[Leo LaSota]

pigpenz, on 2012-March-29, 09:38, said:

NO Leo
I was referring to several hands where you had received 100% on
they were 2 or 3 hands where you opened some real obtuse 1NT openings, say like 4441 13 hcp , just an example
you were able to bring the contract home even though on these hands GIB could have beaten several tricks.

I am just referring to the fact the yes you got 100%, but if GIB had defended differently you would have gotten 0%.
So yes you took your chances and got a good score and you could have just as easily gotten a 0%.

Is this good bridge? I dont know?
you do very well with it and do very well with normal things, alot of people have a hard time with the simple
things of bridge.

Again, I guarantee you the facts are that I average better scores on the hands where I do only "normal things". If you believe that all of my high scores on hands where I make an "unusual" bid are due to the fact that GIB is not programmed to defend a 1nt opener as possibly being on 14 rather than 15 to 17, you are mistaken. It used to be said that 26 HCP were the combined total typically required to bid and make 3nt with 2 balanced hands. Looking at Meckwell in action, they routinely bid and make 3nt on 24 combined HCP. No one would ever argue the fact that this is good bridge. Hands that you may open on 14 and the robot drives to game on 9 are good tests for making 3nt on 23 combined HCP. While occasionally you may make extra tricks primarily because of the GIB's programming, frequently there will not be ways to make these hands or if there are they require very good lines of play for the declarer.

[Leo LaSota]

barmar, on 2012-March-29, 09:52, said:

GIB doesn't know how to take inferences from the play, it mostly bases its simulations on the auction. So if you psyche, it will often go wrong on defense. The kinds of mistakes it makes are often ones that a human player never would (but I think the same can be said for some of its good plays as well). So while you may get lots of declarer play practice, I'm not sure that everything you learn from playing against the robots is transferable to real bridge play.

The off-shape or out of range nt openers sometimes used in robot bridge result in more declarer play practice. What you "learn from playing (many) hands" against the robots is that repeated practice results in improved declarer play for an individual. Of course certain things that work well on a hand against GIB are not "transferable to real bridge play". However, I can speak from my won experience that the declarer play practice against GIB has improved my declarer play in live bridge as well. I can also tell you that I know for sure that participating in ACBL robot tournaments has improved the declarer play in live bridge for many other individuals.

[fred]

pigpenz, on 2012-March-29, 14:55, said:

But isnt that what we are trying to discuss here......is this real bridge in terms of what the ACBL calls bridge?
I am sure the ACBL has hardly any idea what goes on in the robot games, could be wrong but wouldnt bet my life on it.
ACBL is selling a product, Fred is selling a service. Is it anything different?

I am sure Fred could sell an ACBL product where the contract is the same at all tables and lead is the same, then everyone would be
rated on their declarer play....same could be said where there was a game and all you dot to do was be on opening lead against a
redetermined contract. I am sure we would all get better at declaring in one, and get better at leading in the other.....the only thing
that determines what works is wether or not it gets sold.

Probably you were not trying to suggest otherwise, but just to be clear neither I nor my partners see ourselves as "salesmen" as far as our relationship with either ACBL or BBO members are concerned.

Our job is to try to develop bridge-related products and services that we think bridge players will enjoy. We see ACBL as our most important business partner. It is not smart business (nor is it our style) to try to "trick" ACBL into giving away their masterpoints for events run on BBO that we secretly think are unworthy. We really believe that we are doing is good for bridge and we believe that the best way to get the ACBL to agree is to provide them with enough (accurate) information so that they can also see the light.

It is easy to try to sell a product you believe in.

Most of those in positions of power at ACBL know anywhere between a reasonable amount and a lot about our robot games (and some of these people play in them regularly). That is partly because we have made a serious effort to provide such people with information and also because many of them have made a serious effort to study the information that we have provided.

I think that the whole discussion about "real bridge" misses the point for a couple of reasons.

First, as I have mentioned before in this thread, marketing considerations should not be discounted especially at this time in history where bridge in America could really use a boost. It is fortunate that the powers-that-be at ACBL seem to be understanding this point in increasing numbers.

Second, "real bridge" is not a static entity. The game evolves in various ways and on occasion new forms of the game are developed that demonstrate the potential to challenge and entertain large numbers of bridge players. Sometimes old rules have to change or new rules have to be codified in order to support such forms of the game. Organizations like ACBL (not to mention clubs, online sites, etc.) that refuse to adapt to what the players want will not be successful.

Yes it is sometimes hard to know where to draw the line, but the same goes for other aspects of bridge (such as system regulation) and life in general. All we can do is hope that intelligent, well-informed, and well-meaning people are responsible for making these decisions.

