Join Bridge Winners
All comments by Jim Perkins
1 2 3 4 5 6 7 8 9 ... 87 88 89 90
You are ignoring the author of this comment. Click to temporarily show the comment.
^^^^^^^^
7 hours ago
You are ignoring the author of this comment. Click to temporarily show the comment.
“We play a strong club with unusual responses. Specifically, 1 - 1 is not a negative response, but a positive GF with 4 or more s. In general, all responses to 1 opening are transfers. Please ask about our alerts. <Pause> We also lead low from a low doubleton against suit contracts both on opening lead and in the middle of the hand.”

“What is the suggested defense to your responses to 1 opening?” “X of the suit bid shows that suit or requests a lead. Bidding the suit shown is take out of that suit.”

Especially in club games, I don't think it is in the spirit of competition to spring an unusual system on opponents, but we want to practice our system against live competition. I am leaning toward less disclosure at Regional Tournaments and higher, but for now we still do it. I feel like at higher levels of competition we have every right to expect opponents to inspect our convention cards. Convention Card which we ALWAYS* have on the table.

* It has now become part of my partnership duties to go to BW before the session and print 2 cards from our current file and bring them to the game.
7 hours ago
You are ignoring the author of this comment. Click to temporarily show the comment.
LTC is almost certainly easier to learn than HCP. And not quite as useful, but a close enough approximation to start.

It does become highly erratic if no good trump fit exists.
8 hours ago
Go
You are ignoring the author of this comment. Click to temporarily show the comment.
I did have an uncompensated chess tutor tell me that if I found myself prone to overlooking “long distance” attacks, I should opt for closed positions. Of course, the same mental flaw that overlooks long distance attacks, tends to become bored and “not fun” when faced with closed positions. Let alone ones that I set out to create.

Sigh.
8 hours ago
Go
You are ignoring the author of this comment. Click to temporarily show the comment.
If the poker community is any guide, it will be 6 months or more and we will still have people not understanding what is being meant by the Game Theory term “optimally.”
8 hours ago
Go
You are ignoring the author of this comment. Click to temporarily show the comment.
I would say that all the “never gonna happen” comments here sound almost identical to comments being made in chess forums in the early 1990s, and in poker forums 2000-2005. I am not that familiar with Go, but I am sure that the experts in that game made similar claims.
9 hours ago
Go
You are ignoring the author of this comment. Click to temporarily show the comment.
<<A computer can only play as well as it is programmed.>>

This may have once been true but with AI and recent advances in AI it is likely not true in the sense you are saying it. An AI program does not need to be told what the OPPs methods are. Rather it can “infer” methods from its own analysis of a large library of hands. Either a library of hands bid by bridge players generally. Or hands bid by this partnership specifically.

Just as humans do, the AI can develop specific defenses against specific agreements.

In the long run, I believe that AI will become much more adept than humans in ferreting out (and then responding to) agreements, disclosed and otherwise.
9 hours ago
You are ignoring the author of this comment. Click to temporarily show the comment.
We try to give the OPPs as much information as we have without being annoying. We do alert possibly heavy 1M - 4M auctions. We alert possibly heavy 1M - P auctions.

Some, however, are simply annoyed that we don't bid “normal.” And others are annoyed that we have taken the time and effort to have agreements, pretty much at all.
11 hours ago
You are ignoring the author of this comment. Click to temporarily show the comment.
xx
Oct. 21
Go
You are ignoring the author of this comment. Click to temporarily show the comment.
Maybe if I were as expert in Chess as Kit is in backgammon, I could give a better description. ;-)
Oct. 21
Go
You are ignoring the author of this comment. Click to temporarily show the comment.
Well, they don't have the “relevant parameters.” Probably.

But if the proper inputs are known and provided, of course computers can provide the outputs.
Oct. 21
Go
You are ignoring the author of this comment. Click to temporarily show the comment.
I am not a Chess Expert, but the tactic is to make a move that you know your opponent considers a blunder and to which you (correctly, you hope) know he will respond with an actual blunder that you will subsequently exploit.

Meanwhile, had your opponent correctly analyzed the position your move would actually have been a blunder.

At its most basic level it has earned the name “idiot combination” – you hang your Queen because when your opponent takes it she is checkmated in 5 moves. Meanwhile, if your opponent ignores the offer she can checkmate you in 3.

But the leveling can go much, much deeper than that. Well beyond my chess abilities. Which are even less than my bridge abilities so that should tell you something. :-p
Oct. 21
Go
You are ignoring the author of this comment. Click to temporarily show the comment.
Maybe not in complete information games like Go or Chess (although I believe that Chess, the human variety, has its share of bluffing), but bluffs are a huge part of poker (much moreso than bridge) and the computers both bluff and ferret out bluffs much better than humans.

As I understand the current state of the software.
Oct. 21
Go
You are ignoring the author of this comment. Click to temporarily show the comment.
I will start with: “Natural” bidding is without a doubt vastly inferior to bidding designed around or at least taking account of the “space” principle which itself is without doubt inferior to some other artificiality.

That is,

For any given opportunity to call, and paying attention to only the space principle:

50% of hands should take the cheapest action
50% (25% of original) of remaining hands should take the next cheapest action
50% (12.5% of original) of remaining hands should take the next cheapest action
50% (6.25% of original) of remaining hands should take the next cheapest action, etc.

You could substitute “level” for “action” or you could create paired bids if levels are too blunt.

Human memory issues are a complete drag on human designed and utilized bidding systems.
Oct. 21
Go
You are ignoring the author of this comment. Click to temporarily show the comment.
Why not?

But as feeding the relevant parameters . . . one assumes that a human is still required to define relevant and parameter.

At least for today.
Oct. 21
You are ignoring the author of this comment. Click to temporarily show the comment.
Ignore opponents bidding. Not optional, but we can play. Tonight.
Oct. 21
Go
You are ignoring the author of this comment. Click to temporarily show the comment.
I always imagined an AI match as was quoted for some famous (and obviously egotistical) player from the old days: “Who is your ideal partner?” “Um . . . Me.” “. . . OK. . . . Who would you choose as opponents?” “Um . . . the same partnership.”

However, with AI it can be done without creating UI problems.

Probably won't solve the squabbling problem.
Oct. 21
You are ignoring the author of this comment. Click to temporarily show the comment.
@ Richard F: I like your approach. I am not sure it works for us due to other useful agreements that while not working on this hand, work on other hands.
Oct. 21
Go
You are ignoring the author of this comment. Click to temporarily show the comment.
I am still waiting for the brute force/AI bidding program that is not coached by humans. I have to believe that there are far better approaches to reaching optimal contracts than we have found so far.
Oct. 21
You are ignoring the author of this comment. Click to temporarily show the comment.
Richard: you can say on the one hand that we hold only 22 hcp and only 2 Aces. For myself, However,I feel like I dropped the ball somewhere since both partners have controls in 3 suits, we have a double fit, both partners have shown interest and above all that, slam is a very good proposition.

So I asked for feedback and I see that many, not including you, suggest an auction that was suggested by partner at the table.
Oct. 21
1 2 3 4 5 6 7 8 9 ... 87 88 89 90
.

Bottom Home Top