BBO Discussion Forums: 1C (strong) - 1S as double negative... - BBO Discussion Forums

Jump to content

  • 3 Pages +
  • 1
  • 2
  • 3
  • You cannot start a new topic
  • You cannot reply to this topic

1C (strong) - 1S as double negative...

#21 User is offline   nullve 

  • PipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 2,326
  • Joined: 2014-April-08
  • Gender:Male
  • Location:Norway
  • Interests:partscores

Posted 2024-October-01, 16:08

 effervesce, on 2024-October-01, 06:32, said:

		2D/H/S/NT = polish-style transfers, non-acceptance is natural


What are Polish-style transfers?
0

#22 User is offline   Zelandakh 

  • PipPipPipPipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 10,717
  • Joined: 2006-May-18
  • Gender:Not Telling

Posted 2024-October-01, 18:27

After a great deal of playing around, what I decided is that if I wanted to play SPs, the better solution is to play that 1 is "DN or GF with ". In effect that allows you to have 1 as GF except a specific rebid, and of course you then arrange it so that the key sequence becomes 1 - 1; 1 - 1, thus getting you back to 1 DN but allowing Opener to pull specific difficult hands out along the way. I think that works quite well, much better than the direct 1 DN systems. You could probably achieve something very similar in a RR scheme but as I don't have a system of that type in my arsenal, I have not played around with the ideas at all within such a context. In the end though, what I really thought is that the gains to be had by rearranging things in this way are distinctly minor. For my system (where 1 is 15-17 /bal or 18+ any) switching to SPs simplifies some things but makes others more complicated. So while I have kept some notes around for a couple of different options that I feel still have some potential, I have not (as of yet) made this change. In other words, the proper take away from the subject is that SPs just allow you to make some gains and some losses but are not generally a fix. In designing a system you should select the set-up that works most harmoniously with everything else, because the response scheme of itself, assuming it is set up efficiently, is not going to be the breaking point.
(-: Zel :-)
1

#23 User is offline   effervesce 

  • PipPipPipPipPip
  • Group: Full Members
  • Posts: 885
  • Joined: 2007-March-28

Posted 2024-October-01, 20:08

 nullve, on 2024-October-01, 16:08, said:

What are Polish-style transfers?

canape transfer.

first transfer bid is shorter than second bid. Ie 1C-1S-2D-2H-3C shows longer clubs than hearts.
Ming

--Always remember you're unique. Just like everyone else.
0

#24 User is offline   foobar 

  • PipPipPipPipPip
  • Group: Full Members
  • Posts: 518
  • Joined: 2003-June-20
  • Gender:Male

Posted 2024-October-01, 21:55

 effervesce, on 2024-October-01, 10:26, said:

If you have 1C - 1D as GF then neither of you have described any shape. In my experience, the better the opponents, the more likely they are to bid.

It's been several years since I took a look at the 1 GF scheme, but it in addition to the above, the reverse relays were awkward. So effectively, we have three different schemes, each so similar, but yet with slightly differing semantics:
  • Relay scheme after SP response (and bespoke relay breaks if needed)
  • Relay scheme for responder's hand over 1
  • Relay scheme for opener's RR


I would much rather have a single ~symmetric relay scheme that can be used over most openings.

 effervesce, on 2024-October-01, 10:26, said:


It would be much better to have a 1C - 1D system where the balanced hand asks the unbalanced hand's shape and strength imho.


Since all GFs are compressed into 1, opener's RR is just making a guess. Maybe one way to get around it might be to put some balanced GFs into a 1 GF response in addition to 1 GF (a la Charron relay), but this gives up the 1 SP.
0

#25 User is offline   foobar 

  • PipPipPipPipPip
  • Group: Full Members
  • Posts: 518
  • Joined: 2003-June-20
  • Gender:Male

Posted 2024-October-01, 22:13

 Zelandakh, on 2024-October-01, 18:27, said:

For my system (where 1♣ is 15-17 ♣/bal or 18+ any) switching to SPs simplifies some things but makes others more complicated.

