junyi_zhu

I reckon that if you make less than 2 mistakes per board (bidding and play) then you are world-class :) :) :) You may even be a world-champion B) B) B) I don't mean double-dummy errors or even sophisticated errors -- I mean ordinary face-to-face practical errors that most players would accept if simply explained. -- like daisy-picking in the bidding -- or failing to make a play that could help prevent partner misdefending. At Bridge, few errors cost. Some actually gain because of a lucky lie of the cards or compensating opponent error. For a season, our team conducted a detailed post-mortem after every match. Collectively we never chucked less than 100 imps even in 24 and 32 board matches B) Some of these matches, we won, in national competition. On a few occasions, opponents resigned with boards to play. In 1 or 2 matches, compensating errors meant that the score-card deceptively showed us conceding less than 2 imps per board. Although, of course, we should have lost much more :) :( :( I have watched and commented enough top level matches to claim that it's not even close to 0.4 errors per board for world class players at top level competitions. Most hands in bridge are routine and there exists a huge gap between intermediate and world class players. Intermediate players blow at least one trick per hand(some may not cost because errors can cancel out). For world class players, I agree with Mikeh, it's less than 0.1 erros per board.

I don't think so: how many boards did you play against/with GIB ? How much thinking time did you allow GIB ? A team of four with about 500 ACBL master points would beat a team of four gibs without much difficulties in a 64 board match, IMO, if both play sayc, under current BBO's slowest setting of gib. GIB has way more bugs than you can ever imagine. And it's certainly a wrong judgement of Zia to give up his famous bet when gib was invented.

If bridge is an art of partnership, it's just a big nonsense to rate individual's game (unless in individual tournaments). If you rate partnership strength, it's extremely simple, about the same as chess.

I have done that and posted some results of that a while ago. But the system is limited to play and ignores bidding completely. It is less significant as one might think. WC's and gib produce error rates of about 0.4 errors per board. Intermediates make about 0.8 and beginners reach 0.9+. I had a few pickup partner with rates larger than 1, but i did not play enough boards with them to make the results mean something in a statistic. World Class and GIB make 0.4 errors per hand in card plays? That's just an insult of world class players, hahaha. GIB is generally weaker than intermediate players, not even in bidding, but also in card plays. It may make some tough double dummy contracts, however, it blows way more tricks in simple situations and it blows way more tricks in redoubled contracts than you can imagine.

I might rebid an offshape 2nt here because of the poor club suit and stiff HK, then it's rather straightforward: 1C 1H 2N 3D(transfer, gf) 3N(no 3H or 4S) 4N(invitation) pass, (seems an easy pass and 4nt can be high sometimes) If I decide to reverse: 1C 1H 2D 2S(gf, extra length in H in our agreement) 2N(natural) 3C(natural) 3D(trunky diamonds, can't be 5-6 here) 3S(cue) 3N(very minimum, can't bypass 3N) 4H(cue, a slam try, 4D rkc can be ok, but slam try rates to be better than rkc, cause your club isn't very good either, the one with CJ should initialize it:), also 4H should show exact 3 clubs cause if you hold 4 clubs, you want to bid 4C to show it ) 4N(let's play 4NT, since club is only 5-3 fit and my club sucks big time...) pass

even the design of play is quite wrong. Bridge is game of partnership, a game of hidden information, not a game for double dummy. That means one has to develop a strong artificial intelligence to understand what's going on at the table. That includes "what do opp's bids mean", "how to set the correct constraints and generate random deals fast enough", "what does the signal of partner tell? how does it affect your play?" "what's opp's style of bidding, how to adjust to it?" "what are opps' defensive signals, are they reliable or not, what do they tell you about their hands?" "what type of defence I should make and what effects it'll have on partner? can it simplify and clearify the situation and partner may thus have a clear read of the situation?" It's a very complicated issue. And bidding should never be based on a dozen of randomly generated hands that may have already set the wrong constraints. Also, the true money comes from some rare events, like "how to guard against some rare distribution and improve your percentage of play", all those rare events can hardly be generated in a dozen of deals. So Zia's bet is still very sound and in a the near future, computer program just has no chance to defeat human players. It really requires a big leap in AI I believe for computer program to have some real chance against humanbeing.

I was shocked to read this comment. How can one compare a program without defensive signals, basic understanding of opp bidding, tons of bugs in bidding system and basic design flaws in bidding and defence with expert or advanced or even intermediate human players?

If you want to often find the best games or slams in a double dummy base, opening offshape 1nt isn't the right choice, cause 1nt takes up space and it's often impossible to show a 5-4-3-1 or 4-4-4-1 shape after 1nt openings. However, bridge is often not a double dummy problem and 1nt may help you to make some 3NT or 4M that are easy to beat if you bid your hand accurately. Still, off shape 1nt can sometimes lead you into wrong game contracts like 6-1 fit 4M when 3nt is cold, or 3nt when 5-3 fit 4M is cold. Anyway, offshape 1nt bring more randomness into the game, which can be a good choice for underdogs in a KO match.

I play 4D to show both fit and no C control. With only D fit, I never bypass 3nt.

One basic problem of not opening distributional two suiters or three suiters is that if you don't open it at one level, you'll have a hard time to show them later when the bidding gets high. One basic problem of opening distributional two suiters at one level is that if you open them and find no fit, you may get too high and partner may not be able to make successful penalty doubles. So generally speaking, it's a problem of hand evaluation indeed. For fewer losers hands, you can often make a lot of games or slams when you find a fit and when you don't find a fit, you may often get high. Modern bridge players tend to believe that they may eventually find a fit more often and their opps may not be alerted enough to penalize them when they overbid, including not doubling or not defending accurately. Also, 2/1 is good for this aggressive opening purpose cause opponents seldoms know you have stretched when you are in a gf sequence and may not make enough penalty doubles.

C4 and C5 makes a difference here. If partner plays C4, he has three clubs remained in his hand, if he plays C5, he's more likely to hold 654.

this is a simple H lead for both IMP and MP. H lead is the least likely to blow a trick and when you hit partner's long hearts, it's also good. if you really want to attack, S or D don't have to work out better than hearts. The same is true for C. Baically, you can expect two different types of 3nt bid, one is based on scattered HCP, the other is based on a long minor suit, for the first case, you really don't want to attack aggressively cause you may wait and take all declarer's losing guesses often. For the second case, you want to attack, but still, you don't know which suit to attack and H is still a very possible spot to attack. So H is really the percentage action and simple natural lead in my opinion. Change the suit into HQ432, it actually becomes less clear, cause this time you really need to hit partner's good hearts otherwise, H lead may blow a trick. So it may still be a good IMP lead, but less obvious for MP.

seems the basic design for gib's bidding is wrong. it's based on around 10 randomly generated hands, and the constraints for these hands can be very wrong. Also that simply leads to random bids, I have tested gib, in the same situation, it gives some completely different auctions, including a choice between pass 6M and raising it to 7M or raising 2nt to 3nt and bidding 6D with Jxxxx over 2nt opening. I somehow think this is the major barrier to attract more players to play money bridge. Some of my friends quit money bridge basically due to this problem. If computer player plays a decent game, I am sure many really don't mind partnering with them even if they lose some money. It's just like playing in ACBL, you pay a table fee, partnering with gib, you also pay a table fee and sometimes you may win good amount of money. I heard Jack's bidding is based on rules which is definetely a superior design comparing with gib's, because there is little randomness and jack's behavior in bidding can be predictable if it's truly based on rules.

this was posted almost one year ago, and we still don't see the new feature yet.

You can play a raising structure here: 1x double: 2x: simple raise(can be weak if you play those transfer bids) 2x+1: GF raise. 2x+2: invitational. 2x+3: some type of splinter or anything else you like to play. 3x-1: constructive raise with at least 9 trumps. 3x: preemptive.

I think gib is better than average club players and I don't mind partnering with them. Honestly, I feel less tortured partnering with gib. The only thing is that I need to adjust my game strategy. If gib screw up somebody, it's not that gib sucks, but that somebody should try to adjust their games I believe. Anyway, it's a creative idea to introduce gib into on line bridge gambling.

wow, glad to know that you also like table tennis!

wow, glad to know that they are going to make a movie of ender's game. I am a big fan of ender's series.

well, I am also at U of Utah and a phd student here. Welcome!

I think I read somewhere that the standard deviation of bridge deals is about 5.5 IMPs/board. So in a 64-board match, you are looking at about +44 IMPs. So you would need to win by ~88 IMPs for it to be significant. Do you have any idea how much the edge a team of 4 experts would have over 4 average players in terms of IMPs/board? Final piece in the jigsaw! Expert (I'm talking super expert here) against average players I think is like 1.5 IMPs/board. So that means the average players would win a 64-board match about 1.5% of the time. That's a huge advange though. Drop that down to 1.0/board and the average players win 7.3%. I really doubt it. in a 64 board match, the winning percentage for average flight B players against a world class team is very close to zero.

This is really an easy 2C to me. 2C can be as good as bad 16, if partner holds two aces and C fit, he should keep the bidding alive.

4S is wrong I believe. For this 4S, it shows exact 4-2-2-5, pure hands, and around 14 HCP. Also, AAKK are probably too strong for it. a typical holding is: SAQJx Hxx Dxx CAQJxx or AKxx xx xx KQJxx AKx xx xx AKJTxx is just too good for 4S bid. Also, you may not even belong to sp slam if partner happens to hold SQxxxx Ax Axx Qxx

Glad to know that France women's team won their first venice cup. Congratulations, especially my friend Benedicte Cronier(bbo id, patapon). Junyi

My solution is to rebid 1nt with 5-4 minimum. So now your 2D shows some extra, or 5-5. (you may get to 3nt with 5-5 minimum facing a invitational 2nt hand, so be strong to play this method.) After that, you can give up invitational 2nt and play 2nt as gameforcing. If you want to invite, you can start from 3C to invite in either S/H/D.(opener bids 3D to show min and bid his hand naturally otherwise) Over 2NT, opener can bid 3C as a waiting bid, 3D to show long D, extra, 3H: long H, extra, 3S: 3 sp and extra. Also, 3D/H/S by responders can be all treated as gf.