Fred Gitelman
Bridge Base Inc.
www.bridgebase.com

[barmar]

pigpenz, on 2012-March-29, 14:55, said:

I am sure Fred could sell an ACBL product where the contract is the same at all tables and lead is the same, then everyone would be
rated on their declarer play.

This used to exist in f2f bridge -- they were called Par Contests. Bridge World occasionally reprints articles about these events from their archives.

[pigpenz]

barmar, on 2012-March-29, 17:16, said:

This used to exist in f2f bridge -- they were called Par Contests. Bridge World occasionally reprints articles about these events from their archives.

yes isnt that how the intercollegiate par hands used to be? I may be wrong but wasnt there just an article in bridge world
how sometimes the best plans dont always work, as far as the par was concerned on those who used to set them up.

Leo....I am a hand junkie on results what I was trying to say is.....there are alot of hands where you get 100% but somewhere in the play of the
hand GIB had an oppurtunity to set some of those, esp 3NT contracts....so some of the 100% could have been very easily 0%.

Fred...yes you have been good at marketing bridge. whatever works will get supported??? I guess that means numbers, not sure?

[barmar]

I doubt that "whatever works" is the criteria -- that sounds like a variant of the "slippery slope" argument.

We think that "best hand" bridge is close enough to "real bridge" that it makes sense to treat it that way. On the other hand, there's no plan to try to get a sanction for Bridge Bingo.

I think it's clear that reasonable people can disagree on where to draw the line. I don't dismiss your opinion, I just don't agree. I hope you respect my opinion.

If ACBL decides to withdraw our sanction for best-hand bridge, I expect we'd switch these tourneys over to random. We'd probably still run best-hand tourneys, they just wouldn't be able to award ACBL masterpoints.

[Leo LaSota]

pigpenz, on 2012-March-29, 17:26, said:

yes isnt that how the intercollegiate par hands used to be? I may be wrong but wasnt there just an article in bridge world
how sometimes the best plans dont always work, as far as the par was concerned on those who used to set them up.

Leo....I am a hand junkie on results what I was trying to say is.....there are alot of hands where you get 100% but somewhere in the play of the
hand GIB had an oppurtunity to set some of those, esp 3NT contracts....so some of the 100% could have been very easily 0%.

Fred...yes you have been good at marketing bridge. whatever works will get supported??? I guess that means numbers, not sure?

There are many hands in live bridge that I take "more tricks than I should". I would really appreciate it if you or some other "hand junkie on results" would review a large set of hands played and see whether or not the "normal hands" average better than the "unusual bid" hands. You and others may be surprised by what you find, but I guarantee you that the average is better for the "normal" hands.

[pigpenz]

Leo LaSota, on 2012-March-29, 17:50, said:

There are many hands in live bridge that I take "more tricks than I should". I would really appreciate it if you or some other "hand junkie on results" would review a large set of hands played and see whether or not the "normal hands" average better than the "unusual bid" hands. You and others may be surprised by what you find, but I guarantee you that the average is better for the "normal" hands.

Leo....it has nothing to do with you....just this afternoon was looking at a result....and see the distribution of results i wondered why did these 3 people make 5 and everyone else only make four.....for some reason the people that all made the overtrick all opened 2nt on a singleton. Its just one of those weird things sometimes that you see when you look at the results.....what makes the bots defend differently or lead differently from one bid to another.....if you do look at the results you will see that there is a difference....so some defences or opening leads are def influenced by what the opening bids are.

now as to declarer falsecarding play one card to another in the play of a suit I cant tell on that.... playing a card or not when playing a suit def triggers different defenses by GIB...say GIB leads suit and you have kxx op xxx GIB sometimes for no reason will switch suit if you play or dont play honor card.

[mycroft]

No, Leo, it's not about you or what you do. I don't even care about the results you get (except that I am very impressed with your results at this game!) It's that one of the strategies that is always posted for Robot games is "bid NT first, on anything that looks like it will end in NT". But, the *system you're disclosing to the opponents* isn't that. So either you're psyching repeatedly, or playing a system that you're not disclosing. In real life, that gets you sent to Conduct and Ethics; in Robot games it works, because "everybody's doing it". But it *works* at least partly because the robots don't defend as well when they're defending 15-17 NT and you're playing 13-18 NT, frequently 4441. They misguess queens, misguess suit breaks, *because you're not playing what you say you're playing*. And neither is anybody else.

That - and not anything else - is what makes this game "not bridge" to me.

