1994 Toronto Calcutta
By: Fred Gitelman

The following conversation occurred early in March 1993 at the Regal Bridge Club. The conversation was between Irving Litvack and John Gowdy. Irving Litvack is the proprietor of the Regal and is responsible for organizing the Canadian Invitational Pairs each year. John Gowdy, a well-known local bridge player and philosopher, had just finished playing in the Canadian Invitational Pairs.

Gowdy: I must say, Irving, you did a first class job organizing the event. If it wasn't for the poor play of my partner, Fred Gitelman, I would have really enjoyed the weekend. I think there is something wrong with the form of scoring, however.

Litvack: I am glad you enjoyed playing, John. We really go out of our way to make the event attractive for local players. I am pleased we succeeded in your case. I don't understand what could possibly be wrong with the scoring system. IMP pairs scoring is used in all of the big invitational pairs events, from the Sunday Times Invitational in London to the Cavendish Pairs in New York. The ACBL now has a Nationally rated IMP pairs event and more and more Regionals are adding IMP pairs to their schedules.

Gowdy: In some ways IMP pairs is just like matchpoints. In both scoring methods you compare your results with all of the pairs that hold your cards. In matchpoints, each such comparison results in either 0, 1/2, or 1 matchpoint for both sides. Top on a board is the number of pairs that play any given hand minus one. In order to win a comparison you simply have to beat another pair's score. It does not matter if you beat the other pair by 10 points or 1000 points. All you can get from any given comparison is one matchpoint. Every deal is equally important.

The same is not true in IMP pairs. At IMP pairs your results are IMPed against all the other pairs holding your cards. In a quality field, most deals are relatively flat and do not count for much. Every once in a while a deal comes up (typically in the slam range) in which there are many possible results each of which yield vastly different IMP scores. It is as if you were playing matchpoints and all of a sudden the director announces: "For this board only, top will be 300 instead of 12".

Litvack: What's your point?

Gowdy: It is very much luck of the draw which pairs you will play against on these key deals. If you happen to play a top partnership on a difficult slam deal and they do the right thing there is nothing you can do. You have to just accept your loss. The same can happen at matchpoints, but at that form of scoring every deal is equally important. Not so at IMP pairs. Being at the right table at the right time becomes very important.

Litvack: Bob's my uncle! What can we do?

Gowdy: As I see it there is only one solution. Make all of the boards equally important. We can either remove the exciting deals from the event or remove the boring deals. The computer program that makes the hand records can be set to do either. I suggest that we program the computer to make as many slam and wild distributional hands as possible. The players will have more fun that way.

Litvack: All right, you've convinced me. Next years hands will be as exciting as possible. We may have to throw in the occasional flat board so that the players do not work out what we have done. I don't think it would go over too well if the players found out what we were doing so let's keep this conversation to ourselves. Thanks for the advice. I hope you can find a better partner for next year's event.

Approximately one year later (in March 1994) Geoff Hampson, Canada's brightest young star, had this conversation with Irving Litvack at the Regal:

Hampson: Congratulations on another superbly run event, Irving. If it wasn't for the pathetic play of my partner, Fred Gitelman, I would have thoroughly enjoyed myself. I have never seen so many exciting deals.

Litvack: I am glad you enjoyed the event, Geoff. Don't be too depressed by the results. You did manage to beat all of the South American pairs. I didn't notice the deals being any more exciting than usual.

Hampson: You mean to tell me that you played in the event and you didn't notice how many unusual distributional hands there were?

Litvack (starting to get uptight): Nope.

Hampson: All right, I guess I'll have to spell it out for you. Let's start with the hand with 10 solid spades:

Litvack: Now that you mention it, I did hold that hand. It went 2 (weak) on my left, 2NT by Roy (Hughes).

Hampson: The auction started the same way at our table. Luckily, Fred and I have methods here. I was able to bid 4 as a transfer to spades followed by 4NT, Roman Keycard Blackwood. When Fred showed two Aces I bid 7NT trusting him to be able to find a 13th trick. I am sure the bidding made Fred nervous, but he quickly claimed as he held:

Litvack: Roy and I reached the same contract through a delicate cue-bidding auction. I was surprised to see that we won a sizable number of IMPs on what seemed like a flat board.

Hampson: I have never been dealt a 10 card suit except in your Ghoulie games. Fred tells me that the odds of being dealt a ten card suit are more than 60000 to 1. If this was the only unusual hand, I might have dismissed this occurrence as a coincidence. I guess you didn't notice but there were also a large number of 8 card suits being dealt. Fred's computer says that the odds of being dealt an 8 card suit are about 200 to 1. In the 116 hands that we played, Fred and I were dealt 5 eight card suits. In three cases the suit was a club suit that was either completely or almost completely solid.

