I'm confused by the advice I've been getting on doubling from the column of a certain expert (Frank Stewart). In another column, he advised doubling South's one heart as West with this 14 point hand, where both sides are vulnerable. That barely makes sense to me.

But then Stewart advocates passing as West with the following 14 point hand, even when no one is vulnerable.

Spades A952
Hearts QJ7
Diamonds AK62
Clubs 84

I would much rather double with the above hand (good spades and diamonds) than the other hand (good minor suits). Perhaps tongue in cheek, Stewart suggests that the "advancer" (partner of the doubler) mostly refrain from bidding two clubs over opener's one heart (or one diamond), and partner's double, favoring spades and diamonds/hearts instead. This is unless the club suit were at least 5 cards in length, or a four carder headed by a much better honor than other four card suits. That is, xxx in diamonds would do better opposite the above hand than xxxx in clubs, giving the doubler some leeway in having only two good unbid suits, rather than three; that is both majors or one major and one minor.

Are there any experts other than Stewart who would pass rather than double with the above hand? Or should I double more, rather than less, often than Stewart recommends.

  • As before, "Are there any experts..." is a strange question, and unanswerable, you really just want to know if in general this makes sense or not.
    – Joe
    Jan 24, 2022 at 21:29
  • Did RHO open 1H?
    – ruds
    Jan 24, 2022 at 21:32
  • @ruds: South opened 1H. I am (role-playing) West.
    – Tom Au
    Jan 24, 2022 at 22:39

5 Answers 5


The major advance in competitive bidding over the last 30 years has been increasing understanding that what really matters in deciding how much you should compete is how distributional all the hands are, not how strong your hand is.

First, we have the Law of Total Tricks. It says that, on most hands, the number of tricks they take in their contract plus the number of tricks we take in our contract is equal to the number of cards they have in their trump suit plus the number of cards we have in our trump suit. It's not entirely accurate, but nowadays even good intermediate players like myself have a good idea of what adjustments to make.

Now let's think about the scoring implications. Suppose there are 16 total tricks, and you are deciding whether to be in a 3C contract over their 2H contract. Let's assume both vulnerable and undoubled for now. If they take 8 tricks and you take 8, then 3C is -100, better than -110 for 2H. If they take 7 tricks and you take 9, then 3C is +110, better than +100. Now if they take 9 tricks and you take 7, 3C is -200 which is worse than -140. However, if your partnership has somewhere between 17 and 23 of the 40 hcp available, chances are you'll score better by competing. It does not matter whether you are making 3C or going down 1; either way you do better by competing. (If both sides are nonvulnerable, the case for competing is even stronger.)

Now suppose there are 15 total tricks, and you have to make the same decision. If they take 8 tricks and you take 7, then 3C is -200, worse than -110. If they take 7 tricks and you take 8, then 3C is -100, worse than +100. It gets even worse if it's more unbalanced.

The conclusion is - don't worry too much about how strong you are; rather worry more about how good a fit they are likely to have and how good a fit you are likely to have. When you have 3 hearts and 2 clubs, it's much more likely there are only 15 (or only 14) total tricks. Having a 3rd club and one fewer heart makes it much more likely there are 16 (or even 17) total tricks.

  • OK, I'll need to rethink the implications of LOTT, not having done so before. I was taught (50 years ago) to double on "strength" rather than length. Specifically, 14 hcps with three cards in the opening suit (hearts) 13 hcps with two, 12 hcps with one, and 11 hcps with a void.
    – Tom Au
    Jan 24, 2022 at 22:30
  • One of the considerations on this hand is that QJx (or even Qxx) in the opponent trump suit is a negative adjustment on total tricks. Jan 24, 2022 at 23:28
  • You answered my other, related, question by saying that double was a lesser evil than pass. Wouldn't Kx of hearts in West also hurt the total trick count?
    – Tom Au
    Jan 24, 2022 at 23:43
  • No - if the finesse is on for you, then Kx is a trick for you both on offense and defense (unless they have 10 trump), and if the finesse is off, it's not a trick on either. With QJx, it's quite likely to be a trick for you on defense but not on offense, because partner has a doubleton and ends up forced to ruff your winner. Jan 24, 2022 at 23:46
  • Actually, I now believe that this hand is weaker than the other one under my own "rules." 14 hcps is "borderline" here because this hand has three cards in the opener's suit (Including two honors). I would still double, but "pass" now looks reasonable to me. With Kx of hearts in the other hand, I need only 13 hcps to be "borderline." With 14, double is better than pass. Thanks for your help.
    – Tom Au
    Jan 24, 2022 at 23:58

