story once only;
  • the principle of substantive discretionary bids;
  • the principle of fast arrival;
  • the curious principle of the dog in the night-time.
Each call you make, be it a suit you bid, notrump, pass, double or redouble tells partner something about your hand that he didn't already know.  Unless you're deliberately psyching, every bid you make should continue to refine the description of your hand.  This is especially true of non-forced bids, for a forced bid may have to be made on non-ideal cards.

Let's examine the sub-principles one by one.  That there are no undos is self-evident.  It's incontrovertible.  To take an extreme example, partner open 1♣ and you hold ♠KQJ5 5432 Q86 ♣54.  You decide you like spades better than hearts so you bid 1♠.  Partner now rebids 1NT.  If you now bid 2, your partner will now "know" that you are 5-4 or 5-5 (or more extreme still) in the major suits.  No matter how you bid from here on, partner will never believe you are 4-4.

"Don't tell your story twice" is what every novice bridge player is told again and again.  Let's say you hold ♠KQJ8752 5 Q6 ♣542, non-vulnerable versus not.  You deal and open 3♠ and LHO bids 4, partner passes and RHO passes too.  You now bid 4♠.  You've told the same story twice.  You had a reasonable 3♠ call that pretty much described your hand.  Partner decided not to act over 4.  You have no reason at all to bid again.  Partner might have ♠4 QT85 A875 ♣K976.  Between you, you have a likely one spade, two or even three hearts, one or two diamonds and possibly a club on defense.  On offense, you have 6 spades and 1 diamond.  Enough said.

The principle of substantive discretionary bids ("the principle of stuff", for short) is my own rule and I admit that I'm still looking for a pithy name for it.  I've never seen it espoused per se by any expert, but nevertheless, I think it probably is considered to be plain common sense.  It basically says that if you make a non-forced (discretionary) bid in a suit, then you show some values in that suit.  The kind of values you'd like to be led to if partner gets on lead.  So, any time you make a discretionary bid (where pass is a logical alternative) you show "stuff".  Let's say that your hand is ♠Q8752 5 A6 ♣AQJ42, not-vulnerable vs. vulnerable.  Partner deals and passes, your RHO opens 1 and you bid 1♠.  LHO makes a negative double and partner passes.  RHO now bids 2.  Do you bid 3♣?  It's reasonable, maybe not automatic but generally OK.  If partner does happen to get on lead, you'll welcome a club lead.  But suppose your hand is instead ♠AQJ42 5 A6 ♣Q8752.  The auction goes the same way.  Should you bid 3♣ now?  Some might but I wouldn't.  You barely have any "stuff" in your club suit.  Instead, with this hand, I might bid 2♠ at my first opportunity (although admittedly it's heavy for a jump even opposite a pass).

Occasionally, you will be dealt a very good hand except that your long suit is rather weak.  You simply have too many points to pass.  So an overcall is not truly discretionary because you feel you have to do something.  Partner will expect a better suit, but the overall strength of the hand should help to ameliorate any issue arising.

Let's not get confused with normal constructive bidding here.  Suppose the auction went 1♠ pass 1NT pass.  Everyone on this planet would rebid 2♣, regardless of which of the two hands they hold.  They would simply be "patterning out". Their rebid is not discretionary, especially so if the 1NT was forcing (regardless, it's incumbent upon opener to rebid a suit with such an unbalanced hand). No inference could be made by partner that you have better cards in one black suit versus the other.

The principle of fast arrival is too well known to require any long-winded elaboration from me, but let's just put it in this way: if you have two ways of getting to a particular contract all on your own, the indirect route shows the stronger hand.  Let's take a simple example.  You are dealt ♠AQJ742 5 K6 ♣Q852 and partner opens 1.  You bid 1♠.  Partner now bids 1NT showing 12-14 hcp.  You know you want to be in 4♠, but you have two ways to get there: bid 4♠ immediately or go through new minor forcing.  The principle of fast arrival says that if you jump straight to 4♠, you don't expect partner to bid again – he's limited his hand quite narrowly.  But if you bid 2♣ and then over the likely 2, 2, 2NT or 3♣ rebid you now bid 4♠, you are saying that you would be interested in slam if partner has good working cards and/or a spade fit.   Perhaps you would need to have at least ♠AQJ742 5 K6 ♣A852 to use this route.

