If a contract depended on say, a finesse, most good players would look for plays that would give a better than 50-50 chance of succeeding. (Unless the finesse was marked by the bidding or early play.)

Most good players would make, or consider making, a safety play to guard against a 4-1 split (except maybe in matchpoints). This occurs with a greater than 28% frequency (or 14%+ if it was relevant with the singleton on one side and not the other).

But I read of a situation in a tournament where a world class declarer took an inordinately long time to study the play of a hand, to the point where his opponents called the tournament director (TD). When the TD asked the player why, the player answered that he had just worked out a way to make the hand against a distribution with only a 2% chance of occurring (a 5-0 trump distribution with the 5 "offset" a finesse).

He looked like a genius when the relevant opponent admitted to having that hand. Apparently, there was little time to play the game out, but the TD awarded the point to the player after the player described his game plan and the opponents conceded, showing down their hands. All the other declarers on that board went down.

Is that unusual? Do good players routinely study ways to guard against 2% chances over the board? Or do most players have a higher threshold, say in the 5%-10% range?

3 Answers 3


2%? I don't think most would worry about 2% on most hands, because most hands you'd be losing a higher percentage of the time to protect against the lower.

But yes, in this kind of case it's not uncommon to protect against 5-0 splits. While it's not common, 2% is 1 in 50 hands, meaning it might come up a few times in a weekend; the difference between a highly experienced player and a top-tier player might only be a few hands, and those would be some of them.

It really comes down to the hand. If the hand is played at IMPs (where making the hand is paramount, rather than overtricks), and it's solid for 4-1 or 3-2, then there's no reason not to explore 5-0 split lines. I've certainly done it a number of times before, and I'm hardly a world class player.

I suspect your story is a bit of an apocryphal one, given the description of the actions after the tournament director was called are inaccurate to what would really happen at a tournament (slow play involves slow play warnings, and eventually board penalties, but would not modify the results of a hand; the tournament director would not "award the point" either, it would be up to the opponents to accept the claim or not, if a claim was indeed what was made). But the actual details - that a world class player considered his line of play for several minutes in order to find a safe line against a 5-0 split - are entirely reasonable, if a bit boring compared to the dramatic telling.


In an IMPs situation, you can calculate the IMP odds to determine what the right thing to do is. Let's work an example.

The declarer is in 4H vulnerable. Line of play 1 makes 5 most of the time, but x% of the time goes down 1. Line of play 2 always makes 4. Let us assume that your opponents are in the same contract will always take the safety play. Line 1 wins 1 IMP for the overtrick (100-x)% of the time, but loses 12 IMPs for going down x% of the time. 1 * (100 - x)% = 12 * x% when x = ~8%. So if that's the situation, you should only take line 1 if it has less than 8% chance of going down.

Most situations are a bit more complicated but the principle is the same.

  • 1
    I don't think that's what OP is asking. OP is asking how often people bother to spend the time to figure out a line for a 2% likely play, not how people compare different lines of play with different probabilities.
    – Joe
    Commented Jan 9, 2017 at 16:37
  • You can judge whether thinking about a 5-0 break is worthwhile by figuring out the odds of the break and the likely number of tricks a safety line would take.
    – ruds
    Commented Jan 9, 2017 at 16:41
  • 1
    Yes, you certainly can, but OP is asking if people do, not if it is possible to.
    – Joe
    Commented Jan 9, 2017 at 16:42
  • 2
    @Joe: I see this answer by ruds as an answer to the title of the question. The other question, "do good players do that", is a pointless question IMO. Answer is, yes depending on the situation. Now what does that get us? Anyway...
    – Aryabhata
    Commented Jan 26, 2017 at 21:25
  • 2
    @Joe: If you are thinking of the question having an answer like 2%, then the question is completely nonsensical/subjective. The only sensible interpretation of the question IMO gives ruds' answer as a reasonable answer. Anyway, Tom has the knack of asking strangely worded questions :-), so I will drop out of trying to come up with a reasonable interpretation of his question this time.
    – Aryabhata
    Commented Jan 26, 2017 at 23:16

Many, many years ago, in the 35th European Championships, played in Birmingham in 1981, Hungary and Britain played a hand which showed the extremes to which some players go to ensure the contract.

(I have since been able to find the hand in the EBU archives and have updated my answer accordingly.)

Dealer South, Game All.

North: SAK5     H6     DAQJ1075 CA82
South: SQ109875 HKJ987 DK       C3

In the closed room, after an initial pass by South, the bidding was as follows:

    West       North           East          South
   (Rose)   (Dumbovitch)     (Sheehan)    (M. Kovacs)
    Pass        1D             Pass           1S
    2H          3H             Pass           4S    
    Pass        6S             All pass.          

Kovacs took two rounds of trumps, leaving the King in dummy, played the King of diamonds from hand, with the high trump in dummy securing entry to the remaining diamonds and the slam against a 5-1 split in diamonds.

In the open room the Hungarian West doubled, rather than bidding 2H, at least hinting at a 2-suiter. Collings, sitting South for Britain, also found himself in 6S facing a club lead.

He won the Ace of clubs in dummy, followed by the Ace of spades and a low heart. When East played low, Collings could afford to lose the heart trick to West. He could then ruff a heart, draw trumps and take the diamond tricks in his own time.

The diamonds were 6-0. The safety play succeeded and 17 IMPs went to Great Britain. (From EBU Quarterly, August 1981, page 15)

In general, as ruds explains, you estimate the expected value of the safety play. As a rule of thumb, you play safely at rubber and IMPS. At match points you try for overtricks if you are in a contract that most of the room will be in. If you are in a slam that you think very few people have bid, then you would also play safely to ensure a top or near top.

  • That hand's a great find. I only wish I could upvote more than once. Commented Feb 6 at 2:27

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .