semeai

Nice one. Slightly risky. I guess a related falsecard then is that with some of Jx, Tx, Jxx, or Txx in the same spot you drop the J/T (sort of like dropping the J in your first example).

See my comment. I guess it does have to be forcing to be GCC legal, which Mr. C may not like. On the midchart, it doesn't even have to be forcing: "5. Any strong (15+ HCP) opening bid."

1. Yes, it's legal. See item 2 under Opening Bids on the GCC: "FORCING 1NT OPENING BID (15+ HCPs) indicating a strong hand, balanced or unbalanced" It does need to be alerted, not announced. The explanation should make sure to inform the opponents it may be any distribution. Announcing "15-17" would be misinformation. No need to pre-alert I think, but it wouldn't hurt. No comment on 2.

It depends who you're playing. Anything positive means you've done better than your opponents (at least compared to the field of other BBOers playing the hands). I have the sense somehow that 1 imp/board (in the very long run) corresponds roughly to being one "skill level" (int/adv/exp) above your opponents.

Diamonds were sinking fast, but now they're making a comeback!

Let's say the range for the 2♠ bid is A points (low end) to B points (high end). Then the probability I'm bidding 3♥ (general game try) is: Max(0,Min(1,1-.5(10-B)-.1(7-A))) *** More seriously, this looks like a game try to me. I hope my matchpoint sensors aren't improperly aligned. Added: There is a slight advantage of the game try, relative to our intuition. We aren't in 2♠, so there's no worry about losing the safety of the 2 level in exchange for a game try, and I'm fine competing here at all white.

To have a definite example, let's say we have 26 HCP and are missing A K K ♥Q J J, with the ♥Q being the card of interest. Let's say East opened and has 12+ points, i.e. is missing ♥Q or J or JJ or nothing. Then there are 26-10 = 16 remaining non-heart non-high-cards. Finesse: Qxx-xx (6 ways) times ♥Q-AKKJJ (1 way) times 10-6 remaining cards (16 choose 6 ways) Qxxx-x (4) times ♥Q-AKKJJ (1) times 9-7 remaining cards (16 choose 7) Drop: xxx-Qx (4) times -AKK♥QJJ (1) times 10-6 remaining cards (16 choose 6) xxx-Qx (4) times J-AKK♥QJ (2) times 9-7 remaining cards (16 choose 7) xxx-Qx (4) times JJ-AKK♥Q (1) times 8-8 remaining cards (16 choose 8) We have 16 choose 6 < 16 choose 7 < 16 choose 8 (even split is most likely) so without even adding anything up we can see the drop is a winner. If you consider AKKQ♥Q as the missing high cards instead, though, you find that the finesse is still better.

An interesting read. I enjoy his articles. This is a pretty good rule but not completely general. It works when the total length in the unknown suit held by the opponents is 5 or 6, but fails in a way for boundary cases for total lengths 4 or 7, as well as for some more extreme lengths. To be more precise with the exceptions: If the unknown suit would be 2-2 when the 4-1 break occurs, then he says play for the drop. In fact, if you can play one top honor before committing to the finesse, it's slightly favored. [More extreme example: If the unknown suit would be 1-1 when the 4-1 occurs, then it's always correct to play for the finesse, even if you must commit on the first round.] If the unknown suit would be 4-3 when the 4-1 break occurs, then he says finesse. This is correct if you can play one top honor before committing, but if you have to decide on round one you should play for the drop. [More extreme example: If the unknown suit would be 5-4 when the 4-1 occurs, then it's always correct to play for the drop, at least just doing the math (some of the cases for the drop involve 7 or 8 card suits).]

Seems like Heron's just pointing out that a slightly longer time limit may be good for the human declares games. It will still be more skill based with another 5 minutes on the clock. A 10 minute time limit would require a lot of skill, but it wouldn't exactly be pleasant.

This quote is a bit unfortunate. In Bill's defense, he did acknowledge the problem two paragraphs prior in the parenthetical portion of: Also, this clarifies a complaint of Justin's: It's pretty standard for books on bidding systems to make some somewhat overblown throwaway claims about the superiority of the system. I don't think we should fault Bill that his book has some fun numbers to look at (in a short chapter, fairly deep in) and even mentions the problem of bidding judgement (though not other issues brought up here) before concluding a chapter with a throwaway overblown claim.

The argument for empty spaces is cute actually: For definiteness, say one hand has 8 empty spaces and one has 7. Take the 15 cards and shuffle them and deal them out, 8 in a row for one player and 7 in a row for the other. It's equally likely the Queen is in any of the 15 slots, so the ratio is 8:7. [The trick is that it's really easy to see if you use the order you deal the cards in. If you use combinations ("a choose b") then the computation is a bit messy and has a bunch of cancellations but comes out the same, of course.]

I didn't mean to suggest a play; I just don't like large numbers when small ones are at hand. I agree that whatever we think of the likelihood of LHO switching at trick 3 with the two holdings is more important than the prior probabilities. Added: Also, the above vacant spaces (ie including spades) aren't quite right because of the problem that one tends to lead a long suit.

