Improving the 2-over-1 Bidding System


When I learned bridge, my aunt taught me old-fashioned Goren, which was played by most everyone in the casual games on my college campus. Then I discovered duplicate and a host of "revolutionary" ideas -- limit raises, negative doubles, weak two-bids -- that were relatively simple improvements.  

The next big change was the 2-over-1 system, which gained popularity so fast that it became difficult to find a partner who wanted to play any other system. That was fine with me, as I thought 2-over-1 was even easier to adopt than "duplicate standard". 

I soon learned that although the basic premise of 2-over-1 is simple enough, the auctions can become quite complicated and some types of hands are almost impossible to describe. Those flaws -- and the superiority of standard methods -- were the focus of Frank Stewart's article in the May Bridge Bulletin. Arguing the other side was Larry Cohen, who made the case that 2-over-1 is the more accessible, effective choice for players of all skill levels.

Whichever view you support, there's no question that 2-over-1 is the dominant system in club and tournament bridge. Its fans believe its has clear advantages over standard bidding, but most recognize that the system is far from perfect.

Can it be improved? The first step is understanding the source of the problems, which fall into three categories: System definition, user error and design weaknesses. 


System definition

A common misconception is that 2-over-1 has a universal set of clearly defined meanings for early responses and rebids. It would be convenient if you could say "2-over-1?" to a new partner and be confident that you're playing the same system, but you can't. 

Lawrence or Hardy or ... ?

There are several ways to structure 2-over-1 systems and it's helpful to be aware of the variations. A number of authors -- including Mike Lawrence, Max Hardy, Marty Bergen, Paul Thurston and Eric Rodwell/Audrey Grant -- have published books and software that outline their versions of the system.

The two most widely used approaches are from Lawrence and Hardy, who differ in their recommendations for some of opener's rebids. In general, Lawrence emphasizes value rebids that define opener's strength. Hardy includes more rebids that show distribution, but do not pinpoint strength.

Even simple sequences can be misinterpreted if you're playing one style and your partner is playing the other. After the auction 1S-2C, for example, does opener promise stoppers in both unbid suits if he rebids 2NT? He does if he's a Lawrence player, but not if he learned the system from the Rodwell/Grant books.

Misunderstandings about "game level"

Your partnership can agree that "forcing to game" means any game, including minors, but most 2-over-1 pairs believe that's taking the system name too seriously.

Since responder will consider any good 12-point hand a game force, the partnership's combined strength may be only 24-25 high-card points when opener holds a minimum. That's (barely) enough to make 3NT, 4H or 4S. A minor-suit game requires considerably more power, and when you don’t have it, you need to be able to stop below the five-level. That’s why the practical definition of “forcing to game” is “forcing to 3NT or the four-level”.

This can be an issue in auctions where you don't have a major-suit fit and you've been unsuccessful in finding stoppers for notrump. Unless one of you has shown extra values earlier in the auction, a bid of four of a previously bid minor suit may be just a retreat that can be passed.  


User error

Saving too much space

One of the big benefits of 2-over-1 auctions is that neither partner has to jump to show forcing values. The bidding can stay low, allowing more room to exchange information below game level and evaluate slam possibilities. 

That benefit can become a pitfall if an auction has what Bridge World magazine calls "two temporizers and zero describers". This happens when one or both partners are so intent on conserving space that they make every rebid at the lowest level possible, no matter what their strength. The result is that they learn a lot about each other's distribution, but very little about suit quality and high-card strength.

Bridge Bulletin columnist Frank Stewart called this problem his "biggest gripe" about 2-over-1 and cites it as proof that the basis of the system is flawed. Most of these failures, though, can be blamed on the bidders, not the system itself.  

Two-over-one bidding is most accurate if you agree that responder does not jump with extra values, but opener may. In most 2-over-1 auctions, responder is the "asker" (temporizer) and opener is the "teller" (describer). When responder has slam aspirations, he uses forcing, low-level rebids to give opener room to provide more information. This is the principle of fast arrival -- responder's fast rebids (jumps to game) show minimums; slow (low-level) raises and notrump rebids tend to show extras.

Opener, however, makes value bids whenever possible -- he bids higher with stronger hands, lower with minimums. Rebids that show extra values include jumps in his suit or notrump and (in Lawrence style) reverse rebids. The sooner opener can make one of these bids, the easier it is for responder to determine the partnership's combined strength. 

Suppose you open 1H with  ♠A74  AQJ1097  6  ♣KQ3  and partner responds 2D. Those who have fallen in love with the space-saving features of 2-over-1 will rebid just 2H, which conceals their strength and suit quality. They've also increased their chances of scoring +680 when the field is making +1430. It's worth using up an extra level of bidding if it gives partner extra information, so the value bid of 3H is perfect.

It won't always be possible for opener to show his exact strength with his first rebid. After 1S by you, 2C by partner, you have to settle for a minimum rebid with hands such as

     ♠AJ8643  42   AKQ  ♣K4  

     ♠KQJ65  AQ854  Q2  ♣A 

Neither hand has a strong message to send about where to play, so don't waste a level by over-emphasizing a weak suit or jumping in a new suit. Make your rebid at the two-level and plan to show your extra strength later.

