xxhong

This is simple. A 3H in gib's system shows a better hand (about 10-12 points) and the responder's hand is too strong to pass 2H. Therefore, 2S is the only bid available.

No, I am talking about software improvements. I think this is the right place to post because it is closely related with gib weakness and money bridge. Of course, if you think suggestions for the software is more appropriate, you can transfer the post there.

Suppose you know opp is likely to make a contract that gib would often go down, do you quit in the middle of the hand? In a perfect world, gib would make it. However, in real world, gib may go down in many makable contracts due to the early bidding distortions, which gives gib the wrong constraints to do the simulations.

We certainly live in different worlds. The whole post talks about gib weakness due to wrong constraints from the bidding and the abuse of this problem by quitting. BBO should at least provide some comments on this issue. Otherwise, there will be more players doing that in money bridge, which would just ruin the whole game. I am a frequent money bridge player and the frequency of opp's quitting when I declare a vul game and slam, or even in the gf bidding process has been increasing in my observation. This is very annoying and I seriously suggest some actions against this kind of cheating behaviors.

Now it is getting more and more frequently that human players quit when our side holds a strong hand. Last night, two players quit in the middle against me. In one hand, my gib opened 2C and I bid 2D as waiting. In the other hand, I declared a vul game. A few weeks ago, a player, xxxxxxx, "proudly" told me that he takes advantage of this quitting in the middle and it is an "art" to know when to quit in the middle(I don't really think bbo should continue the service to players like him). The basic idea is simple. In many situations, gib may make declaring mistakes based on the wrong bidding information provided by the human opps. In those cases, quitting in the middle is very profitable, because if you don't quit, your human opp declarer don't make such kind of silly mistakes. Also, if your human opp is a stronger declarer than gib, you should almost always quit when he declares and your hand is weak and not involved much in defensive decisions. There is a very very simple fix of this problem. If one player is disconnected, a gib should be summoned to replace his seat until he returns or until the end of this hand (Of course, some waiting time can be issued when one player is disconnected before a gib is summoned to make it better to those who don't have very stable internet connections but can return in one or two minutes after the disconnection). This would guarantee nobody can take unfair advantage of the gib arbitrage, which is so obvious a bad treatment if your opp intentionally quits and you always obey the rule to stay.

Can gib really play some reliable bridge as its description? One hand: (gib) KQxxx Kx Qxx Kxx (mine) Ax AT9xxxx KJ9x - The bidding: 1S(gib) p 2H(me) 3C(human opp) 3H(!) 4C 4N 5C Doulbe(claiming playing DOPI) Here, gib should never raise to 3H with 2 hearts a horrible hand. Then later, it lied again, with one KC, it should pass instead of doubling. Example 2: Gib Qxxxxx x xxx AQT Mine AKJx QJx Kxx xxx The bidding went: 1H(opp's gib) 1S(my gib) 1N(opp's psyche) Double(it shows 18-19? This is just ridiculous and opens up a lot of psyche possibilities without good reasons) 3D(opp's gib, it shows 12+ points, another nonsense) Double(Another lie, this double shows 2 diamonds exactly, but gib holds 3) 3H(human opp holds 4 hearts and 2 jacks) Pass(which is forcing, but the double shows 18-19 and forces to game, the pass intended to hang them more if human's gib bids more above 3H) pass(opp's gib's value is very limited) Double( lied again, double shows exactly 2 hearts) pass 3N(3N played from my side looks good, it turns out to be a very good contract after H or D lead. ) Double(opp's gib now got hooked) 4S(! fine, gib doesn't like playing 3NT, and rebid 4S with a bad suit) pass pass(slam is remote) Double(opp's gib try to penalize it) p p (Redouble, trying to maximize the profit) p 5C(Cuebid! this is nonsense, if 4S redouble made 6, the score is higher) After a 1NT psyche, gib made so many bidding mistakes in just one hand. There are some basic logical bugs in the code for sure. Also, after the psyche, it just makes no sense to do any simulations, all simulations are based on the wrong assumptions. The correct strategy is to only use the input from partner and opp's gib as reliable constraints to do the simulation, not opp's psyche. And the pull of the redouble is really really very very bad. As I said again and again, the programmer should really play more bridge with gib to understand and then solve all those problems. Gib in many cases is free to deviate one or even two cards from its system description if the flawed simulation suggests something else. The flawed system and simulation then in turn opens a million profitable psyche possibilities.

