[DavidKok]

To be clear, I would personally open an 11-13 balanced hand with 4♥ with 1♥ always as long as I consider the full hand to be of sufficient playing strength to meet that description, regardless of suit quality. Of course some 11's are really 10's, 9's or even worse, and the quality of the major suit(s) does play a role in this, but there's no specific rule other than evaluation of the full hand.

I play 1M-2♣ as a game forcing relay, rather than inv(+) relay. The strong notrump allows for a semiforcing 1NT, through which responder is saying "if you have 11-13 balanced I want to play here, I may have other intentions facing any other hand type".

On theoretical grounds I think canapé and game forcing relays are a natural match. If responder has a strong hand they are often interested in opener's long suit to help evaluate their hand and decide the safety level, so a generic "tell me your long suit, if any" asking bid is useful. Personally I don't like game forcing relays in many situations so I've kept it to a minimum here, and in fact I think standard 2/1 is also plenty good as a start, but in theory it makes a lot of sense and this is also the structure of the original Cottontail Club (all of 1♦-1♥, 1♥-1♠ and 1♠-2♣ were relays, though these were overloaded to be 4-way relays which I think is not good). But this is one angle in particular that I think might be worth exploring at some point, though I'm generally more concerned with the competitive auctions than the constructive ones.

[straube]

When I design systems (or just continuations) I always look at the frequencies of responses as a check. Yeah, I know it's not supposed to be Fibonacci. So take a look at your 1D opening. The 1M responses are usually (understandably) 5+ cards in length. So ought to occur roughly half as often as standard responses but 1N semiforcing is like 7-12 or so? Lots and lots of 1D-1N responses. I don't have a good generator atm but I bet you do. Why don't you run your responses (in terms of frequency) for 1D and post them here?

I think you don't like the ambiguous 1D opens (and they have their drawbacks in comp for sure) but after frequent 1H responses their rebids add a lot of clarity. I think of 1H as a relay almost. Even the 1S response is not too difficult to handle. It is relay-esque.

Look at your 1D-1H, 1S rebid. If this is 5S/4+D it's just very astonishing to show such a specific pattern so low. Standard symmetric reveals that around the 2D level. For me, this is 1S rebid is a big red flag that something isn't right. For contrast, isn't your 1H-1N, 2C 5/4 either way?

Draw everyone's attention now to awm's and sieong's double-barreled invites over 1M (precision 5-cd opening)...
https://www1.dal13.s...445#entry904445

which is so strong and at the same time confirmed for me that 5-cd majors are the way to go. I just don't know how you'd go about having double-barreled invites (roughly 10-11 and 12-13) over your canape system. His structure was a huge step forward for precision 5-cd majors though I don't know of anyone but me who ever played them.

btw I didn't perceive awm's replies to you as "hostile" or "gotcha". He's given me loads of criticism and help over the years. Anymore when I post a new idea, I start with the notion that there is likely something wrong with it, maybe it's not even new, and I'm more likely to get criticism than approbation.

[DavidKok]

The general idea that frequencies are supposed to decrease (approximately) exponentially with base 2 (natural) or 1.618 (relay) is a rough guideline at best, and I think it is helpful to consider why we might wish to deviate from it. In the case of my 1♦ the 1NT response is much more common than 1♥ and 1♠, which I think is somewhat fine. However, having to respond 2♣ or 2♦ with strong hands without a 5cM is a weakness, and it would be better to put these in the 1♥ response instead. This matches closely with a traditional style of canapé systems. If the 1♦ opening denies a 4cM we can either, as I did, have responder's bids (almost always) promise five, or we can go the other route and have 1♥ promise three and 1♠ show heart shortage, anticipating opener's canapé. This style is 'easily' upgraded into a relay approach: 1♥ as a generic asking bid on hands that feel comfortable they can handle any rebid, while bypassing 1♥ shows hand types with specific problems (e.g. heart shortage). The advantage is we get to start a (possibly game forcing) relay auction several steps lower, while the downside is that we will struggle to show hands with long hearts and the structure is more complicated. For now I've chosen to stick with the natural approach until problems arise. As mentioned before I'm primarily concerned with contested auctions, and I don't wish to optimise the constructive auctions for now (though I've added some relatively standard bids and rebids in the google doc linked earlier).

The 1♦-1♥; 1♠ auction is suboptimal for sure, and you may well wish to use that as an artificial bid (maybe a toggle to split ranges or heart length?). I'm fine with the natural use for now though. In general I want to make sure I have a comfortable way to show all hands first, and only optimise the structure second. The inferences compared to standard are enough of a change that I think focusing on the basics makes sense. As an aside I think the high frequency 1♦-1NT is really good, not really bad. We immediately bid to our likely best contract on all the hands where we can tell that's going to be best, and the opponents are left in the dark about declarer's distribution.
With 4♦5♣ the systemic opening bid is 2♣, so 1♦-1X; 2♣ shows 5(+)♦4(+)♣ with the clubs never being longer.

[straube]

DavidKok, on 2024-February-24, 05:01, said:

As an aside I think the high frequency 1♦-1NT is really good, not really bad. We immediately bid to our likely best contract on all the hands where we can tell that's going to be best, and the opponents are left in the dark about declarer's distribution.

