kfgauss

We appear to be South and thus white on red, actually. I'll bid just 5♣, trying to help partner decide whether to sacrifice (my 6-5 shape is something extra at these colors, of course). I realize this doesn't get rid of things like the queen ask and actually gives them extra room, but partner can come in with 5♠ (or more) over the response to make things a bit harder, and I'm not willing to commit to 6♣ which really commits us to 6♠ and I'd like to introduce my minor to help partner out, so 5♠ is out. I guess this may help the opponents in various sorts of decisions, but the decision of whether to sacrifice with 6♠ over 6♥ seems to be the most important one. Andy

Oh -- sorry, totally missed that I didn't give that trick. Partner followed with ♣8 and we play udca. Andy

You're West and lead ♥4 against 1NT at Matchpoints (South opened 1N and was allowed to play it): [hv=d=n&v=n&n=s962hq763d864cq92&w=saj5hk8542dq7ckj7]266|200|Scoring: MP[/hv] Your ♥4 goes to the 3, 10, A. Declarer returns ♣3 and you hop ♣K and return the ♥5 (if you wouldn't falsecard, fine) to the 6, J, 9. Partner returns ♦10 to declarer's J, your Q, and dummy's 4. What now? Also, if you would defend differently earlier, feel free to comment on that. Andy

Forgive my novice question but with ♠ 2-3-4 and ♥ J 10 9 what is the reply to cappelletti? Over a Cappelletti 2♦ you bid 2♥, but over a (Landy) bid of 2♣ for the majors, you'd bid 2♦ asking partner to bid his better/longer major so that you're guaranteed a 5-3 fit (ie in case partner is 5-4). Andy

This comparison to forcing pass systems raises an interesting point: if the "fert" 1♠ bid over a strong 1♣ means that now your pass shows values, is it no longer "purely destructive"? This is surely technically true (it now has a purpose other than destruction). To have this be valid you'd really have to make sure you only passed with good hands, of course. Andy

The question "is it ok if I claim on a squeeze" seems to be a reasonable one, if not entirely supported by the laws (which I don't know in this case). If your opponents refuse, you don't seem to gain any information other than something like the fact that they don't want to have to think through a squeeze and check that it works. It seems potentially irritating to the opponents if you claim and they don't want to think about it and you have to get the director to come over and think through the squeeze, so asking them first seems like a kindness. BBO is of course different from live bridge in that you can claim and show all your cards to the opponents while not seeing anything yourself, so perhaps on BBO when asking such questions you should always show the opps all the cards. Andy

I think 25% is an overstatement, the combined chances of ♦KQ being together and/or an endplay should be < than 50% (I'll gladly leave to the math junkies the task figure out the correct number). As Roland pointed out, having the ♦9 (or even the 8, in some cases) would increase a lot the chances of success of an elimination and throw in. This is sort of irrelevant to the main point of this thread, but as I claimed this number, here's why: ♦KQ together is ~50% (it's a bit less because once you have one honor, you have a bit less space for the other, but not by much). If this is the case and there's no diamond lead from nothing (surely the chance of a heart lead from the King, which is actually fairly likely when it's offside, will pick up at least whatever I lose due to a diamond lead from nothing). Whenever ♦KQ are together and you find the ♥K onside (and trump are not 4-0), there's an endplay: draw trump, eliminate ♣ and ♥ and lead a diamond to the 10. Andy

For example, the modified queen ask, for when partner's a better reader of body language than you are: Step 1: I don't have the Queen and don't know where it is Step 2: I don't have the Queen, but my LHO does Step 3: I don't have the Queen, but my RHO does Step 4: Yes, I have the Queen Extra modifications can be made if you have more room: (this version usually for when you have all the keycards, so you know you don't lack the A or K of trump) Step 1: I don't have the Queen and don't know where it is, or I know where it is, but it's behind the AK and not doubleton Step 2: I don't have the Queen, but my LHO does, but I don't have the J Step 3: I don't have the Queen, but my RHO does, but I don't have the J or 10 Step 4: LHO has the Queen and I have the J Step 5: RHO has the Queen and I have the 10 Step 6: LHO has the Queen and I have J10 Step 7: RHO has the Queen and I have the J Step 8: The Queen will fall doubleton Step 9: Yes, I have the Queen Andy

