Join Bridge Winners
All comments by Eric Kehr
1 2 3 4 5 6 7 8 9 ... 21 22 23 24
You are ignoring the author of this comment. Click to temporarily show the comment.
I can show and (at least 4-4) with a 2 bid, so I do that.

I'll probably find myself bidding 3 on the next round as well.
3 hours ago
You are ignoring the author of this comment. Click to temporarily show the comment.
Surely this hand is worth a slam try. It is incredibly unlikely that 3NT makes and 4NT doesn’t. If you can’t make a slam try at MP over 3NT without the ability to stop in 4NT If partner is unsuitable, then you ought to change your methods.

This is the second hand in as many days like this.
Dec. 12
You are ignoring the author of this comment. Click to temporarily show the comment.
I think the optimal approach is to use some sort of coded responses which depend on what was opened and what was replied.

For a simple (i.e. largely natural) system, I think reverses need to guarantee some sort of extras, so a catch-all bid of 2M is needed for a lot of hands with only 5M.
Dec. 11
You are ignoring the author of this comment. Click to temporarily show the comment.
It would be interesting to here from the OP what sort of hands Pass as West, and what hands bid immediately.
Dec. 10
You are ignoring the author of this comment. Click to temporarily show the comment.
I'm not sure whether the issue is balanced v unbalanced; it's more whether West is strong enough that he can afford to make one move to look for a slam.

Here, there can't be many hands where the limit is exactly 9 tricks in NT.
Dec. 10
You are ignoring the author of this comment. Click to temporarily show the comment.
IMO, West should move again after 3NT. That doesn't preclude playing in 4NT if East's hand is very unsuitable for playing in , so I believe this is right at any form of scoring.
Dec. 10
You are ignoring the author of this comment. Click to temporarily show the comment.
Deleted
Dec. 9
You are ignoring the author of this comment. Click to temporarily show the comment.
What is the probability that you or partner forget the system?
Dec. 8
You are ignoring the author of this comment. Click to temporarily show the comment.
I do wonder how important pscychology, table presence etc really is.

Obviously imperfect humans with a good psychological grasp will beat equally technically skilled humans who lack that. But would they be able to beat technically perfect or near perfect players who were psychologically “naive”?
Dec. 8
You are ignoring the author of this comment. Click to temporarily show the comment.
Partner played this hand and took the straightforward finesse, which is certainly the best play for 13 tricks, and might well be the best play for 12 as many have said.

As far as I can see the contract is unmakeable as the cards lay - West had Jxxx East had KJ9x.

But it did occur to me that in some sense the finesse is unnecessary in that if it would make, then you must be able to pick up the (as long as you can read the position!). But I wasn't sure whether playing small from dummy at trick 2 (or maybe even running 4 first), then A, then running the black suits, was in practical terms as good a shot at making the contract as anything, as I have no idea how to assign a probability to picking the position.
Dec. 8
You are ignoring the author of this comment. Click to temporarily show the comment.
What was common practice?

Bidding the higher suit so you have a retreat into the lower (as here); bidding the lower suit so you have a retreat to higher suit (as in the case I referenced); or bidding the suit partner has length, in so you don’t need a retreat (as in both cases)?
Dec. 8
You are ignoring the author of this comment. Click to temporarily show the comment.
Or is it 2 which gives a retreat position to 2?

Isn’t there a hand where one of the Blue Team overcalls 2 on a hand with both minors because, so some say, that was the style at the time allowing an escape to 2 if doubled?
Dec. 8
You are ignoring the author of this comment. Click to temporarily show the comment.
Can't you come down to JT A opposite 7 32 and then set up the ?

ETA: As Oren says!
Dec. 7
You are ignoring the author of this comment. Click to temporarily show the comment.
That's a good point. However, he did win those matches quite easily. i.e. it is likely that players quite a bit weaker than he would have also have been competitive.
Dec. 7
You are ignoring the author of this comment. Click to temporarily show the comment.
On board 16 in the other match (on page 5) when the contract is 6, I don't understand playing on first, rather than setting up the and then drawing trumps - no matter what one makes of the lead.

What distributions does it gain on?
Dec. 7
You are ignoring the author of this comment. Click to temporarily show the comment.
I think most people would accept the second one.

It's fair enough to force it to follow the rules of the competition; not quite as fair to force it to play some sub-standard system it thinks is pathetically weak ;).
Dec. 7
You are ignoring the author of this comment. Click to temporarily show the comment.
AlphaGo was fed a database of games; but AlphaGo Zero was not. And nor was the very recent AlphaZero which won at Chess and Go and Shogi.
Dec. 7
You are ignoring the author of this comment. Click to temporarily show the comment.
Mike, my guess is something like this:

For years in the West, chess was seen as the pinnacle of “brain games” (partly because Go was not all that well known over here) and a lot of effort was put into making computers good at chess. And the AI community was always saying how close they were to getting computers to master chess, and the naysayers (like David Levy) were always saying how much harder the challenge was than the programmers thought.

Eventually, the computers did get better than humans, and Go had become more well known in the West, and the naysayers started saying things like “Well, you may have conquered chess, but that is because of the superior tactical ability of the machines. They are still strategically weaker than top GMs. But you'll never be able to master Go, because that is far more strategic.”

So the focus shifted to Go. And for years the naysayers looked to be right. But then AlphaGo and its successors proved that computers could play Go better than the top humans.

Interestingly, now that computers were better than humans at Go, a lot of the naysayers started shifting their attention to the superiority of human programming to the self-taught algorithms: “This method to let the computer teach itself will work on something like Go which is largely strategic, but it won't work on chess because of the large tactical element. The self taught programs will be beaten by the best human programmed programs because of the latter's superior tactical ability”.

Now it is true that there are naysayers from the bridge community, but the chess community is much larger. And also Demis Hassibis was a top junior Chess player. So it is natural that they next looked at chess.

ETA: Another (and possibly most important) point, is that at this stage it is easier to demonstrate their superiority at Go, Chess, and Shogi, where the strongest players are other programs, and they can play a match of any length at any time, than it is to arrange a match against humans.
Dec. 7
Eric Kehr edited this comment Dec. 7
You are ignoring the author of this comment. Click to temporarily show the comment.
I don’t think you are right about the strength of chess computer programs in the 70’s.

David Levy made a famous bet in 1968 that no computer would beat him by 1978. And he easily won that. And he was an IM, not a GM, and so well outside the top 100.

He wasn’t beaten in a match by a computer until 1989.
Dec. 7
You are ignoring the author of this comment. Click to temporarily show the comment.
You don't program it. It programs itself. Somehow.
Dec. 6
1 2 3 4 5 6 7 8 9 ... 21 22 23 24
.

Bottom Home Top