Join Bridge Winners
All comments by Christopher Monsour
1 2 3 4 5 6 7 8 9 ... 125 126 127 128
You are ignoring the author of this comment. Click to temporarily show the comment.
Well, you don't know LHO's actual hand. On some other holding he might well bid 4.
an hour ago
You are ignoring the author of this comment. Click to temporarily show the comment.
Possibly people from the East Coast have a genetic memory of the days before transfers?
an hour ago
You are ignoring the author of this comment. Click to temporarily show the comment.
Craig, I really like that setup with a strong NT. I usually play weak NT, but next time I am playing strong NT in a serious partnership, I will likely suggest implementing something like what you play.
an hour ago
You are ignoring the author of this comment. Click to temporarily show the comment.
You are assuming facts not in evidence. In two-session events at sectionals and regionals I've seen directors simply turn every NS to EW and vice versa and have one of them move up a section so they face different at-the-table opponents, but the players are all matchpointed against the SAME pairs they were matchpointed against in the first session.
13 hours ago
You are ignoring the author of this comment. Click to temporarily show the comment.
So do you put the invitational 4-5 hands through 1NT-2-2-2 or 1NT-2-2-2? Is the other of those two sequences a puppet to 2NT so you can show 5-5minor light invites?
13 hours ago
You are ignoring the author of this comment. Click to temporarily show the comment.
Well, it depends on the rest of the system. If I am playing with Dan, then 1NT-2-2-2 is a game-forcing shape-ask if I'm not a passed hand, and five spades invitational (with no implication of four hearts) if I am a passed hand.
14 hours ago
You are ignoring the author of this comment. Click to temporarily show the comment.
Using the ACBL as an example of how to run a bridge game is like using the current US executive as an example of how to run a country.
16 hours ago
You are ignoring the author of this comment. Click to temporarily show the comment.
“Pick a slam” doesn't help when you don't have a slam force. You don't need a way to set clubs here, but you do need a way to unset spades. Otherwise you will be unable to make a “pick a slam” bid later.

For what it's worth, it's also nice for 4 to be natural if it makes a future 4NT bid into six-keycard Blackwood.
16 hours ago
You are ignoring the author of this comment. Click to temporarily show the comment.
The real question is whether jumping to 5 would be a void splinter, exclusion keycard, or 2 of the top 3 clubs honors with strong spade slam interest.
21 hours ago
You are ignoring the author of this comment. Click to temporarily show the comment.
If you want to play this as garbage (which in most systems makes the most sense) I think it should show equal majors or longer *spades*. I want opener correcting to 2 with equal length in the majors for several reasons:
(1) I want the lead coming up to the strong hand as often as possible when there may be a large strength discrepancy;
(2) I want as often as possible to conceal what may be our only high cards;
(3) I want to conceal the misfit when opener is 2-2 in the majors; and
(4) I'd like the opponents to have to worry that partner may have four spades when he signs off in 2.
Nov. 18
You are ignoring the author of this comment. Click to temporarily show the comment.
We may need to suggest slam would play better in s. Clubs are not set, though, since if we stop in game at matchpoints it should be 4 (and maybe even at IMPs). This doubles as a cue-bid since we wouldn't be suggesting clubs as trump on a Q-high suit.
Nov. 18
Christopher Monsour edited this comment Nov. 18
You are ignoring the author of this comment. Click to temporarily show the comment.
Playing Precision the 3M preempt is more valuable since it can be wider ranging. Frankly, over a takeout double I think loading the weak end of the mixed raise into 3M and the strong end into Jordan 2NT makes the most sense and leaves all jump shifts available for fit-bids (or WSJs, but I've never understood the appetite to make a preempt in one of RHO's suits).
Nov. 18
You are ignoring the author of this comment. Click to temporarily show the comment.
Playing the same boards is important, but I would have thought the arrow switches even more important!
Nov. 18
You are ignoring the author of this comment. Click to temporarily show the comment.
Sounds reasonable as long as the director's compensation and the entry fees are reduced accordingly!
Nov. 18
You are ignoring the author of this comment. Click to temporarily show the comment.
I would say that the interesting thing about whether bidding and play is more important is that if you split bidding and play into individual and partnership pieces, you would probably find that bridge is something like 50% partnership bidding, 30% individual play, 10% individual bidding, and 10% partnership play (all defense of course), so even though bidding is more important, *as an individual* play is more important. So your talent at playing the cards may be more important in finding a partner, because after all the partnership can develop and improve on the partnership aspects, but from your partner's perspective he's pretty much stuck with whatever your individual talents happen to be. (Of course, if you are unwilling to work on partnership bidding, your good card play won't help you keep a good partner for long.)
Nov. 18
You are ignoring the author of this comment. Click to temporarily show the comment.
You're right that you probably don't need 1000 resamples. When I use bootstrap, it's usually to get estimates of things like the 98th percentile, so I'm used to needing a lot of resamples. I do think though that you want something more than stability from one resample to the next as a stopping rule. You either want to derive theoretical properties (or look them up in a standard reference like Davison and Hinkley) or at least look at empirical stability across ten or so different sets of resamples and not just two. This isn't as time-consuming as it sounds, since you can tune the number of resamples on a very small number of deals, and once you've found the right number, consistently apply it to the rest of the deals without checking again each time.
Nov. 18
You are ignoring the author of this comment. Click to temporarily show the comment.
Michal, yes, such a program exists. It's called resampling the defenders hands 1000 times for each deal and running the double-dummy solver on each resample. It's just very compute-intensive.

It doesn't take into account how likely the lead is to be accurate, or whether a given declarer will guess right on a deal that involves a guess, but those are somewhat subjective evaluations anyway.

If you wanted to get fancier you could have the computer play the deal out against itself. But then you'd need to simulate for each defender also and at every trick, and the computer still wouldn't make a discovery play to save its life. So if you want that much realism you're probably better off letting a commercial bridge-playing package play each resampled deal against itself. That probably isn't that much more time-consuming than the double-dummy solver, but it still won't replicate human play (though it will likely be a closer approximation) and it will be harder to know what the answer represents.

I say this is all compute-intensive, but it's also quite parallelizable, so certainly realistic for someone with the skills and the motivation.
Nov. 17
You are ignoring the author of this comment. Click to temporarily show the comment.
If it's IMP pairs, it's extremely important!
Nov. 17
You are ignoring the author of this comment. Click to temporarily show the comment.
Michal, I was just trying to lay out the different steps in the logical chain. Each may need separate testing. (I do believe, though, that I made the exact point that some misbid grands have a 40% chance.) Many of the specific numbers I mentioned were I hope obviously just educated guesses that could be refined with further study.
Nov. 15
You are ignoring the author of this comment. Click to temporarily show the comment.
I tend to think they don't know which bid was a misbid, just that they misbid. And if there was competitive bidding, they don't know whether they misbid or the opponents' preempt worked. Similarly, when they see they made one less trick, they don't know whether they misplayed, misguessed, or the opponents just made a good lead.
Nov. 14
1 2 3 4 5 6 7 8 9 ... 125 126 127 128
.

Bottom Home Top