Finally, we have the curiously named "curious principle of the dog in the night-time" (dog principle for short).  I admit to naming this one too.  The allusion of course is to Sir Arthur Conan-Doyle's short story "Silver Blaze".  Inspector Gregory asks "Is there any other point to which you would wish to draw my attention?"  Holmes: "To the curious incident of the dog in the night-time."  Gregory: "The dog did nothing in the night-time."  Holmes: "That was the curious incident."

Let's say that your partner opens 1 and your right-hand opponent bids 1♠.  You cue-bid 2♠, which is of course forcing to 3.  LHO doubles.  Partner passes.  That's to say he does nothing.  But nothing is something in this context.  Partner had the choice of calling 4, 3, 2NT, 3NT, redouble, or three or four of some minor suit.  But he did none of these things.  His pass is forcing on you to do something (you've already told him that we want to play at least 3), but he is leaving the door open for you to do something below the level of 3 if you wish, such as to make a game try.  Most probably, partner has no extra distribution but most probably he has extra high-card points.  With neither of these, he would simply bid 3 (fast arrival) effectively closing the door to any game tries you might otherwise have made.  Of course, different partnerships might have different agreements, but I think this would be the standard agreement without any discussion.

Now, for an example of the principle in action: Every bid tells a story.  Let's say that you pick up the following hand: ♠AQJ9 754 KJ6 ♣K92.  What you have is a nice balanced 14 count.  You open 1♣ and partner bids 1.  RHO now bids 1.  You have a bit of a problem.  Should you bid 1♠ (and perhaps suggest at least four clubs), or should you make a support double?  Or fib a bit and bid 1NT?  Or should you pass?  I think the latter is too wimpy.  And I think 1NT is too dishonest.  Let's say that you bid 1♠, and later you show support for diamonds.  You are showing better than a minimum hand (fast arrival).  With, say, ♠KQJ9 754 KJ6 ♣Q92 (in my opinion, not really an opening hand at all playing standard American or 2/1), you would reluctantly perhaps have to make the support double and potentially miss your spade fit (although if partner's hand is good he will check back somehow).

Now, suppose that over our 1♠, partner bids 1NT and RHO bids 2.  Do you double now?  If you have the 14 point hand above, you have already described your hand reasonably accurately (partner may expect a fourth club but will realize that you might not have had a perfect 1NT rebid after the heart overcall).  If you double here, you're telling partner that you have a balanced hand with defense in both red suits (you won't be surprised if LHO takes a preference to hearts so you'll want to have some defense there too).  And considerable extra values, given that you would have opened with 1NT if you had a balanced 15-17.  Perhaps this hand: ♠AQJ9 KJ4 KJ6 ♣K92.

Thus, if the auction goes as described, partner will really expect you to have a rather good defensive hand.  If he should chance to double 2 with ♠T2 AT2 QT84 ♣T765 (maybe he shouldn't!), he's going to be awfully disappointed if you show up with ♠KQJ9 754 KJ6 ♣Q92.

Another example comes from a common situation:  ♠KQT92 74 KJ6 ♣K92.  You open 1♠, LHO bids 2, partner passes, RHO comes in with 2♠.  Do you double?  Of course not!  Partner already knows you have five spades (assuming you're playing 5-card majors) and is planning to lead them whatever you do.  On this occasion, it's true that you have nice ones.  But double here shows extras.  Possibly just a sixth spade, but generally speaking you show more than a minimum opening.

