smerriman

I don't recall the last time I was disconnected from BBO; it was probably several years ago, and I don't have the highest quality internet. So what you are experiencing definitely doesn't affect everyone and so can't be solely an issue on BBO's side. The exception is playing robot challenges / tournaments, which can give an error if you are playing at exactly the time of day that all the daylongs finish and are being calculated (?). No disconnection, just some table errors. Edit - I'm running a poll on BW. Check back soon for the results..

I'll probably regret breaking my resolve and replying to your post, but anyway. I've told you many times that you cannot look solely at the description, see your hand matches, and think it's OK to bid. The description is meant to show what the bid guarantees, NOT the criteria for making the bid; they are two unrelated things. GIB's definition of 2♣ in its code is either a balanced hand with 22+ HCP, or a very distributional hand with 19+ HCP. It is not capable of using "or" in descriptions, so it combines the lowest common denominators. To GIB, your hand is therefore a completely impossible 2♣ opener, despite the fact it matches the description. You have to actually understand the purpose of bids before making them. 6NT is an awful bid, don't get me wrong. But so is 2♣. You've said numerous times in the past you're just going to open 2♣ with unsuitable hands whatever others think, which is your choice; GIB will respond by jumping to slam, whatever you think; that's its choice. GIB's never going to change; you can if you want this to stop happening.

Correct. And of course, there are plenty of occasions where it may look irrelevant to you, but does actually impact bridge logic (e.g. the 2 and 3 from KT32 does make a big difference if you're the opening leader and play 4th best leads, which the robots do give some weight to). But it affects GIB either way. This wouldn't make sense, since paying attention to what hands each person can hold based on the bids they did / didn't make in the auction is a vital part of bridge logic (even if the robots don't do it very well at times). What (I think) I was referring to by that line was not having dummy's played cards involved in the seeding, but it's possible that's done already.

If you're playing in an individual robot tournament, where robots are in the other three seats, no; the robot will always play identically if you play identically. In general, it's still making choices based on random numbers, but the random numbers are synchronized at different tables, so it's guaranteed to come up with the same choice at every table. I assume this is the hand you're talking about. 6 people made 1Nx, and 6 went down, but if you look closely, there were numerous differences by the humans in the first 4 tricks. 3 people played the ♠A to trick 2, while nobody else did, giving them a different set of random numbers from that point onwards. They all went down. Of the rest, 3 people played the ♦9 at trick 1, while nobody else did; they all made. Of the rest, 2 people played the ♦6 to trick 1, but they played different diamonds at trick 3, leading to different random numbers and one making, one going down. That leaves 4 people, all of whom had identical play to the first three tricks. You were the only one to lead the J of diamonds to trick 4, separating you off from the rest. All of the others lead the 3 of diamonds, but the one that went down followed with the 9, while the two that made followed with the 6. Of course, it may seem silly that small changes in what card you play affects the results, and in many cases there is no bridge sense to why it caused the robot to change its play; it just came up with a different coin toss. But in your example, there were no occasions when the cards were identical and the robots differed - it was always the human differing first.

Nah, it has 3♠ as the strong heart raise, and 2NT then 3♥ as the heart invite. This has popped up in a previous thread somewhere and it only does it when holding heart support AND 5 spades. Not in the old database but most likely a clash with a rule where 2♠ is natural.

I think AI is a long way off "figure out if playing spot card x vs spot card y is likely to give interpretable and helpful information to my human partner and more than what it costs by giving the same information to my human opponents" yet. To be honest I'm not sure if it'll ever get there, but if it does, it will be fun to see..

Can't disagree with that :) Though I would point out that my (and NookAI's) experience is that robots who signal accurately are considerably *worse* bridge players, because they are far too easy to exploit. So it depends whether you're wanting competition, or a learning tool.

Agree with mycroft. You could replace 'GIB robots' in your post with 'any weak players' and exactly the same is true. Read literally any of AL78's posts about the quality of bidding and play at his club, and the horrible bad habits that people there have picked up that keep seeming to work against him. Or kibitz any human table in BBO's main bridge club.. Obviously nobody should be told to learn how to bid and play from GIB, just as they shouldn't learn how to bid and play from these humans. But saying this means they are a disservice to the game completely ignores the huge positives that they bring, allowing people quick access to the game at any time they like, without the bullying (and much worse partners) that comes with many other forms of it. And with the right mindset, it is possible to learn a lot playing with GIB, even if it's of the form why not to do what GIB always does. And I would go further and say mycroft's point a) is far more damaging than what you "learn" from robots. Seeing people say "oops" solely because they go down in a contract that double dummy says is makeable (and vice versa) drives me insane, and means they never notice or learn from the true mistakes. But again, it's a good tool when used correctly, but easily misused.

No, hands still match the 1NT opener; I'm just saying that you can't look at one specific bid and argue that GIB should completely ignore another one, or that anything is really an 'error' when you've contradicted yourself. I don't know how GIB resolves contradictions like this; but the variance that GIB has built in for dealing hands 'close' to what you have seems to disappear, because everything is too far away from what you've shown. So all dealt hands end up as extreme maximums to get as close as possible. As Helene mentioned, all 17 counts include the Ace. Might look further at this in part 2 of my reverse-engineering GIB post, if I ever around to publishing part 1 after many months of planning it..

