All comments by Phillip Martin
You are ignoring the author of this comment. Click to temporarily show the comment.
Thinking about what to do is not in itself a problem. In fact, it's usually a good idea.
Jan. 18
You are ignoring the author of this comment. Click to temporarily show the comment.
I've never understood this argument. If I'm your partner, I trust you to do what you think is right. Why would I want to play with someone who does what he thinks is wrong just to humor me?
Jan. 16
You are ignoring the author of this comment. Click to temporarily show the comment.
WIth the tools available today, t's very easy. You can find courses on Coursera or Udemy. Here is one I took some time ago. Neural networks is only one section of the course, so it may be better to find a course dedicated to the subject. But this course does happen to be quite a good overview of neural networks and other machine learning techniques. https://www.udemy.com/course/data-science-and-machine-learning-with-python-hands-on/
Jan. 16
You are ignoring the author of this comment. Click to temporarily show the comment.
Lots of agreements are possible, but I believe one agreement you should avoid is playing double shows a singleton in the bid suit. Double should show some hand type where there is a chance partner will choose to pass it. Similarly, pass should show some hand type where there is a chance partner will want to double.
Jan. 16
You are ignoring the author of this comment. Click to temporarily show the comment.
Estimating the IMP expectation is hard enough. Now you want me to estimate the area under the normal density function as well? If the match wasn't infinitely long to start with, it will be now.
Jan. 13
You are ignoring the author of this comment. Click to temporarily show the comment.
If you think a line of play has a positive IMP expectation, you take it. Doubled contracts can be hard to assess, since your potential gain and loss is affected by whether or not the same contract is doubled at the other table. But you judge the risk vs. reward as best you can and play accordingly. Why would anyone want a teammate to do anything else?
Jan. 13
You are ignoring the author of this comment. Click to temporarily show the comment.
If opener has a take-out double of diamonds and can't double, he risks the auction ending there. Selling out with shortness in the opponents' suit is generally not a good idea. But if he wants to double to show diamonds and can't, it may not make much difference. He (or partner) may be able to double something else for takeout later.
Jan. 12
You are ignoring the author of this comment. Click to temporarily show the comment.
Applying restricted choice does not require you to assume your opponent is choosing his card from QJ at random. In fact, the whole point of applying restricted choice is to avoid making any assumptions about how your opponent plays. When you see an honor and when the only possibilities are Q, J, or QJ, then playing for the honor to be singleton is guaranteed to work approximately 2/3 of the time. If you try to guess how an opponent plays with QJ and play accordingly, you can tie this result if you guess correctly, but you can't do better. And you do worse if you are wrong. Just forget about this “but in practice people don't at play at random” distraction. It's irrelevant.
Jan. 10
You are ignoring the author of this comment. Click to temporarily show the comment.
Of course it's a consideration. A priori either opponent is more likely to hold one honor than to hold both. A more important question is whether other considerations are more compelling–specifically what do they know about my hand and why are they leading this suit at all?
Jan. 10
You are ignoring the author of this comment. Click to temporarily show the comment.
The 4, but not for any of the reasons listed. It's because I have nothing I want to encourage partner to shift to. If I lead a club (and I'm not sure I would, by the way), it's because I think clubs is where our tricks are coming from.
Jan. 9
You are ignoring the author of this comment. Click to temporarily show the comment.
In reference to Edgar's point above, related by Paul, the “I don't want to win that way” argument cuts both ways. Once Lowenthal revoked, allowing a vulnerable game to score. Our opponent wanted to waive the penalty, but John was adamant. He insisted that he be penalized. “I don't want to win that way,” he said.
Jan. 9
You are ignoring the author of this comment. Click to temporarily show the comment.
First, I think you overestimate the frequency of having a takeout double on this auction. On many hands where you have a takeout double, you would have acted already. Just make the 3 the 3 on this hand, for example, and you would have overcalled 2. If you were dealt a penalty double, however, you are unlikely to have acted on the previous round. Second, frequency is not the only consideration. Another consideration is what you gain by taking a given action vs. being forced to take the second-best action. On this hand, for example, if you can't make a takeout double, you should probably pass. So, to evaluate the advantage of being able to double, you must estimate the expected gain of doubling and expected gain of passing, subtract the latter from the former, then multiply by the frequency. Do the same for a penalty double. Then compare the two. This is not easy to do, and I'm not at all sure how such an analysis would turn out. In any event, it's not enough simply to argue you are dealt one hand type more often than another. If it were that simple, we wouldn't disagree so much in these situations.
Jan. 9
Phillip Martin edited this comment Jan. 9
You are ignoring the author of this comment. Click to temporarily show the comment.
I have trouble with this answer. If you are going to forgive your opponentsâ€™ mechanical errors in club games, you should always do so. Choosing to forgive errors for some opponents and not for others gives the former an edge that has nothing to do with bridge. In addition, to make how they treated you in similar circumstances the criterion for forgiveness strikes me as petty.
Jan. 8
You are ignoring the author of this comment. Click to temporarily show the comment.
With 12-14, you can pass. With 18-19, you can bid 2NT as you were planning to do. So I believe this should be played as an off-shape strong NT.
Jan. 8
You are ignoring the author of this comment. Click to temporarily show the comment.
If double is defined as takeout, I have the perfect hand for it. So I'll pass. Once partner realizes I'm not going to double even when I have the perfect hand, maybe he'll agree to play this double as penalties.
Jan. 7
You are ignoring the author of this comment. Click to temporarily show the comment.
East might well have been interested in slam, so the bad trump break just means they make four instead of the five they would expect on normal breaks.
Jan. 5
You are ignoring the author of this comment. Click to temporarily show the comment.
Any time declarer has two stoppers in whichever red suit I lead, I'm probably not beating this, since he rates to have at least seven tricks in the black suits after driving my club ace. It seems I'm more likely to find him with a single diamond stopper than with a single heart stopper, so I am surprised so many choose the heart lead.
Jan. 3
You are ignoring the author of this comment. Click to temporarily show the comment.
You have no unauthorized information. You can do whatever you want.
Jan. 3
You are ignoring the author of this comment. Click to temporarily show the comment.
If he has both queens, yes. So you must assume he's missing one them.
Jan. 2
You are ignoring the author of this comment. Click to temporarily show the comment.
Cashing the spade ace may cost a trick in some situations.
Jan. 2
.

Bottom Home Top