BBO Discussion Forums: This surprised even me - BBO Discussion Forums

Jump to content

Page 1 of 1
  • You cannot start a new topic
  • You cannot reply to this topic

This surprised even me

#1 User is offline   wbartley 

  • PipPipPipPip
  • Group: Full Members
  • Posts: 290
  • Joined: 2010-July-23

Posted 2023-February-02, 08:07

https://tinyurl.com/22earn64

Anyone have a plausible explanation for this?
0

#2 User is offline   helene_t 

  • The Abbess
  • PipPipPipPipPipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 17,197
  • Joined: 2004-April-22
  • Gender:Female
  • Location:UK

Posted 2023-February-02, 08:22

View Postwbartley, on 2023-February-02, 08:07, said:

https://tinyurl.com/22earn64

Anyone have a plausible explanation for this?

We have sometimes seen GIB defaulting to pass when there was a hole in its database. For example, it used to pass a 1NT opening if it had a hand that was too strong for 6NT but too weak for 7NT.

I thought that bug had been fixed, but I can't think of any other reason why it would pass here. It seems unlikely that the simulations would show that pass is the winning action.
The world would be such a happy place, if only everyone played Acol :) --- TramTicket
0

#3 User is offline   johnu 

  • PipPipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 5,030
  • Joined: 2008-September-10
  • Gender:Male

Posted 2023-February-02, 12:39

Quote

This surprised even me


Sometimes I LOL, sometimes I pound the keyboard in anger, but I'm never surprised :)
0

#4 User is online   smerriman 

  • PipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 4,023
  • Joined: 2014-March-15
  • Gender:Male

Posted 2023-February-02, 13:16

View Posthelene_t, on 2023-February-02, 08:22, said:

It seems unlikely that the simulations would show that pass is the winning action.

Based on past threads, wbartley only ever plays with basic robots, which don't simulate. Advanced robots simulate, see that passing is insane, and thus never pass.

To elaborate on the actual issue, the database basically says to bid 4 you need help in spades that this hand doesn't qualify for, and that you should stop in 3 with at most 12 HCP.

But both of those, as well as 2NT, 3NT, and even 3 are marked as permitted bids for advanced robots to choose between.

I reported this one myself on the forum ~ 6 years ago and it has come up in other threads since, but when I discovered it was a basic robot problem only, I realised that there isn't really any upside in them fixing it - plugging this type of bidding database gap is one of the main reasons the advanced robot algorithm exists.

It's possible that the database gap is even deliberate, since it makes some sense that this is a situation where you would only want to decide by simulation anyway, and basic robots disabling half the algorithm was never really an intended feature of the code.
0

#5 User is offline   thepossum 

  • PipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 2,565
  • Joined: 2018-July-04
  • Gender:Male
  • Location:Australia

Posted 2023-February-02, 17:55

I think the problem is that South's bid was not forcing enough. - sorry it is forcing to 3 isnt it :)

I find the game try bids rather confusing though, and inconsistent
You never know for sure what kind of try and how they are interpreted - asking for help or something else
0

Page 1 of 1
  • You cannot start a new topic
  • You cannot reply to this topic

1 User(s) are reading this topic
0 members, 1 guests, 0 anonymous users