Stephen Tu

Don't blame Fred; he (or rather BBO) bought GIB from its developer Matt Ginsberg. And even Ginsberg probably isn't solely responsible for this rule, as I believe he got the original bidding database from Meadowlark bridge (Rod Ludwig).

Why are you blaming your aggression? Blame the incompetence of your allies in the EW field. It shouldn't matter how many you open, E/W should not be missing game, or blowing the trick if they find 3nt. That's outside your control, all you can hope for is that these E/W are equally generous when you reach their table and get them as opps instead of allies. It may not even out in a single session but will over multiple sessions. I see this in lots of your posts; you get bad scores because your comparison field is abysmal (and often tiny therefore more random, since random outliers represent huge chunk of percentage on a board), or get bad scores because your partner did something rather nuts. Just worry about what are bad decisions by you, and try to get better partners or try to train up current partners. But I agree 3H looks normal to me.

https://web.mit.edu/...tter/index.html If you want to use different movement / # of rounds / # of boards you may have to figure out how to run the bridgemats program that generated the above:https://web.mit.edu/...mats/index.html which requires a Perl interpreter (I use https://strawberryperl.com/ on windows), and a postscript viewer to view the output. Might view the following old thread for reference: https://www.bridgebase.com/forums/topic/72739-pete-matthews-software-to-create-table-guide-cards/page__view__findpost__p__869166

It's very normal, not having established any fits yet, for double to be "strong hand, DSIP".

Just pretend you are in a cone of silence. Bid what you would have if you were screened off from seeing or hearing anybody, only saw the bids, no alerts/non-alerts, no correct/incorrect explanations, no unusual breaks in tempo. Assume partner remembered agreements. If you have multiple plausible actions, choose among ones that are *not* suggested to be more likely successful based on anything (alert or not, explanations, tempo, anything other than bids themselves) that suggests partner has forgotten what bids mean, except for impossible sequences. (Occasionally, partner makes a bid that can't exist systemically. One where you can figure out from the bid alone that something has gone awry, even without any missing/wrong alert/explanation. Only in those cases can you try to wriggle out.) Here, 1nt-2c!-p-2s, with partner alerting, explaining as 2c as showing diamonds, and bidding spades anyway, is clearly not impossible; partner is showing a spade suit of his own. So bid as if this happened (probably passing with most hands unless exceptionally good support for spades, I don't think 2S is normally forcing over a diamond showing call). Do this whether or not they alerted, because you didn't get to hear the alert or non-alert. Do this whether or not they said 2c showed diamonds, or if they said 2c showed majors, because you didn't get to hear that either. If they just alerted, and no one asked and no explanation was given, wouldn't you just assume partner remembered but prefers spades to diamonds? Whether the proper alert is technically deemed AI or not is irrelevant, because you are supposed to bid the same way regardless. You are less constrained if alert was proper, and no incorrect explanation given, because you certainly don't have any UI in that case, and can choose to do pretty much anything. I don't know what info you have from just an alert, all you know is partner thinks 2c is conventional. Since it is, you don't have anything useful to help you even if you wanted to break rules, either you trust partner remembers you showed diamonds but wants to bid spades anyway, or maybe you think partner thinks it's some other convention because they forget all the time, but how is that ever going to change your action anyway?

I don't think describing 4S as an underbid is accurate. I'd describe 4S more as inaccurate description/misbid vs 4c, rather than a strength difference. Both bids show very strong hands and force to game opposite a min negative double, and partner is allowed to go on, just one bid is showing a strong 19+ bal/semibal hand (and probably that 1 or 2 people around the table on this auction are bidding on not very much and shouldn't be taken too seriously), and the other is more accurately showing a distributional powerhouse (though this one more distributional / fewer high cards than normally expected), that would help partner evaluate better.

The problem is that you took the robot "out of book". This kind of auction isn't really supposed to exist, because hands with spade support are either supposed to raise immediately over the double (with 4 cd support, to 2S/3S/4S, or make a splinter, or bid 4c with a 4/6+ type of hand), or redouble (support, 3 cd spades exactly). So this sort of delayed, deny support initially, then show it later is not really supposed to ever happen, so it falls back to some default rule that it must be 3cd support, but North has no idea how strong it is (see explanation, up to 17 total points) because this sequence isn't programmed into the normal bidding book, because South was supposed to raise to 2 right away, which would show a minimum opener with support. There's no "pass then raise later to show that I opened a bit lighter then minimum on something that maybe shouldn't be opened". If you just raised immediately over the double, this probably limits your hand more and North maybe doesn't bid so much. Also maybe things like game tries with the North hand aren't really defined on this sequence, so it feels no options than just blasting game.

