spaderaise

I'd like to play a "human-declares" game with 3 robots, using deals I upload myself. (That is, when my robot-partner becomes declarer, we swap places and I play the hand -- as in the robot tournaments.) Is it possible?

Hmm - thanks for the idea - when I try this for the only player I happen to know to have played in them (sliynk), I get an empty list and "Some tourneys not shown for security reasons".

Is it possible to get hold of hand records from old tournaments? Specifically, I'm think of past weekly free instant tournaments. Ideally, I'd like to find a way of playing these against the bots. I enjoy doing this before watching videos of people playing the tournaments (eg Peter Hollands' excellent series). Any suggestions very welcome!

I also have Bridge Baron, on my iphone, and I haven't seen the phrase "Half double stopper" in that version, but it does use the phrase "Half or double stopper", which is still fairly odd. But then its explanations of bids are often risibly bad. It seems to have a very limited language in which to explain meanings. For example, a Michaels overcall (1H) 2H will be explained merely as "5+ spades"; apparently the "5+ diamonds or 5+ clubs" information is simply too complicated to convey.

If my calculations are right, then jefford's numbers correspond to (as a gain for the side bidding 1N): non-vul: .059*3 + .17*5 + .393*5 - .378*6 = 0.724 IMPs / board vul: .059*5 + .17*7 + .393*6 - .378*10 = 0.063 IMPs / board i.e. rather better in 1N than 3N when non-vul, and very close when vul.

OK, I'll have a go. South is not in possession of unauthorised information (let's assume - the answer might be different if for example North had reacted visibly when South didn't alert 2C). Since North appears to show clubs and spades with longer clubs, some people holding the south cards might consider removing 3S or 3SX to 4C, but failure to do so is hardly enough to "provide evidence of an unauthorised understanding", so I wouldn't rule fielded misbid. North, presumably, is in possession of unauthorised information, because he has seen the lack of alert of 2C. Let's work on the basis that he believed that 2C was showing 5-5 in the majors. (If he believed something different, this could change the answer). His logical alternatives over 3D, believing that he's shown 5-5 in the majors, seem to be Pass and 3S. Is one of them demonstrably suggested by the UI? If anything, I think Pass is suggested over 3S by the UI. Bidding 3S will (in fact) show clubs and spades with longer clubs, and from the north cards it looks very likely that partner will have more clubs than spades, and may correct back to clubs. This looks highly dangerous for NS, while a pass is reasonably safe. So the UI doesn't suggest 3S over Pass at North's second call. I don't see that he has any logical alternative to Pass at his third call. So North's actions look OK. So I'll rule no adjustment.

Diamonds are not 0-7 because both opponents played a diamond to the first trick.

I think you missed that West led a diamond - so he definitely has more than zero. Ruffing the small diamond after the first or second round of trumps is safe.

(birthday problem) Here you can find it on a Cambridge University maths exam paper from 2009: http://www.statslab.cam.ac.uk/~rrw1/prob/ProbabilityTriposQuestions.pdf And here you can find it on the first problem sheet for the current first-year probability course for maths undergraduates in Oxford: http://www.maths.ox.ac.uk/courses/course/19623/material You'll find it in university courses and textbooks the world over. And why not?! It's an important and surprising example and plenty of undergraduates won't have seen it before. And even for who have already seen it, there's plenty of depth to the question if you want to generalise it. The fact that a bright 13-year old might understand it doesn't mean that it's not worth a bright 19-year old thinking about it too.

It sounds like you (and aguahombre) disagree with his logic, but what's the problem? Is it particularly worse to put strong 5-5 hands through 1C-..-2D rather than through 1D-..-3C? Of course, the 1C-..-2D sequence becomes less well defined, but in compensation the 1D-..-3C sequence becomes better defined. Arguably this is better since it keeps the auction lower on more hands, and increases the definition of the auction that has reached the 3-level at the cost of reducing the definition of the one that has reached only the 2-level. Am I missing something obvious?

I am looking at Blackshoe's diagram, and still hearts are not South's longest suit. Also, there is no indication that South was in receipt of UI before bidding 4H. I guess that iviehoff meant that North's 5C bid was the "unauthorised panic".

Indeed. Of course. I am not arguing that the two need to be interpreted the same way. The bridge regulation is ambiguous - there are two sensible ways to interpret it. I already said this above, and many others had already said it. I posted in response to David, who said that he was unable even to imagine how somebody could interpret it as applying to a double of a no-trump bid. I really didn't expect it to be contentious!

A double is alertable if it asks for a suit other than the suit doubled. A gift is taxable if it is made to a person other than the spouse of the giver. <Statement A> holds if <C, specific object of type B> is an <object of type B> "other than" <description D> where "description D" specifies precisely 1 or 0 (according to circumstance) objects of type B. <Statement A>: Alert is required <type B>: suit <C>: suit requested <description D>: "the suit doubled" <Statement A>: Tax is payable <type B>: person <C>: person to whom gift is made <description D>: "the spouse of the giver" The question is: does the regulation apply if the number of objects described by D is zero rather than one? Campboy's reply to you is on the money. Asking "what if the gift is made to a dog" is asking "what if C is not of type B", hence "what if the thing requested is not a suit". OK?

