smerriman

See: https://blog.bridgebase.com/2020/05/22/how-do-bbo-points-work/ - there are section awards and overall awards.

Like I said, it's a very commonly occurring bug in GIB, and obviously nothing akin to real bridge. Avoid opening 2♣ wherever possible, and you'll avoid experiencing it.

It makes some sense that cheaper minor followed by continuing over a non-forcing bid can be used to show a strong hand. But on that condition, there are plenty of bids less than 6NT which accomplish that. Just that none of these sequences are defined, so after its first couple of bids, it will always just add the points and jump to 6NT. While 2♣ is of course wrong here, I've been burned by this too many times when holding legitimate 2♣ openers to have much sympathy for GIB.

Yes. GIBs bidding after 2♣ is extremely buggy; unless you have a strong balanced hand, or a 20+ one suited hand, opening 2♣ will rarely work well. It'll always jump to 6nt on its second or third bid with values like this, no matter what you do. 2♥ bust is a much better system, and would be easier to program too.

I'm not sure if you quoted the wrong post, since I didn't ridicule anyone / call anyone names. My question to pgcto was an honest one; how would having access to the data benefit you? Other than reducing messages to the BBO staff (which is not me), I can't think of one.

Like I said, if declarer has J (doubleton or trebleton), the K is right. But if West has the J, low is right. Here's a copy of your diagram, with a slight tweak, and the same play up to that point: [hv=https://www.bridgebase.com/tools/handviewer.html?https://www.bridgebase.com/tools/handviewer.html?bbo=y&lin=pn|South%2CRobot%2CRobot%2CRobot|st||md|4SK9543H42D862CAKQ%2CSA87HAJ7DQJ75C962%2CSQJT62HQ963DC8743%2CSHKT85DAKT943CJT5|sv|N|ah|Board%202|mb|1d|an|Minor%20suit%20opening%20--%203%2B%20!D%3B%2011-21%20HCP%3B%2012-22%20total%20points|mb|1s|an|One-level%20overcall%20--%205%2B%20!S%3B%208-17%20HCP%3B%209-19%20total%20points|mb|2s|an|Cue%3A%20limit%20raise%20or%20better%20--%204%2B%20!D%3B%2011%2B%20total%20points|mb|D|an|3%2B%20!S%3B%207-9%20total%20points|mb|3h|an|4%20!D%3B%204%2B%20!H%3B%2021-%20HCP%3B%2014-22%20total%20points|mb|p|mb|3n|an|4%2B%20!D%3B%204-%20!H%3B%204-%20!S%3B%2013-20%20HCP%3B%20stop%20in%20!S|mb|4s|an|3%2B%20!S%3B%207-9%20total%20points|mb|p|mb|p|mb|D|an|4%2B%20!D%3B%204-%20!H%3B%204-%20!S%3B%2013-20%20HCP%3B%20stop%20in%20!S|mb|p|mb|p|mb|p|pc|DQ|pc|ST|pc|D4|pc|D8|pc|S6|pc|D3|pc|SK|pc|S7|pc|D6|pc|D5|pc|S2|pc|DT|pc|SQ|pc|D9|pc|S4|pc|SA|pc|HA|pc|H6|pc|H8|pc|H4|pc|H7|pc|H3|pc|H5|pc|HJ|pc|S3|pc|S8|pc|SJ|pc|HT|pc|C3|pc|CT|pc|CA|pc|C2|pc|S9|pc|C6|pc|H9|pc|DA|pc|S5|pc|C9|pc|HQ|pc|C5|pc|CK|pc|DJ|pc|C7|pc|CJ|pc|CQ|pc|D7|pc|C4|pc|DK|pc|H2|pc|D2|pc|C8|pc|HK|]400|300[/hv] Click through to the same trick and you'll see East has to play low. GIB takes both cases into account (no matter whether someone might have played differently earlier with that holding). If its simulated deals involve enough cases when West has the J, it will play low. K is clearly the right play for humans, primarily due to inferences based on earlier play - like why declarer didn't cover with the 9. But GIB doesn't ask those questions, and when you give it a losing option, its simulations may stumble upon too many of the failing cases to flip the balance.