Now that you are back, it will be great to learn more about your system on a different thread (hint, hint :D).

 Zelandakh, on 2024-October-01, 18:27, said:

After a great deal of playing around, what I decided is that if I wanted to play SPs, the better solution is to play that 1 is "DN or GF with ". In effect that allows you to have 1 as GF except a specific rebid, and of course you then arrange it so that the key sequence becomes 1 - 1; 1 - 1, thus getting you back to 1 DN but allowing Opener to pull specific difficult hands out along the way. I think that works quite well, much better than the direct 1 DN systems.

IMPrecision uses a variation of this idea (1 is either super-positive or DN).

 Zelandakh, on 2024-October-01, 18:27, said:

In other words, the proper take away from the subject is that SPs just allow you to make some gains and some losses but are not generally a fix. In designing a system you should select the set-up that works most harmoniously with everything else, because the response scheme of itself, assuming it is set up efficiently, is not going to be the breaking point.

Indeed; we have examples of several 1 systems that seem to do just fine at the highest levels with a traditional 1 response. Tarzan Club, which uses a 1 negative with a 15+ 1 has done exceedingly well.
0

#26 User is offline   DavidKok 

  • PipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 2,635
  • Joined: 2020-March-30
  • Gender:Male
  • Location:Netherlands

Posted 2024-October-02, 11:53

 Zelandakh, on 2024-October-01, 18:27, said:

After a great deal of playing around, what I decided is that if I wanted to play SPs, the better solution is to play that 1 is "DN or GF with ". In effect that allows you to have 1 as GF except a specific rebid, and of course you then arrange it so that the key sequence becomes 1 - 1; 1 - 1, thus getting you back to 1 DN but allowing Opener to pull specific difficult hands out along the way. I think that works quite well, much better than the direct 1 DN systems.
This is similar to my attempt at folding semipositives into symmetric relay, I ended up at 1 being a double negative or a semipositive(+) hand with hearts.
0

#27 User is offline   nullve 

  • PipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 2,326
  • Joined: 2014-April-08
  • Gender:Male
  • Location:Norway
  • Interests:partscores

Posted 2024-October-02, 12:40

 effervesce, on 2024-October-01, 20:08, said:

canape transfer.

first transfer bid is shorter than second bid. Ie 1C-1S-2D-2H-3C shows longer clubs than hearts.

Just to get a feeling for this:

1-1
?

a) 18 hcp, 3613 (2, intending to pass 2, right?)
b) 18 hcp, 3514
c) 18 hcp, 4513
d) 18 hcp, 5413
e) 18 hcp, 3145
f) 18 hcp, 3154
0

#28 User is offline   foobar 

  • PipPipPipPipPip
  • Group: Full Members
  • Posts: 518
  • Joined: 2003-June-20
  • Gender:Male

Posted 2024-October-02, 16:28

 DavidKok, on 2024-October-02, 11:53, said:

This is similar to my attempt at folding semipositives into symmetric relay, I ended up at 1 being a double negative or a semipositive(+) hand with hearts.

This sounds like great idea for someone really wanting to play SPs, since it's always +0, right?
0

#29 User is offline   effervesce 

  • PipPipPipPipPip
  • Group: Full Members
  • Posts: 885
  • Joined: 2007-March-28

Posted 2024-October-02, 16:39

 nullve, on 2024-October-02, 12:40, said:

Just to get a feeling for this:

1-1
?

a) 18 hcp, 3613 (2, intending to pass 2, right?) 2D tf hearts correct
b) 18 hcp, 3514 2D tf hearts
c) 18 hcp, 4513 2D tf hearts
d) 18 hcp, 5413 2H tf hearts then bid spades
e) 18 hcp, 3145 probably 1NT, dont really want to bid to 3m
f) 18 hcp, 3154 probably 1NT, dont really want to bid to 3m


Like I said, its a crapshoot option after 1C - 1S.