Overuse of the Forcing Notrump

Many players dislike responding 2C to a 1S opening with   ♠Q4  AJ43  A62  ♣Q764, but the alternative -- a Forcing Notrump -- can leave you with awkward choices later in the auction. 

If you respond 1NT with this hand and partner rebids 2C, 2D or 2H, you have no forcing bid available. Your only option is a unilateral jump to game, which prevents you from learning anything more about partner's hand. You'll also have trouble communicating your strength if partner makes a jump shift or if the opponents enter the auction.

You can avoid these problems by limiting your Forcing Notrump range to a maximum of a bad 12 points. When you hold game values, establish the force right away by choosing a 2-over-1 response, even with a weak suit.

"Auto" cuebidding

Some 2-over-1 auctions will be so efficient that you'll be left with bidding space you don't need, and there's often a temptation to fill it just because it's there. Here's an auction that causes problems for many pairs:

  You       Partner
    1S          2C
    2H         3H

You hold  ♠KJ1082   KJ85  A72   ♣6

Partner’s raise to 3H is slow arrival, suggesting at least moderate extra values. You have room to make a "free" control cuebid on the way to game, so you could try 4D, just in case partner was planning to head for 6H.

To partner, though, 4D will sound like a serious move toward slam. He'll cooperate by cuebidding 4S or 5C and you’ll be propelled to 5H – if you can stop there – with what may be an ordinary fit and only 26 high-card points. 

The first partner to offer a cuebid in this type of auction should be promising something extra (a strong 14+ points). With a dead minimum, just raise to game and let partner decide if his hand warrants a try. If he has enough to make 6H opposite the hand above, he’ll bid again.


Design weaknesses

Many of the problems encountered by 2-over-1 pairs are caused by confusion about how to interpret system bids and make the best use of auction space. Other problems are inherent in the system itself. Not all are fixable, but many players have fine-tuned the structure with new treatments and gadgets that have proven effective. Here are some of the system's limitations and ideas for solutions.

Forcing Notrump woes

All bidding methods involve tradeoffs, and the big one in 2-over-1 is the loss of an easy way for responder to show invitational strength. Most of these hands are handled by starting with the Forcing Notrump, which is the "necessary evil" of 2-over-1 and can be blamed for many of the system's flaws.

Every 2-over-1 player has landed in at least a few ridiculous contracts after Forcing Notrump auctions where opener rebids a minor. The 2C and 2D rebids often give responder an uncomfortable guess because they're only "semi-natural" (each could be a 3-card suit) and they don't pinpoint strength (opener could have up to 17 high-card points). It's not uncommon to be playing a poor fit at the 2-level when your only making contract was 1NT. 

The most basic advice for responder in these auctions is "Try not to pass", especially when you hold a good 9 points or more, which could be enough for a game. Keep the bidding open by taking a preference to partner's major with a doubleton, even if you have fair support for his minor. With stoppers in the unbid suits, stretch a bit to rebid 2NT with 10 points. Avoid bidding a new suit unless you have 6+ cards or a strong 5-carder. Accept that some deals will be  misfits where you have to settle for what could be a 3-3 fit.

You may want to consider adopting some of these methods for improving (or avoiding) Forcing Notrump auctions:

Semi-forcing Notrump: This change helps you locate "real" minor-suit fits and stay low when opener has a dead minimum. Your 1NT response is still 5 to 11 points, but opener is allowed to pass with a balanced 12 or "bad" 13 points. If he bids a new suit over 1NT, he promises 4+ cards or extra values.

One drawback is that although opener will be balanced when he passes 1NT, responder might not be. You may struggle in 1NT when a suit contract would play better -- and would have been located if opener had been forced to bid again.

If you adopt the semi-forcing treatment, you'll want to modify other parts of your 2-over-1 structure. Unless you don't mind missing 5-3 major-suit fits, you can no longer use 1NT to start the description of hands that have support for opener's major.

A problem hand with this structure is a balanced 14-point opener such as  ♠AQJ54  AJ10   64   ♣Q84. If you open 1S and partner responds 1NT (semi-forcing), he could have up to 11 points, so you can't pass. A 2C rebid, though, defeats the purpose of playing the Semi-forcing Notrump, which is to avoid the 3-card rebids. One way to accommodate these hands is to switch to a 14-16 range for your opening 1NT.

Bergen Raise modifications: 2-over-1 provides an inelegant method for inviting game when you hold three cards in opener's major -- after a 1S opener, a hand such as  ♠Q74  A3   K9743   ♣J102 . To describe this 3-card limit raise, you must start with 1NT, then jump to 3S at your next turn. This delay can make it difficult to communicate your strength and support if the opponents overcall or partner makes a jump shift.

