[mikl_plkcc]

---

What actually did the robot think?

[smerriman]

mikl_plkcc, on 2025-January-29, 14:36, said:

What actually did the robot think?

If East had paid attention they would have noticed their first double promised 5+ diamonds, and their second double promised 16+ HCP, so the robot would ask what actually was East thinking?

It just can't handle everyone else at the table psyching from its understanding of bids.

[jillybean]

Never my what what on the bots mind, what was North thinking when he let the bot play in 6NT undoubled?

[mikl_plkcc]

What if the robot escape to 7♦ and made it?

[johnu]

smerriman, on 2025-January-29, 14:43, said:

It just can't handle everyone else at the table psyching from its understanding of bids.

If robot believes East, than making 12 tricks in 6NT isn't nearly as good as beating 3NT by 7 or 8 tricks. Or if robot believes North, then if North can make 3NT, then how is 6NT going to work out. Does GIB make those types of score analysis?

[barmar]

johnu, on 2025-January-30, 05:19, said:

If robot believes East, than making 12 tricks in 6NT isn't nearly as good as beating 3NT by 7 or 8 tricks. Or if robot believes North, then if North can make 3NT, then how is 6NT going to work out. Does GIB make those types of score analysis?

If it's allowed to do simulations, it does. That's basically how the sims work -- it deals a bunch of possible hands, runs double dummy analysis, and determines the expected score for each action (passing the double versus bidding on). The action with the higher average score is chosen.

But if it's just getting the bid from the bidding DB, sometimes this isn't always taken into account.

[smerriman]

And in this case, the rule is "if you're balanced, have no fit with partner, and have enough combined points for 6NT, bid it and specifically skip the simulation process".