Join Bridge Winners
All comments by Csaba Czimer
You are ignoring the author of this comment. Click to temporarily show the comment.
Other: South should have bid 2N instead of 3
Jan. 27, 2018
You are ignoring the author of this comment. Click to temporarily show the comment.
Steve we play relay style, most of the time we explore responder's shape (and range heading to a slam), thus opener rarely shows anything (opposite a balanced responder he may choose to show his suit(s) and shortness). As we only have 1M and 2m showing 8-12 HCP and shape (1N = 13+ any, 2 and up are weak hands with a long suit), most of the time there is not enough space to show 6-5s below game level, they must be bid as 6-4 or 5-5.
Anyway, even if North had had the opportunity to show his clubs, I think that he would have bid NT at the first opportunity to protect his heart jack.
Jan. 21, 2018
Csaba Czimer edited this comment Jan. 21, 2018
You are ignoring the author of this comment. Click to temporarily show the comment.
Our 2 reponse promises but denies 4-card majors. With a 4+ major we show the major first. We play relay continuations, but we can't normally show 6-5, thus we must show it as either 6-4 or 5-5.
Jan. 21, 2018
You are ignoring the author of this comment. Click to temporarily show the comment.
(deleted)
Jan. 20, 2018
Csaba Czimer edited this comment Jan. 20, 2018
You are ignoring the author of this comment. Click to temporarily show the comment.
2 was not an option, we reserve it for extreme hands. Partner's only other choice was pass, asking my hand. It has the disadvantage that may wrongside NT. I don't think 1N was a bad bid, it just took me in the wrong direction this time (still was thinking about stopping in 4N after 4).
Jan. 20, 2018
You are ignoring the author of this comment. Click to temporarily show the comment.
our 3 shows any solid suit with no A/K/Q outside. Shortness and length can be asked:
- 3N: strong offer to play
- 4: I know your suit, which is the short one? Next relay: is it void? Next relay: ask for length of the solid suit.
- 4: bid your suit! Then the same relays as after 4.

With outside strength we bid a transfer-like positive response (showing 8-12) or 1N showing 13+ any.
Jan. 20, 2018
You are ignoring the author of this comment. Click to temporarily show the comment.
Another vote on “bid 3N”.

Usually it is 1 sure stopper which may be better to play from opener's hand. Opener may have Qx. Responder has Ax(…), perhaps Kxx(…) and enough strength or tricks to play in 3N.

(although if an unknown partner bids it I would think it is a splinter)
Jan. 12, 2018
Csaba Czimer edited this comment Jan. 12, 2018
You are ignoring the author of this comment. Click to temporarily show the comment.
When I play standard (2/1 more or less) in an established partnership then I prefer 2 as
(a) GF with or balanced hand
(b) INV with 3(rarely 4)-card support

Opener's bids:
2: any minimal hand (cca 11-13). Now 2M can be passed, any other bid is NAT and GF

all other rebids promise some extra (cca 14+)
- 2//N/3: NAT 2M is 6+ suit, 2N is 5332 or 5M422 with 17+)
- 3…3N: diamond side suit (3 is 55+, 3M is short in , 3oM is short in this one, 3N is 5M-4-2-2 and 14-16)

However, I prefer relay precision :)
Dec. 22, 2017
You are ignoring the author of this comment. Click to temporarily show the comment.
you are right, line A is the best one.
Ask for a free copy of http://home.planet.nl/%7Enarcis45/suitplay/help/mainhelp.htm">Suitplay
Dec. 22, 2017
You are ignoring the author of this comment. Click to temporarily show the comment.
Our 2 response denies 4-card majors, thus not interested in majors (when playing standard). Many pairs in Hungary plays this way.

Im my relay precision 2 is natural, 5+ suit, nonforcing (cca 9-12) and denies majors too.
Nov. 26, 2017
Csaba Czimer edited this comment Nov. 26, 2017
You are ignoring the author of this comment. Click to temporarily show the comment.
I seriously doubt that the AI self-experimenting learning came up with a system where some opening bids are not used.

The opening bids table is very suspicious for me (I guess some human involvement), and can't imagine, that it is optimal (those 1 and 1N openings smell).
Maybe just the AI should play some more years (or centuries) with itself.

