iandayre

I think you are playing results too much Lycier. On the prior thread you want to be in slam with the DK onside. (you are right that 2S was an underbid). But on this hand, you don't want to be in the nearly 90% (with no opposing bidding) grand slam? And with the opposing bidding it is nearly 100%. If you went down it was declarer error.

Given the description of the 2NT rebid as 11-13, one would assume it is not game forcing. Descriptions do indicate forcing to 3NT when that is what is intended. So I see no problem on Hand one, opener should rebid 3D which will probably be passed. On the second hand, responder is certainly too strong for a non-forcing 2NT. Playing 2/1 with the exception of the auction 1D-2C, which some do, responder would have to rebid 3NT. But this is a very difficult hand, no game is good but I am not aware of a system that can stay out of one. Possibly it could go 1D-2C-2D-2NT (if GF)-3S and with the inability to bid 3NT N would bid 4C and pass S's 4D. But even that contract will often be defeated. Let's think about the second hand without the SQ, making the 2NT rebid reasonable in a non-GF world. Opener with the misfit should still rebid 3D rather than 3NT. You overbid both hands IMO Lycier.

Certainly GIB's criteria for bids needs work. Any human player as N would have bid 4C over 3S, and S could bid 4H showing support with no extras. But GIB has not yet been programmed to understand that it is possible to have fits and two suits, so 4H would be taken as a cuebid, leading to disaster. Let's get to work on this programmers. Every bid description should be re-evaluated. There are so many that have nothing to do with reality.

I appreciate you trying to help Antrax but I couldn't begin to follow your directions if my life depended on it. I have never heard of an "HV Tag" and I have no idea what you are telling me to do with it, if I did know what it was.

Interesting question whether passing 2D or correcting to 2H was more likely to keep you out of your Spade game. Mind you I have sympathy with the pass.

Yes, it is a support double situation for GIB, so the second round pass denied 3 or more diamonds. I believe most human partnership only use support doubles over 1M responses. I have played that double in that auction (specifically 1C P 1D 1S Dbl) shows 4 hearts without the strength to reverse. This is one of two conventional calls sometimes referred to as Snapdragon.

I agree with Steve. GIB's random playing of honor cards and high spots helps declarer frequently, and very rarely helps its defense. It should be deleted from the programming.

Apparently GIB always considers only the next bid, not the rest of the auction. Clearly, doubling again, intending to pass partner's 4H call, which will happen 90+% of the time, is nonsensical. I would never have doubled the first time, much less the second. GIB simply can't evaluate its own long suits.

As BBradley suggests, the problem was using Blackwood in the first place with no control in an unbid suit.

I have tried to post the hand but I can't figure out how to do it. I saved it in my own file but I don't see how to transfer it to the BB. At IMPS South held Jxxx, AKx, AQT, Kxx. North held K9x, Qx, K, AQJTxxx. At most tables, including mine, South opened 1NT, 2NT trf to C, 3C, 3D shortness, 3NT, 4NT Blackwood for C, 5C 0 or 3, 6C. At these tables W lead from xx of Spades, AQ of S and a ruff, down 2 for -8.6 IMPS And I thought I might not be high enough. But at one table the auction proceeded the same way through 4NT, and opener showed only 2 Key Cards with 5H. N still bid 6C but got a different lead, making 13 tricks for +13.4 IMPS! A 22 IMP reward for misresponding to Blackwood! This isn't a complaint about the bots, but it is interesting sometimes how a very minor change in the auction will cause them to make a different opening lead, often leading to a very different result. A couple of pairs bid 6NT which of course also fails on a S lead, but it was not found. Actually yes, one complaint about the bots. A human N, finding the partnership off 2 Key Cards, would bid 5S, an unbid suit, asking partner to correct to 5NT. The bots don't seem to know that bid, I have seen several instances where it could have been used, but was not.

Sure hands could be constructed where pass leads to the only plus, but they are strongly against the odds here. One should always assume that a takeout doubler holds a small singleton in the opponent's suit until further clarification. Also, the advancer's hand is good enough that a plus will usually be quite achievable on offense

4D?? Another candidate for the strangest bid ever made by GIB. Why can't it just jump to 4H over the double?

Definitely a description bug. One you should catch, as a minimum response to a takeout double can't possibly promise any values.