1S 2C 2H 2S 4C(cue bid, showing extra, 15+HCP, 5-4-2-2 shape, 2 or 4 KC)4D(cue) 4H(cue) 5C(cue, slam zone, denies SQ) 5N(choice of slams, we have no cuebids available, we may miss one KC and SQ, but DQ, SJ and HQJ should justifies this bid, if partner thinks that we are in the slam zone) 6N This hand is no easy.

I would bid 2D with north's hand. The rest is usually simple: 1S 2D 2H 2S 3D(now the pattern is clear) 3S(4 or more spades) 4C(serious slam try) 4D(cue) 4H(cue) 4N(RKC) If you go 4N after 4D, it is also not too difficult for north to know how valuable his DKQ is after south tell him they have all KC and HK. Usually, 2/1 is a better choice than J2NT when you hold a reasonable side suit, because 2/1 often offers your more information at low levels.

Mostly east, who has an easy 4S bid as it appears. West's hand is tough to bid, 3NT, 4S or pass are all possible spots. It is a matter of how aggressive your opp bids. Against aggressive players who frequently open 3C with balanced 6 card suit, un-balanced 5 card suit, it looks like a good time to pass. Playing against sound 3C openers and with conservative partner, it is probably ok to bid 3NT. Of course, sometimes, 4S can be the top spot when partner holds 4 or 5 S and the distribution isn't too terrible. quote name='TWO4BRIDGE' timestamp='1306461807' post='549306'] E/W missed 6S+1 slam 3CX-3 - 11.2 IMPs instead of + 4.0 for slam [hv=pc=n&w=skt74h6da972caj96&e=saqj653ha9dkqj3c5&d=n&v=e&b=9&a=3cdppp]266|200[/hv]

This is probably due to the lack of definition after the bidding develops after a takeout double and a suit bid. A simple rule of thumb would be 2S to be 5+ S, constructive, but nonforcing, 3S as gf with 5 or more spades IMO. I doubt the programmer has the clear definition here. This 3NT bid is more like a simulation result, where simulation shouldn't be allowed here IMO. The basic idea of bidding improvement is that you can make a bidding sequence to find the right hand type and the right contract, others can only rely on statistics over a large range of hands, you make a lot of money. I think, in GIB's world, many of its bidding relies too much on the rough estimation of a very large range of hands.

Well, it is an over optimistic expectation IMO. In 2008, it played 6 games, won 3, tied 1 and lost 2 against good human players head to head in limit holdem. However, it has a lot of problems facing multiple opps and facing opps who often change strategies. In No limit game, this problem will become even more severe, because very strong players frequently change strategies in their game. Also, the challenge of computing resource in no limit game is much larger than in limit games.

Try another partner then. It is ok to either open or pass this hand. It is just ignorant to say something like it would be passed in expert consensus.

No AI can beat good human players in a deep stack no limit holdem game. All the draw pokers, limit holdem and omaha are much simpler in that sense. No limit poker is about observation, psychological and strategical pattern recognition and assigning the right constraints to all the possibilities. All those things are very very difficult to program.

What I mean is under normal game conditions. If the beginner decides to go all in in every hand, he has some higher equities, about 1:9 in that single session, still one order of magnitude lower. Also, those who decide to go all in in every hand are clearly losers in long term, that still means pretty much close zero chance in his whole poker career to become a winning player. This is like throw a dice and you are always a 1:9 under dog, it won't take many sessions for you to go broke.

