Dumb Luck Slam

Seven trumps and 29 points. Continue reading

This hand verified the old saying that it is always better to be lucky than good. I was sitting West in a pairs game, and we were vulnerable. After North’s pass my partner opened with a club. Holding the following array, I figured that we probably had game.

A K J     A Q 9 7     10 9 8 6 4     5

Faced with the choice between hearts and diamonds, I chose 1. I expected to bid diamonds on the next round. My partner, however, gave me a raise to 2, and I signed off in 4. To my shock, he waited a few seconds before proffering the 6 card. What in the world could he have? Maybe he discovered that his doubleton in diamonds was actually a second heart suit. I have done that.

I waited anxiously as North decided on a lead. Her choice was the K. This was what I saw in dummy.

7     K Q 3     A J 7     A J 10 9 6 2

Well, this should be easy. We are only missing six trumps, the Q, and KQ combinations in both minors. What could possibly go wrong?

My approach was ill-conceived, but it worked. I took the first trick and immediately finessed the spade, which worked. I dropped dummy’s two diamond losers on my ace and king of spades. I then drew two rounds of trumps in the dummy and then led the J. North took it with her queen and returned a trump, which allowed me to claim the rest: three spades, four hearts, one diamond, and four clubs. In actual fact, I ended up ruffing a good club on the thirteenth trick. Maybe we were underbid.

The reason why this was a poor approach was that North could have ruined me by simply leading a diamond back. I was lucky that she held K5, not an attractive holding from which to lead.

This is how I should have approached the hand:

  • Do I need to make this, or will down one be an acceptable score? The answer is clear. No sane team would bid a slam with these cards. Everyone else will be in a comfortable game. I must take whatever risks were necessary to make the bid in order to avoid getting a 0.
  • Which hand should be the master hand? Well, it never hurts to count tricks. There are three or four trump tricks, two or three spade tricks, and two aces. So, I certainly need four tricks in one of the minor suits. After the first trick I have a certain club loser. Therefore, I cannot afford to attack diamonds at all. That means that the dummy, even though it only has three trumps, must be the master hand. It also means that the spade finesse is essential in order to make the contract. I absolutely must have three spade winners in order to discard the dummy’s losing diamonds.
  • Is it necessary to draw trumps? I must concede a club to North (presuming that she did not lead a singleton king). But what if South should ruff in first? If North holds up, and South is able to ruff the third club trick, then I will be set if North started with four hearts. Similarly, if North wins the second trick and then leads another club for South to ruff, I will be set if North started with four hearts. That possibility can be eliminated by taking out two rounds of trump. If both opponents follow, I need not worry about South ruffing in, and if one of them has five trumps, I never had a chance to make it.
  • Which trump should be led? The key is to preserve two entries to dummy. There are two considerations: (1) I must end up in dummy so that I can start on the clubs; (2) I must preserve an entry in trumps. So, a low trump should be led to the hand. The finesse at that point is a 50-50 play, and playing the Ace commits declarer to playing for the drop, which is slightly less than a 50-50 play. So, it is a little better to take the finesse. If it works, a low heart should be led back to the board.
  • If the trumps split three-three, it is clearly safe to start on clubs. At some point North will take the king, or South will ruff in. If North takes the king first, declarer can win any return, draw the remaining trumps with dummy’s queen, and then play on trumps. If South ruffs, North can overruff, return to the dummy with the A, and then resume the clubs. When North takes the king, declarer wins any return, and goes to the board in hearts.
  • If South has four trumps, and North takes his K before South ruffs, he will be caught in a trump coup in clubs as long as declarer has not exhausted his entries. If North has four trumps without the jack, declarer’s only hope is that she has exactly one fewer club than the dummy. In that case, declarer can trump the last club with the ace, and take the last trick with dummy’s Q.

So, the contract fails if North has the Q. If you take the finesse, it fails if North has the J. It also fails if anyone has five or six hearts. If you do not take the finesse in hearts, it fails if either opponent has four hearts. If you do take the finesse, it fails if North has any holding of four hearts and less than four clubs. Not a great slam.

On hands like this one, you can bid poorly and play poorly, and still get a good result. On others you can do everything right, and still get the shaft.

