Join Bridge Winners
All comments by David Caprera
1 2 3 4 5 6 7 8 9 ... 30 31 32 33
You are ignoring the author of this comment. Click to temporarily show the comment.
So, I'll be honest. I did not remember this board when I read yesterday's installment where we screwed up the defense. It was with some trepidation that I waited for part two. Knowing we would be the last table, I did not rush to see our result and was relieved to see that we didn't get them both wrong.

With the report on the final board, I think I can say that I enjoyed the article more than most others did. Thanks Oren.
14 hours ago
You are ignoring the author of this comment. Click to temporarily show the comment.
Less 2% for a club ruff on the go and another bit less than 1% for cross to the heart ace and a club ruff. And a smidgen for the diamond ruff. And the heart ruff.

It isn't difficult to construct deals where East-West take the first five tricks. I guess they should have stopped in a part score.

When someone says, “Slam was 50% on a finesse, it never is.”
Dec. 11
You are ignoring the author of this comment. Click to temporarily show the comment.
I open 3C. 4C would put us past 3N.

3S in competition we play as not forcing but I believe that to be a minority view. When auctions become competitive, you need to show your suit.

Ax is a raise, forcing or not.
Dec. 11
You are ignoring the author of this comment. Click to temporarily show the comment.
Not a fan of 8-13, too wide a range, but it didn't really affect the result. With 11 cards in two suits, West should take another call. But it is not an easy hand and I am reminded of Walter “Big Train” Johnson's admonition, “A singleton king is a balanced hand.” West's distribution is 4=2=2=7.
Dec. 10
You are ignoring the author of this comment. Click to temporarily show the comment.
I have always thought that 2-4-6-8-9-7-5-3 ranking of cards (2 most encouraging, 3 most discouraging) made the most intuitive sense (perhaps encrypted even better), but they never caught on.
Dec. 9
You are ignoring the author of this comment. Click to temporarily show the comment.
I have always liked pigeon coups.
Dec. 7
You are ignoring the author of this comment. Click to temporarily show the comment.
I had a partner 15 years ago who did VERY WELL in the finance business using AI to value $100M mortgage portfolios. Games of complete information (chess, Go, checkers) don't present the same problems that does bridge. Nor is “partnership” an issue. As I have come to appreciate it, the real problem is “incomplete information.” While one element is probabalistic, that can be handled.

One problem, difficult for AI, is assuming an opponent made a mistake (and with what probability). I am not aware of a computer having ever found a Grovesnor Goofball.
Dec. 6
You are ignoring the author of this comment. Click to temporarily show the comment.
Yes, totally random. Actual position was J87 and QT. I played the 9. LHO was 5305. He might have shown majors with 5404. But the whole thing was too bizarre and I didn't work it out.
Dec. 3
You are ignoring the author of this comment. Click to temporarily show the comment.
My understanding was that Mulberry was a Paul Soloway creation. We may both be right.
Nov. 25
You are ignoring the author of this comment. Click to temporarily show the comment.
Nor should they. Yes, there needs to be disclosure. But please don't tell an AI how to evaluate hands. This is truly an opportunity to resolve so many aspects of the game. I would love to know how AI solves hand evaluation - is it close to K-R or does the weighting of a hand need more value for suit length and pattern?
Nov. 24
You are ignoring the author of this comment. Click to temporarily show the comment.
The idea of hcp must have been programmed in. AI would do much better at hand evaluation than 4-3-2-1.
Nov. 24
You are ignoring the author of this comment. Click to temporarily show the comment.
It was probably 1976. Chicago at the Palmer House. Michael and I won our first sectional pairs event. In that era, the seeding “committee” was Gerald Caravelli. Top seeds were tables 3 and 9, but we didn't know that. Having won Friday afternoon, we returned for Friday night and I proudly approached the seeding desk where I announced that, “We are defending champs.” Without missing a beat, Gerald said, “Congratulations” as he gave us entry 1A North-South.
Nov. 23
You are ignoring the author of this comment. Click to temporarily show the comment.
I agree that the problem should define…P-Dbl-2N. Without lebensohlish 2N available, it will be difficult for responder to catch up if he passes.

I don't like defending 2H at matchpoints.
Nov. 23
You are ignoring the author of this comment. Click to temporarily show the comment.
Is the pizza at New York West any good? Any recommendations for pizza place that stays open late?
Nov. 23
You are ignoring the author of this comment. Click to temporarily show the comment.
Playing as you suggest is what I suggested to my wife. Similarly, (1D)-2N-3D-3H (any min)-?, and now it may be better to play 4C is hearts and 4D is clubs (game ordered), but there is a memory issue and potential car crash lurking. Annie is still considering it.
Nov. 19
You are ignoring the author of this comment. Click to temporarily show the comment.
So, you have no game invite in hearts? (1C)-2N-(P)-3H being anything between a yarb and a balanced 12 count with three hearts seems like too big a range.
Nov. 18
You are ignoring the author of this comment. Click to temporarily show the comment.
3H is best raise. 14-15 shortness. 3S would then ask, responses low-high.
Nov. 18
You are ignoring the author of this comment. Click to temporarily show the comment.
Many who play transfers after 1M-(DBL), play that 1M-(DBL)-3M-1 as mixed and a direct 3M as preempt. It comes with a cost. If 1M-(DBL)-4m is a splinter, then the combination leaves responder with no way to show hearts and diamonds. In competitive auctions, showing fit one way or the other may be more valuable.

I know this will sound blasphemous, but the least valuable bid of the above choices may be 1M-(DBL)-3M. Particularly the practice of bidding 3M with any garbage hand that has four trump. It is the proverbial gun fight of a Smith and Wesson versus a pea shooter. If you lighten up your mixed raise (5-8) and your limit (9-11), and forgo the trash preempt, you can show a fitting raise with 3M-1.
Nov. 18
You are ignoring the author of this comment. Click to temporarily show the comment.
I trust West played the J.
Nov. 15
You are ignoring the author of this comment. Click to temporarily show the comment.
4H shows a diamond raise for us. The problem sequence is hearts and diamonds. I like the idea that 3S showed support, but only after a Kokish sequence.
Nov. 14
1 2 3 4 5 6 7 8 9 ... 30 31 32 33
.

Bottom Home Top