[xeno123]

Figure one or other of the GIBs should have been swearing at its partner's bidding over this uncontested auction where they ended up in 4♠ with a 3-2 fit:

---

I'm guessing the 3♠ bid was some sort of probe for NT.

Down 4 in this contract was a top - none of the other 10 players in this contract doubled either. Only one EW GIB pair found the cold (as the cards lie) 5♣ contract.

[Bbradley62]

Holy mackerel! So much wrong with this auction and the explanations:

• 2♣ should show 5+♣, not simply biddable (which could be only four)
  
• East isn't strong enough to reverse.
  
• The explanation of 3♠ can't right. Does it imply that East is trying to find a Moysian, since if West had four spades he would have bid them already?
  
• Sometimes, GIB's understanding of a bid is completely different than the explanation, because the understanding and the explanation are derived independently. But I don' t think I've ever seen one GIB agree with the explanation (3♠) while another one didn't.

[1eyedjack]

Until I read the explanation here I hadn't realised that the 2H rebid shows a reverse (ie extras). Reason being that 2H rebid following an inverted minor raise does NOT show a reverse, and I often rely on that pitiful crutch because a 2N rebid is so wide ranging while remaining non-forcing, and the 2N rebid means the same regardless of which 2-minor response you get to a 1D opener.

I am unconvinced that I will never encounter a hand with only 4x Clubs with which I would not want to respond 2C. Have to think about that.

Presumably the OP's "top" on the board resulted from superior defensive card play, because all E-W would bid the hand identically to the same spot. Perhaps some Souths took action over 1D?

[Bbradley62]

1eyedjack, on 2014-October-25, 20:59, said:

I am unconvinced that I will never encounter a hand with only 4x Clubs with which I would not want to respond 2C. Have to think about that.

If you have a 4card major along with your 4 Clubs, you'd bid the major over partner's 1♦. Otherwise, if you don't have enough diamonds to raise, you'd have a no trump hand. (Note that West was a passed hand here, so he's not a 3334 that's too strong to reply 3N.)

1eyedjack, on 2014-October-25, 20:59, said:

Presumably the OP's "top" on the board resulted from superior defensive card play, because all E-W would bid the hand identically to the same spot. Perhaps some Souths took action over 1D?

It's also possible (I didn't check) that OP played this hand in the MBC, not in a tournament, in which case many of the EW seats were occupied by humans, not by robots.

[Cthulhu D]

1eyedjack, on 2014-October-25, 20:59, said:

Presumably the OP's "top" on the board resulted from superior defensive card play, because all E-W would bid the hand identically to the same spot. Perhaps some Souths took action over 1D?

Not for sure - GIB uses simulations to determine what the is the best bid in a number of places, and those sims are seeded with a random number generator. It's possible that different seeds result in sufficently different auctions to generate a different bid.

[1eyedjack]

Cthulhu D, on 2014-October-26, 07:00, said:

Not for sure - GIB uses simulations to determine what the is the best bid in a number of places, and those sims are seeded with a random number generator. It's possible that different seeds result in sufficently different auctions to generate a different bid.

No, that is not (generally) possible. There have been a number of other threads about this.

It is possible in an Instant tourney, or in an MBC event, where different versions of GIB may be sitting in the same seat at different tables. But in a true "robo dup" and indeed in an Instant tourney if you disregard your own table, there is no scope for variation by robots presented with (precisely) the same initial conditions (including order of play).

You are correct that different seeds may well generate different actions. But GIB uses the same seed at each table. This is I think desirable.

[Bbradley62]

Well, just because there was random speculation... this was, in fact, a RoboDupe. Most Souths doubled 1♦, leading to different results than were achieved at OP's table.