Help
This was one of two consecutive boards with GiB as partner.
Perhaps partly my fault, a minor splinter over 1NT is a bizarre convention which I have no intention to lose sleep over: if I missed some obscure developments to show a maximum and/or clubs then mea culpa. I hoped 3NT would limit damage and be neutral, but 4♥ promising (if I use the UI) 4+ clubs, 4- "biddable" hearts and 12+ total points (does that include the shortage?) seemed discouraging for slam in clubs or NT. Any guesses (or certainties) what would have happened if I had bid 4NT or 5♣ here? I passed.
37 out of 39 were in 6NT making.
Page 1 of 1
From bad
#1
- Group: Advanced Members
- Posts: 7,481
- Joined: 2014-February-18
- Gender:Male
- Location:Italy
Posted 2022-December-05, 16:42
#2
- Group: Advanced Members
- Posts: 3,899
- Joined: 2014-March-15
- Gender:Male
Posted 2022-December-05, 19:02
Agree having a minor splinter is pretty silly in the first place.
To GIB 4♣ over 3♦ shows a maximum with club support, which would have worked out well as it would have then followed up with Blackwood and then put you in 6♣, which you could correct to 6NT if you thought that was best.
After 3NT, GIB's book bid is 4NT quantitative, which would also have worked out well. But GIB simulates to see if something else will work better.
One problem here is that it thinks you'd only accept a quantitative 4NT with 17 HCP rather than some 16s.
In this auction it's extremely rare for you to actually hold a 17 count in the first place. Since it doesn't run that many sims when bidding (under 20, if the old version of GIB is similar), most of the time a 17 count never even shows up. (Maybe influenced by a lack of 4♣ the first time?)
So in almost all cases, all roads will lead to you responding 5♣ - because whether it bids 4N, 4♥, or even 4♠, it assumes you'll correct to the best partnership fit, which is 5♣ in virtually all cases, including this one (preferring a 4-3 heart fit to 5-4 clubs was a little odd, though not as odd as GIB's choices of course
).
In reality, if you bid 5♣, it will then resimulate and raise to 6♣, so you would have gotten to slam that way too. But it doesn't know that when choosing a bid over 3NT - the "book bid" when it extrapolates the auction would be to pass 5♣ - so its decision making at that point won't work out to be very accurate.
When I ran this with the old version of GIB, its most common continuation over 3NT was.. 5NT invitational to 7! But not because it wanted to play grand, just because it figured slam was worth a shot, and since those 17 counts weren't showing up, 5NT and 6NT tied for the best result, and the former just happened to be earlier in the database which broke the tie.
Next most common, but almost as regular as 5N, was passing 3NT.
Occasionally it bid 4N, when the 17 counts did appear.
And on a rare occasion, it bid 4♥.. because a couple of the simulated hands had you holding 3253 shapes, where it expected you to correct 4♥ to 4♠ (everything else ending up in 5♣), and the database tell it to jump to 6N over 4♠, and on those hands 6NT made. I guess that's what would have happened in your case.
To GIB 4♣ over 3♦ shows a maximum with club support, which would have worked out well as it would have then followed up with Blackwood and then put you in 6♣, which you could correct to 6NT if you thought that was best.
After 3NT, GIB's book bid is 4NT quantitative, which would also have worked out well. But GIB simulates to see if something else will work better.
One problem here is that it thinks you'd only accept a quantitative 4NT with 17 HCP rather than some 16s.
In this auction it's extremely rare for you to actually hold a 17 count in the first place. Since it doesn't run that many sims when bidding (under 20, if the old version of GIB is similar), most of the time a 17 count never even shows up. (Maybe influenced by a lack of 4♣ the first time?)
So in almost all cases, all roads will lead to you responding 5♣ - because whether it bids 4N, 4♥, or even 4♠, it assumes you'll correct to the best partnership fit, which is 5♣ in virtually all cases, including this one (preferring a 4-3 heart fit to 5-4 clubs was a little odd, though not as odd as GIB's choices of course
).In reality, if you bid 5♣, it will then resimulate and raise to 6♣, so you would have gotten to slam that way too. But it doesn't know that when choosing a bid over 3NT - the "book bid" when it extrapolates the auction would be to pass 5♣ - so its decision making at that point won't work out to be very accurate.
When I ran this with the old version of GIB, its most common continuation over 3NT was.. 5NT invitational to 7! But not because it wanted to play grand, just because it figured slam was worth a shot, and since those 17 counts weren't showing up, 5NT and 6NT tied for the best result, and the former just happened to be earlier in the database which broke the tie.
Next most common, but almost as regular as 5N, was passing 3NT.
Occasionally it bid 4N, when the 17 counts did appear.
And on a rare occasion, it bid 4♥.. because a couple of the simulated hands had you holding 3253 shapes, where it expected you to correct 4♥ to 4♠ (everything else ending up in 5♣), and the database tell it to jump to 6N over 4♠, and on those hands 6NT made. I guess that's what would have happened in your case.
#3
- Group: Advanced Members
- Posts: 7,481
- Joined: 2014-February-18
- Gender:Male
- Location:Italy
Posted 2022-December-06, 15:16
Thanks.(Cannot upvote from this device)
Page 1 of 1