Antrax

I've taken the liberty of summarizing your post.

Again, if you can afford it, it shouldn't happen if you rent an advanced robot (assuming you're using the web client)

I bid 2♠ and ended up in 4 anyway, and made 11 tricks! Hurrah.

Err...

Checkers is EXPTIME-complete and it was solved recently. Don't confuse generalizations of games with finite real-world games.

[edit] Odd, I was replying in another thread.

It happens occasionally (you only notice it when you're damaged by it). The robot plays by creating random hands consistent with what it's seen so far. If you're unlucky, the robot might randomly choose weird hands and take a strange line of play. This is mostly applicable to the basic robot - the advanced one is considerably less likely to suffer from this.

Probably an abbreviation of the Southern pronunciation "Padner"

Given the 1♣ opening: 1♣1-♠ 1NT-2♣* 2♦**-3NT * "Checkback stayman", asks for 3 card spade support or a 4-card heart suit ** No major fit, minimum for the 1NT rebid range After E shows no spade support and a minimum, W knows they have two balanced hands totaling 30-31 HCP, so 3NT is the best spot.

What's the problem you're trying to solve?

"Relaxed" is pretty much what you want.

One comment about Stephen Tu's post: I think part 2 of "Learn to play Bridge" should be tackled after reading books such as "Bridge for Dummies". About card play, Gitelman's Bridgemaster 2000 is well worth the investment.

You either don't, or play some form of Stayman that caters to it.

Not so obvious to me :( What's wrong with that?

Can I nominate EarlPurple for newcomer of the year? His join date says 2003 somehow.

[edited nonsense out. Still too early]

Is there a typo in the bidding, or should we really find a way to make 12 tricks on this hand?

The iOS version seems to be 3.64.

TBH, what I did was just move to four-suit transfers, which gives responder another bid to describe his hand. I'm not a great theoretician so I can't really comment about your idea.

The very simple method I had when I played like this had opener bidding 4m if he's interested in slam and 3NT if he's not. Once 4m agrees that we're going to play in that minor, you can cue bid on the way to 4NT to ensure you won't get too high. It's not a very good method (wastes a lot of space) and it might not really be SAYC - I was taught all sorts of things as "SAYC" that turned out not to be so.

(Technically that's a DDoS with an extra "D" for "distributed", but nobody does the old-fashioned version anymore)

Have you read the explanation for the cue bid? It often shows a higher minimum range than is required for your bids up to now, so GIB simulates it unlikely you won't have a club control.

It's an easy fix for a human. Not sure how easy it is for GIB (AFAIK it has no rules when playing the hand, only simulations). More to the point, is this really the problem with GIB, that psyching 1NT with a 2NT opener makes it give up an overtrick? Assuming you have limited time to work on it, wouldn't you rather fix something else?

It's explained later:

Yes, it's a no-win play but that's not how GIB thinks. GIB knows declarer has ♣xx left so it doesn't matter what it pitches, so it pitches whatever. The thought of the ♠ always being a safe discard unless declarer revoked and the ♣A being safe only if declarer counted his HCP correctly doesn't enter its head.