EdwardPiwowar / BBA

2 stars 0 forks source link

From the horror cabinet #644

Open ThorvaldAagaard opened 1 week ago

ThorvaldAagaard commented 1 week ago

0E8E82B5993CCADFD884F9DC53A7 image 7C looks wrong The penalty double is just bad

0577FF3A88476C5F45D7E435FC6E image

With a known 10-card fit East should show trump Q, then at least 6S can be bid

0CDE169755DC253E3A36B2284CAB image Table 2: Rebidding 3D with 4-card support is wrong

0120039B805DCE36776472803608 image

6S??

105E761F904388AD460F26A1BCED image

East should show the 18-19 bal

1C1565D7CE28ED2F4BE92043A3E9 image

Asking for aces, but then pass only missing an ace so 7 could still be on

1E57D8F29CC1EA5EFC7586228018 image

The splinter is perfect for East at table 2, so missing the slam is bad

1D75637DAE2E0E0960BFBCF01502 image

Really not worth a slam try on the east hand, so pass to 3N

1625F55DC7B09C490B04EF2610AE image

4N is just to optimistic

This was the first 26 boards of 433 I recorded long time ago for Soloway Jump shift

ThePokerDude commented 5 days ago

B4 0E8E82B5993CCADFD884F9DC53A7 South's hand is a typical semi forcing. If we do not have a semiforcing opening we should open with 1s. North should respond with 3nt to 3s

B11 0577FF3A88476C5F45D7E435FC6E

With a known 10-card fit East should show trump Q, then at least 6S can be bid Definitely - this needs to be hard coded. Not sure how you implemented the trum queen ask. The most simple one - you check for quuen trump in the hand of the responder to the question.

A more advanced one would be to keep track of known trumps to each of the bidders. East knows of 10 and west knows of 9 So for the queen trump ask the more advanced code would look like: if KnownTrumps >=10 then true else if Trump queen there then true else false end if

B14 0CDE169755DC253E3A36B2284CAB

general rule as opener if you have fit choose a bid which shows fit immediately only as responder it is allowed to go for a delayed fit showing

B15 0120039B805DCE36776472803608

Table 1: 3s instead of 6s related to B14 "only as responder it is allowed to go for a delayed fit showing" 1s p 2d p 2h p - now you bid 3s (if 2s could bo dubleton). As you are forced to game you show 3+ support in spades and extra values.

Table 2: the jump to 4h is wrong as north should not exclude the possibility of 3nt after 1s p 2d p It is better to bid 2h instead of 4h - a hand with no extras in HCP

B21 105E761F904388AD460F26A1BCED

west's first bid is wrong - if fit not guaranteed (only 4c) and 5+ cards in onother suit -> bid other suit so insted of 2c west shoud bid 1d or 2d if 2d shows a strong hand

Afterwards 1c p 2c p - here 3nt 18-19 balanced

B22 1C1565D7CE28ED2F4BE92043A3E9

Table 2 1s p 2c 4h 4nt p 5c p

West passes (probably) because west considers the option of 0 key cards I propose that you implement a HCP plausiblity check. 2c game force means 12+HCP Together we have 14+12 (or more) = 26+ HCP -> maximum of 14 HCP missing HCP of missing key cards = 15 (3 aces + Kc) -> partner cannot have 0 key cards

B23 1E57D8F29CC1EA5EFC7586228018

As 4h is a calcualted bid the calculation after splinter bids seems wrong To decide if to continue (cue bid or kcb) or to sign off you should check a weighted point count the weight factors should be 1 for cards outside splinter suit, 0.8 for ace in splinter suit, 0.3 for king in splinter suit, 0 for Q and J in splinter suit. In this situation:

14/15 weighed points -> continue <14 -> sign off Off course this model could be refined but I think it could be a good starting point for splinter evaluation

B24 1D75637DAE2E0E0960BFBCF01502 Table 1 When partner shows 0, so we miss 2 we need to pass to 5d The calculation behind the 6d bid is wrong. If we have asked for aces, we have to include the information for our decision/calculation. For a slam the "missing aces" count has to be 1 or less.

B26 1625F55DC7B09C490B04EF2610AE Table 1

Here I see two different problems. Let's say asking for aces is ok - 1st - we should implement 4d as kcb instead of 4nt 2nd - if blackwood is only possible with 4nt then if trumps are clubs or diamonds we need 5nt as sign off

here after 5s we had to bid a hopeless slam with 2 aces missing. If 5nt would be sign off, we would have good chances making it.

EdwardPiwowar commented 3 days ago

B11 0577FF3A88476C5F45D7E435FC6E There is a condition for 9 trump and Jack, but there is no condition for 10 trump Next BBA image

EdwardPiwowar commented 3 days ago

B14 0CDE169755DC253E3A36B2284CAB Next BBA image

EdwardPiwowar commented 3 days ago

B15 0120039B805DCE36776472803608 Currently image But I think we have an agreement for 4H if 55

EdwardPiwowar commented 3 days ago

B21 105E761F904388AD460F26A1BCED Next BBA image

EdwardPiwowar commented 3 days ago

B22 1C1565D7CE28ED2F4BE92043A3E9 I will leave this hand because the bidding is different now, but I will check how the hcp plausiblity check works. image

EdwardPiwowar commented 3 days ago

BBA will calculate 6 for a slightly different or slightly stronger hand. image In this hand 35 to 53 SD dummy is weak.

EdwardPiwowar commented 3 days ago

Currently BBA stops on 5D. RKC at a minor is always a risk and difficult to implementation. image

EdwardPiwowar commented 3 days ago

B26 1625F55DC7B09C490B04EF2610AE Crosswood is implemented but 5N is currently king ask.