Litvack: How many 9 cards suits were you dealt?

Hampson: None, and I think you are avoiding the issue. I thought that this hand was an interesting bidding problem:

Nobody was vulnerable and your LHO opens 1. Partner bids 2, Michaels, and RHO leaps to 5. What do you bid?

Litvack: I think 8 card suits are made to be trump. I would bid 6.

Hampson: Suppose you pass and your partner re-opens with 5, presumably showing a huge hand with longer hearts than spades. What now?

Litvack: How can I not bid a slam?

Hampson: Well your partner didn't promise a diamond void but you certainly have a very good hand considering your partner bid all the way to the 5 level himself. Suppose you bid 6. It goes Pass, Pass, Double.

LHO leads a club (the double was Lightner) and dummy tables:

Yes, 7 is laydown but you must make 6. RHO ruffs the club and returns a diamond, putting you in dummy. To make the hand you have to cash the A and ruff a heart and then lead and pass the J. RHO started with 2380 distribution (yes another 8 card suit) with the doubleton 10. Fred bid only 5 over 5 so he wasn't put to the test in slam. The way things were going for us, I doubt he would have gotten it right.

Speaking of 6-5 hands, the odds of being dealt one are about 75 to 1. Fred and I held no less than 12 6-5 hands in the 116 boards that we played. Rather against the odds wouldn't you say?

Litvack (trying to avoid the question): I heard that you bid and made 7 on one of the other 6-5 hands.

Hampson: Fred actually did well on that hand:

Fred played 7 from the short side after a weak no trump opening and a transfer auction. A spade was led to the A and Fred cashed two rounds of trump. RHO began with 3 trumps and Fred was at the crossroads. He could either draw the last trump and rely on the club finesse or try to ruff out the Q. What would you do?

Litvack: I would think about it for a long time.

Hampson: That goes without saying. Fred thought about it for a long time as well. Eventually he opted to finesse in clubs. This was necessary as LHO had 4 clubs to the queen. Computer analysis later suggested that Fred's line was superior by about 6%.

Litvack (still trying to avoid the issue): What happened to you with:

Fred held that hand. He opened 1 as is his style. Meckstroth thought it was right to open 2, but Hamman also opened 1. It went 1 Pass Pass back to him. Having never had this sort of problem before, Fred made up what he thought was a simple solution - bid what you think you can make. He bid 4 trying to convey the message that he could make either 4 or 5 in his own hand. I did not quite get the message as I passed with:

6 requires only a 3-2 club break and 6 is even better. In 6 you can sometimes succeed when clubs are 4-1 by drawing 2 rounds of trump and starting clubs. If the player with a singleton club has only two hearts, you can score a club ruff in the dummy. Fortunately for us, LHO had a singleton club and 3 hearts so slam had no play. Hamman suggested that you should re-open by cue- bidding 2 and then bid 4 at your next opportunity.

Litvack: One of our better results happened on a 6-5 hand:

South opened 2 on his 6-5 showing clubs and a major. Roy overcalled 2 and North cue-bid 3. South, with several spade stoppers, bid 3NT. Roy led the A and we took the first 6 tricks. North/South are cold for 6, of course.

Hampson: Of course. Unfortunately we got to 7. Fred opened 1 as South and Gord Chapman, West, overcalled 2. I was North and made a negative double. John Sabino, East, now made the excellent bid of 5. Fred bid 6 and I raised him to 7.

Litvack: You didn't give him much slack.

Hampson: True, but we were out of the event by this point. The one and only way we could make any money from the event would be to win a session award. This was one of the first boards of the last session and I judged that although bidding seven might be a bit of a stretch, we really needed to create some action if we were to win the session.

Fred ruffed the opening heart lead (first hurdle) and thought for some time. He didn't look very happy. Eventually he played a diamond to the jack, which held. At this point Fred sat up in his chair. This was just the kind of break we needed and it seemed like he thought he was making the hand. Unfortunately, the 5-1 spade break and 4-2 diamond break combined to make the hand unmanageable. 7 had to go one down. At least we won some IMPs due to the result at your table.

Litvack: That's a very sad story. I hope you can find a better partner for next year's event.

Hampson: That shouldn't be too difficult. I hope you can find a more normal set of hand records.

Litvack: I'll see what I can do.

These conversations never actually happened. The odds that I quoted are real and the above deals (plus many more unusual ones) were played in this year's Canadian Invitational Pairs. I strongly believe that computer generated hands conform to the expected statistical distribution. The events that I have described are either a statistical anomaly or the result of a programming error. I am not complaining. I thoroughly enjoyed the deals in the event. Thanks to Irving Litvack for putting together an excellent tournament, to Geoff Hampson for being a patient partner (most of the time), and to John Gowdy for being John Gowdy.