Join Bridge Winners
All comments by Arend Bayer
You are ignoring the author of this comment. Click to temporarily show the comment.
Isn't this a case where Larry Cohen and the Bridgewinners team clearly know what the right agreement is for their target audience? (A pickup partnership between two advanced, but not expert, players.)

I would be completely happy with some amount of dictatorial decisions - after all, I can't think of any more benevolent dictators than this team! You could still make a post “We propose the following decisions; if you strongly disagree with any of them, please comment and we may put them up for vote.”
Feb. 14, 2018
You are ignoring the author of this comment. Click to temporarily show the comment.
I think 6 is worse than 4N. Slam cannot be good opposite two keycards and no K. (Obviously slam cannot be good opposite a hand that would respect a 4 sign-off either, but at least at that point slam was still a theoretical possibility.)
Feb. 11, 2018
You are ignoring the author of this comment. Click to temporarily show the comment.
If 2 denies a slam try, responder can often jump to 4M without giving away unnecessary information.
Meanwhile, to have a slam try opposite a passed hand, opener probably needs some shape, so he may as well start to show it right away.
All in all, I think “standard” gets it right in this auction (2 = any game try, bids above 2M = slam tries).
Feb. 9, 2018
Arend Bayer edited this comment Feb. 10, 2018
You are ignoring the author of this comment. Click to temporarily show the comment.
Paul, I don't know what Jan discussed with her partner. But I have a suspicion - 3NT would suggest playing 3NT from opener's side, whereas 3 would suggest playing 3NT from responder's side…
Feb. 9, 2018
You are ignoring the author of this comment. Click to temporarily show the comment.
I have a hunch that we'll soon get a chance to outvote David and Richard soon and make BW 2/1 use serious or non-serious 3N :)
Feb. 5, 2018
You are ignoring the author of this comment. Click to temporarily show the comment.
Mike, I don't think money could buy you an expert with better credentials or expertise than Greg.
Jan. 27, 2018
You are ignoring the author of this comment. Click to temporarily show the comment.
I am curious why you chose not to include the Bayesian analysis in your original article. Isn't the Bayesian factor the single most precise quantifiable information we can obtain from analysing fresh data?
Jan. 24, 2018
You are ignoring the author of this comment. Click to temporarily show the comment.
Shouldn't 4 suggest a hand like this? A 4NT bid with good spade values.
Jan. 9, 2018
You are ignoring the author of this comment. Click to temporarily show the comment.
I am open to other preferences in established partnerships. But for BW standard, I would strongly prefer if 15-17 balanced hands never open 1M.
Jan. 9, 2018
You are ignoring the author of this comment. Click to temporarily show the comment.
I tried to keep the footnotes brief but accurate, sorry if I failed in that. 1D is unbalanced, except that a 5332 hand with five diamonds may choose to open 1D. (More likely to open 1D with a good suit, and more likely with 3=3=5=2.) 1C contains all other balanced 11-13 hands without a 5-card major, and all 17-19 balanced hands.
Jan. 8, 2018
You are ignoring the author of this comment. Click to temporarily show the comment.
So double still shows four hearts, and you give up on the natural 1NT bid?
Jan. 7, 2018
You are ignoring the author of this comment. Click to temporarily show the comment.
“Do we play support double when opponents bid a natural 1N?” seems a good question to poll for the Bridgewinners system.
Jan. 3, 2018
Arend Bayer edited this comment Jan. 3, 2018
You are ignoring the author of this comment. Click to temporarily show the comment.
I think it is logically consistent to award a PP without a score adjustment in this board. Say you judge that over 2H, only 3H and 4H are logical alternatives. Then 4H is clearly suggested by the UI, and it is reasonably to conclude that every experienced player should be aware of that. If you also deem South to have a clear 4H bid over a 3H raise, then there is no basis for a score adjustment, but there is a basis for a PP.

(I disagree with this judgement, since passing 2H seems clear to me, so it is at least a LA.)