If anyone has better continuations Im happy to hear them.
Ming

--Always remember you're unique. Just like everyone else.
0

#30 User is offline   DavidKok 

  • PipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 2,635
  • Joined: 2020-March-30
  • Gender:Male
  • Location:Netherlands

Posted 2024-October-03, 14:00

View Postfoobar, on 2024-October-02, 16:28, said:

This sounds like great idea for someone really wanting to play SPs, since it's always +0, right?
I think not quite. There was some finagling with the both majors hand type in my structure though it's at +0 but a bit awkward, and also I think the three-suiters (4x1 or 5440) are higher (though instead I opted for partial resolution instead), but other than that I believe so.

1-?
  • 1: DN or SP(+) 4(+) unbal not 4(+)
  • 1: SP(+) 4(+) unbal not 4(+)
  • 1: SP(+) 5(+) no 4(+)cM not 5(+), or 5-10 or 14+ bal
  • 1NT: SP(+) 5(+) no 4(+)cM not 5(+)
  • 2: SP(+) (54)+ majors either way
  • 2: SP(+) any 4441 or 5440
  • 2: SP(+) 55+ minors
  • 2+: CR4SH 11-13 balanced

0

#31 User is offline   nullve 

  • PipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 2,326
  • Joined: 2014-April-08
  • Gender:Male
  • Location:Norway
  • Interests:partscores

Posted 2024-October-04, 08:26

View Posteffervesce, on 2024-October-02, 16:39, said:

Like I said, its a crapshoot option after 1C - 1S.

If anyone has better continuations Im happy to hear them.

I'm still not sure what your continuations actually are.

Quote

e) 18 hcp, 3145 probably 1NT, dont really want to bid to 3m
f) 18 hcp, 3154 probably 1NT, dont really want to bid to 3m

So the 1N rebid doesn't really show a balanced or semibalanced hand after all. Then what do you do with

g) 18 hcp, 4153, small singleton
h) 18 hcp, 4144, small singleton

Do you rebid 1N or do you force to 3-level (1-1; 2-2; 3) at least with g)? If 1N, then your 1N rebid is starting to look like it could be part of a structure based on Woolsey, as in the example structure I sketched above.

I also wonder if you sometimes use the Polish-style transfers on GF hands. If you do, then it seems pretty extravagant to keep the 2 rebid as an ART GF, since your partscore bidding suffers immensely (IMO) as a result. If not, do you ever need to be able to transfer to clubs/diamonds via 2/3 instead of just jumping to 3/3? (Freeing up 2 could be useful.)
0

#32 User is offline   Zelandakh 

  • PipPipPipPipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 10,717
  • Joined: 2006-May-18
  • Gender:Not Telling

Posted 2024-October-04, 15:30

View Posteffervesce, on 2024-October-02, 16:39, said:

Like I said, its a crapshoot option after 1C - 1S.

If anyone has better continuations Im happy to hear them.

For the PC/AUC opening style, the obvious structure is to play

1NT = 15-18(19) bal
2 = 15-18 nat
2 = art, strong
2M = 18+ nat
2NT = (19)20-21 bal
3m = Acol 2

and increase the top end of the 1 opening slightly to compensate. It is probably also a good idea to play a natural 2NT opening in this case too, allowing the 2NT rebid to be stronger and for the artificial 2 to become a game force. By this point though, you are making enough compromises to shoehorn in the 1 DN as to remove most of the potential efficiency gains of the system structure, so you are just adding complexity for little overall benefit.

If the opening 1 is 18+, which seems to be the basis of the post above, then one very simple solution is just to play Transfer Walsh at +/-6hcp and not worry about relays at all. I would not be at all surprised if that ends up being more efficient than a 1 DN scheme. You can achieve something similar in a more typical strong club setting too with a little tweaking, for which the Italian Nightmare system is probably the best known example.
(-: Zel :-)
0

#33 User is offline   kwiktrix 

  • PipPip
  • Group: Members
  • Posts: 48
  • Joined: 2011-June-06
  • Gender:Male

Posted 2024-October-10, 14:14

 awm, on 2024-September-24, 14:18, said:

 kwiktrix, on 2024-September-24, 14:00, said:

As the individual who probably provoked this thread, here are my thoughts:
1) if the 1D response includes junk hands, you are letting the opponents in the auction cheaply - ESPECIALLY 4th seat - when responder has junk.
2) when responder has balanced junk, unless you have a monster 22+ balanced hand or a strong shapely assortment, you do not want to be declaring. 1N is the response for any balanced hand from 15-21, 2C to find the junk hand’s major (if any - or to show a 2suiter), any other bid to play.
3) its value is not constructive, it is obstructive. More importantly, it allows the other non-GF responses to be semi positives.