Two Nightmarish Hands

Literally. They kept me up all night. Continue reading

I was up almost all night fretting about two nightmarish hands that led to our downfall in the last round of the Swiss in Johnston. There was actually a third hand that produced equally dire results, but I was the dummy on that one, and it seemed so routine at the time that I paid it little attention.

I was sitting West.


Hand #5 (North dealer, NS vulnerable):

Here was the bidding at our table:

North East South West
Pass Pass 2 2
Double 4 Double Pass
4 Pass Pass Pass

 

The result was one overtrick.

At the other table the bidding was much simpler.

North East South West
Pass Pass 2 Pass
2 Pass 3NT Pass
Pass Pass

This was South’s hand:

A K A K x x A J A J x x

With 24 hcp and a five-card suit, 3NT was not an unreasonable bid. Unfortunately West will invariably lead a diamond, East will play the queen, which forces the ace. West has two entries, and five certain diamond tricks. Even with the monster shown above and North’s five hearts South would be lucky to garner seven tricks in notrump.

So, my aggressive bidding had the unintentional effect of scaring them out of a disastrous notrump contract. On the other hand, if South had bid only 2NT, North would not raise to three. He/she would either pass (only a little better) or bid Stayman or a transfer, which would lead to 4.

Knowing that most experts do not like bidding 2 with a two-suiter, I would have strongly considered bidding only 1 with the South hand. That would surely have forced West’s hand and inevitably led to a heart contract, although maybe we would have stopped short of game. Ideally, it would have gone like this:

North East South West
Pass Pass 1 1
Pass 3 4 Pass
4 Pass Pass Pass

Of course, if North had bid 4, I would not have been happy. I would be forced to retreat to 5. Who knows if North would dare to correct to 5?


The other disastrous hand, board #7, at first appeared to be a fairly interesting play problem.

I held this hand:

A x x x A 10 x K x x x x

I do not remember why I thought that this agglomeration was worthy of an opening bid, but I tabled the 1 card. My partner responded 2NT with the following:

K J x x K 9 x J A K x x x

He did not splinter in diamonds because we play that a splinter shows minimal game-forcing values, and he had more than that. When I rebid 4, he reluctantly passed.

I was delighted when North led the Q, since that almost certainly meant that he had the jack as well. I drew two rounds or trump, which flushed out the queen, took the marked finesse in hearts, conceded a diamond, and claimed twelve tricks.

I should not have been so hasty. I should have begun by leading low to the J. That would have given North the opportunity to play his ace in fear that I might have the KQ. If that did not work, I should have taken two rounds of trump and then tried to set up a long club, which is an 84% play. Only if that failed should I have assumed that North led from the QJ. Of course, it was overwhelmingly likely that he had the jack, but there was no reason to take a chance. The other plays are what Jay Stiefel calls “can’t cost.”

We learned that the opponents with our cards at the other table bid the slam and made it on the lead of the A. How could they know that they had such a magical fit?

They could not have used Losing Trick Count. My hand has eight losers, and partner’s has six. LTC says that we only have ten tricks.

They could not have used Bergen’s adjusted point count. My hand has only 12 declarer points. I probably should not even have opened. Partner’s has 18 dummy points. Even if I add a point for the doubleton in my hand, we are two points short of the 33 that Bergen recommends.

At the other table East used Blackwood and then bid the slam even though she learned that she was off one key card and the queen of trump. She probably did not realize that she was actually missing all four queens and two jacks as well. Aaaargh!

Sue and Ilene Make a Slam

Don’t ask how. Continue reading

Sue has often shared with me stories about the slams that she has made. I always asked the same question: “Did you bid it?” Prior to yesterday she always sheepishly admitted that she did not. This time was different, and that is an understatement.

Sue was holding this hand on Friday when she saw her partner, Ilene Mahler, open the bidding by playing the 2NT card:

6    K 10 2    K 9 7 6 5 4 3    J 7
Sue told me that she had bid 2 to transfer to clubs. I explained that she could not have made that bid because Ilene had already bid 2NT.

“Oh,” she said. “I must have bid 3.”

“What do you and Ilene play that that means?”

“Transfer to clubs.” She had looked on her convention card, but I noticed that she had been looking at the 1NT section, not the 2NT section. I directed her attention there, and she reported that the line was blank.

“So what did Ilene bid?” I queried.