Unfortunately, even fairly experienced players are often confused about their obligations after UI from a mistaken explanation or lack of alert. I think that's simultaneously an argument for a high bar (“They genuinely didn't know.”) or a low bar (“We need to educate players.”) for assessing PPs in such situations.
Dec. 18, 2017
You are ignoring the author of this comment. Click to temporarily show the comment.
Sorry, go is far more complex to learn than the “language” of bidding. The reason that it is difficult for AI to learn bridge is that you have to interface with humans, and we humans have strange ideas about disclosure of partnership agreements.
If every BBO partnership had their agreements available in machine-readable form, then I am sure you AI could easily learn to play bridge much better than humans.
Dec. 7, 2017
You are ignoring the author of this comment. Click to temporarily show the comment.
I think this is a great idea. However, with all respect to Carl, I don't think he quite thought this through to the proper conclusion.

I see many players making odd choices in their second bid (such as making a short suit game try after 1S P 2S in a suit where there aren't short, or not giving correct responses to stayman) that can be quite successful against robots, but where human players would quickly figure out what is going on. Clearly we should also let robots specify the second round bid. I've also seen players make cuebids in suits where they have no control, or bid blackwood despite having a void, just to stop their partner from bidding a grand - all tactics that would be silly against humans. My solution is to have robots controls the entire auction.

I have also seen players postpone obvious claims for all but one of the remaining tricks, only to see one of the robot defenders make an obviously wrong discard and letting the human declarer get away with all the remaining tricks. Clearly, the robots should take over play as soon as every reasonable line and defense would lead to the same number of tricks.

For now, that would let humans choose the opening leads, and likely play the first few tricks of the hands. Once the crazy actions in the auction and the late play have been eliminated from the game, we could have another look at actions in the beginning of the play. If they turn out to be a proper measurs of skill, fine, if not, we can always let the robots take over even more.
July 14, 2017
You are ignoring the author of this comment. Click to temporarily show the comment.
Han, you only say so because you don't get invited to the HoF induction dinners!
April 10, 2017
You are ignoring the author of this comment. Click to temporarily show the comment.
At the table I led a club. On reflection, that seems questionable against weak NT/4-card majors - dummy is likely to have 5 clubs.

Dummy had Axx x QJ98 KQTxx, declarer KQTxx KQxxx xx x, leaving partner with xxxxxx Kxx Axx. Double dummy, your lead doesn't matter. Single dummy, partner reasonably played me for a club singleton.
Jan. 28, 2017
You are ignoring the author of this comment. Click to temporarily show the comment.
Isn't the main concealment advantage of Puppet after the opening lead (and hence invisible to your study)?

Consider the auction
1. 1N-3C-3D-3H-3N versus
2a. 1N-2C-2D-3N or
2b. 1N-2C-2H-3N.

*After* the opening lead, the defenders know dummy's shape (duh). They also know whether declarer has four hearts in the stayman auctions, but not in the Puppet auction.
Jan. 17, 2017
You are ignoring the author of this comment. Click to temporarily show the comment.
Thanks, I had suspected this is the case, and now I have found the threads:
A post by Frances Hinden explaining her methods:

http://www.bridgebase.com/forums/topic/25446-hmmmm/page__view__findpost__p__281456

A thread where this issue is discussed again:
http://www.bridgebase.com/forums/topic/27265-pass-or-redouble/
Jan. 10, 2017
You are ignoring the author of this comment. Click to temporarily show the comment.
@Andy: I can think of a few differences between enforcing revoke penalties and calling for a ruling on the hand in question.

1. When my opponent revokes, I am 100% certain that all of the authors of the relevant law fully intended the penalty to be applied in the specific situation that happened at my table. I am somewhat doubtful everyone on the committee that agreed to this 8-trick rule really intended to make it illegal to open an Acol-Two or equivalent with the hand being discussed here.
2. When I revoke, unless I am playing against a beginner (new to duplicate bridge), I can be almost certain that they know they are entitled to some redress, and that all they need to do is to call the director and they shall receive it. Against most opponents in the EBU, I could get away with opening this hand with 2C.
3. While not everyone agrees with automatic penalties for revokes, most bridge players will at least find them acceptable. I think most would disagree with outlawing a Benji 2C opening on AKJ9xxxx - xx AJT.

(If I understand some of your other posts correctly, you are saying that 2. did not apply when this hand actually occurred. That certainly makes a difference in my view.)
Jan. 10, 2017
Arend Bayer edited this comment Jan. 10, 2017
.

Bottom Home Top