iandayre

You hold Q4, KJ764, A9, AT53. Unfavorable and it goes 1C, P, 1NT to you. 2H or Pass?

No problem here. 2C over 1H and 2S over 1S. Tougher: Axx, x, AQxx. Qxxxx. Still raise 1S to 2, but 1H creates a problem. I would open 1D here, just can't stomach rebidding 1NT with a small singleton.

I am pretty sure that "exhibit #3459" is a joking comment indicating he has seen the problem many times before. Since you mention Bergen, I'll tell you what he recommends - Pass a forcing 1NT response with a minimum balanced hand that would pass the delayed limit raise. I have been playing this way for years, and while nothing works 100% of the time, I am way ahead of the game. Bergen further recommends that the jump shift to the other major shows any splinter raise. First step asks, show up the line. This leaves immediate 4C and 4D calls open, which he uses as: 4C - Balanced 3 card raise, stronger than 3NT (say 16-18) and 4D - Stronger preemptive game raise. Immediate 4M raise weaker. Jumps to 4 of the other major are preemptive, natural, to play.

The best article I have ever seen about this issue was in the ACBL Bulletin, by George Rosencranz, very early in my duplicate career (late 70's). I have been playing it with regular partners since. It uses the first 4 of Stephen's preferences. I am 100% in agreement that all 3NT bids, fast or slow, show stoppers. The only difference with Stephen is that the slow cue is played as showing a stopper - implying slam interest since 3NT was not bid. I recall once in a strong Swiss team match, I used that action and we reached 7D, an easy make. Our expert opponents languished in 3NT. The same article, which ran over several issues of the Bulletin, covers all aspects of defense against Weak 2's. It is still valuable almost 40 years later.

I made this observation some years ago. I believe BBO did at least comment on it. GIB does raise to 3H occasionally I have since learned, but it is definitely overly aggressive in raising to game on this auction. The actual hand is a clear-cut 3H call.

As far as I know, the word "penalty" does not occur in any GIB description. I find it curious that GIB uses the concept of "takeout" quite often but "penalty" is outside its vocabulary.

OK, the descriptions don't come first. Basically you are saying that the programmers do the best they can, and when they are done and can see the descriptions of various calls, they see how close they came. About right? Let me ask this. GIB does pretty well at No Trump bidding. It consistently opens with 15-17 HCP and relatively balanced distributions. Let's say they wanted to change this to adopt a 12-14 NT. By no means am I suggesting this should happen. But in general terms, how would this be approached?

Stephen, I would like to understand this better. I completely understand that there must be great difficulty in programming GIB to bid accurately. Isn't that a different issue? Let's take a simple example: the many situations where the bidding description calls for 25+ or something close for a particular action. There are many, many others. Doesn't the description come first, then the programmer tries to get GIB to bid according to that description? If the descriptions are not in accordance with sound bridge logic, GIB's bidding will be poor even if programmed perfectly, no? As for Jack's comment regarding Ginsberg's bridge pedigree, I was going by Fred's message to me, now I think about 15 months ago. I am sure you read it. I know nothing of him other than from that.

GIB doesn't understand the concept that you should get out of the auction ASAP on misfit hands. Often it trots out new - often 4 card - suits at the 3 level when any experienced player would have passed, even with a void in partner's rebid suit.

The current GIB version is 35. This can be found in the top pinned posting in the GIB Robot Discussion area. The original author's name is Matthew Ginsberg. He sold the program to BBO quite some years ago. His focus in creating the program was on card play. Unfortunately he had intermediate (at best) knowledge of bidding. Thus the myriad poor descriptions of bids in GIB. BBO slowly makes improvements but there is still an immense distance to go, and little financial incentive for BBO to speed the process. Issue reporting mechanisms are: 1) This Forum and 2) While playing in a tournament you can click on "Send Robot Report" to tell BBO about GIB issues.

I blame South. Negative double the first time possible. Having not done so, must show some life (3S) Opening 1NT isn't bad, but even if S invites, how can N accept?

It should be noted that NMF/Checkback are hardly new. They were being played in the 60's, when a Bridge World editor described it as a "petty little odious bid". For a time the convention took the name PLOB based on that comment. The convention was quite well established when I began playing duplicate in 1977.

Disagree. The actual hand probably lacks sufficient overall strength. Make it a 6-7 count and 2S seems a big improvement over selling out at the 2 level with a known majority of strength.

Right of course, it's not at all recent. It's just that I'd expect to see less of it as the years go by.

I am wondering - are beginners today taught invitational jump rebids by responder? After 1C-1S-1NT, rebids such as 3C, 2NT and 3S are invitational. This has been standard for at least 4 decades, but every time I play a tournament I see "intermediate" players making these calls with game-going hands, and presumably wondering why they were passed out below game.

The double does not show Hearts. Fortunately, GIB does not play one of the worst conventions ever devised, stolen bid doubles.

Few worse slam hands for GIB than this one. http://tinyurl.com/gosc4t6 Over my splinter 3H, GIB cue bids, allowing us to stop at 5S. Other players bid only 3S with my hand instead of splintering (not unreasonable). GIB goes off the rails. For starters, it bids BW with the C void. After the 1 or 4 reply, N assumes 4!. But in that case, why not just bid 7S???. It asks for Kings, a truly pointless exercise, then signs off at the 6 level. Of all GIB's weaknesses, I don't think there are any situations where it is quite so futile as when bidding 5NT. To top it off, the opponents cashed the correct two aces against me, making 5. ALL of those in 6 were allowed to make after a Club lead, ruffed.

With most good human bidders, once cuebidding starts, you continue to show controls held. It doesn't work that way with GIB. Each additional cue bid shows additional strength. Your second cue bid of 5C showed 22 total points! With GIB you must sign off in 4S over 4H.

I wasn't referring to the description. I said the actual, 11 HCP hand is a solid responsive double.

As long as there have been computer-dealt cards in live tournaments, there have been players who believe they are reviewed and modified by tournament organizers. An similar and equally ridiculous idea that won't go away. But Jack, why the need to download hand files to learn your percentage? BBO provides this information without extra work on our parts.

The only correct bid by S was accepting the transfer, which never should have happened anyway.

Agreed GIB should pass 4D. The double, though, seems like a perfectly normal responsive double to me.

I agree at the 3 level. At the two level, they are an excellent tool and should be retained. GIB actually handles them fairly well, given its inability to evaluate properly opposite a splinter. If you bid 2S over 1D with AQJTx, KJx, AQxx, x, then follow with 3(or 4 if necessary) Clubs, GIB won't promote 3 small nor de-evaluate KJx in that suit.

Perhaps so, but GIB transfers with a 5 card suit 100% of the time. You cannot sensibly pass 2S.

GIB likes to make calls like 3H AS IF they meant a general game try. Even its own GIB partners rarely if never interpret them that way. Lots of luck for the human ones.