Yes, let's hope so. Funny, I don't believe I have ever seen it pass a 1 level response in competition, no matter what piece of junk it opened.

It has become standard that a jump to 3NT over an inverted minor raise shows 18-19, i. e. a hand that would have jumped to 2NT over a 1M response. With a 14 count that wants to be in game, you bid a major then 3NT your next call. With the actual hand, you should show your stopper with a 2S rebid, promising no extras. The proper GIB rebid is 2NT (not sure that is what it would actually do) and you would of course pass. I certainly would agree that GIB should pass 2NT on the actual auction as well.

Nowhere in the GIB convention card descriptions are Negative Free Bids listed. Therefore, one should safely be able to assume that after an opening bid by partner and an overcall, a new suit is forcing by an unpassed hand. But that's not the case, as my GIB partner opened 1C today, I got a 1H overcall and I bid 2D, passed out. I had a full opening bid and missed an easy game, the only compensation being that most of the field had the same result. What about this, GIB developers?

It takes some bridge judgement to decide when to run out from partner's suit. Unfortunately GIB seems to have the judgement exactly backwards. Here are two hands from an ACBL MP Robot duplicate today. GIB held 9, T9xx, Txx, AJxxx. In third seat it passed partner's 1S opener and LHO balanced with 1NT. Those partners (I was not one) who rebid 3S were treated to a 4H runout! This is a very common GIB error and it must stop. On the second hand, none vul, GIB East held KT8xxxx, Txx, Void, KJx. Partner W opened 2D, P, P, Dbl, P, P. 2S anyone? No, with a void in partner's suit and the opponents having shown a clear preference for defense against that contract, GIB left partner in 2DX. Minus 1100! Imagine, it runs out with hand 1 but not with hand 2.

The particularly irritating part was that those who made the novice error of reopening with 3D instead of double reached at least game, sometimes slam. I suppose one might consider passing the double, but it's a offbeat action with an unbid 6 card suit partner has indicated support for.

The particularly irritating part was that those who made the novice error of reopening with 3D instead of double reached at least game, sometimes slam. I suppose one might consider passing the double, but it's a offbeat action with an unbid 6 card suit partner has indicated support for.

Stephen Tu couldn't have been more right in his recent comment. The last thing GIB needs is new conventions. It needs better judgment, and better definitions of bids, in common auctions. GIB held Ax, AQx, T9xxxx, T9. 1S by me, 2H overcall, P, P back to me for reopening double. Surely it must find something other than a simple 3D call promising nothing! With KQxxx, x, AKJx, Kxx I considered a raise to 4D but that was described as showing 20+ HCP. So I passed and soon claimed the dreaded +190. A Club holds it to 11 tricks. I suppose the hand doesn't provide an easy bid, with 2NT, 3NT and 4D all possible, but 3D is terrible.

Stephen Tu couldn't have been more right in his recent comment. The last thing GIB needs is new conventions. It needs better judgment, and better definitions of bids, in common auctions. GIB held Ax, AQx, T9xxxx, T9. 1S by me, 2H overcall, P, P back to me for reopening double. Surely it must find something other than a simple 3D call promising nothing! With KQxxx, x, AKJx, Kxx I considered a raise to 4D but that was described as showing 20+ HCP. So I passed and soon claimed the dreaded +190. A Club holds it to 11 tricks. I suppose the hand doesn't provide an easy bid, with 2NT, 3NT and 4D all possible, but 3D is terrible.

I would open both hands 1D. I don't like rebidding 1NT with a singleton in any form of bridge, less so with GIB which will rebid a weak 5 card suit quite often. Max really said to open 1C when the D's are stronger? Seems like it should be the other way around - open the D when they are strong, otherwise the C. I knew Max Hardy at the table, one of the nicest gentlemen ever to play as well as a very accomplished writer and player.

I would open both hands 1D. I don't like rebidding 1NT with a singleton in any form of bridge, less so with GIB which will rebid a weak 5 card suit quite often. Max really said to open 1C when the D's are stronger? Seems like it should be the other way around - open the D when they are strong, otherwise the C. I knew Max Hardy at the table, one of the nicest gentlemen ever to play as well as a very accomplished writer and player.

Certainly 4NT should be for the minors. Funny hand, if 4S were making you would have had a great result going down undoubled in 4NT.