One issue - if you double, partner bids clubs, now what? There are people who play Equal Level Conversion, where 2D now doesn't show extras, and that's fine as far as it goes (but swap the minors, and you don't get to play that, and when you do have the 20-count with diamonds, you're stuck). One of the Emperor's rules: Thy takeout doubles doth promise the other three suits.*

Other issue - three-card hearts is a big warning sign, having quacks in them even more so. If they have a heart fit, partner is short, and will feel strongly encouraged to compete, "knowing" their fit is better than it is. If they don't, most likely the first three tricks go HA, HK, H ruffed. So much for your QJx.

I think what Stewart is trying to do is convince the "double with 13 flat, I've got too much to pass" people to stop (because it's bad, especially for newer players) and perhaps to start listening to Marty Bergen instead: "Points Schmoints. Shape rules."

It's interesting that he presents two flawed hands and asks "which one should you double with?" back to back. As I said on the last one, double isn't a good bid - apart from all the discussed problems, you're likely getting a spade lead from partner if opener isn't declarer, and that's really not good; and you'll be tempted to not lead a spade into an auction like 1H-X-2H-2S; 3H-p-p-p (well, after the obvious high club is discouraged. Of course, partner will want a spade lead holding several hands he doesn't actually want a spade lead on, assuming you're Qxx or xxxx). But it's better than this hand. At least, that's Stewart's argument.

Now, let's split it down the middle and give you AT84 Q5 AK842 73...(To give you an idea how nasty this is, it's basically why ELC was invented. Other pairs sacrifice the natural 1NT overcall for this hand).

* ObDisclaimer: some of the rules on that page are not intended to be taken completely seriously. But not all. And not this one.

  • jeff-goldsmith.org seems to be locked?
    – jrw32982
    Apr 22, 2022 at 19:16
  • Fixed - thanks. Unfortunately, the owner of the site has met The Gentleman at the End.
    – Mycroft
    Apr 24, 2022 at 23:55

This is a matter of partnership agreement. Most takeout double advice is to require three in each unbid suit. The issue with this double is what partner will do with a hand like

Spades 84
Hearts T932
Diamonds JT8
Clubs Q932

You're going to end up in 2 clubs with 17 points between you and a 6 card fit. Some partnerships would prefer to take that chance, to avoid missing either better part scores (particularly over 1H, when you might play 1S) or even better results. I've definitely seen people play both ways, but to be honest I wouldn't suggest someone at your level play it.

Think of it this way. Your concern seems to be mostly that you'd prefer to favor the majors; but in the hand where you have 3-2-5-3, 1S is available, and if you do end up in a short fit, at least you only need 7 tricks! On the other hand, in this case, the only thing "majors" really gives you is the possibility of 110 if you can make 2 - while 2C in a 6 card fit is the bigger risk, and requires an 8th trick.