AKA an 8:7 ratio in favor of QT when spades 4-4 and a 9:6 ratio in favor when spades 5-3 (ratio of number of vacant spaces for the Queen).

Yes, it's sort of a strange statistic with double dummy being used. Better might be to use the GIBson trick as a definition of "cold": Analyse hands (similar to DD analysis) pretending that the defenders get to construct their hands as they go; ie they have a pool of 26 cards and can take any card to play at any time, except that they can't play a suit they've previously shown out of. [Might want to forbid ruffs on opening lead as well or the statistic will be pretty meaningless.] Setting this up would take a decent amount of work I suppose as you'd essentially be writing your own (modified) DD solver as part of it.

Excellent point. I missed that the spade encouragement happened before the diamond pitch (they were in the same sentence and partially on the same trick --- they must've happened simultaneously and independently!). Given that I think it's basically 50-50 without the info from East's pause, this should make the drop better.

Your collaboration distance (as partner or teammate) from Meckwell might be interesting too. Meckwell to Hamman to Lall gives us an in on the forums.

Do you (or ulven) have data on this? From the Fantunes Revealed thread and book, it seems as though their 2♣ opener does well, even though it's similar to what you describe. Maybe comparing "all precision 2♣ openers by pairs that allow 5 clubs - 4 other" and "Fantunes 2♣" we can get deduce something about the contribution of Fantunes bidding judgement, such a commodity in the other thread. (This would involve subtracting off the contribution of their card play, which we at least have a suggested number for in that thread.)

Indeed, though I suggest these are small corrections. Playing your cards up the line really is quite natural. More significant: I also did not take into account the fact that East thought for a bit about the ♦3 discard (I suppose this argues in favor of the finesse), or whether there's any chance this West would feel 12 HCP makes for a more comfortable overcall (quite unlikely, as these are really clear overcalls, but there may be some intermediates who would pass with "only 10 HCP"). I suggest that what you think of these determine your choice on this hand.

The odds are not as people have been suggesting. Let's suppose, as seems likely, that these opponents always play up-the-line in side suits unless they are signalling attitude. 1. West is A108 KQJxxx ?6 xx. He isn't A108x KQJxxx 6 xx because he would've played ♠10 then ♠x. Hence East is Q742 xx ?543 xxx. The two ?'s are the Q and 7 of diamonds. 2. Since the opponents play up-the-line, there's a difference in which spot West shows us (i.e. West's lowest diamond). As the simplest example of this, if West's lowest spot were the 7 (and you agree with part 1), we would be certain he held the ♦Q. In fact, in this situation, the odds are even. Note that in many cases, you'd know West's last card. For example, if West played the 3 and East played 5-6-7, then you'd know that East held the Q765 and West 43. Monty-Hall-type problem to illustrate this: Monty has a car, a goat, and a chicken behind three doors. You choose one, and he controls the other two. Then he says "I'm going to show you the door I control with the smaller thing behind it," and he opens one of his doors and it has the chicken behind it. What are your odds now? [suppose you believe him, and you believe he was always going to do this same thing of revealing the smaller object. You also believe cars are bigger than goats, which are bigger than chickens.]

The alert would have been UI too --- he had understood it! If this had happened behind screens, what would North have thought of the situation? Is it possible the auction itself (i.e. with or without an alert, and correspondingly with or without South understanding), when played without screens, forces North to take the incorrect action? Maybe guessing partner did understand it and guessing partner didn't understand it, on the given auction, are both LA's.

I came to the same solution as r_prah: Cash ♦A now so we don't get endplayed later. If partner has ♠K, we're always setting them. Partner may well have nothing. The encouraging signal may just be "please oh please don't go looking for strength elsewhere."

In order to have the comparisons with other players, BBO gets 16 (I think) different anonymous players to play each hand. Sounds like you were grabbed twice for the same board, presumably because BBO didn't realize you were actually the same anonymous person.

I tried Richard Pavlicek's ancient RP Deal Finder program (DOS-based!), found here, which searches through a library of more than 2 million pre-double-dummy-analysed hands. I'll interpret "cold" as double-dummy-makeable. I stopped the program at around 50000 deals. One side having a makeable grand slam of any sort was 1.4% of the 50000 deals and either side having one was 2.8% of them (these numbers looked pretty stable by the time it got out near 50000 deals).

Did you log out and then log back in between the identical hands? My guess would be that the server treated you as two distinct anonymous players, the second of whom hadn't seen the hand yet as far as the server knew. If that's the problem, hopefully it won't be too hard for BBO to fix.

Was it this post? It looks like bluecalm is talking about the play-of-the-cards-vs-double-dummy standard deviation as 3.5 imps/board. Later in the same post bluecalm mentions the bidding-(with-double-dummy-play)-vs-par-score standard deviation as 6 imps/board. The par score is e.g. more often doubled than real contracts, so that number is presumably a bit high for more practical concerns. Your guess of 6 imps/board sounds pretty reasonable, but of course someone with actual data coming along would be best.