The bot miscounted the clubs

[thorvald]

[hv=url=https://www.bridgebase.com/tools/handviewer.html?bbo=y&lin=pn%7CHuman%2CRobot%2CRobot%2CRobot%7Cst%7C%7Cmd%7C1SAK975H95DA2CAJ95%2CSJT86HAJ76DQJT6C4%2CSHQ842D7543CQT862%2CSQ432HKT3DK98CK73%7Csv%7CB%7Cah%7CBoard%207%7Cmb%7C1S%7Can%7CMajor%20suit%20opening%20--%205%2B%20%21S%3B%2011-21%20HCP%3B%2012-22%20total%20points%7Cmb%7CP%7Cmb%7CP%7Cmb%7C1N%7Can%7CBalancing%20notrump%20--%2011-15%20HCP%7Cmb%7C2C%7Can%7C4%2B%20%21C%3B%205%2B%20%21S%3B%2021-%20HCP%3B%2015-22%20total%20points%7Cmb%7CP%7Cmb%7C2H%7Can%7C4-%20%21C%3B%204%2B%20%21H%3B%202-%20%21S%3B%205-%20HCP%3B%205%2B%20total%20points%7Cmb%7CP%7Cmb%7CP%7Cmb%7CP%7Cpc%7CS2%7Cpc%7CSA%7Cpc%7CS6%7Cpc%7CD3%7Cpc%7CSK%7Cpc%7CS8%7Cpc%7CD4%7Cpc%7CS3%7Cpc%7CS5%7Cpc%7CSJ%7Cpc%7CH2%7Cpc%7CS4%7Cpc%7CD5%7Cpc%7CD8%7Cpc%7CDA%7Cpc%7CD6%7Cpc%7CS7%7Cpc%7CST%7Cpc%7CH4%7Cpc%7CSQ%7Cpc%7CC2%7Cpc%7CC3%7Cpc%7CCJ%7Cpc%7CC4%7Cpc%7CCA%7Cpc%7CH7%7Cpc%7CC6%7Cpc%7CC7%7Cpc%7CH6%7Cpc%7CH8%7Cpc%7CHT%7Cpc%7CH5%7Cpc%7CH3%7Cpc%7CH9%7Cpc%7CHA%7Cpc%7CHQ%7Cpc%7CDQ%7Cpc%7CD7%7Cpc%7CD9%7Cpc%7CD2%7Cpc%7CDJ%7Cmc%7C6%7C]399|300[/hv]

 

Bidding a 4-card ♥-suit with 5-card ♣, and the explanation should probably be less than 4♣, and not 4-♣

 

I have noticed that after passing an opening bid the bot tends to make stupid bids with max ie. 4-5 HCP

[TylerE]

4- is GIBerish for 4 or more. It's a fit showing advance.

[helene_t]

My guess is that including points for the spade void, North is too strong to pass and too weak to bid 3♣. So only 2♥ (or 2♦) are left.

 

Maybe the problem is that North has too few HCPs to bid 3♣ while too many TPs to pass.

[smerriman]

The robot thinks that pass, 2♥, and 3♣ are all viable options here.

 

(It's true that a basic bot will always pick 2♥ amongst those options - partially for the reasons Helene said; it's too strong to pass. But this is completely unimportant here and not something that needs fixing, since the advanced robot will consider all three options).

 

And the simulation results in it picking 3♣ the majority of the time, pass almost all of the rest of the time, and 2♥ only on rare occasions. So no issues here as far as I can tell.

 

As always, the number of simulations run during bidding is small, so you'll occasionally get odd results. One of the times it came up with 2♥ was when it only simulated 17 hands, found in most of them South would correct 2♥ to 2♠ or 3♣ so it made no difference, and on one occasion only 2♥ happened to allow the opponents into the auction where they found themselves in 3♦ and you profited from them going down. On another occasion, 4♥ happened to be making in a Moysian fit, and that was the only way to bid it.

 

Edge cases like this are always to be expected when Monte Carlo simulations are involved.

[thorvald]

4- is GIBerish for 4 or more. It's a fit showing advance.

 

4- is 4 or less

 

4+ is 4 or more