There's a huge difference between the two scenarios. In the second case, North has one trump left in dummy. If declarer has the J, clearly playing the K is best. But if declarer doesn't have the J (no assumptions are made about what card West will return from what holding), clearly playing low is better, as West can knock out the last trump and prevent declarer from ruffing what would likely be a diamond loser. While GIB will still play the K more often than not, giving it a losing option sometimes works. This is a useful strategy for stealing extra tricks, like when you have an open suit, cashing a side ace may cause GIB to think that suit is equally likely to be open too, and guess wrong.

1♦. Maybe there are rare hands where that gets passed out and we have game, but I'd rather take that risk in order to drastically simplify being able to describe my hand.

This is a limitation on iOS's side. If it's moved into the background, they suspend the app shortly afterwards and do not allow it to run any code at all. (Things like music are handled via other means). So from BBO's perspective, there is literally no way to distinguish someone that switched apps from someone who disconnected. BBO could potentially change how long it takes for someone who disconnected to get logged out, but I suspect it's small for a reason.

But aren't those only difficult if you choose to bid them? That is, if you bid 2♣ with all invitational hands (or weak with diamonds) and 2♦ with all game forcing hands, then that seemingly already covers everything you were able to do with NMF. The only downside is being unable to sign off in 2♣, but a big upside is having more space when game forcing. It's definitely a lot more complex when you want to build in all of the other bids, but that doesn't seem to be a necessity when comparing to NMF.

I don't know Acol either, but if 3♠ shows that hand, what does 2♠ show? In Standard American, with an invitational hand and 3 spades, you bid 2♠, so there 3♠ is forcing. But maybe this is different in Acol due to 4 card majors.

Actually, this is more interesting than I thought, and it looks like it does affect advanced GIB too. At the point GIB has to choose a discard on the king of diamonds, there is still a 100% single dummy line (the one nullve showed, discarding both spade honors). So even though there were easier routes, it's fine for even advanced GIB to get to this position. If I recreate that situation multiple times with the older version that I'm using to test - with the single dummy GIBson engine enabled - it's discarding the K about half the time and making the contract.. but throwing the club the other half and going down. But it's not because the sample is too small and it believes throwing the club will always work - it understand that line of makes if the diamonds are 4-4 or the club ace is with the short diamonds, so that logic is right. It's doing so because it doesn't believe throwing the king is a 100% line - when it lists out holdings that its chosen line works for, there is always one where it believes it will get 11 tricks. So when those 'failing' cases outnumber the above failing cases, it chooses the club. Every time I re-run it it's coming up with different holdings where it believes it can go down.. why it can never seem to find the right line I have no idea; you'd think it'd come up with it once in a while. (Actually, eventually on about the 20th attempt it came up with SK being 100%). So easy to replicate too, so would be straightforward to debug :( Not that that would make it trivial to fix; the single dummy algorithm described in Matt Ginsberg's paper takes a fair bit of understanding, and I could understand it getting stuck in local minimums etc. But still, definitely improveable.

Was this basic GIB or advanced GIB? Makes total sense if it were basic GIB due to the simplistic way it works, and you get what you pay for, but shocking if it were the paid robot which has proper declarer play.

The last GIB update was 3.5 years ago, and the last update that involved its play logic was 8 years ago (with that only being forcing it to cash an ace against 7nt). So I wouldn't hold your breath :(

Even with the code that GIB has built in to allow fluctuations, when I ran this through an older version of GIB, it generated 290 possible deals and in every single one of them South had AK of spades for its 3NT bid. It therefore makes no difference what card it plays, and chooses one at random. GIB is well known to trust the bidding too much. Yes, you'd think it'd be very logical when down to so few cards left to try dealing all distinct hands, even if you give the other ones a very low weight to break ties only. But it doesn't.

