sfi

I start by looking at the hand, the vulnerability, the type of scoring, and the quality of players involved. Questions I am interested in include: Would a correct explanation have been likely to change RHO's choice of 3H? Was partner's pass of the double influenced by the explanation?

Before looking to change the tournament format, it is probably worth identifying the problem you are trying to solve. The round-robin format has received extremely strong support in the past two polls in addition to this one, so it's not clear what its perceived deficiencies are.

There is certainly an element of randomness coming out of the group stages, but I suspect that's hard to avoid with 16 board matches. It feels long enough that if you play reasonably well you can reliably make it into the knockout stage, which is the most important criterion. That being said, a longer group stage may work well. TBH, anything that avoids a Swiss-style format is a good thing, and the details mostly determine how many matches everyone plays before people start being eliminated.

Thanks - good match. And kudos to the organisers.

Agreed. I've had several sets determined largely by who is on the right side of the robot doing something ridiculous. That wasn't even the only double-digit swing the robot created in that match. P.S. I really like the current group format. Everyone gets a reasonable number of games and one bad set doesn't knock you out.

Set 2: sfi 23 - 9 stephen tu http://webutil.bridgebase.com/v2/tview.php?t=ARDCHALLENGE:3bb4b58f.d0c9.11e6.b596.0cc47a39aeb4-1483347339&u=sfi

Set 1 from the final is tied at 14 each. You think I would have learned to not trust partner's defence by now... http://webutil.bridgebase.com/v2/tview.php?t=ARDCHALLENGE:56b37009.d034.11e6.b596.0cc47a39aeb4-1483283389&u=sfi

Yeah - I can imagine that one was particularly annoying. But thanks for the game.

The point is that the analogy to a live bridge club is not necessarily accurate. Some people use it more as an internet version of home bridge or of a practice session where you want to discuss hands. For instance, this evening we had three of the four players talking to kibitzers while playing as well as the players themselves discussing the hand - some in the same room and some over skype. None of which skewed the results or the competitiveness of the session. Any number of these points would be inappropriate for many of the games on BBO, but they are extremely valuable to us. The fact that BBO provides options to suit a range of different uses is a positive, and limiting those options would be a step backwards.

Set 4: sfi 38 - 15 phil http://webutil.bridgebase.com/v2/tview.php?t=ARDCHALLENGE:4a157f29.cef8.11e6.b596.0cc47a39aeb4-1483147647&u=sfi Most of the big swings across the 64 boards seemed to be down to particularly bad robot decisions, which was unfortunate.

Set 3: sfi 13 - 3 phil http://webutil.bridgebase.com/v2/tview.php?t=ARDCHALLENGE:e76a9d94.ce4e.11e6.b596.0cc47a39aeb4-1483074896&u=sfi

It really isn't, and that is the essence of your problem. I don't say this lightly because I think most systems are reasonable, but forcing a 3H bid on this hand is simply bad. Bidding has moved on since Stayman wrote about it in the 1940's, and no good pair would give up all that bidding space on hands with slam interest. For that matter, almost no bad pair would give up the bidding space either. You really are by yourself when you choose 3H, and predictably you got into trouble for exactly the reasons you might expect.

Set 2: phil 35 - 19 sfi http://webutil.bridgebase.com/v2/tview.php?t=ARDCHALLENGE:c2298ecb.cbf7.11e6.b596.0cc47a39aeb4-1482817565&u=sfi

Without any specific agreements 2D is fine. However, there is a trend towards playing 2C as natural or balanced, which means 2D and 2H can now show 5 card suits. This sort of agreement helps find trump suits for slam early and means that you have a way to emphasise the balanced nature of your hand. These agreements work best playing a system where a 2/1 bid is forcing to game. And of course, you need to have these agreements with partner.

I am not in a position to play for the next day or two, so will reject and send you a new challenge.

Sure, 2D is forcing. But if it is game-forcing then I can safely bid 3S now to show a good hand with 3-card support. If not, I may need to do something else. What that is depends on our agreements, but this hand is pretty good and the auction has improved it. Even playing standard american, a lot of people agree that 3S is forcing on this sort of auction, just to cater for hands like this. There is a lot of merit in that, since an opening hand opposite a 2/1 with a fit is likely to make game more often than not, but you need to agree that with partner. Similarly, if it is forcing to at least 2NT, then 2S is still unlimited since I can't pass the bid. That means I want to encourage partner. On the other hand, if 2S is non-forcing our chances for slam aren't great and I will probably just raise to 4S. Once again, have clear agreements on what is forcing and what isn't is important, and it's hard to give a general answer without knowing this. Finally, if it were on BBO with a random player and no agreements, I would just bid 4S here and not worry about the slam. At worst it's going to be a reached by a bunch of other people.

On hand 1 I am going to choose 3C and bid them again if it makes sense to. Despite the suit we have a lot of losers and partner will surely make some noise if game is on. On hand 2 I take a quick peek at our system card before working out what to do. In particular, what level does 2D commit us to? This information is fairly important at this point in the auction.

The WBF doesn't forbid much, and almost nothing constructive that isn't an opening bid. Having lots of alerts doesn't need to slow the game down - most time is spent on trying to work out what to do rather than querying the opponents' system. Besides, ruling that a pair with lots of alerts is wasting time would be an odd decision, to say the least.

True, the time when they can get off is when they are actually in the most trouble. It's worth the person in the balancing seat being aware of this and doubling if they can handle the later auction anyway. For instance, they might be able to bid NT with a stopper.

Over the past couple of years my regular partnerships have switched to having a takeout double available after all strength-showing doubles & redoubles. This includes: 1C - (X) - XX (1NT) - X 1NT - (2D = artificial) - X and so on. It makes constructive bidding much better. Since your side has already shown values and partner is on the lookout to make a takeout double, you don't often lose the ability to penalise them anyway.

QF set 4: Mkgnao 45 - 19 sfi http://webutil.bridgebase.com/v2/tview.php?t=ARDCHALLENGE:d05dc8f8.c559.11e6.b596.0cc47a39aeb4-1482090021&u=sfi

QF set 3: sfi 29 - 14 mkgnao http://webutil.bridgebase.com/v2/tview.php?t=ARDCHALLENGE:6e445e66.c465.11e6.b596.0cc47a39aeb4-1481985060&u=sfi

QF set 2: sfi 35 - 28 mkgnao http://webutil.bridgebase.com/v2/tview.php?t=ARDCHALLENGE:c4a5d7c0.c2e4.11e6.b596.0cc47a39aeb4-1481819848&u=sfi

The most useful thing would be for the robots to simply deal with the psych in a sensible way - evaluate their hand and trust their partner. You know, the same way people deal with it in real life. There's no need to do something silly like banning them.

LOL, what do you expect the robots to do? Cancel the hand?