“She said 4!” Even though they were playing transfers, Ilene thought that Sue’s bid indicated a strong spade suit. If I had been in Ilene’s seat, I would have alerted the bid, and then, when the opponents asked what it meant, I would have responded “Nothing.”

“And what did you do?”

“I said 4NT, just hoping that she would bid something else or pass, but she took it as Blackwood. She had three aces, so she bid 5!”

“Did you mention your seven diamonds at that point?”

“No, I bid 5NT, but she took that as asking for kings. She had one, but she accidentally bid 6. My heart sank, but she quickly called a ‘finger fault’ on herself and changed it to 6. I was so relieved that I passed.”

So, not only did they reach the best contract. They managed to keep the strong hand concealed!

The play was easy. They had to lose the A, but it was almost impossible to lose anything else. The best part was that 6NT would go down against best defense. They easily got a top board.

Sue asked me how I would have bid it. I looked at Ilene’s hand.

A J 8    A Q 7 5    Q 8 2    A K Q
So, IMHO Ilene’s hand was slightly too strong for an immediate bid of 2NT. I would have opened 2 and then bid 2NT after Sue responded 2. I would have immediately bid 4 (Gerber) with Sue’s hand. When I heard about Ilene’s three aces, I would have crossed my fingers, rubbed my rabbit’s foot, and tried 6.

That is, I would have bid 6 if my partner had responded 4NT to my Gerber bid. The reason that I make that distinction is that the Gerber convention, which I might use once a year, seems to engender counting lapses in many people. By a strange coincidence I used it twice in the last two weeks. On one of those occasions my partner, who held a hand similar to Ilene’s above, responded 4 instead of 4NT. I took the plunge anyway, but with Sue’s hand I would have bid 5 after a 4 response.

On the hand in which my partner made a mistake I was castigated by one of the opponents for asking for aces with a worthless doubleton. Well, admittedly, it is a little dangerous, but when my partner has shown a balanced hand with over half of the deck, I think that it is reasonable to employ Blackwood or Gerber. The alternative is to bid one’s controls. However, when one partner has one or zero controls, and his partner has a lot, it is sometimes not feasible for the weaker hand to use that approach. In short, control bidding works better when the controls are split more evenly, or when the strong hand is the captain.

Incidentally, I play that 4 is Gerber if and only if two conditions have been met: 1) The first bid or the last bid by the partnership must have been in no-trump. 2) The last bid must have been 2NT or less. That is, 4 was a jump.

LTC v. Bergen (Real World, Part 2)

A horrible slam. Continue reading

The opponents were vulnerable when my partner dealt me this unimpressive collection on one of the last rounds of an equally unimpressive session at the Hartford Bridge Club:

A 9 7 4 2   10 6   Q 10 8 6 3   7
My partner opened 1, and I quickly responded 1. After his jump to 4 I paused to assess the situation. His rebid generally showed a hand with twenty or so points and five losers. My hand had only six HCP, but there were only seven losers. Losing Trick Count (LTC) analysis said that we could make six. I was most worried about diamonds, so I lied and bid 5. My plan was to try 6 if he bid 5 and to stop at five if he did anything else. Sure enough; he bid his diamond control, and I went straight to 6.

LHO made the very passive lead of a diamond. This is what my partner set down on the table:

K J 8 5 2   K 8 7 4   A K   K Q 6
 Could I complain? Well, yes, his opening bid was a bit strange, but he does have nineteen points and five losers. Nevertheless, against best defense this slam was down one off the top, and it would require a bit of luck not to be down two or even more. However, I realized that with a diamond lead it was still theoretically possible to make the contract if both spades and diamonds split, and the A was on side. The fact that LHO did not lead the A even gave me a little encouragement. However, RHO pitched a diamond on the second round of trump. I ended up down two, but it could actually have been worse. Every card was wrong. The opponents could have taken two hearts and a club off the top. If they did, and I made the percentage play of the drop in spades (as I did), I would have only managed nine tricks.

So, what went wrong with LTC? Well, the basic problem was clubs. Partner’s King and Queen were so worthless to me that I ended up pitching them on the diamonds. Add a little bad luck to that, and you end up with an LTC calculation that is off by three tricks!