This isn't really consistent. If you do not want to be declaring, isn't it good to be letting the opponents into the auction? If you shut them out, you'll be declaring...

The intent is to let them into the auction under duress. As an example:
I open the bidding 1 with 15HCP. My partner has 3HCP, for a total of 18HCP. LHO is irrelevant to this discussion - they will overcall irrespective of responder's intended bid - so LHO passes. This leaves RHO with an opening hand and they will bid constructively.
  • Scenario 1 - we include SPs into 1 (0-9) - RHO has a free 1M overcall available and we don't know whether responder is SP or DN.
  • Scenario 2 - we define 1 as DN - RHO has a free overcall available and LHO knows that they have the balance of power.
  • Scenario 3 - we define 1 as GF or DN - RHO has a free overcall available but can imply that their partner has the remaining HCP, but if not, will gladly interfere at the 1level to disrupt our clarification auction.
  • Scenario 4 - we define 1 as DN - RHO has a free overcall available at the 2level and LHO knows that they have the balance of power. But they do not have a simple 1level start.

Quote

Of course, it's not really true that you don't want to declare -- you often have around half the values despite the weak response. If you have a fit somewhere it can be your hand for a partial or even a game. You don't really need that much strength; give opener AKxxx x AKxxx Ax and you have good odds for game if partner produces queen-fourth of spades and out. Even opposite xxxx xxx xxx xxx it's a worthwhile game to bid at IMPs (you really just need spades to divide). Hands like this (where you just need four-card support for your major) are not unusual. But it's unpleasant to be forced to play these hands at the three-level when partner has no fit for the major.

I don't want to declare 1N if I'm balanced 15-19. I'm not making the contract. I only want to declare if I have a suit (which I will bid instead of 1N), or if responder has a suit. And responder will tell me if they have a "suit" by showing a DN with shape (Charron has every bid from 1N-->7N available for unbalanced SP and shapely DNs.
Should responder have their own 5card suit after 1(P)1(P)1N, they will bid it. We will miss the magic 44M fits that anyone opening a strong 1N will also miss (although I'm contemplating using 4suit 4+ transfers).
Your example of AKxxx x AKxxx Ax will be passed out at 1S using 2/1. With Precision (and 1 includes DN), the auction will go 1(P)1(1), then what?

Quote

But who are you obstructing? Opponent in 2nd seat has already passed over a strong club, so he's probably balanced (or really weak). I guess you've kept 4th hand from doubling 1 or bidding 1 (if he was doing one of those things). But you've also greatly obstructed your own partscore (or light game) auctions if you have a fit, and if you have no fit then keeping 4th hand out could easily come back to bite you.

See the Scenarios in the first paragraph above...

Quote

This last is the one real advantage. But there are other ways to avoid bidding 1 on semi-positives that don't come with such a high cost.

Not sure of the relevancy. Charron uses 1 as either GF with as Primary Keysuit, or SP BAL. All other responses (with the exception of 1 and 1) show unbalanced SP, or in some instances, unbalanced DN hands.
0

#34 User is offline   nullve 

  • PipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 2,326
  • Joined: 2014-April-08
  • Gender:Male
  • Location:Norway
  • Interests:partscores

Posted 2024-October-11, 05:23

View Postkwiktrix, on 2024-October-10, 14:14, said:

I don't want to declare 1N if I'm balanced 15-19. I'm not making the contract. I only want to declare if I have a suit (which I will bid instead of 1N), or if responder has a suit.