Is this worth not being able to retransfer sometimes? That is, one can play that the next three bids above accepting the transfer show concentration (or weak doubleton, your choice), with 2N as a stand in for M-1 (ie the suit below your major), and never bid 3M-1. (Superaccepts that don't want to make one of these bids bid 3M). I find these sorts of choices very hard to quantify and make. Does anyone have opinions on whether being able to show both is worth giving up the retransfer sometimes? A side note/question on retransfering: whenever one can retransfer, I like to play that bidding 3M is an invite. This should be used sparingly, as we don't like responder to become declarer here, but do people have opinions as to how often we should be using this invite? (e.g. "never" or "only when we have several tenaces ourselves" or "whenever we have an invite") Andy

First of all, 6S isn't ridiculous. The contract as it is is 25% (♥K onside plus ♦KQ together -- there's an endplay), and we've all been in worse slams. It's unfortunate that partner's doubleton is the same as your own. If we swap ♥ and ♣ in partner's hand, then it's basically just on the ♥ finesse and so is 50%. If we're simply going to be counting points (which is ok at a beginning stage -- though beware that counting points isn't always best), then we should notice that 32-34 isn't always 33+, which is the amount we should have for 6S. (Side note: we should count distribution points too because we have a fit, adding 1 point for the heart doubleton, but then a correction is to subtract that 1 point because we should only count points for either shortness or honors in a given suit, so the Jx is only 1 point, so we only have 12, not 13 -- so I get 31-33, but not including the correction is ok too and doesn't really matter here.) This suggests that we should invite slam in spades. A 5♠ bid seems like a nice simple way to do that. Then partner with his minimum (19) can pass. (I agree with Roland's comments on your initial 2D bid [if we've decided this hand is only worth 12 points, a limit raise in spades describes it nicely], and also that 3NT typically shows 18-19 because a 1NT opening is 15-17 and a 2NT opening is 20-21. Notrump ranges are an exception to the "min = 13-15, med = 16-18, max = 19-21" framework unless you play that 1NT = 16-18, which isn't SAYC and I don't think is what Fred is teaching.) Andy [edit: cleared up discussion of distribution points above]

I would just leave the suit length blank when no specific info has been given regarding it. Perhaps others disagree, but this seems a reasonable method. For example, when you transfer, nobody needs to see that your 2D bid caps the number of diamonds you hold at 8 (if only 5+ hearts is guaranteed) or maybe 6 (if hearts > diamonds is guaranteed). Obvious negative inferences (ie if I have 5 hearts I can't have 9 diamonds) aren't something you're required to tell your opponents. I also would ignore the possible outcomes. Perhaps if there's some specific situation where it's particularly useful to know you could fill it out then, but then it will be noticed as it's different from the default setting. Andy

This seems to me to be a bad idea. I fear it will encourage silly results in the last round by pairs that are "in contention" for the low prize. An alternative: try charging double and then giving everyone half back if they finish the tourney. Maybe this is bad business as it'll look like your tournament charges twice as much (I understand you're not necessarily in it for the business, but this could be a suggestion to other TD's too), but one hopes that people will understand what's going on and that this won't harm business too much and will lead to more people finishing. I don't know how big an issue this is, though, as I don't really play tournaments very often. Andy [edited for typo]

Statistics really isn't part of pure math programs these days. It would usually only be required coursework if you're getting a degree in "applied math" (or a math degree with an "applied" focus) or if the department doesn't make any distinction between pure and applied math. This means that most of the students graduating (in pure math) from the top math schools have usually never taken a statistics course (unless they took one in high school, which is reasonably common but would be very basic). That said, most of them will understand Bayes Theorem anyways (informally, perhaps). Feel free to feel about this as you will. Andy

There's a great article on mixed strategies in cardplay by John Swanson at www.johnninaswanson.com/articles.html called "A Hand for Deep Blue." He analyses the position Axx opposite KJ9xx and determines that a mixed strategy is required for the defenders regarding when they should falsecard the 10. It's a short article, so I won't summarize -- just go read it. The position has a nice history too (see the notes below the article), with a writeup by Edgar Kaplan in the Bridge World after it came up in the Vanderbilt and led to an accusation of cheating (Kaplan's claim was that the accused made the right play, actually). Andy

