Short Suit Bids at Auction Bridge

An Analysis of the Requirements for Bids on Suits of Three or Four Cards

R. F. FOSTER

SUITS which can be considered as short, for bidding purposes, may not be so considcred short for leading or play. In the bidding department of the game it has so long been the custom to look upon five cards as the minimum for an original call in a suit, that any bid on only four cards in a major suit, or three or four in a minor suit, often calls lorth something like an apology from the bidder to his partner, as if he had done something quite unusual.

In the play of the hand, especially in the opening leads, four-card suits are not regarded as short, and suits of three cards only are frequently selected as better than longer ones. It is only in bidding—upon suits of less than five cards—that we have now to deal.

REASONS FOR SHORT-SUIT BIDS

ITT WAS the consideration of the question ol II suit distribution, in the player's own hand, that first called attention to the advisability of bidding four-card suits. In all hands of thirteen cards that offered a choice between a majorsuit bid and no trumps the suit was always given the preference if it was one of not less than five cards, and had some winning cards in it. As an original call it was regarded as better because safer. It is only lately that players have come to regard major suits of only four cards as better than no-trump calls, when the three other suits are unequally distributed.

This modern idea is based upon the experience that hands with missing suits, singletons, or two two-card suits, are not good risks as no-trumpers, and the best players now recognize only three suit distributions that are favorable to no-trump play. These arc:

4-3 3 3,occurring 1 05 times in 1,000

4 4 3 2, " 215 " 1,000

5 3 3 2, " 155 " 1,000

Total 47 5 " 1,000

The distribution of 5 4 2 2 is called a borderline bid, depending on the high cards in the two short suits. This comes 106 times in 1,000 deals.

The remaining four or five hundred deals in every thousand all show missing suits or singletons, so that about half the hands dealt are bad no-trumpers, simply on account of the suit distribution. T he result of this consideration has led modern players to prefer a good suit of even only four cards as a bid, in all hands which have no five-card suit, but have the danger signal of a singleton or missing suit. Take the following example:

Z dealt. Those who bid no trumps on his cards, and were left to play it, lost the odd trick against easy aces. Those who recognized the unfavorable distribution and bid any one of the four-card suits escaped this.

The most usual opening bid was a spade, overcalled by A with two diamonds, Y and B passing. The dealer then bid his hearts, and if overcalled was helped by his partner, Y. I'hey scored three odd and four honors. Some started with a club, and afterward bid the major suits. At two tables A and B went too far with the diamonds and were doubled and set 200 points, less their honors.

THE STRENGTH REQUIRED

rip HERE are many hands that are not strong II enough even to consider bidding notrumps, but which contain a good four-card suit. When the hand as a whole contains enough high cards in three different suits to suggest a no-trumper, there is no question about the support that the four-card suit bid will get from the other and perhaps shorter suits. The hand just given, counted by the double-valuation system, is worth eight tricks.

It is when the four-card suit suggests bidding it on its own merits, and not as part of a notrumper, that we have to consider just how strong it should be, and what outside support it should have, if any, to make up for the shortness in the suit itself. If it is admitted that any five-card suit, headed bv what would be estimated as four tricks on the doublevaluation system, such as ace and king, is a

good bid, what is required to make up for the missing fifth card?

We have to consider the probability of the adversaries holding numerical superiority in the trump suit, and the strength in plain suits required to meet or offset that contingency. Some elaborate calculations have been made for the distribution of the remaining nine cards of the suit in which the bidder holds exactly four, and it has been shown that his partner should hold three or more about two-thirds of the time, so that there is no particular risk on that account. With less than three he will deny the suit.

According to the calculations made by the late W. H. Whitfcld, Professor of Mathematics at Cambridge, England, and for years the Card Editor of the London Field, four cards in the trump suit possessed no material trick-taking value when considered from the point of length only, but each trump over and above four was worth a trick. This is easily understood when it is considered that a fifth trump takes the place of what would otherwise be a losing card, in the bidder's hand.

If this is true, the player who bids on a fourcard suit that is headed by four trick values, such as ace and king, with an outside trick anywhere, has just as good a bid as if he held five trumps and no outside trick. The thing looks like an exchange of values. If the adversaries have the extra trump, they have not the extra outside trick. If they have the outside trick, the bidder has the extra trump.

If we accept the principle that the partner will deny the suit if he has less than normal assistance, say three small trumps, we have to deal only with cases in which he has exactly three, and find the probability of one or other adversary outholding him. This chance has been found to be about two to one in their favor, which is exactly the reverse of their chances when the bidder has five and his partner three. It is then two to one against either of the adversaries having as many as four trumps.