[barmar]

mycroft, on 2012-March-30, 10:14, said:

No, Leo, it's not about you or what you do. I don't even care about the results you get (except that I am very impressed with your results at this game!) It's that one of the strategies that is always posted for Robot games is "bid NT first, on anything that looks like it will end in NT". But, the *system you're disclosing to the opponents* isn't that. So either you're psyching repeatedly, or playing a system that you're not disclosing. In real life, that gets you sent to Conduct and Ethics; in Robot games it works, because "everybody's doing it". But it *works* at least partly because the robots don't defend as well when they're defending 15-17 NT and you're playing 13-18 NT, frequently 4441. They misguess queens, misguess suit breaks, *because you're not playing what you say you're playing*. And neither is anybody else.

That - and not anything else - is what makes this game "not bridge" to me.

But the fact that you're only fooling robots is what makes me say "It's close enough".

If you did this playing with humans, I'd call foul because you're using an illegal way to get an advantage over those other contestants. But the robots aren't contestants, they don't have the opportunity to win that you're taking away from them.

Note also that this strategy has little to do with whether the robot tourneys are "best hand" or not. I suppose there's a bit of safety in bidding light, since you don't have to worry about an opponent having a huge hand that they can punish you with. On the other hand, you also can't preempt them.

[rduran1216]

I play about 1000 hands against robots per month. I've seen others who take advantage of the robots by psyching or by making calls that are outside of known structure. Is it comparable to live bridge? In some respects yes, but the dbl dummy play of robots on defense combined with their horrendous errors on other hands makes it so that u know how to beat the robots.

Check my results against them, I mean you definitely learn how to beat the bots. Its really not the same as live play, but thats neither good nor bad IMO.

[pigpenz]

barmar, on 2012-March-30, 10:41, said:

But the fact that you're only fooling robots is what makes me say "It's close enough".

If you did this playing with humans, I'd call foul because you're using an illegal way to get an advantage over those other contestants. But the robots aren't contestants, they don't have the opportunity to win that you're taking away from them.

Note also that this strategy has little to do with whether the robot tourneys are "best hand" or not. I suppose there's a bit of safety in bidding light, since you don't have to worry about an opponent having a huge hand that they can punish you with. On the other hand, you also can't preempt them.

this may be true, but I have heard of people getting barred for capricious bidding cause they are distorting the results.
On the post I had done earlier on a tournament Leo was playing in, 18 board tourney 9 hands he opened 1NT and what was it he only had balanced 15-17 hand once. Thats half of the tournament.

I dont know is that fair for the rest of the field?

so by the fact your only fooling robots, makes it "close enough".

It's an intersting debate I dont know what the answer is.

[mycroft]

Bridge is a game of full disclosure, and you are expected to try to beat the opponents by playing your system better than the rest, and the opponents are expected to try to beat you with full knowledge of what you are playing.

Whether or not "it's just robots", a game where you try to beat the opponents, in any part, by lying about your system better than the other competitors, is Not Bridge. I know it's a challenging game, and I have respect for those who play it well, but it's Not Bridge.

[hrothgar]

mycroft, on 2012-April-02, 12:26, said:

Bridge is a game of full disclosure, and you are expected to try to beat the opponents by playing your system better than the rest, and the opponents are expected to try to beat you with full knowledge of what you are playing.

Whether or not "it's just robots", a game where you try to beat the opponents, in any part, by lying about your system better than the other competitors, is Not Bridge. I know it's a challenging game, and I have respect for those who play it well, but it's Not Bridge.

Would you argue that a player who makes a psyche is not playing bridge?

[barmar]

If the robots want to know what my bids mean, they should just ask.

[hrothgar]

barmar, on 2012-April-02, 13:20, said:

If the robots want to know what my bids mean, they should just ask.

I make it a practice to announce, regardless of whether or not they ask...
Rude little bastards never reply. Never so much as a thank you or a have a nice day...

Oh wow! Jut had a sick thought.
Program the robots to bicker with one another...

Have both robots run their own sims and have one robot make snarky comments when the other chooses the "wrong" action.
That would make it MUCH more like bridge.

Better yet! Have the robot's miss-analyze boards and offer bad advice...

It will be just like the Newton individual!

[Phil]

WD Hrothgar.

[barmar]

Maybe we should replace GIB with ELIZA:

"Is it because you have a 5-card spade suit that you came to me?"
"Tell me more about your clubs."

[BunnyGo]

barmar, on 2012-April-02, 14:12, said:

Maybe we should replace GIB with ELIZA:

"Is it because you have a 5-card spade suit that you came to me?"
"Tell me more about your clubs."

"Psychoanalyze GIB" New e-macs command?

[hrothgar]

BunnyGo, on 2012-April-02, 15:22, said:

"Psychoanalyze GIB" New e-macs command?

GIB analyzes us all the time.
Only seems fair