For item (3), that is 1m-1M;2M, a simulation should probably just look at balanced and semi-balanced hands with a doubleton (ie not 4333) -- or at least, include this as an item as well. Also, it should differentiate between hands where the doubleton has an honor and hands where it doesn't (and perhaps when it does, what the honor is and maybe exclude hands with two honors). My current style is to raise with 3 cards on balanced hands whenever the doubleton is xx, and sometimes when it has an honor and the hand looks suitish. Andy

Gib always seems to pull its counterpart's penalty doubles when I'm bidding against them. Not sure whether this is an internal misunderstanding or just bad judgement, though. :) Andy

Flame asked for suggestions about inv 4M 5+D hands and I gave what I think are some useful structures over 1C-1D;1N to help opener continue to bypass 1M and bid 1N when balanced. Clearly either way is fine. Perhaps to you "Walsh" means always bypassing unless GF and maybe that's the original/"official" structure, but clearly Flame had some doubts about the invitational hands and wasn't asking a textbook question about Walsh. An essential point of Walsh -- being able to rebid 1N after 1C-1D when balanced -- is preserved in my suggestion. Sorry if you seem to disagree with my ideas (ie about Drury too). I really mean no harm and am just getting ideas out there. Andy

I think invitational hands do a bit better bidding 1D if you can handle it over 1NT (which partner will bid even with 4 in a major, of course), but certainly either way is reasonable. Bidding 1D helps in slam exploration and also in finding the right partscore (and helps a bit in evaluating your hand for game) and in competitive bidding when your fit is actually in diamonds. Not bidding 1D helps prevent the auction from going 1C P 1D 2S; P 3S ?, but perhaps partner can double 2S at least a fair amount of the time when he'd accept an invite, or perhaps you can double 3S now. In any case, make sure you play 1C P 1D 1S; X = 4 hearts (any hand). Here's a 2-way checkback structure over 1C 1D 1N structure I've played that's designed to deal with the invitational hands as well as the GF hands: (taken from my system notes with the side comments and all) 2C forces 2D (signoff or inv), 2D forces 2H (GF) 2C 2D = signoff 3C = signoff 2M = inv 4M 5+D no 3C (to parallel the direct 2NT) 2C 2D 2M = inv 4M 5+D 3C ("slow shows doubt of strain") 2D 2H 2S/2N = GF 5+D 4H/4S respectively (try to rightside 4M?) 2NT = inv balanced, no 4C no 4M 2C 2D 2NT = inv 5+D 4+C no 4M, more NT oriented 2C 2D 3C = inv 5+D 4+C no 4M, more suit oriented 2D 2H 3C = GF 5+D 4+C 2C 2D 3D = inv 6+D no 4M 3D or 2D 2H 3D = both GF strong suit, direct = less doubt 3M = GF at least 5M 6D 2D 2H 3M = splinter for clubs Andy PS It seems I didn't define 2C 2D 3M, which I guess is an autosplinter, or the 2 extra ways to bid 3NT. Presumably 2C 2D 3NT and 2D 2H 3NT are a bit slammish -- probably the former has 5 diamonds and the latter doesn't. Also I guess we never bid 1D with 4M-4D GF -- minor tweaks could rectify this if you want to bid 1D with such.

I like to be able to shade in 3rd seat (and to open those 5332 10 counts with spades in 4th seat) and think it's a loss not to. I agree there's a loss to the natural 2C bid, but I don't think it's great. I bid 1NT or 3C on these hands, depending on how nice my club suit is. (Aside: with some partners I play 3m as a fit jump here and 2NT as the club hand.) Here are some comments on your other points: (1) & (2) Feel free to play 2D showing 4 trump, or play some 3-level bid showing 4 trump if you wish and only bid 2C with 3 trump. (3) Drury has so much room and one can play all sorts of game tries over it. An example: 1M-2C; 2D = unsure about game 2M = no game 4M = yes game other = slam try Then, over 2D, bids by responder: 2M = 8+-10 other = game try (use whatever method you like) with a true limit raise (4M = if there could be game, there is) Then, over 1M-2C;2D-2M, use your normal game tries again. (4) A point I'll add here is that I've considered scrapping 2C as drury and using 2D instead, at least over 1S (then 1S-2D;2H is like the old 1S-2C;2D, etc). There's a weak two in diamonds, so you're much less likely to want to bid 2D than 2C naturally. Also, if you're playing 2/1, you've already given up these bids in 1st/2nd (for the purposes of the non-GF hands we're talking about) -- not really much of a point, but thought I'd bring it up. Andy