I think you misinterpreted the output of the calculator. The combos from one line to the next are not all of equal frequency and can't just be totaled, you have to look/use the percent column. For example, 4-1 hearts, 7-3 clubs there are 600 combos of these particular suits, but you have to look at the percentage column, it makes up 4.76 percent of hands. But 4-1 hearts, 6-4 clubs is 1050 combos making up 12.5% of hands. The latter combos are each worth more, because the first is multiplied by just the 4 ways to deal a singleton spade out of the 4 remaining, while the 2nd is multiplied by 6 ways to deal 2 spades. 1050/600 * 6/4 = 2.63 12.5% / 4.76% = 2.63 So my advice would be if you want to use that calculator to total based on the percentage column, not the combos column.

Not quite sure how you are calculating this.Aren't there 10 * 14c9 = 20020 combos where West is 3-1 in the reds, and 5 * 14c8 = 15015 combos where West is 4-1 in the reds? By my calcs:hearts 4-1, ck onside, 8580 out of the 15015 combos of x=4=1=x deals, need to hook (others down regardless). hearts 3-2, CK offside, 7150 out of the 20020 combos of x=3=1=x deals, need to go up CA (others anything works). So hook is favored. I think this holds as long as West finds the 2 pitches of hearts from 2-1 in the reds at least half the time. In real life, my opponents were proverbial LOLs with not a lot of awareness, and gave me no such difficulties. East led low on the third diamond to force West to ruff, which might be clever, but with this opponent maybe more likely just random. West, despite looking at HJxxx and the CK, led a heart (??!) and avoided immediate disaster when partner held stiff T, but of course I had no problems after that. But I was wondering what the right play was against better defense, I didn't know how much the odds of 3-2 heart break were swung by the 6-1 break not having studied it before, now I know it's swung only a little by a 5-2 break but quite a bit when 6-1. I think I am convinced that finesse club is best against good opps on the proposed defense. Other questions worth thinking about: - What should East's strategy be? Should he underlead 3rd round holding CK, to force partner to ruff but then also maybe dissuading declarer from finesse? Or should he lead high honors and trust expert partner to ruff good trick to lead the club? Should he only underlead 3rd round *not* holding CK trying to convince declarer not to hook? - What's the strategy if the diamond split were 2-5 rather than 1-6? The pure math swings strongly now to playing for 3-2 hearts, but also, isn't it easier to be awake to ruff and lead C from 4 hearts and CK, than to be aware to ruff and lead C from like HTxx and CJxxxxx?

[hv=pc=n&s=sqj532hakq4dt42c4&n=sakt6h9853dj75caq&d=s&v=e&b=3&a=1sp2n!p3c!p4sppp]266|200[/hv] 3C = club shortness. Matchpoints West leads a small diamond, which turns out to be a stiff. East plays Q,K,Ace. On the 2nd diamond west pitches a club. On the third diamond, west ruffs and leads a club. Thoughts?

I've seen this suggested before, but I don't think the early cheap trump setting is of much value. I'd rather have some scheme that helps opener evaluate vs responder's stiff or not to see whether they have too much wasted opposite that, to still get back to 3nt despite having a decent fit, when you have secure double stop and too many wasted values. So shortness asking, or maybe showing wastage opposite a stiff, or denying wastage opposite a stiff, etc.

Most common these days I see 1nt-3d to show this sort of hand, when playing 4-way transfers using up the 2s/2nt responses. Unfortunately there aren't standardized followups to 3d. 1nt-2s as minor suit stayman (followed by 3H on your example hand to show 55 with a stiff heart) used to be more common. Certainly 1nt-3H could conceivably show the same thing, with whatever you now used 1nt-3h for moved into a different bid.