Gib plays very few penalty doubles. One of the gib requirements to double at high level is when it assumes that its side has a game somewhere. Most low level doubles by gib are takeout or card showing. I think gib should play penalty doubles when opps voluntarily bid something at 3 level or higher after a long sequence, here takeout double makes no sense. However, gib usually makes wrong constraints and assumes that opps may make it or have a better spot to run, that's why you rarely see gib doubles with trump stacks.

You certainly misunderstand no limit holdem. No limit is the most complicated game format of poker, especially for deep stacks. Limit holdem is pretty much a math game. The reason why you think no limit is easy is because of the game format of the tournaments. In the late stage of many tournaments, the ante and blinds are so high, that players don't have a deep stack at all and even the chip leader also has a very shallow stack. In that case, players have to gamble to avoid being eaten alive by the blinds and crap shooting happens. Beginners simply have zero chance to beat experts in a deep stack no limit holdem. Still, in the early stage of no limit tournaments, experts have much higher chances to survive.

I think the common sense meaning of the double should be penalty. Partner has made a limit bid and denies support for your suit and shows some value in opp's suit. Opp may overcall some 4 card suit and raise with 3. A general rule of penalty double from opener is that it should be a penalty double when responder has limited his hand, with a few exceptions.

Looks like a 4S bid.

Easy pass. It is an HCP trap. This hand defends well against H contracts and plays poorly facing many 10-11 HCP hands. If I really want to bid, I'd bid 3S.

Gib's 2H shows 14 or more points. Another ridiculous convention.

Young players make a lot of money from Poker by their own ability or luck. Almost all young players lose money in almost all bridge tournaments in US. Also, among those who make money in bridge tournaments, most money are paid by rich clients, not by the game itself. In some sense, many professional bridge players are like tools of rich people in US to prove that the rich can win in bridge. Many with high respect to themselves wouldn't choose this career path, although bridge is a great game. The running format of Poker actually wins more respect from me. IMO, Bridge began dying when rubber bridge is dying and great players became the tools of the rich.

Partner may hold something like x x AQJx KQJTxxx and in a gambling mood to wish for a diamond lead and one ace from you.

I think there is one fix. If you takeout double at one level opening bid, gib now doesn't take your strong suit bid to show length in unbid suits. That's the only fix in takeout doubles. Example: 1C x p 1S p 2H now shows strong hand with H and doesn't show support in other suits. When the bidding is high or when opp have bid two suits, the same thing keeps. Example: 1D p 1S x p 2C p 2H: this still shows a good hand with long H and 4+ clubs. Or 2S x p 3C p 3H, this also shows support in unbid suits. All these cases should be fixed. Another hand problem is 1c p 1S 1N: this is the so called sandwich 1NT to show 5-5 in unbid suits. So if you hold a strong balanced hand, you are completely out of bids, because double would also show 4-4 in unbid suits. This is one of the biggest nonsense in gib system, which opens a lot of psyche possibilities when you hold a weak hand when white, because opp can be completely out of bid, holding strong balanced hand or strong one suiter. Really, the programmer should play more with gib to understand all such kind of weakness in the bidding design. Now BBO pretty much ignores the QA part of GIB and let users do that job. In some sense, users who report important bugs should be paid IMO.

1C 1D 1H(one round forcing) 2D(gf, 5+ D, sign offs and invitational hands go through 1S and 1N) 3S(void, 4D, doesn't have to have extra) 3N(sign off) 4C(cue, extra) 4D(6+D) 4H(cue in H) 4N(even number of KC, SA excluded) 5C(CK) 6C(CQ) 7D

Rain, I did a lot of testings. Seriously, Gib plays no counting signals. In many situations, when counting signal is very very important to decide how many rounds defenders need to duck, gib plays very randomly. You simply mixed the situations when gib plays high-low with those situations gib prefers to play the highest card it thinks that doesn't cost a trick. About 50% of times, that happens when they have even number of cards, which really doesn't mean they have even number of cards. Gib does have some regular patterns in defending, including always inserting the lowest of its honor sequence at 3rd seat, always tending to cover an honor at early stage of the play, but it doesn't play any counting or altitude signals, unless bbo has modifies the code and added the counting signals recently, which I don't think is the case.