I later (the game ended shortly before 11 p.m.) remembered that Bergen’s method was better on contracts above the four level. My research had revealed that Bergen’s superiority was usually derived from the fact that LTC’s assessment was too conservative. I decided to reassess my hand using Bergen count. For the starting count Bergen would have added to my six HCP one point for my fifth spade, and one for my fifth diamond to a total of eight starting points. Once partner supported my spades I could claim two for my singleton, one for my doubleton, and one for my long second suit. To him my hand would therefore be worth twelve points if I was declaring spades. Even if partner had had his twenty, we would still have been a little short of the thirty-three that he recommends for a slam bid. I deduced that Bergen would say to pass. At least his method says to pass; my understanding is that Marty himself held the green card in great disdain.

But wait. Partner actually could claim twenty-two dummy points! He has two quality suits and a doubleton in diamonds. In fact, if the suit that my partner actually opened, hearts, had been one of the quality suits, and if the suit without quality had been the one suit that he never mentioned, clubs, I would have had a pretty good play at making the hand. Maybe my bid was not as stupid as I thought when I saw the dummy.

It may be worth noting that we do better if my partner declares the hand. It would have been theoretically possible for him to garner eleven tricks.

LTC v. Bergen (Real World)

Diagnosing some slam hands. Continue reading

In last Saturday’s club game my partner and I missed a slam at each of the first three tables. We later missed still another slam, but we salvaged one on the very next hand. This may sound like a recipe for a disastrous session, but in fact we enjoyed a 65+% game.

Hand #16: My hand (16 HCP; 6 losers):

♠AKJ864 JT AK2 ♣T5
Partner’s hand (12 HCP; 8 losers):

♠T75 A832 65 ♣AKJ3
I opened a spade; partner put in the game force with 2♣. I rebid my six-card suit. Partner bid four spades.

The opponents held a motley collection of four queens, the K, and the J. The finesse of the Q♠ works, but so does the drop. You can ruff the diamond at any time. The club finesse works, and there is no reason not to take it. Three of the seven pairs only took twelve tricks, but it is hard to see a strategy that fails to produce thirteen.

Two popular methods for assessing the viability of slams are Losing Trick Count and Bergen Points. These two approaches were described here.

Losing Trick Count predicts that we would only take ten tricks with these cards. It is easy to understand how the spade loser disappeared, but only one finesse was even available. Even if it had lost, twelve tricks would be easy if the opening lead were anything but a heart. This hand seems to be a good example of the type that LTC systematically undervalues.

How about Bergen Points? My hand upticks to 18 using Bergen’s starting points. I can add two points for the long spades and one for the quality of the spade suit. I have to discount the JT doubleton, at least for the time being. Partner earns an extra point for the quality club suit, and he gets to add a point for his doubleton. That gives him 14 dummy points. We are getting close.

The final adjustment goes to my hand. When he shows support for my suit, I get to add another point for the sixth spade and one point for the doubleton. Together we have 34 Bergen Points, and we should definitely have bid 6♠!

How could we have done it using our bidding methods? I think that my partner’s bidding was fine. Even with the two extra points that Bergen allots him he does not have much to brag about. If anyone is going to go on past game, it would have to be me.

What if I had cue-bid the A after he signed off with 4♠? If I think of my hand as a 20-point powerhouse with two flawed suits rather than as a level-two hand with too many losers, it seems a natural bid. If partner then bids five hearts, I know that he has the ace of hearts, but I can never learn about his club holding. Some play that bidding the heart ace implies that he also has the club ace (because I skipped clubs), but we had never discussed his. Maybe we should.

If I had bid Blackwood instead, however, I would have learned about three cards — the two aces that he had and the trump queen that he was missing. That would tell me that at worst the slam was probably hinging on a finesse. If I had an optimistic view of the hand from the point count, I probably would have gone.

The other alternative would have been for partner to bid 3♠ rather than four. That would have made it easier for me, but it would have overstated his values.

Hand #22: My hand (9 HCP; 8 losers):

♠T54 AQ72 K753 ♣32
Partner’s hand (21 HCP; 3 losers):

♠AKQJ3 4 A4 ♣AK976
Partner, the dealer, faced a very difficult decision. His hand met the 4×4 criterion for opening 2♣, but there were two pretty good reasons to open 1♠ instead. Many experts never use 2♣ for two-suiters. Moreover, one of partner’s suits is clubs, and it is hard to show clubs after a 2♣ opener. At any rate he opened 1♠. I bid 1NT. He jump-shifted in clubs. I vacillated between 3♠ and 4♠. Which would be stronger in this situation? I was not sure. I picked the former, and partner signed off in 4♠. I thought about going on, but I figured that my partner had no fewer than five losers.