This would seem to indicate that the odds are against four-card suit bids as such, but when the suit is backed up by enough outside strength to restore the balance of power, bids on four cards are just as good as bids on five.

FIVE TRICK VALUES WANTED

While a hand containing only four trick values will justify a bid on five cards, on account of the trick value of that fifth trump, it is evident that when that extra trump is missing we must still have its equivalent; that is, five tricks, to justify a bid on only four cards. In my investigations of these four-card suit bids I have found that all such bids on fivetrick hands, are just as safe as the conventional bids on five cards in four-trick hands.

Modern bidding tactics lay it down as a principle that if there are four trick values in the hand, and a five-card suit, there should be a bid, unless the distribution is unfavorable, such as all the strength being in very short major suits. When we come to look up the authorities we find they do not agree on the four-card suit bids; and some of them seem to be simply guessing.

(Continued on page 106)

(Continued from page 79)

Whitehead, on page 24 of his Auction Bridge Standards, says that for all practical purposes four cards headed by A K Q are as good as five cards headed by A K alone. Under the head of what he terms "Four-card Compensated Units" he lays down the rule that any four-card suit headed by four tricks, at the double valuation, is a good bid if there is a quick trick outside.

Milton C. Work, in his 1924 edition, goes more into details, and gives the various outside strength required according to the strength in the suit bid. He starts by agreeing with Whitehead that four cards to the A K Q are worth 5 trick values, and need no outside tricks to justify the bid. He then gives the three combinations; A K, A Q J, and K Q J, which are each worth 4 trick values, and demands at least 2 trick values outside, bringing the total to 6, instead of 5 onlv. Proceeding to the combinations of A Q 10, A J 10, and K Q 10, which are each worth three trick-values, he demands 4 outside, bringing the total to 7. Going on to the still weaker combination, K J 10, worth only 1 1/2 tricks, he asks only 4 1/2 outside, bringing that total to 6 again.

Why certain combinations of cards in the suit named in the bid should require a total strength of only 5 tricks in one case, 6 in others, and 2 in others; and then, as the suit gets weaker, go back to 6 only, is not explained, and neither facts nor argument are offered in support of such inconsistency.

If the double-valuation system has one point in its favor above all other systems it is its regularity and its proved ability to produce the results predicted by its theory. This theory is that the combined hands will win, on the average, the full number of tricks that they are supposed to be good for. Therefore if five trick values is a good bid on one kind of a four-card suit, it should be good enough for any kind of a four-card suit, provided at least two of these tricks are in the suit named. My examination of a large number of deals which indicated good four-card bids has convinced me that it is.

BIDS ON THREE-CARD SUITS

I am unable to find any writer who even suggests bidding on less than four cards in a major suit, and I believe I am the only writer that advocates bidding minor suits of less than four cards. The distinction between the two classes of bids is that the major suits are bid with a view to having them for the trump; minor suits always have in view the hope that they " ill lead to something better, or as a preliminary to a secondary bid in a major suit. Take such a hand as six hearts to the king ten, with ace and two small clubs. Any modern player will bid the clubs, as an approach to the hearts. Why would it not be an equally good club bid with only three to the ace king:

M\rule for bidding minor suits of only three cards is to restrict the bid to hands that contain a good secondary bid, or that wt^uld support either major suit if it were bid by the partner. Take this distribution as an example:

Any of the four players might be the dealer, and any but A would have a bid. B has five values for a spade bid on four cards. If Y were the dealer he would certainly bid a heart on four values to five cards, and Z would certainly assist him if B overcalled him. Reverse the suits and let Y have a spade bid and B two hearts, Z would equally assist the spades.

This being the test, willingness to support the partner, I fail to see the objection to bidding a diamond on Z's cards as dealer. He has five tricks in hand and can support his partner in any suit. But if we take A's hand, there is no justification for an original bid in clubs. What would he do if his partner bid hearts: He can neither support the suit nor deny it. If B started with a heart and Z overcalled, A would pass. Therefore he should pass as dealer.

ANSWER TO THE NOVEMBER PROBLEM

This was the distribution in Problem LXXVII:

Hearts are trumps and Z leads. Y and Z want all the tricks. T his is how they get them:

Z starts with his smallest club, which Y wins with the queen. Y takes out both of A's trumps. On the first trump lead B will discard the spade ten and Z the club king. If B discards the ten of diamonds on the second trump lead, Z will shed the jack of diamonds and Y will lead the five.

If B now discards the spade king, two of Z's spades are good and he can discard the smallest one. But if B discards a club instead of the diamond, Z will get rid of the club ace and all Y's clubs are good.

Z must be careful to discard the high club, and not the spade, on the second trick, or B will be free to shed a club on the second trump lead and must then make a trick with the king of spades.