But the generally undefined / impossible 3N shows 25+ HCP (game opposite partner's 0); the text description is just a rough attempt at resolving the contradiction. All simulated hands have South holding the ace as Helene suggested. The bidding is more a system failure than anything else; GIB has no way of inviting with this hand, so the transfer is used for anything with 0-8 HCP. Much simpler when you can superaccept.

The decision between ♣T vs spade isn't even *close* in the old version of GIB; it throws a spade every day of the week. Got no explanation for how BBO have messed this one up since then at all.

Would suggest providing specific details - does "can't chat" mean they can't find where the chat interface is, or they can't type into the chat line, or they can type in and not submit, or submit and it appears to send but others at the table can't see it, or .. lots of options here :)

That quantifies declarer advantage, but not what I was referring to - which was whether declarer advantage *changes* based on whether you're leading into the strong hand vs the weak hand.

Recent post on BW about how someone was thinking about a hand, the opponents called the director upset about the slow play.. and suddenly they were disconnected from BBO and had to log back in to see they had been subbed.

You, by johnu.

The simulations do show that it never costs to overtake the heart. Sadly they also show that it never costs to duck :( Not because it makes any assumption that your play of the queen shows the jack; even the trick before that, every single simulated hand show you had to hold both queen and jack to make up your bid. Not getting GIB to relax assumptions about your bids, at least late in the hand where you have enough time to test every alternative, is one of GIB's biggest flaws :(

With 4NT available if you wanted to play in a minor, I don't think you need a second bid specifically designed to stop on a dime of 4m. 4m is the wrong place to play if you're making 11 tricks, wrong most of the time that you're making 9 tricks, wrong any time it pushes them into a making game..

He said they both bid the same, so the auctions were identical? Really not sure what you're getting at.. from both your comments I think you've totally misread the post. And you cut off the key part of the sentence which says 'yet East had one more discard against my opponent'. This discard changes absolutely nothing about the hand in a bridge sense, just that it forces a new set of simulations, which is precisely what Thorvald was saying by the first half of the sentence. He was asking whether one simulation was more likely than the other to get to desired outcome. In hand 1, no. In hand 2, yes.

Apologies for cutting your post short, but the issue is at the first hurdle since GIB plays the double as takeout (as do most humans). You can see this from the description. Takeout up to 4♥ is the most common agreement (and often higher, e.g. definitely 5m for me) - what are you going to do in South when you have both majors and shortness in clubs, for example, which will be pretty common? You simply can't afford to not have a way to get into the auction. If I knew you couldn't find your fits over 4 level preempts I would be bidding them very regularly! If you reconsider North's position from the context of responding to a takeout double, you'll probably realise why it now thinks it holds a very strong hand. Now, maybe it should bid 4NT to show a two suiter, though I don't think it has that defined. But it's going to be interested in slam one way or the other. I still agree that its initial pass is not great though, but GIB is too points based to figure that out.

? He didn't. First one is just random, nothing to do with what you did. The second one makes sense though; GIB knowing more about the hand is precisely why it plays the ace. East can never tell if one card would put you to more of a guess than another card if both are double dummy equals. It bases its choices solely on the rare cases where one is actually better than the other. These rare exceptions can easily swing the result either way, and there are definitely less exceptions later on. Of course, East can beat you by ruffing a spade at trick 2, so it's not without risk..

I'm a bit confused. You said you "passed" and conceded 13 tricks, implying as normal that you're South and were just switched in the bidding diagram. Why would you want to do that? I can understand not liking GIB's initial pass, and not liking South's double, which both look wrong to me, but having done both of those, 5♣ seems reasonable to me.. obviously, North can't pass, and 4♥ seems very much an underbid. The main reason you're not making slam is due to South having too many clubs.

Low from Qxx, high from xx(x) regardless of support for me. More important for partner to know if we have an honor, and I'm not leading the queen except in unusual circumstances. But I know some hold to the principle of leading the honor. Against a suit, count is more useful in general, so that's when it changes based on the bidding; low from xxx but high if we've supported and can give the honor info instead.

Is 5♣ an option? Not sure if it'll help since it seems to come down to whether we have a diamond loser or not, but I may still choose it if I can't find a better way to find that out. 2♠ followed by 3♦ over 3♣ is my other choice though that's not always going to help if partner has xx, assuming they accept with Qxx. Definitely interested in the best solution here.

Just for reference, I recall this poll on BW a couple of years ago on a similar (albeit not identical) auction which showed agreements were well divided.

Thus is definitely true, though on the flipside the NT opener is very tightly defined, so you could argue that having the more variable hand hidden could be tricky too. But I think the hidden honors works out better. But I think an even greater thing that DD fails to take into account is the opening lead. This is always where the biggest swing between DD and real life comes into play, and I suspect that leading into the strong hand is much riskier than leading into the weak hand. DD ignores this entirely, though it's hard to quantify.