Join Bridge Winners
All comments by Danny Sprung
1 2 3 4 5 6 7 8 9 ... 69 70 71 72
You are ignoring the author of this comment. Click to temporarily show the comment.
Passing as West has 2 ways to lose. Notice that we might not even make 3NT on a bad day. One should always be a little leery of NT when you have an extra stopper in the opponents suit.
Dec. 10
You are ignoring the author of this comment. Click to temporarily show the comment.
What else could east do? Poor shape, medium high cards, double positional heart stopper, poor honor structure.
Dec. 10
You are ignoring the author of this comment. Click to temporarily show the comment.
Bridge has a language (bidding) that is far more complex than machines are able to process at the current time. I think.

If the translations I see on the internet between 2 very common languages are any indication, there is still a long way to go in that regard.

Yes, it is true, there are only 38 ‘words’ in our language, but, as usual, context is everything.
Dec. 6
You are ignoring the author of this comment. Click to temporarily show the comment.
No, I think you are spot on. Let's do some math. I'll assume Money is Chicago, so slam is worth 1370/1430. Assume the contracts that fail are down by a trick and a half on average.

Further, let's assume that the opponents will double you when you are going down, and let you play it when you are making. It is more complicated than that, because there is the 7 level to consider. Ignoring that:

If each contract is 20% to make, you have a positive expectation of +6 by bidding, -154 by passing.

I think all of those assumptions are quite pessimistic.
Dec. 6
You are ignoring the author of this comment. Click to temporarily show the comment.
Yes. It might be a doubleton, if I am reading the methods correctly. I guess it could be a singleton, in some 4441 patterns.

Why wouldn't it be alertable?
Dec. 5
You are ignoring the author of this comment. Click to temporarily show the comment.
Last time was 12927.
Dec. 4
You are ignoring the author of this comment. Click to temporarily show the comment.
Double playing negative free bids are alertable in my opinion. Clearly unexpected. Just my opinion.
Nov. 29
You are ignoring the author of this comment. Click to temporarily show the comment.
Cover cards vastly overvalue AK doubleton
Nov. 27
You are ignoring the author of this comment. Click to temporarily show the comment.
Don: I was told directly by a BoD that it was Philly.
Nov. 26
You are ignoring the author of this comment. Click to temporarily show the comment.
The Bridge Committee of the BOD gave the charts a serious perusal during the board meetings in San Diego.

They had some concerns with our proposal:

Application of when the Open+ chart is allowed.
Approval process of suggested defenses for artificial preempts.
A different presentation of the charts.

Some minor bridge issues. I haven't seen the list of these issues yet.

The plan was for the subcommittee to amend the charts for a vote in Philadelphia.
Nov. 26
You are ignoring the author of this comment. Click to temporarily show the comment.
Ah, the old double invite, I see.
Nov. 24
You are ignoring the author of this comment. Click to temporarily show the comment.
That looks backwards to me. One should be accepting game tries more aggressively than slam tries.
Nov. 24
You are ignoring the author of this comment. Click to temporarily show the comment.
Ed: If you would force a declarer to lead the 2 from AKQJ2, I think we will have to agree to disagree. You are basically saying that ANY legal play is rational. If the laws wanted it that way, they could have phrased it as such.
Nov. 21
You are ignoring the author of this comment. Click to temporarily show the comment.
Richard: I disagree. If declarer has AKQJ2 of a suit, he shouldn't have to lose a trick to the stiff 3. That would be irrational.

Much different than David or the OP position.
Nov. 21
You are ignoring the author of this comment. Click to temporarily show the comment.
Good problem, Ron. A lot depends on your agreements.

I think 2NT is the best rebid, if you and your partner are on the same wavelength, that 2NT merely describes hand type, and does not guarantee a stopper. Partner needs to checkback for a stopper with a normal raise to 3NT. We can then wiggle out of NT when it is right.

The other calls, 3 and 3 are obviously flawed. You have what could be a good hand for a 5-2 spade fit, but the tap is coming in the wrong hand. 3 is probably best if you can't bid 2NT, since your suit is so strong, it figures to be the safer suit partial.
Nov. 18
You are ignoring the author of this comment. Click to temporarily show the comment.
That would be true if some boards were ‘more difficult’ than others.

It is true that playing all pairs is analogous to your example. Skipping the weakest pair in the room materially changes your expected percentage. Skipping certain boards does not.
Nov. 17
You are ignoring the author of this comment. Click to temporarily show the comment.
John: Having one range for Michaels is a bit dangerous. I'd surely Michaels over a minor, not vulnerable with any excuse.

Changing the vulnerability, and forcing your side to the 3 level should change you range.
Nov. 12
You are ignoring the author of this comment. Click to temporarily show the comment.
I had almost this exact hand at a recent sectional. I passed, and was soon +300, My counterpart overcalled 1NT and was +120.
Nov. 8
You are ignoring the author of this comment. Click to temporarily show the comment.
The main difference between bridge and all of the other games mentioned is the bidding. It is more like learning a language, than the brute force understanding of various strategic moves, hidden information (poker) or not.

Obviously, one could try to teach a program 2/1 or Meckwell lite, or whatever, but no matter how much time and effort is used to do that, there will still be countless sequences where there is no ‘definition’, so the computer will have to use first principles.

Since that is the case, better to let it develop its own language. And, since whatever language that is will have to interact with everyone else's language, this is a massive undertaking.
Nov. 7
You are ignoring the author of this comment. Click to temporarily show the comment.
I doubt we'd ever program ourselves into obsolescence. I am sure we can learn a LOT about our game from the sheer power of well programmed bots.

Ask any Chess/Backgammon/Go player how much they have learned from the neural nets.

Have ‘we’ learned much from the bridge programs. We have some questionable value from the Bird/Anthias work. We have some vague input from some world class players doing simulations by hand.

We are nowhere near having the kind of bridge knowledge the rest of these games have.
Nov. 5
1 2 3 4 5 6 7 8 9 ... 69 70 71 72
.

Bottom Home Top