So, take care with those frivolous bids or doubles that "can't possibly come to any harm".  You might get away with it when the next player bids something, but partner may take some later action in the fond illusion that you actually had something for your earlier action.

Friday, January 22, 2010

Grendel comes to grief

It was a cold wintery night in the land of the Geats.  The fires in the great hall were heaped high with well-seasoned logs and the flames leapt up into the chimneys.  An unusual truce had been called and two visitors had peacefully entered the gates of Heriot as honored guests of King Hrothgar.  The occasion was Grendel's mother's birthday, January 21st, and with her son, they had come for a friendly game of bridge with Hrothgar and Beowulf.  A card table had been set up not far from the main hearth and around it sat these four very intense players. Grendel's mother was in the hot seat so to speak with her back to the fire.  Well-armed guards were stationed around the table in case there were to be any tricks not made up of four cards.

Still in the first rubber, Grendel had made a routine 4 game when this hand came up:




The auction went as follows:

p1NT 1p2
p2 2p2NT
p3X 3 p
pXX!all pass

1) forcing
2) the "impossible" 2 showing an excellent fit in clubs and game interest
3) "SOP" double: save or penalize

Grendel and his mother were playing Precision with a game-forcing 2/1 major structure, hence the 1NT forcing bid.  After Grendel's 2rebid, his mother saw fit to use the impossible 2♠.  Whether the hand truly warrants it is debatable, but it doesn't come up that often and she wasn't going to let an opportunity to use it go by. Grendel naturally put on the brakes with his 2NT call and his mother than signed off in 3. But they reckoned without King Hrothgar. Bravery was never lacking on the part of Hrothgar and the favorable vulnerability and the fact that the opponents had announced a good fit in clubs but were still only at the three level prompted a "save or penalize" double, a cousin of the "BOP" (balance of power) double and part of the DSIP family ("Do something intelligent, partner"). In other words: take out into a pointed suit with an offensive hand, pass with a defensive hand. Beowulf was happy to pass (the best they could do would be to make 1, now no longer an option.  Grendel's mother of course, thinking originally they might have game was not going to let her big chance go by: "Redouble!", in a voice of thunder.

Beowulf lost no time in leading a trump. Grendel won in dummy with the 8 and called for the heart. Hrothgar played a nonchalant 6 and Grendel stopped to think. Recalling a recent hand with a similar holding and never making a trick in the suit at all, he decided to try the K. "Glad to see you think I'm good enough to duck the Ace", commented the King good-naturedly. Meanwhile, Beowulf quietly took the K with his A and led another trump. Grendel ruffed another heart and paused for more thought.

Things weren't looking too good, but there was still hope. If Hrothgar had started with Txx and four hearts and only two clubs, he'd be endplayed on the fourth heart and would be forced to lead a diamond. Grendel confidently called for the ♠2 and finessed the nine, endeavoring to create the necessary extra entry to hand. If the Hideous Hog or even the Abbott was playing the hand, he mused, the nine would win and nine tricks be almost assured for game, rubber and 370 point bonus to boot! Alas, for our anti-hero, the nine lost to the ten, the last trump came back and there was nothing for it but to concede down 2 for 1000 points above the line for the hosts.

"You should have made that simple hand," said Grendel's mother with quiet but definite venom. "You're the real dummy, not me. Forget about your trashy hearts. With diamonds so favorably placed, you can make 2 spades, 2 diamonds and five clubs. Do you expect me to bail you out of trouble even at the bridge table?"

Wednesday, January 20, 2010

Fourth suit forcing

As mentioned in Forcing Raises, I've recently been re-reading Barry Rigal's Test Your Bridge Judgement.  I never knew that fourth-suit-forcing was such a versatile bid until I first read his various scenarios.  Actually, he recommends that if used at the two-level it is forcing for one round only, while if at the 3-level, it's forcing to game.  That makes reasonable sense, although I'm not quite ready to understand the nuances of that treatment.  Forcing to game is certainly the simplest and the easiest on the old ticker!