You seem to hold two very unusual beliefs:

1) 1N rarely makes with 15-19 BAL opposite a DN.
2) 1N is a good contract only if it tends to make.

View Postkwiktrix, on 2024-October-10, 14:14, said:

Your example of AKxxx x AKxxx Ax will be passed out at 1S using 2/1.

Reality check:
Spoiler

This post has been edited by nullve: 2024-October-11, 18:57

0

#35 User is offline   kwiktrix 

  • PipPip
  • Group: Members
  • Posts: 48
  • Joined: 2011-June-06
  • Gender:Male

Posted 2024-October-12, 08:03

View Postnullve, on 2024-October-11, 05:23, said:

You seem to hold two very unusual beliefs:

1) 1N rarely makes with 15-19 BAL opposite a DN.
2) 1N is a good contract only if it makes.


1) I'm not sure that I'd use "rarely". But over a 10000 hand DD simulation for a 15-17HCP notrump vs DN = 0-5, 1N is made only 32.9% of the time, with a weighted average trick count of 5.9. Weighted average MP score is -29.3NV and -92.4VL (Details below).

15-17 1N Range
Points	Count	ScoreNV	ScoreV
0	6	-350	-700
1	14	-300	-600
2	72	-250	-500
3	344	-200	-400
4	1125	-150	-300
5	2251	-100	-200
6	2898	-50	-100
7	2137	 90	 90
8	909	 120	 120
9	204	 150	 150
10	37	 180	 180
11	3	 210	 210
			
WtAvg	5.88	-29.26	-92.44

Although I meant 15-17, I stated 15-19. Here are the data for that expanded range. WtAvg tricks = 6.3. Weighted average MP score is -3.4NV and -54.3VL (Details below).

15-19 1N Range
Totals	Count	ScoreNV	ScoreV
0	4	-350	-700
1	9	-300	-600
2	52	-250	-500
3	267	-200	-400
4	820	-150	-300
5	1821	-100	-200
6	2674	-50	-100
7	2308	 90	 90
8	1424	 120	 120
9	488	 150	 150
10	102	 180	 180
11	27	 210	 210
12	4	 240	 240
			
WtAvg	6.26	-3.35	-54.28

2) Vulnerable - absolutely.
0

#36 User is offline   PrecisionL 

  • PipPipPipPipPip
  • Group: Full Members
  • Posts: 987
  • Joined: 2004-March-25
  • Gender:Male
  • Location:Knoxville, TN, USA
  • Interests:Diamond LM (6700+ MP)
    God
    Family
    Counseling
    Bridge

Posted 2024-October-12, 10:34

Many of us would be interested in the same analysis for 17-19 1NT rebid playing a strong club opening. Posted Image

Edited12/6/24: 14-16 1NT opening assumed.
Ultra Relay: see Daniel's web page: https://bridgewithda...19/07/Ultra.pdf
C3: Copious Canape Club is still my favorite system. (Ultra upgraded, PM for notes)

Santa Fe Precision published 8/19. TOP3 published 11/20. Magic experiment (Science Modernized) with Lenzo. 2020: Jan Eric Larsson's Cottontail . 2020. BFUN (Bridge For the UNbalanced) 2021: Weiss Simplified (Canape & Relay). 2022: Canary Modernized, 2023-4: KOK Canape.
0

#37 User is offline   mycroft 

  • Secretary Bird
  • PipPipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 7,489
  • Joined: 2003-July-12
  • Gender:Male
  • Location:Calgary, D18; Chapala, D16

Posted 2024-October-12, 13:30

Don't want to take away from kwiktrix, but did want to try out my skill at ramping these up with anntzer/redeal.

Tricks  Count   Sc. NV  Sc. V
0       2       -350    -700
1       7       -300    -600
2       23      -250    -500
3       113     -200    -400
4       595     -150    -300
5       1325    -100    -200
6       2495    -50     -100
7       2641    90      90
8       1889    120     120
9       717     150     150
10      156     180     180
11      35      210     210
12      2       240     240
WtAvg: (if first entry is not 1.0, something is very wrong)
1.0     6.6388  23.018  -14.747

So, looks like a wash (but better, by about 40 points, NV rather than V). Note: I bet a lot of the extreme hands are "0-5, but a very long suit". Maybe I should try pulling out "North has 5cM or 6cm or short clubs".

