Saturday, October 15, 2011

Trial bids

After a major is raised to two, most pairs have agreements about the meaning of a new suit: a "trial" bid, or "game try."  These days, the most common agreement seems to be the "help suit" game try.  Personally, I'm not a fan, because it's so hard to know what kind of help partner needs.  And, if responder retreats to 3 of our major, the defense know that they can fairly safely start with the trial suit if they have no other compelling lead.

Personally, I like the power of a double fit and so, traditional second suit bids make more sense to me.  On the other hand, I like short-suit tries too, because it is so easy for responder to know which points in his hand are working and which aren't.  With 7 or more "working" points, it's probably reasonable to bid game.

With some partners, I've played a rather complex system of two-way game tries: the first step is a prelude to a short-suit try (after partner's relay), the third and fourth steps are long suit tries and the second step is everything else (including the "power" try and the short-suit try we skipped with the first step sequence). 

A number of years ago, I read in the Bridge World about Eric Kokish's scheme for two-way tries.  This idea made a lot of sense to me.  Now, I have been able to agree that system with several partners.  It's simpler than the other two-way scheme and really has all of the advantages, while concealing the nature of opener's (declarer's) hand as much as possible.

After 1–2–, 2♠ asks for the lowest ranking suit in which responder would accept a game try (2NT substitutes for spades).  Otherwise, 2NT, 3♣, 3 are short suit tries (NT again substituting for spades).

After 1♠–2♠–, things are simpler: 2NT asks for partner's lowest-ranking suit with help, and 3x is a short-suit try in x.

Of course, if responder names a suit where he has help but opener isn't impressed, he can make a counter-try in a higher-ranking suit, at the expense of revealing more about his hand.

Here's a hand from the club this week where we didn't get to the optimum contract, playing help suit game tries.  After 1–2– and holding ♠AK92 A8542 K85 ♣K, Len asked for help in diamonds with 3.  After all, he didn't need much help in spades.  With ♠QT75 T96 JT94 ♣A3, I reasoned as follows: my spade honors were likely useless (since partner skipped over spades) and my diamond honors might simply be too slow to be really helpful.  I therefore bid 3 only.  Well, I was right about the diamond honors but of course wrong about the spades.  We made four for a score of 66% (it wasn't the easiest of contracts and many of those who did bid game went down).  

Playing Kokish with this hand, opener doesn't have to be the one to choose which suit to make his game try in.  Assuming that he doesn't fancy a short-suit try in clubs, he can simply start with 2♠: where can you help?  I would respond 2NT (I have help in spades) and we'd be off to the races.  If I bid 3♣, he could still ask about diamonds.

With Kokish, the distinction between "help" suit and "second" suit is less of an issue because the only kind of help that responder will show is of the form honor third or better.  He will never show "help" with x or xx.


  1. Hi, Robin,

    I have been using a system of amorphous game tries which not only conceals the nature of the hand of opener/declarer, but also permits opener to show game tries based upon true two-suiters and hands that need help in trump quality. The method discloses to the defense even less of the hand of opener/declarer than does Kokish game tries, I think.

    The system might be called Nagy game tries. It is described by Marshall Miles.

    After the game try bid of the next cheapest suit, responder bids the lowest suit in which he would reject a short suit game try. The inference is that responder has substantial non ace values in suits he has skipped and no substantial nonace values in the suit he has bid. This should tell opener what he needs to know.

    Meanwhile, bid of a new suit by opener is natural five card suit (meaning that responder can evaluate his hand in the same way he might opposite a partner that had made a Michaels call showing a specified two-suiter) and re-raise to 3M by opener is an ask about trump quality. There are other wrinkles, too, but the above describes the core of the system.

    Overall the approach seems to combine undisclosed short suit game tries and help suit game tries with long suit game tries and trump suit game tries, in a fairly natural approach.

    Like any game try method, successful use does depend upon the judgment of the users. Responders need to understand what hands would reject a short suit game try (hands with wasted seoondary honors or matching shortness) and what hands would accept short suit game tries (hands with no wasted secondary honors or matching shortness, particularly those hands with a fourth trump). And openers need to have the ability to "put" responder on secondary honors in suits that he skips and then evaluate how such honors complete or fail to complete his hand.

    Although Kokish and Nagy share many of the same positive attributes, where I think Nagy is superior is that opener never shows the suit where he has a short suit game try. That concealment makes the defense tougher.

    The hand you and Len held presents sort of an iffy game, not surprising given that opener has a six loser hand, and not unexpected since the value of stiff honors is nearly also hard to ascertain. But if opener did choose to try for game, he would learn quickly that responder has the SQ, and that is useful information to opener. 1H-2H-2S-2NT would be the start. Meanwhile, although declarer knows about the SQ, the only missing high honor, even savvy defenders can't specify the spade holding of responder (could be KJx, for example).

  2. With respect to all of this science about game tries, perhaps we should remember the advice of Kit Woolsey. Woolsey, who uses tons of science in many auctions (competitive auctions, choice of game auctions, anything in slam investigating) has something like this to say about game tries: You bid game, then you try to make game!

  3. Thanks, Jeff, for letting me know about Nagy game tries. It's possible that they are technically superior to Kokish. I don't know. But what I do know is that Kokish game tries are, after maybe 15 years, getting to the fringes of the mainstream; so knowing how to play them can be helpful when developing a system with a new partner. I doubt if the same is true with Nagy.

    I've played against Kit Woolsey a couple of times and there's no doubt that he's a very fine player. I recently had the pleasure of experiencing some of his more scientific treatments first-hand at the Warwick tournament. I'm happy to say we came out ahead on the exchange.