But what about this sequence, with silent opponents, and playing (or aspiring to play) "expert standard"?

Pass 1 1 1♠ 2♣
In my opinion, 2♣ was clearly the fourth suit and therefore forcing (and not necessarily promising four clubs).  Obviously, it couldn't be forcing to game by a passed hand.  But opener was still completely unlimited and I thought would be keen to know that I had 11 hcp with five hearts (and no enthusiasm to bid 2NT from my side of the table).

Incidentally, with some of my other partners, I play XYZ in this sequence where 2♣ is unambiguously forcing and artificial.  But what of 2 by a passed hand (in that context)?  Perhaps that should be natural and, obviously, non-forcing.

In fact, I could make 3NT on 23 hcp and the actual defense (I didn't – shame!).  But my partner clearly thought that 2♣ was natural and raised me to 3♣.

What do you think?

Thursday, January 14, 2010

Coping with delayed competition

One of the interesting aspects of club bridge, even local tournament bridge, is that people do unexpected things late in the auction.  There are well-thought-out systems for coping with direct overcalls, including the negative double, forcing (and also non-forcing) new suits, fit-showing jumps, cue-bids, etc.  You're not likely to do well in tournament play without having a good understanding of these methods.

Then there's the somewhat less frequent interference after opener and responder have both bid.  Intervention here needs to be more circumspect, but it happens frequently enough that we have methods to cope with it: support doubles, etc.

Once we've got through the first round of bidding and the opponents have passed, we might reasonably expect a clear field to ourselves, apart from  the occasional lead-directing double, etc.  But there are some players who, either didn't sort their hand properly the first time, or they have a death wish, or some other obscure reason to decide that now, after the opponents have exchanged copious information, would be a great time to introduce that suit that they didn't think could be bid in an earlier round.

Occasionally, this will be a good player who has decided to "await developments" with a hand that is hard to bid initially, maybe a two-suited hand with around 9 high-card points, or perhaps something like: ♠65 5 AQ62 ♣AT9754, as described by George Rosenkranz in his wonderful book Tips for Tops.  In this case, holding these cards fourth-in-hand at favorable vulnerability, he sat back and listened to the following auction: 21 – 2NT2 – 33 – 34 – 45 passed around to George.  1Flannery, 2game Forcing shape enquiry, 34513, 4any extras? 5no.  George now bid his clubs, was doubled and played the hand for down two, thus saving 320 points.

George knew what he was doing, but most of the people who come into the auction late do not have the right hand for it at all. Here's a case in point: my hand (vulnerable vs. not): ♠AJ84 T A8742 ♣J52.  My partner dealt and opened 1, which didn't thrill me.  RHO passed and I bid 1♠, obviously.  LHO passed and partner rebid 1NT.  At this point, RHO decides to come into the auction.  Let's see, her partner obviously has nothing much, the opponents have no fit and we don't even know if my hand is going to try for game.  But she bid 2♣, notwithstanding all the good reasons not to.  So what am I to do?

What would I have done over a pass?   Most probably, I would have passed.  I don't have five spades and 2 would be game-forcing check-back the way we play.  I could make a uni-lateral decision and bid 2♣, forcing 2 and then pass it, but that's pushing the envelope a little for a normal match-point situation.  Aha, thought I, this interference has allowed me to bid a perfectly natural 2 which presumably won't be considered forcing, since it is in competition.  My partner thought it was forcing, however, and bid 2NT, despite having KJx of diamonds, which went an inglorious one down (3 would also have been down 1 while 2NT should have been down 2).

This led me to start thinking: should it be forcing? and if so why?  I posed the question to a good player friend and he said that he thought it would not be forcing without discussion.