This auction was not one of our finest moments, but we beat the two pairs that went to seven, and the two pairs that somehow found a way to lose two or three tricks.

Bergen values partner’s rock-crusher at 24 starting points. I have ten dummy points. Even before partner starts adding in distribution after he discovers our spade fit, we are in slam territory. In fact, Bergen would probably have joined the two pairs who bid the unmakeable grand. For all I know, he would have made it, too.

Hand #31: My hand (17 HCP; 5 losers):

♠Q6 AKQJ852 KT8 ♣Q
Partner’s hand (12 HCP; 8 losers):

♠A743 9743 A2 ♣KJ8
I opened a heart. LHO liked her seven-count well enough to overcall a five-card spade suit headed by the jack. They were even vulnerable! I will try to keep this information in mind for future reference.

Partner bid 2♠. I did not think that I had enough to do anything besides sign off in 4.

Three pairs out of seven bid and made 6♠, which can be set with a spade lead. Two defenses found the killing lead.

LTC says that we should stop at four or five. In theory that is correct, but getting the twelfth trick only required the LHO to set down the unprotected A♣ (or any other non-spade) at trick one.

Bergen values my hand as worth 22 points as declarer. Partner has 13 points in support of hearts. From the perspective of Bergen points this one is a no-brainer. The best possible result is 6NT played by partner. I doubt, however, that too many people are going to ignore an eleven-card fit.

At this point I gave our partnership the Wienie Award. We had only played eleven hands, and three times we had been able to claim unbid slams after the first few tricks.

Hand #27: My hand (10 HCP; 7 losers):

♠AQ842 A94 9652 ♣7
Partner’s hand (17 HCP; 6 losers):

♠K93 Q8 KQJ ♣AQT93
No one found this slam either. After two passes LHO opened 1. Partner overcalled 1NT (15-18). I transferred to spades, and we settled for the game in spades. My hand’s only exceptional feature was the singleton in clubs.

Partner had to bring in the club suit in order to score twelve tricks.

I had only 11 starter points; I was not thinking about slam. Partner gets a point for his long club suit and one each for the quality of his minor suits. All told, he has 19 dummy points. When I learn of his spade support, I can add three more points (two for the singleton plus one for the four diamonds). That gives us 33 Bergen Points, and we should have bid slam again!

Maybe this pair of hands would look like a slam to Marty Bergen, who might have consider partner’s hand as too strong for a 1NT overcall. However, my four-card suit is the one that LHO bid, and my singleton is in my partner’s long suit. Adding points for these features seems dubious. The only positive intangible is the fact that the points are all sitting between us. We needed both of those long club tricks to make twelve tricks, and that required RHO to have no clubs higher than the 10. The transportation was tricky, too.

In fact, only three out of seven pairs found all twelve tricks. One played in no trump (making four), two made only four spades, and one five.

Hand #28: My hand (10 HCP; 5 losers):

♠74 KQ6543 __ ♣AJT87
Partner’s hand (17 HCP; 7 losers):

♠KQT9 AJ8 AT ♣K954
I had no scruples about opening my shapely ten-count. Partner forced to game with 2♣. I rebid my hearts, and he supported. At this point I was not stopping short of slam, and we made it easily.

I was shocked when I discovered that we were the only pair that had bid the slam. LTC says that this one is in the bag as soon as partner shows support for hearts. Bergen would value my hand at 18 declarer points! The challenge for him would have been to avoid bidding the grand without the A♠.

This was a really interesting set of hands. We only bid and made one of them, but our total score was still above average. Someone who used LTC exclusively would have done better than we did, provided that they did not get carried away on hand #22. If Marty Bergen could have controlled his tendency to see thirteen tricks where there are only twelve, he would have been the overall winner.

Here is the final scorecard. The edge for Bergen Points in slam-oriented hands seems even greater in practice than in theory.

161012101012221313101×10,3×12,2×1312311112101×10,2×11,3×1211271112101012281212121012

Hand # LTC Bergen Us Field Possible