The system is already changed -- we have Drury. Drury is a wonderful gadget with an enormous amount of room. Great 8's and most 9's should be bidding it in my opinion. Also note that 1NT by a passed hand should only be semi-forcing, and so you should take the "bad raises" out of it and just raise to 2M with those. This is perhaps a bit more extreme than standard, but I think it's fairly standard for at least decent 9's and basically all 10's to be bidding drury. [Edit: Totally missed your point that you don't like Drury. It's perhaps ok to get to the 3 level with those 9 card fits (I assume you don't open on 4 in 3rd/4th very much), but what about the 3-card limit raises? It'd be quite a travesty to end up at the 3-level with a fairly balanced pair of hands with 10 points each.] Andy

For game try purposes, yes I'd prefer clubs to diamonds (though I'd make a short suit try if I had one since help in diamonds is fairly good too). For the purpose of helping partner compete, I don't feel I should tell partner about my clubs -- my secondary honor is in diamonds and my primary honor is in clubs so this wouldn't be a very pure double fit. Andy

I passed on this one. Partner didn't use drury -- passed hand raises to 2M are weaker than ones by unpassed hands. Fearing them competing in spades is an interesting point. I don't think we play preemptive re-raises, though, and I'm not willing to make a game try to shut them out (though it will work quite well at doing that when pard rejects). When they compete to 2S, we could bid 3C over that to do what you suggest, Richard, but I don't think I want to emphasize clubs over diamonds on this hand. I'll just bid 3H when the time comes. Andy

Your relay background is showing through. It's a bit more holistic than you make out. It suggests a group of hands (things like balanced 10 counts -- where if they don't have a club honor, they'll need a bit more strength and/or better stoppers) and basically asks the question: opposite this group of hands, should you be in 3NT (or should you pass or correct to 3C). This is somewhat a combination of 1) and 3). Clearly it's still a question, but it's not a simple question (an example of its complexity is that different partners will bid 2NT with slightly different groups of hands and so you'll have to evaluate slightly differently opposite each group). There's somewhat of a standard for the group of hands 2NT shows, though, (one would hope this is based on some impression of what the optimal group to do this with is) and so we can come up with reasonable answers to this bidding problem. Examples of sorts of questions we need to ask ourselves are: if the clubs run, am I going to have enough side suit tricks quickly enough to make 3NT? if the clubs don't run, will I have enough extra strength to pull through? Andy PS Not that it's necessarily relevant to my comments above, but I bid 3C, but that was probably too pessimistic upon reflection, especially vul at imps. I have the impression that it's right to bid 2NT here with only 8 or so points with the club King, and a reasonable 10 without it, say xxx Kxxx Qxxx Kx or Kxx Kxxx Axxx xx. Even these hands have some play for 3NT though -- especially if the hearts have some spot cards.

Sorry for my hasty analysis. My guess about the size of the effect I mention (the "overcounting" effect) in the real case (ie 52 cards) is quite wrong. It's actually reasonably large. Say we have 4 clubs and righty tells us he has 4 diamonds exactly (be aware that this "exactly" clause increases the effect: go back and look at the 8 card example and you'll see that the effect is basically doubled if righty had told us "I have exactly one diamond"). Let's compare the chances he that he has 2 clubs or 4 clubs (I'm just simplifying the conditions by only allowing these two possibilities). If we have 4 clubs and 4 diamonds, he's odds on to have 2 clubs by odds of 80:49 (that is "2 clubs":"4 clubs"). If we have 4 clubs and 2 diamonds, he's odds on to have 4 clubs by odds of 49:44 (that's "4 clubs":"2 clubs"). So, giving us more diamonds makes it likely that RHO has fewer clubs by a fairly large amount it seems (a more complete analysis would be better, but I'm apparently lazy). That in turn implies that more diamonds in our hand means pard is more likely to have support, and also that LHO is more likely to have a stack (just not for Gerben's stated reason). Andy [Edit: One should notice that restricting things to the case where RHO has exactly 4 diamonds is really making it so more diamonds in your hand means more room in LHO & pard's hands, which isn't really true, and this is inflating the effect. In the toy model in my last post, this effect and the real effect have exactly the same -- hence the comment above about doubling. I should really check to see how they compare in the real case, as they needn't be the same size there too.]