Let's think it through.  If I want to force, presumably because I have five spades and invitational or better values, I could cuebid 3♣.  If partner has a minimum, and three spades, he will bid 3♠, otherwise he will bid 3NT or 4♠ with a maximum.  What if he doesn't have three spades or a maximum, though?  He could bid 3 with tolerance for diamonds.  But that's a bit risky.  So, 3♣ is just a too pushy for a forcing bid.  Therefore, I there's a good case that 2 should be forcing (not to game, but for one round).  It fits with the general principle that responder's new suits are always forcing.

Far more flexible however (see the previous blogs on doubles), is the simple double.  Since we have an unbid fourth suit available, we could consider it to be an "action", "BOP" or two-way double.  But, according to my doubles rules, double must be penalty since 1NT was "to play".  Also, a BOP double tends to be made by the player sitting under the bidder.

I didn't really feel that J52 was good enough to make a penalty double.  But given that we couldn't play in 2, for reasons described above, it was going to give us our best possible result of +100.  As it was, we were -100 for 2NT down 1 (actually we should have been down 2).

So, what's the moral of this story?  Dealing with crazy people at the bridge table can be tricky.  But here's a suggestion:  the only forcing bid in this sequence is 3♣ and it is forcing to game.  2 should not be forcing.  The sequence simply doesn't admit an invitational hand, though if partner does have three spades and takes a preference to spades (he might do it with two though), and we could then invite with 3♠.

What sort of hand do you think my RHO held for her late entry into the auction? ♠T95 AK752 Q ♣KQ76!

Sunday, January 10, 2010

No bad no-trump hands

The catch-phrase of the inimitable Barbara Woodhouse was "no bad dogs".  I've recently come to apply that to the evaluation of no-trump hands.

First, let me observe that I am an inveterate up- or down-grader of hands for the purposes of playing in a suit contract.  I believe in Aces, good intermediates, combined honors and all of the other motherhood-and-apple-pie rules of hand evaluation.  And of course, I've always looked at notrump-oriented hands the same way.  Are the honors in the long suits? etc. etc.

But I've noticed that good players don't bother with this.  If they have the right number of points, they invite or raise as appropriate.  Let's say that your partner opens 1NT or 2NT using whatever range you favor.  Is, say,  ♠A2 KQJ T987 ♣6543 a better or worse hand than ♠Q72 94 AJ86 ♣KT53?  The latter hand would certainly be more suitable if partner opened 2NT and you were entertaining thoughts of a minor suit slam.  But if partner opened 1NT showing 14-16 (the range I like to play when vulnerable or in third/fourth seat), which would be the better hand for purposes of inviting or raising to 3NT?  I used to think that the latter hand would be the better hand.  Now, I'm less sure of it.  Because who knows what exactly is in partner's hand?  Presumably he has a sprinkling of honors in most if not all suits.  If he has xx or xxx or even Axxx, he's going to be reasonably happy with those heart honors if the opponents start leading hearts (or not), however stubby they look in our hand.

Yesterday, in a side game at the Newton regional, I held this hand: ♠KJ6 K42 KJ ♣KJ853.  Nobody was vulnerable and my RHO dealt and opened 1C.  Did I shrink from my duty and pass on account of this aceless hand being too awful?  Of course I didn't.  For no-trump purposes, this was every bit as good as any other 15 point hand.  Admittedly, I didn't know at that point if my partner would have a sufficiently balanced hand to make it work, but here's what happened.  LHO bid 2S and partner bid 3, which since we play Lebensohl in this situation was game forcing.  I briefly entertained the notion of bidding 3NT to have the lead come up to me.  But given that my RHO had an opening bid and likely a couple of the aces that I didn't hold, that didn't seem like a sure thing.

When dummy came down, the GLM on my left said, in the nicest possible way, something like "That's the ugliest 1NT bid I ever saw".  But as we all know, there's the good, the bad and the ugly.  This one was, in a sense, all three because my partner did a great job making 4, for 9/12 matchpoints, even though we are really only supposed to make 3.

I, therefore, had the last laugh.