Opener - (overcall) - cue bid of overcall, is 100% strong support for opener's suit (most commonly limit raise+), because hands without support have negative doubles and new suits (if not playing negative free bids) to force not having support. (Opening bid) - overcall - (pass) - cue bid of opener's suit, is *usually* support for partner's suit, but it might not be depending on: - if you are playing some advanced scheme involving transfer responses (which sometimes shift raises to a different bid) - whether you have some method of forcing with another suit or not (for some people, cue is 100% support, you have other methods of forcing in other suits, like say jump shifting, or playing new suits at certain levels F1, or playing a transfer scheme; for other people, all their new suits are NF, and jump shifts might be fit-showing, so they have to include some non-support hands in the cue bid) Now, in response to the above cue bids, if the bid support is a minor, people will strain to show stoppers in the opp's suit anyway, because reaching 3nt is a high priority even with a minor fit, since 9 tricks is often easier than 11. Less so with major fit, though with flat hands and the opp's suit well stopped 3nt should be offered. Cue bids by *opener* though, after responder has shown some suit of their own, are assumed *not* support, and just show a strong hand often looking for a stopper for 3nt. This is because there are usually a plethora of other ways for opener to show a raise (raise to 2, 3, 4 levels, jump cue as a splinter), but no other way to show just a huge one-suiter of their own. Also cues on later rounds of auction by responder/advancer are often gropes for 3nt, when the cue isn't of the first 2 types mentioned above.

That's because people never use RKC directly after 2c-2d-2nt without setting a suit first. It's inapplicable when a trump suit hasn't been set, 2nt-4nt and 2c-2d-2nt-4nt are absolutely never RKC. After 2c-2d-2nt, or 2nt-?, normally 4nt is played as a quantitative invite (bid 6 if max, pass with min, advanced partnerships may have methods to accept but also look for say a 4-4 minor or 5-3/6-3 minor fit also). 2nt - 4c is Gerber for a lot of players here, at least in countries where South African Texas is unpopular. Though some might prefer 4c to be part of set of bids to find minor suit slams (perhaps just showing long clubs). But it's quite rare to be dealt hands where you just want to have partner answer aces and then you can just count tricks and place the contract. So it's far more common over 2nt to search for a suit fit first, or to set a suit first then use keycard/kickback/minorwood/optional minorwood (e.g. texas transfer then 4nt, 2nt-3S! (puppet to 3nt)-3nt-4c/4d showing a minor one suiter) Gerber is more useful when your suit is solid like KQJTxx and you aren't missing K or Q of your suit, or you are going to chance it on power (~32+ hcp) but don't want to be off 2 aces. Or on some hands, you might start with stayman and then set a trump suit, then RKC (2nt-3c-3H-3s! (setting hearts by bidding other major)-?-4nt). Or one might jacoby transfer at 3 level and show a 2nd suit, then RKC if opener shows fit for the major. Or you might be playing GF jacoby transfers and know right away if opener has a major fit or not (if they only accept with 3, your system not catering to very weak hands being able to sign off in 3M opposite 2 only), then RKC. Only with beginner level players (or "permanent beginners") who have never learned about quantitative invites over natural NTs (or otherwise when no suit has been agreed) where one is supposed to treat 4nt as natural and pass it frequently. Gerber is fine over 2nt if you have the hand for it, it just very rarely comes up. The main thing is sequences like 2nt-3h!-3s-4c should not be Gerber (what if you have a good hand with spades and clubs, you want to find slam in either suit, you need 4c natural; if you just want to rkc in spades just bid 2nt-4h!-4s-4nt, if you just want to ask aces then 2nt-4c!). Or 2nt-3c-3h-4c should also not be Gerber (you have 4 spades 5 clubs or 4 spades 6 clubs, want to explore slam in either suit). Basically Gerber is fine as a direct bid over 1nt/2nt/natural 2nt rebids, but if you are looking for suit slams and need to be able to bid clubs naturally or to show control in clubs, you don't want to be interpreting 4c as Gerber all the time on all sequences. Bad club players want every 4c bid to be Gerber; good players use it on specific defined sequences only, and it just comes up extremely rarely, as usually some other sequence is better (to find suit slam with extra chances vs playing in NT, sometimes to find grand in a suit when 6 is the limit in NT).

If a suit has not yet been agreed, typically 4nt should just be a quantitative invite. Partner either passes or bids 6nt (or maybe 6 of his own suit if it is solid). With some possible exceptions where partner has shown a strong single suiter in a major at the 3 level where it's not possible to agree M then kc (since 4M would be to play), where you can agree 4nt is RKC for the major.