  • Under the ground rules I have laid out, I would bid two diamonds as "advancer," because JT8 of diamonds plays better opposite Ak62 than Q932 of clubs opposite 84. Change the doulber's 2 of diamonds to a low club, and you're technically within the three card rule, but that is not a much better holding than 84 opposite Q932. I would pass as doubler if you switched my diamonds and clubs so I have weak diamonds and strong clubs.
    – Tom Au
    Jan 24, 2022 at 22:11
  • Sure, you could make those agreements. I wouldn't, though, as there's nothing special about diamonds over clubs; they're both minors, and in the all-important case of competitive auctions in the 1-2 level where there's 15-16 tricks for both sides available in total (see Alexander Woo's example), you don't really want to cut yourself out of half of those auctions just for one specific hand layout (4-3-4-2).
    – Joe
    Jan 24, 2022 at 22:16
  • Even if advancer has 2-3-4-4, you'd want them to bid diamonds over clubs because... why? There's plenty of situations where clubs are disfavored, but it's because of structural reasons (mostly, that 1c is a catchall bid for opening hands, and various systems that make club fits harder to find because the club bids are used for conventional responses). In the particular exchange here, clubs and diamonds are equivalent - there's zero reason to construct a system around "try to play in diamonds over clubs".
    – Joe
    Jan 24, 2022 at 22:22
  • The other issue is that with opener having five hearts and the doubler three, the chances of the advancer having four or five hearts is something like 16%. And cutting yourself out of half of that is 8%. I say "spades and diamonds" arbitrarily, another partnership could say spades ad clubs. The other 84% of the time, the advancer will have at most three hearts, meaning either a third spade, a fifth club, or a fourth diamond to break the logjam.
    – Tom Au
    Jan 24, 2022 at 23:51

"Stronger" vs "weaker" is a red herring. You're not doubling with the other hand and passing with this one because the other hand is stronger, you're doing it because, given the information from the auction so far, the other hand is more offensive and this one is more defensive.

Take KQxx xxx KJxx AK as an example. It is clearly stronger than either of the hands mentioned, but I would argue that you should pass over RHO's 1H opening with this hand (over any other suit opened at the one or two level, I would bid NT as cheaply as possible).

Under standard agreements (at least in the US), doubling an opening bid is an advertisement to partner that (a) we have some strength and (b) based on the shape of our hand, we would rather declare than defend. If the hand doesn't have the strength to double and rebid, we promise 3-card support for all unbid suits so that partner can confidently bid their best suit at a level that indicates their strength.

Thus, neither the hand in your question nor the hand mentioned in this answer qualify for a takeout double of 1H. In both cases, you lack 3-card support for clubs. Because of that, you have a lower chance of improving the contract for your side. For example, you may get to a bad club contract, or scare your opponents away from bidding too much. You may even get to a bad contract in spades or diamonds when your partner also has three hearts and LHO scores heart ruffs.

At the same time, you are giving information away to the opponents. If you do end up declaring, by doubling you have given away the location of the bulk of your side's high cards. Your opponents may end up making a contract by playing you for high cards when the normal line of play when you've been silent would cause them to go down. While this is also a risk when you have a normal takeout double, in that case the upsides outweigh the downsides. By bringing more downsides into the picture with your unusual shape, you have changed the calculus.

While there are agreements that would allow you to take out with this shape (others have mentioned Equal Level Correction), they are not without cost. They are more complicated and make it more difficult to bid when you have a strong hand.


The gist of your question. was, if Frank Stewart advocates doubling with this hand that has a weak three card spade suit (and Kx in the opponents' hearts), why isn't the above hand also worth a double with much stronger three card side suits?

But that's not the point. You were close to the mark in a comment when you mentioned that "this hand has three cards in the opponents suit." Such a hand should not be doubled because "three cards in the opponents' suit" plays almost a trick better on defense (but not offense) than one with only two cards in the opponents' suit.

Put another way, you should expect to be doubled and go down at least two if partner is forced to bid with "nothing." That's -300 non vulnerable or -500 vulnerable. Let's say that vulnerabilities are equal, and the opponents' game scores are 420 and 620 respectively. A hand with two in the opponents' suit will often yield them game opposite a (near) Yarborough, while a hand with three in the opponents' suit will usually defeat it. Put another way, the (high) risk is considered by most to be worth it in the first case and not the second.

A hand with three in the opponents' suit would have to be worth 15 points or more to be strong enough for a double. If that includes a stopper in the opponents' suit, the proper call is 1NT. If it doesn't, you're still "behind the eight ball" with 15 and should wait for an even stronger hand to act.

You must log in to answer this question.

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