Thursday, January 7, 2010

More about doubles

If you're new to this blog or didn't see yesterday's article, take a look at Double! before you read this one.

I mentioned yesterday that "three strikes you're out" is one of the penalty double triggers.  Actually, just like in baseball, where not all strikes are counted (it's OK to foul the ball when you have two strikes against you, thus prolonging some games to interminable lengths), the rule could be restated as: the second double of our partnership is always penalty, except that pure takeout doubles don't count towards the total.

Here's a better way to think about it.  It's logical that there should only be one bidding box per partnership, right?  You can never rebid a previously used bid (unless your LHO is asleep) so we don't really need four copies of every bid.  In fact, strictly speaking, we really only need one bidding box per table.  But I digress...

Imagine that between you and partner you have one light green colored double card, one amber (orange) double card, and several red double cards.  The light-green card is for pure takeout doubles.  Light green, the color of leaves in the early spring, suggests all is roses in the garden and all partner has to do is bid his/her suit and all will be well.  There's only one of these per auction.  The amber card is for cooperative (two-way, DSIP, card-showing, action, etc.) doubles.  This one says, careful, it might be better to pass than to bid a poor suit.  Again, there's only one.  The red card is the one we're familiar with.  The opponents are going down, possibly with lead-directing aspects to it.  There are lots of these.  The world record for penalty doubles, by the way, was set in the 1975 Bermuda Bowl between North America and Italy played, appropriately enough, in Bermuda.  Bobby Wolff opened 1 and Giorgio Belladonna made an off-shape takeout double (such tactics generally worked well for them).  Bob Hamman redoubled and Benito Garozzo bid 1.  Eventually, the Italians ended in 2X, down 3 which helped send 14 imps to the American camp.  There were seven rounds of bidding including seven doubles (six by the Hamman/Wolff) and two redoubles.  The squadra azurra went on to win of course as they usually did in that era.  But I digress still further...

Lest you think that I am suggesting a new kind of bidding box, with multi-colored double cards, please rest assured that the idea of the colors is for illustrative purposes only.

The point is that the sequence of our doubles in an auction can go takeout, cooperative, penalty, or it can go takeout, penalty, or it can go cooperative, penalty.  But there can be at most one takeout and at most one cooperative double.

Here are some possible continuations after W makes a takeout double:
  • E passes for penalty: all following doubles are penalty;
  • E makes a forced bid (over pass): W's subsequent double is cooperative;
  • E makes a free bid (or jumps the bidding): W's subsequent double is penalty;
  • E bids over a redouble: all following doubles are penalty.
You may be wondering what should happen after some other kind of double that isn't clearly a pure takeout double (i.e. a double of a natural bid when partner has only passed, or has not yet called).  Here's a summary of what should happen after such doubles (assuming no other penalty-triggering events have intervened):
  • negative double: the next double should be cooperative;
  • responsive doubles (this is essentially a cooperative double): all subsequent doubles are for penalty;
  • conventional doubles (e.g. support, maximal, snapdragon, Rosenkranz, etc.): all subsequent doubles are for penalty [but this is a subject for partnership agreement – I'm just suggesting a simple rule, not necessarily the best rule];
  • lead-directing doubles (these are essentially penalty doubles): all subsequent doubles are for penalty.
After partner makes a double that promises a single (unknown) suit, then an immediate double by you should be to enquire "which is your suit?".

You're probably saying to yourself "but that's how everyone plays!".  Well, it's true a number of experts do play this way (definitely not all), but unless you've really discussed it with your partner (or you actually are an expert), you really can't assume that he or she will be in tune with what you're doing.  Most experts, if shown a bidding sequence culminating in double, will, if there's any doubt in their minds, look at the level at which the double was made.  If it's a game contract (with the exception of an opening bid of 4 of a major, or 5 of a minor), then the double will be deemed penalty.  If below game, the expert will have to look more closely but will often consider it to be a balance-of-power (i.e. cooperative) double at the three-level if conditions are right, and will be inclined to see it as pure takeout at lower levels.  I'm generalizing a bit.  But the point is that the scheme described in these two articles is independent of the level of bidding.  That simplifies matters considerably.