Those who play Bergen major-suit raises can build in a way to show this hand immediately, without going through the Forcing Notrump. After partner opens 1H or 1S, your jump to 3D is the 3-card limit raise. A jump to 3C is a two-way bid -- either a 4-card constructive raise (8-10 support points) or 4-card limit raise (11-12). If opener wants to know which one it is, he bids 3D to inquire. Your retreat to three of the major shows the constructive raise; any other bid confirms the stronger hand.

Impossible 2S: This widely used convention applies after the auction 1H-1NT; 2C or 2D. A 2S rebid here is "impossible" (therefore artificial) because you've already denied spade length. It promises a good high-card raise (10+ points) of opener's minor with at least 4-card support -- a hand such as  ♠954   43   KQJ4   ♣KQ65. Some pairs specify that it also denies a spade stopper.

This allows you to distinguish between invitational and constructive raises. A direct raise of partner's minor (1H-1NT; 2C-3C) suggests a more distributional hand with 5+ trumps and fewer high-card points -- a hand such as   ♠654  5   A764   ♣KJ1084.

Opener's rebid after an overcall:  

   You     LHO     Partner    RHO
    1S       Pass       1NT          2H
      ?

The vagaries of the Forcing Notrump give you a difficult decision when you hold a hand such as  ♠AKJ43   4   Q962   ♣AK10. An "expert standard" agreement is that opener's double here shows extra values and shortness in overcaller's suit. A freebid of a new suit promises five cards. This gives you a flexible way to define your strength and cater to as many of partner's hand types as possible, without getting too high.

Bart and Gazilli: Many experts like these conventions because they allow you to show a wider range of hands after a Forcing Notrump response. Both have complex, artificial structures and are for serious partnerships only.   

Invitational hands with long suits

The basic 2-over-1 structure offers no clear way to describe a response such as  ♠43  K2  J104  ♣KQJ985  after partner opens a major. There are two ways to fill this "hole" in the system:

Out-of-control cuebidding

Earlier, we discussed the questions that arise in this type of auction:

  Partner     You
    1S             2C
    2S             3S
    4D

Partner's 4D is obviously a control bid, but does it promise slam-going values? Or is he making a "courtesy cuebid" on the way to game, just in case you're interested in slam? Is it possible that he thinks your slow-arrival raise to 3S demands a cuebid?

The simplest agreement is that cuebids are not mandatory here, and the first partner to make one promises extra values. To show a minimum, partner would raise to 4H. That will keep you from getting caught up in cuebidding sequences with unsuitable hands, but it wastes valuable bidding space on deals where you hold slam values and wanted to hear a control bid.

Serious 3NT:  You can improve your cuebidding with this convention, which is used in forcing-to-game auctions where you've agreed on a major at the 3-level. A 3NT continuation by either partner says you have serious slam interest and want to hear a control bid (ace, king, singleton or void). If you fail to bid the Serious 3NT when you have the opportunity, you deny extra values. In the auction above, partner's 4D is therefore a "non-serious" cuebid with a minimum opener.

Note that some partnerships reverse the meanings, so 3NT is "non-serious" and a cuebid is the stronger move toward slam. 

Picture Jumps:  These rebids can also help you avoid unsuccessful cuebidding expeditions to the 5-level. In 2-over-1 auctions where you've bid only two suits, a jump to game in the major -- by opener or responder -- denies controls in the unbid suits. For example, if you open 1S and partner responds 2H, your jump to 4H shows a hand such as  ♠KQJ54  AJ72   Q4   ♣83 .

Hand-evaluation problems

The ideal suit for a 2-over-1 response is 5+ cards, but a 2C or 2D response will often be a 4-card suit in a balanced hand. This possibility makes it difficult for opener to evaluate his holding in your suit, and he'll be hesitant to raise with 3-card support.

Natural 2NT response: One solution is to play that a 2NT response to a major-suit opening is a balanced 13-15 high-card points with no 5-card suit. It may contain three cards in opener's major, which you can reveal later. Over 2NT, opener makes natural rebids at the 3-level, allowing you to find 4-4 fits in other suits. 

With this agreement, a 2-over-1 response is always 5+ cards or 16+ points. The only drawback is that you can't use Jacoby 2NT as your forcing major-suit raise. A workable substitute is the cheapest jump shift (1H-2S and 1S-3C), which will require you to develop a new set of follow-up bids.

Handling interference

Opponents' overcalls in your 2-over-1 auctions are relatively rare but can be troublesome. After you've made a 2-over-1 response, there's little need for playing doubles as takeout. The typical overcall is a lead-director with a weak hand, and you need a way to make the opponents pay when they've stepped out too far. A double of an overcall should be penalty-oriented, showing tricks, at least moderate length and (usually) little or no support for partner's suit. Not all agree, so discuss this with your partner.


Design-your-own gadgets

There are several rebids in 2-over-1 that are used so rarely that they could be assigned new, more valuable meanings. You and partner may want to discuss ideas for defining these auctions. They include:


   ©  Karen Walker       Back to Karen's Bridge Library