It would be interesting (and probably my next attempt) to do "score vs par". With two flat hands and about 20 HCP, par should be closer to real "what should happen" than normal.
When I go to sea, don't fear for me, Fear For The Storm -- Birdie and the Swansong (tSCoSI)
0

#38 User is offline   mycroft 

  • Secretary Bird
  • PipPipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 7,489
  • Joined: 2003-July-12
  • Gender:Male
  • Location:Calgary, D18; Chapala, D16

Posted 2024-October-12, 22:10

So I did some tweaking, and did some "North will DTRT over 1NT" (transfer to a 5cM, or a 6cm, or garbage Stayman with any other xxx1 or 4450 or (443)2), and compared against par:

17-19 NT, Responder 0-5, will pull NT
Vul avg result vs par
None -15.170
N-S -66.760
E-W -14.120
Both -71.430

Only ran 1000 hands - par x 4 is time-consuming.

In my tests, some 100-hand sets beat par, at least NV.

Now, we all know about par. So given "declarer's advantage" and "opener's advantage" (and, if the opponents are going to interfere, they're more likely to do so over 1 than after 1-1, and almost never after 1-1; 1NT), I think I would say this is "net zero", probably positive if NV.

Please, please dupe my tests - I know I've made simple stupid mistakes in my sims before, and this one ended up being "not simple".
When I go to sea, don't fear for me, Fear For The Storm -- Birdie and the Swansong (tSCoSI)
0

#39 User is offline   mycroft 

  • Secretary Bird
  • PipPipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 7,489
  • Joined: 2003-July-12
  • Gender:Male
  • Location:Calgary, D18; Chapala, D16

Posted 2024-October-13, 09:28

Okay, now I got myself wandering down the rabbit-hole: why should I care about "score vs par" in "north will pull" hands, but not "score vs zero"? So, yeah. Specific hand results in spoiler, just because it's *long*.

17-19 NT, Responder 0-5, will pull NT, 10 000 hands
Spoiler

Weighted Average:
ScoreNV: 55.180
ScoreV: 28.285

So it seems to be net positive for score (again, DD vs RL. Note that declarer's advantage over DD goes down (especially for the really high trick scores) in suit contracts. But then, it's up on the really low trick scores, too).

Again, no simulation around "would the opponents interfere", which would change the math significantly (and turn the results into a horrible mess, and "opponents' level of aggressiveness/defence to strong club" is almost impossible to simulate, and...)
When I go to sea, don't fear for me, Fear For The Storm -- Birdie and the Swansong (tSCoSI)
0

#40 User is offline   DavidKok 

  • PipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 2,635
  • Joined: 2020-March-30
  • Gender:Male
  • Location:Netherlands

Posted 2024-October-13, 10:03

I do have simulations for strong club openings that take a lack of interference by the opponents into account. Conditioning on LHO not overcalling or jump overcalling changes the odds markedly compared to the a priori case. Of course this is subject to assumptions about the overcall style though, but I think "disregard the possibility of interference entirely" is much worse than making simplified assumptions.

My apologies for not following the discussion thread much - which question do the double dummy simulations set out to answer? I thought it was to give opener a balanced hand in the 17-19 range, responder a double negative (0-4 or so), and then compare 1NT-including-the-response-system with 2-suit (how do we pick the suit?). I'd be happy to run those simulations myself and see what answers I get, but at the moment I don't yet understand what the exact question is.

P.S. The double dummy solver has been crunching 20+ deals/second for me, so it should be fine to run even a relatively large number of deals.
0

  • 3 Pages +
  • 1
  • 2
  • 3
  • You cannot start a new topic
  • You cannot reply to this topic

2 User(s) are reading this topic
0 members, 2 guests, 0 anonymous users