OK, an example (more tomorrow):  World Pairs, 1998.  Berkowitz's hand:  ♠74 KT32 A94 ♣QT76.  Opponents are vulnerable.

all pass
11-15hcp, 2+ diamonds (i.e. nebulous precision opening)
cooperative (BOP)
Cohen/Berkowitz didn't feel that penalty doubles should be on immediately after a support double (I agree with them, but it's a partnership discussion issue). But nothing else had happened such as a jump by our side, or a notrump bid, etc.  So the final double was cooperative. 3♠X went down 2 (500) for a top. Cohen would not have been able to make a penalty double (assuming they would play it as penalties) because although he has something in spades, he didn't know whether Berkowitz had good defensive values or had strained to bid 1 with a distributional hand. Note that Berkowitz was prepared for a bid of 4♣ (Cohen might have had five of them) or for 4 thus showing a real diamond suit.  And he had the perfect holding for a cooperative double: two small, which would come in very useful if Cohen decided to pass, as he did.

Wednesday, January 6, 2010


Is there a call in all of bridge that is more prone to misuse, misunderstanding, mistakes, and several other misses?  4NT is possibly as dangerous, but it doesn't come up very often.

Yet, double is often the most flexible call and can be deadly (to the opponents) when playing with a partner who really understands its use, preferably your clone sitting across the table.  The reason why it can be so deadly is that as declarer, you rue the day you play a doubled contract with extra values on one side (usually your right) and a trump stack on the other side.  Sometimes when the stack and the extras are in the same hand you can throw that opponent in at some point to give up a trick or two.  But that won't happen when the assets are divided.

I've been working for years now on a simplification of the meaning of double which doesn't give up on juicy penalties and can be used on many hands, whatever the level of bidding. It's based on a concept that's been around for a while: Do Something Intelligent, Partner (DSIP).  The basic rules are summarized in a later blog entry DSIP rule summary.

Of course, no double is 100% takeout or 100% penalty.  A partner can take out his partner's penalty double if he has a long weak suit he hasn't mentioned, but this is a fairly rare occurrence, especially playing as above.  A traditional, low-level takeout double can be passed with sufficient length in the suit and playing strength, but again it's quite rare.

When I first started playing these kinds of doubles, we got into all kinds of trouble.  Eventually, I realized that the reason for that is that we were making DSIP doubles on hands which were too distributional and partner would pass when actually it was right to bid on.  Sometimes we did it with insufficient values when competiting in any way was wrong.  Thus, good judgment and extra values are required to make one of these DSIP doubles (and it helps to have an understanding partner too!).  You don't need extra values yourself if you know that your side has the balance of power (a BOP double is really just a special case of DSIP).  You do need a couple of cards in the enemy suit yourself though.  If you don't have those cards, there's almost surely another bid you can make (bid a new suit, support, or rebid your own suit).

Notice that the main difference between this system and "standard" bidding is that the level is not a direct factor in whether or not the double is cooperative or penalty.  Certainly, the level and the vulnerability affect whether you will make a double, but they don't affect the meaning.  But also note that our side's third double is always penalty (three strikes, you're out!).

One area that is somewhat vague is what to do after notrump is bid.  This is a matter for partnership discussion.  I like to play negative doubles after 1NT with all my partners (except when we open 10-12) at both the two- and three-levels.  So, when isn't 1NT "to play" by my definition?  When it's an opening, when it's a forcing 1NT, or when it doesn't promise specific cards.  So 1 p 1NT X is for takeout as normal.  But 1 1♠ 1NT X is for penalty because the 1NT bidder has bid 1NT "to play", i.e. with specific stoppers in the enemy suit, and obviously we'd be raising spades if we had a fit.

