smerriman

Can't get much closer than that - just a reminder, if it's still tied after the last set, you'll play 8 board challenges until you have a winner.

With 2H and 2D, you'll be in the wrong red suit though, won't you, as partner will surely pass 3♦. Or are you happy to be playing in the 5-2 fit? You also mentioned you might have a 9 card diamond fit, but if you're saying West promises at least 3 clubs for the double, the same logic applies to diamonds, so a 9 card diamond fit is impossible. So based on your logic of wanting to declare when partner has 2 hearts, you'll only actually be declaring in an 8 card fit if partner also has exactly 3 diamonds, which seems a very narrow target.

Partner specifically chose to bid spades over either of our suits, and if he does pick one over 3♦, it will probably be the wrong one (even if he knew we were 6-5, it'd look like 6 diamonds). I can't see any positive to bidding.

Found it. So consistently replicated at other tables.. .. in that case I'm baffled, since barmar has definitively stated the basic robots don't simulate during the auction. So it shouldn't be possible for them to ever consistently come up with a different bid from a bidding table..

You can confirm at a practice table what GIB is programmed to do (and consistently does): [hv=https://www.bridgebase.com/tools/handviewer.html?lin=st||pn|human,,~~M9693m3h,|md|1SKT9862HT96DKCJ92,S4HAK432D943CKQ74,SAQ73H7DQJT862CA5,SJ5HQJ85DA75CT863|sv|e|rh||ah|Board%203|mb|2S|an|Weak%20two%20bid%20--%201-4%20!C;%201-3%20!D;%201-3%20!H;%206+%20!S;%2010-%20HCP;%207+%20total%20points|mb|D|mb|4S|an|The%20Law:%2010%20trump%20-%3E%20game%20support%20--%202+%20!S%20|mb|P|mb|P|mb|P|]400|300[/hv] So if this happened in the main bridge club then it must have been some sort of server error rather than related to GIB. What happened at the other tables?

Was this with an advanced robot? Basic GIB bids 4♠ as you'd expect. Maybe an "advanced" robot had a simulation meltdown which happens from time to time..

He says nothing of the sort in the PDF; his advice aligns with Vampyr's above. Of course, it's 100% irrelevant to this hand, since the double was artificial and not for penalty, and of course the passes were due to a bug. Specifically that GIB doesn't know what to do when you make insane bids as a human.

You answered your main question yourself. 3♠ shows an invitational hand with 4 spades, so the reason you 'mess around with 1nt' is to tell partner you have an invitational hand with only 3 spades. Opener knowing whether you have 3 or 4 makes a huge difference in them being able to determine whether to accept or not. This hand is perfect for such a 3 card raise, so 1NT.

[hv=https://www.bridgebase.com/tools/handviewer.html?bbo=y&lin=pn||st||md|1S2356AH479JD57C59%2CSTQH2KAD34QC478QK%2CS489KH68QD269JC26%2C|rh||ah||sv|b|mb|p|mb|1N|mb|p|mb|3N|mb|p|mb|p|mb|p|pc|S4|pc|S7|pc|SA|pc|ST|pc|S2|pc|SQ|pc|SK|pc|SJ|pc|S9|pc|H3|pc|S5|pc|D3|pc|S8|pc|D8|pc|S3|pc|H2|pc|D2|mc|9|]400|300[/hv] You probably wouldn't have expected to be making this after the spade lead. Plain (un)lucky, or is there a case for leading the 8 even if playing 4th best?

Looks like there was definitely a server glitch; have seen it before but only with robot tables. Lin shows the diamond 4 being played twice, breaking the handviewer link. So I doubt it's something you could have resolved yourself.

It's not really an agreement, just logic. You're a passed hand, didn't preempt in spades, and are telling your partner you're willing to play at the 4 level even if they're void in spades - you *must* have diamonds.

It's a pretty tiny mistake, really. The solution is perfectly fine, as is the wording, just that the demo shows you cashing the Ace before the first trump. Of course, it should happen after the first trump, but has no effect on anything else.

It sounds like you have some major computer issues. Have just tested and it takes me literally 10 seconds to go from not having BBO open at all to be logged in.

There is no way any robot could possibly have jumped to 4♠, because 4♠ is a weak preemptive bid. Perhaps you meant that all of the humans sitting in South jumped to 4♠ (which is a commonly seen beginner mistake). Basic GIB works solely by point count; it requires 13 total points to game force with 2NT, and will stick to a limit raise with 12. It counts total points by adding points for shortness, but not when that shortness includes an honor, so it counts this as 12 (but would game force if you moved the club or heart honor elsewhere). But yes, of course its evaluation method is pretty poor as this is an obvious 2NT bid.

Good luck in the final!

You're in a very small minority there. As Kit Woolsey put it: Virtually all 15-16 counts of that shape will be opening 1NT, because there is simply no other option.

No, only advanced robots. Basic robots are unable to 'count tricks' while bidding, only make book bids based on points + distribution.

5♦ is just plain undefined. GIB tends to pass if you make a bid that doesn't exist. Yes, it shouldn't, but really, just don't make undefined bids :)

What Paul said. There would have been another auction to 4♥; they don't have the play records either.

If you are a sub and you finish the tournament, it will be counted towards your TCR. If the person you subbed for returns, your TCR won't be affected. TCR is updated once per day, though it's a rolling average over 60 days (and requires completing at least 10 tournaments in that period) so it can take a number of tournaments to see significant changes.

To be honest, that would be far more complex. The point of having the same seed is that if everything is equal going in, then everything is equal going out. If you try to change the behaviour from the first time it runs to later times, not only do you have to save a lot more information, but it would also mess everything up from that point forward, since the random number generator will be in a different state for later simulations that you haven't saved.

Given the last challenge in fact did not affect the qualifiers, I'm going to minimise the drama and continue the semifinals as per the final results: nige1 vs billyfung2 garant_7 vs icycookie 4x16 board challenges, same format. I won't set an official deadline on this but let's aim for approximately a week.

No, basic bots never simulate during the auction. Both bots simulate during the play (advanced with a higher sample). That's why any bidding bugs are reported with basic bots to see what the 'book' bid is, and can be 100% replicated at a bidding table.

If you're playing with a basic robot, it will make exactly the same bid in the same situation. If you're playing with an advanced robot, it will start with the basic robot's bid, then simulate a number of hands and see whether other bids would lead to a better final result. In a tournament situation, different tables will match the random seeds and ensure the robot runs the same simulations and comes up with the same bid. Anywhere else, like the situation your described, the simulated hands will differ, and therefore the outcome can differ every time the hand is plaeyd.

You've probably read about the system where 2♠ is a transfer to clubs, where opener can suggest playing in 3NT by superaccepting with 2NT, as others have alluded to. Then you can still stop in 3♣ when partner really is weak, but find some light 3NT games when you have a good fit, if not the full points for a normal game. With GIBs system, it's never possible for opener to bid 3NT, because a 15-17 hand opposite potentially 0 points from partner is never enough to 'cover the other suits'.