Probably more posts devoted to this by now than it deserves! - but for what it's worth, the analogy is between the phrase "person other than <description of non-existent person>" and the phrase "suit other than <description of non-existent suit>".

Well, thanks, but you're reading more into it than intended. All it was serving to demonstrate is that a regulation that says "other than the spouse of the giver" can apply perfectly well when there is no spouse of the giver. In the same way, a regulation saying "other than the suit doubled" could perfectly well apply when there is no suit doubled. I don't think that one interpretation is right, and the other wrong, in the case of the bridge regulation. It's ambiguous, as you quite rightly say. I was just mystified that David was (by his own account) unable even to imagine the other reading.

Try this comparison: a tax law that says "A gift is taxable if it is made to a person other than the spouse of the giver". Would you say "I cannot imagine how anyone thinks this applies in any way to a gift made by somebody unmarried"? No, it obviously does apply also to unmarried people. If the giver has no spouse, then any person is "a person other than the spouse of the giver". In a similar way, a natural reading of the sentence above is that if there is no suit doubled, then any suit is "a suit other than the suit doubled".

If you enjoyed this, you should definitely get hold of the book written last year by the same author: Bridge at the Enigma Club. Highly recommended! By the way, the article above is not representative of Peter Winkler's style of writing - it is an English translation of a Dutch translation of his original article....

In the example I gave, partner knows my exact heart holding, and so knows whether the H6 is high or low. Declarer does not know my holding and does not know whether the H6 is high or low. The meaning of this signal depends completely on the information of my heart holding, which is known to partner and not to declarer. In what way does this not fall under the definition of "encrypted"? OK then, so are methods 2(a) and 2(b) that I described in my post encrypted or not? (In each case the entire agreement is disclosed to declarer, just as written).

Making this into a meaningful definition is pretty tough. The grey areas are enormous. Two examples: (1) Spades are trumps. Partner leads a heart and declarer ruffs the first round. (Now partner knows my heart holding but declarer does not). Declarer plays trumps and at my first discard I throw the H7. By agreement, this encourages clubs if the H7 is my smallest heart, and encourages diamonds if H7 is a "high" heart. Is this an encrypted meaning? Partner can read it but declarer cannot - it certainly fits the definition of "determined by information which is available to the defenders but not to declarer". But of course it's a very widely played method. (2) Spades are trumps (maybe declarer's exact spade length is known, eg through a Stayman sequence). Partner leads the A and then K of clubs, and I follow twice. Which of the following agreements about my carding are encrypted?: (a) high-low shows an even number of black cards, low-high shows and odd number of black cards; (b) if I hold an even number of trumps, then high-low shows an even number of clubs and low-high shows an odd number of clubs; if I hold an odd number of trumps, then the reverse. Method (a) sounds pretty straightforward and unencrypted, while method (b) is generally regarded as encrypted. But of course, the two methods are precisely identical. (By the way, it's very standard for a signal to show something about two different suits at once. For example, a low heart both discourages hearts and suggests clubs over diamonds. Or a trump echo shows at least 3 trumps and a shortage in a side-suit. So if there is something wrong with method (a), it should be something other than combining information about two suits at once). A signalling method is (roughly) an agreement about what card to play from any given hand. Typically partner has some partial information about my hand, and declarer also has partial (but different) information about my hand. Furthermore, I don't even know what partial information each of them does have about my hand. Also, some "information" is 100% provable from the play of the cards, whereas other information is not certain but may for example be implied by the bidding, or result from an inference from the play that tends to suggest one layout over another. In the light of all this a watertight notion of whether a signalling method involves "information available to partner but not to declarer" is likely to be elusive. (And of course, there is general agreement that designing signals so that they are useful to partner but not to declarer is an absolutely reasonable aim....)

Natural opening 1NT bids are overwhelmingly played with a three-point range, across a wide variety of systems. I'm wondering about two related things: (1) in the context of a three-point range, how beneficial are invitational sequences? (2) how playable are larger ranges? There's a lot of downside to invitational sequences like a traditional 1NT-2NT. 2NT is an unsatisfactory final contract - often too high or too low. In addition a sequence like 1NT-2NT-3NT gives more help to the opposition than 1NT-3NT; defenders know more about opener's count, and also opening leader knows in advance that the contract might be tight and can use this in deciding between an active and a passive lead. Would you be willing to dispense with invitations over 1NT entirely, leaving responder to choose between pass and 3NT rather than inviting? If you dispensed with invitational sequences, you could find other useful meanings for the freed-up bids - have such systems of responses been developed? If you do play invitational sequences, how about extending opener's range, for example to a four-point range? For example in the context of a first-seat weak no-trump, some of the value comes from preemption, and this would be increased by moving from 12-14 to 11-14, say. Of course, with this comes a loss of accuracy. Maybe the most important effects are in competitive auctions rather than uncontested ones; responder is less well placed to know what to do if second hand interferes. (btw I'm interested in constructive methods here; I know that some people have played, for example, 1NT as 8-14 in third seat when opener's pass is limited to 10 points or less, but that's a different kettle of fish.)

South discards hearts on the fourth and fifth diamonds, and overruffs the fourth heart?