I have lots of examples squirreled away for my upcoming book on this subject (might be a few years yet!) but here's a new example which occurred last night.  At the table, the double that I think was right didn't get made, but that's another story.

Playing 12-14 no trumps, matchpoints. Vulnerable vs. not. Your hand is ♠K3 K7 AJT87 ♣KQ98. The auction went as follows:



What do you do now? Playing the system that I outlined above, double is the most flexible call and most likely to get us to our top spot of 6♣ (or 6).  At the table, the hand shown above bid 4 and the opponents pushed on to 5 at which point, this hand's partner made a cooperative double which this hand passed. We ended up with 500 which was a lot better than going down in 5♠ but obviously not as good as 1370. I honestly doubt whether we'd have reached the slam in any case, given the barrage of interference we suffered, but I hope you'll agree that the double here is the most flexible call assuming you and partner are on the same "page".

Note that, although we were at the four-level, where most partnerships would only have penalty doubles available, none of the trigger events that cause us to shift to penalty doubles had occurred. Neither of us had jumped, there were no earlier doubles, redoubles, cuebids, NT calls, etc. Every reason to continue playing doubles for takeout!

I will be adding more detail and examples in the next day or two: see More about doubles.

Monday, January 4, 2010

Forcing raises

These days most of us are brought up on the notion that a raise of partner's suit is inherently non-forcing.  This is all the more so when there has been intervention by the opponent.  And a preference not only is not-forcing but it warns that opener would be well advised not to bid again.

But re-reading Barry Rigal's excellent book Test your Bridge Judgement has helped solidify an idea that's been rattling around my brain for a while now.  Incidentally, this book is one of the best – and most thought-provoking – bridge books I know.  I would put it in my top ten.

The situation arises in the following circumstances.  Opener bids 1 banana, let's say.  For now, we'll assume that it's a major but the same idea applies (though typically with an extra card in support) for a minor suit.  Partner makes a bid that does not deny three bananas, but is otherwise forcing for one round.  Opener now makes a bid which is invitational and partner now raises bananas.  That raise is forcing.  At least that's what I'm asserting here.

Let's see an example in action (this is actually the one from Rigal's book in the chapter called Split Ends).  You hold ♠A9753 A6 A843 ♣KQ.  You open 1♠, LHO passes and partner bids a forcing 1NT.  RHO passes and you make a generally balanced invitation with 2NT.  Partner now raises to 3♠.  By bridge logic, this is forcing.  Partner has three spades but he didn't bid 2♠ at his first turn.  His 1NT was therefore based on either a really bad flat hand with three spades or a hand that was going to make a three-piece limit raise, i.e. extras.  Clearly, he has the latter hand because with the former he would simply pass 2NT.

So both partners have an invitational hand and therefore each would accept a game invitation.  Therefore, 3♠ must be forcing.  Another way to look at it is this:  opener has made an invitation, not knowing that responder has support.  When that support is now revealed, it is good news, so it's hard to imagine passing.

Here's another example, Distracted Choice from the same book.  You open ♠AJ3 K7653 96 ♣AQJ with 1 in third seat and partner bids a natural 2♣.  You know he has 10-12 hcp and decent clubs but he hasn't denied three hearts.  If you had a minimum hand you might pass.  But with just a little more than that, you raise to 3♣, obviously forward going.  Partner now shows his three-card support with 3, or at least it should be three cards.  Again, partner has given you the best possible news (he could have passed 3♣) so there's absolutely no reason to pass.  We bid 3♠ which at this point obviously just shows some fragment for notrump purposes and partner bids 4♣ and we sign off in 4.  In the story, it turns out that partner only has QJ and it takes all of Barry's skill to make the contract.  I'm afraid I would have gone down with trumps splitting 4-2.

So sometimes, raising partner's suit is non-forcing and sometimes it's forcing.  Are there other situations that I'm missing?  Any other comments?