The reason that affects your TCR is because it is an inconvenience to directors just as much as anyone else quitting - all they see is that someone has quit and have to go to the table to find out what's going on. For what purpose? The only people that post in the forum upset about them believe they are wrong (and they never are). Sure, it wouldn't hurt, but it involves time to build something new into the UI.. and it would have no benefit to the vast majority of users. If you're under a certain TCR requirement (or even close), the best solution is to not quit tournaments :)

Registration has closed - matches will be announced shortly. Sorry everyone for the long break in between - even though it's really not much effort to organise, I've just never quite gotten around to it. Since we haven't done this for a while, I'll make this the most challenging format - MP, non-best-hand, and will allow a full 2 weeks of registration ending July 29, 11:59PM EST. As usual, those who haven't played before, read the following before registering:

Just to summarise: If you have double dummy activated, every time a card is played, it loads in double dummy results on all cards in the next hand to play. If you click next card after it loads, the double dummy score moves with the card to the middle of the trick. If you click next card before it loads (or have it set to trick by trick, so it autoplays the next card for you), it stops loading double dummy and moves to the next hand; thus the score isn't shown in the middle of the trick. I interpreted the OP's post as meaning when they toggle double dummy on, it was missing values on some cards in your hand, as opposed to cards in the middle of the trick. If it's indeed the latter, then that's easily explainable, though pretty minor.

If you are going to analyse your hands, you're definitely going to want to do this. All of your existing posts have proven pretty definitively that your score is often completely unrelated to how well you played (true in general, but even more so due to the high variance at other tables). As you can easily get a bad score for playing well, you'll also have a lot of hands where you made mistakes and got a good score.

There's a stickied thread in the BBO support forum. In brief, just type: [hv=pasteurlhere]400|300[/hv] GIB used to bid 1♦ in this situation but was reprogrammed at some point in the past to follow the more modern route of 1♥ which in itself I have no problem with. But many of the follow-ups that are required for this to work weren't added at the same time, and as a result it will never show its 5 card diamond suit later (or even a 9 card diamond suit) - there are quite a few bugs that have come up with this sequence. The rest of the auction is just a result of the missing diamond logic; it considers you to have extras for doubling, it considers itself to have at most 8 HCP for not bidding anything, and thus decides it has to make some non-minimum rebid.

I think I would bid 2♥, it's an awful hand but it's going to be an equally awful decision next time round if I pass, and both 1♥ and 3♥ feel way too much.

I think there's a distinct chance mikeh forgot [didn't realise, was only mentioned in other threads] you were playing 4 card majors, given all of his references to the number of spades in each hand (your partner likely having 3, you having "only 5" and already shown your hand, etc). So while his post makes complete sense in a 5 card major context, I think it's at least worth a clarification as to whether any of it still applies. Perhaps the other factors mean you shouldn't compete to the level of the fit, but 'grotesque' to bid an extra level when you have an extra trump is surely a misunderstanding.

No, you'd be trying to take 9 cards in a *fit*, as you wouldn't have bid a new suit otherwise as per the above condition. It may be 2:1, but if you gain 1/3 of the time and had support to be at the 3 level anyway, what's to lose?

You don't need to rush - start with a simple 1♠ which tells your partner you have 4+ spades and 6+ points. This is a forcing bid, so you can continue to describe your hand in later rounds of the auction. There are lots of ways the auction can continue. For example, if your partner bids 1NT, the most common convention at lower levels is New minor forcing, where you bid 2♣ and will find out from partner's response whether you want to be playing in 3NT or 4♠. Or if partner continues with 2♣, then you can bid 2♥ as Fourth suit forcing and will also find out if partner has 3 spades for you. But it's never possible to show everything in one bid - keep things simple and allow the auction to develop as you gradually learn more about each others' hands.

Yeah, it looks like while GIB has a definition for xx as South, it doesn't define xx by North. Advanced robots will continue with a jump to 3♠ followed by a 3NT signoff so won't get stuck in this case, though the xx should be defined nonetheless.