[thorvald]

I have seen some situations where the bot leads the 9 from T9, thus making it difficult for partner, when leading thru Jx

See this deal

---

So thru declarer it leads the T

---

But here it leads the 9

Now here it doesn't make a difference, but I have seen this cost contracts because the other bot assumes the T is on the closed hand

[smerriman]

thorvald, on 2022-August-18, 17:18, said:

I have seen this cost contracts becuase the other bot assumes the T is on the closed hand

It doesn't. While robots follow lead conventions most of the time (see here; I now think the reason it's not 100% is due to the flag about the weighting for correct signals), they don't make *any* inferences about past plays by either partner or opponents.

I.e., it does not reduce the simulated hands by excluding ones where a different lead would have been made. (And equally, when a finesse works, it think it's just as likely that it will fail the next time, even if ducking would have been a 100% losing play).

[thorvald]

smerriman, on 2022-August-18, 17:41, said:

It doesn't. While robots follow lead conventions most of the time (see here; I now think the reason it's not 100% is due to the flag about the weighting for correct signals), they don't make *any* inferences about past plays by either partner or opponents.

I.e., it does not reduce the simulated hands by excluding ones where a different lead would have been made. (And equally, when a finesse works, it think it's just as likely that it will fail the next time, even if ducking would have been a 100% losing play).

I remember this combination

Dummy Jxx
Declarer Q

I did play from dummy to Q and Ace and LHRobot returned the 9 from T9xx, and RHRobot used the King that i ruffed and got a discard on the J

But still reading the lead conventions (Thx for linking) I still can't figure out why leading the 9 can be right

[smerriman]

It wasn't right - most of the times GIB doesn't follow the rules, it has no legitimate reason to do so, and ended up working poorly.

It just seems the robot has been programmed to lead a false card a small proportion of the time - that's the only explanation I could come up with.

(Note the Nook AI project was able to destroy WBridge5 due to the fact WBridge5 carded too predictably - maybe that's what the developers were trying to avoid, though I'd rather it lead accurately 100% of the time.)

[thepossum]

I find the difficulty in reading the Bot's leads to be quite a strength

Was that a singleton, small from length, middle of nothing. Nope

I appreciate its not necessarily apropos of the OP but I have noticed a tendency to lead away from a King through an AQ on occasions

More seriously though I have looked at its opening leads a bit and was surprised at how disciplined (???) it appears to be

This is from the point of view of declarer. I keep struggling reading its opening leads in defence

[thorvald]

Now I got an example where the bot plays a card, that is difficult to read for partner

---

The bot leads ♠4 -J ->Q->A

Now the next time it leads the 8 from 987 - hard to see it ever being right, and I ended up giving away the contract