_________

The goal of the bidding is quite complex and is closely related with the play:

1. to find our best contract (where we score the most in expected value)

2. disrupt the opponents' bidding (take away space, preempt), (weak opening (strong pass) systems)

3. give only those information which are more useful to partner for deciding the contract than for the opponents (in deciding their contract and defending against ours). As a part: if we can choose, describe the will-be-dummy rather than declarer

4. give partner useful hints for the opening lead / defence (lead directing overcalls and doubles)

Some bids are fully descriptive, some are multi-purpose (descriptive and defensive at the same time).

I wonder how all these goals appear ara fulfilled.
Nov. 25, 2017
Csaba Czimer edited this comment Nov. 25, 2017
You are ignoring the author of this comment. Click to temporarily show the comment.
“I found this article because I was looking for some kind of app where I can enter our system”

Jim, I wrote a small (~ 4000 lines) JavaSript code that you (or your partner) can practice bidding with. And yes, it can play your system. Well, if you write it first. Currently it uses my http://hanglemezek.hu/csabi/index.htm">Relay precision, it can discover pattern, keycards, uses spiral scan, etc. I also tried how it behaves on slam bidding problems (ATB - Missed slam) that were published here, mostly it easily bid it.

No AI at all, it uses your rules, if none found then it uses some “common sense” rules (it is its weakest part). It revaluates hands in each round of bidding according to the information it got so far. The system is mostly a plain text file, quite human readabable. But I coded some trickier parts of my system (slam bidding (spiral scan) mainly) in JavaScript functions.

It can bid one or two hands (potentially four, but see below)

Drawbacks:
- it is in an early development stage, all kind of errors can occur (not too frequent, so it is perfectly usable for practicing)
- at the moment only one side bids (I designed it to be able to handle opponent's bids, but have not entered any competitive situation so far
- I am very busy with my job(s) in the next few months, thus won't have much time for discussion (until let's say March)

An example of “system notes” that it understands is http://hanglemezek.hu/csabi/1s.bid">here (1sp opening) The whole system is 9000 lines (126 KB) of text.

Some more info is http://bridgewinners.com/article/view/ambitious-computer-partner/?cj=393881#c393881">here and http://bridgewinners.com/article/view/intelligent-auction-to-grand-slam/?cj=403147#c403147">here
Nov. 25, 2017
Csaba Czimer edited this comment Nov. 25, 2017
You are ignoring the author of this comment. Click to temporarily show the comment.
I knew this version:

- What is the difference between uninformed and apathetic?
- I don't …
Nov. 25, 2017
You are ignoring the author of this comment. Click to temporarily show the comment.
@Steve: yes, exactly.

No method can solve everything. I opted for improving the bidding of the gameforcing and slam try hands instead of the weak ones, more IMPs can be won or lost here.
Nov. 23, 2017
You are ignoring the author of this comment. Click to temporarily show the comment.
Let's make it clear: it is a normally dealt board (at least not intentionally crazy).
Nov. 23, 2017
You are ignoring the author of this comment. Click to temporarily show the comment.
With a slam try we have CONFI, without that Stayman then 3N
Nov. 21, 2017
You are ignoring the author of this comment. Click to temporarily show the comment.
My favourite method is puppet to 2 with
(a) 5-4+, weak (will pass)
(b) 5, inv (2N next), not 5-5m, not very pure 5-4m
©* GF with a minor, short int the other minor (3m next)
(d) GF with 4441 or 5m-440 with both minors (3M next)

2 is already promised a major.

* This way 2 then 3m is NAT GF with a major shortness, can be asked which one.
Nov. 21, 2017
Csaba Czimer edited this comment Nov. 21, 2017
You are ignoring the author of this comment. Click to temporarily show the comment.
(I voted mostly South)

If you pass with 14 points and a 5-card spade suit in 4th seat, then I suppose you also pass with a similar 15 HCP hand in direct seat. This way you may have 29 points and let the opponents play undoubled (non vulnerable to make it worse).
Nov. 14, 2017
You are ignoring the author of this comment. Click to temporarily show the comment.
We felt the same and thus tried double as a strong (16+) hand. It was the worst convention I ever played in terms of IMPs per occurance.
Oct. 29, 2017
You are ignoring the author of this comment. Click to temporarily show the comment.
Something like Kx, xxx, AQJTx, xxx

If I can't double on level 1 when there is not much danger to be cought, I definitely can't double for takeout on level 3 when they can catch us easily.
Oct. 22, 2017
Csaba Czimer edited this comment Oct. 22, 2017
.

Bottom Home Top