If a high frequency 1D-1N is best here (it may well be), it makes me think something is likely off in the 1D opening itself.

I'll leave it there. Good luck fleshing out your system

[DavidKok]

straube, on 2024-February-24, 04:38, said:

For contrast, isn't your 1H-1N, 2C 5/4 either way?

I misread this last time, sorry. Actually the situation is even worse, it is 5/4 either way and can also be 1=4=4=4 or 4=4=1=4 maximum (minima with this shape are best off passing 1NT, I think). There's a section of the google document discussing this, it's actually a nifty part of the system and does well. Think of the 2♣ bid as an artificial bid to increase the total number of sequences, that just happens to be limited and contains clubs always.

straube, on 2024-February-24, 04:38, said:

Draw everyone's attention now to awm's and sieong's double-barreled invites over 1M (precision 5-cd opening)...
https://www1.dal13.s...rreled-invites/

which is so strong and at the same time confirmed for me that 5-cd majors are the way to go. I just don't know how you'd go about having double-barreled invites (roughly 10-11 and 12-13) over your canape system. His structure was a huge step forward for precision 5-cd majors though I don't know of anyone but me who ever played them.

Thank you for this link, I had not read this thread before. It's quite a read and the structures suggested therein are similar to what I put in my canapé system so far, though the considerations are different. The hand type I've wrestled with the most is when partner opens 1♠ and we have less-than-GF with long hearts. In standard 2/1 this hand type would have to choose between 1NT (semiforcing) or 3♥ (intermediate jump shift), possibly missing a good 2♥ contact when opener has a balanced minimum. I've since switched to the transfer style approach - 1NT SF, 2♣ GFR, 2♦ shows hearts, 2♥ is a light invite, 2♠ is competitive - which I think is an improvement. However, I still don't know how to use the 2♦ bid best, and am looking to explore this further. Also responder can of course just blast game, my opening range is not as wide as the one suggested in that link.

[straube]

DavidKok, on 2024-February-24, 05:57, said:

I've since switched to the transfer style approach - 1NT SF, 2♣ GFR, 2♦ shows hearts, 2♥ is a light invite, 2♠ is competitive - which I think is an improvement. However, I still don't know how to use the 2♦ bid best, and am looking to explore this further.

I used to use this as well (e.g. 2D transfers to hearts). double-barreled is far better imo. Of course double-barreled (like what I was doing) requires a relay structure and it must be organized such that a relay break shows a strongly invitational hand with 5 of the other major.

[DavidKok]

I'll reread the thread, I thought the double-barreled approach relies on opener not being balanced that often and caters to opener having 4(+)♥ on the 1♠ opening. I am not sure how to translate that to my setting, where 1♠ denies exactly four hearts and is frequently 11-13 bal.

[straube]

DavidKok, on 2024-February-24, 07:58, said:

I'll reread the thread, I thought the double-barreled approach relies on opener not being balanced that often and caters to opener having 4(+)♥ on the 1♠ opening. I am not sure how to translate that to my setting, where 1♠ denies exactly four hearts and is frequently 11-13 bal.

No, one of its benefits is solving the 5332 11-13 hand opposite 12-13. I could never handle that with my former approach because I didn't want to rebid a minor fragment with a max. Double-barreled solves this. Also handles responder's single-suited hands. Very weak? Immediate jump shift. 10-11? Bid 1N and hope for a rebid. 12-13? Bid the 3rd step and show the minor or relay with the 2nd step and then break.

My point in bringing it up was not only how much better this is for 5-cd major limited systems but how it reinforced for me, my own bias of "Gee, 5-cd majors must be right" because I can't imagine anything like this over 4-cd majors.

Having the fifth card makes the strong invite (e.g. 1S-2D) workable because you can often subside in a 5-2 fit (sometimes missing a superior minor suit fit) when opener is limited.

Adam, could you try explaining this better?

[DavidKok]

foobar said:

1) Can you elaborate more on your thoughts about AEC vs. SCUM? As I see it, the latter swaps the 1D with the 2M openings. So, the SCUM 1D opening is overwhelming likely to be 11-13 balanced, and is likely to be most resilient of the lot, since the only other variation (single suited minor can be easily sorted out for the most part).
2) It seems that 1M openings that show (mostly) 5+ shouldn't have issues with 3-card raises as long as there's a provision that the 1N response can never contain such a hand.
3) As Adam noted, SCUM, AEC, and IMPrecision contain the 11-13 hand to a single opening (in roughly descending order of purity), but your proposal spreads a common hand type across three different openings. It will be interesting to look at specific examples where this approach works better than the corralling solution (especially w.r.t. to SCUM and AEC).

• Comparing AEC and SCUM, I think the 2M openings in SCUM are a serious issue. If we don't locate a fit immdiately (with responder having support for the major or having both minors) we are likely too high or in no man's land. Also giving up on weak 2's is a notable cost. The SCUM 1♦ opening is more resilient, but I think as a full opening system it does worse. Also I don't love putting single suited minor hands in with 11-13 balanced. Raptor hands are often 5431 or 5422 so somewhat close to balanced, while I'm not sure if single suited minor is.
  
• I agree, the SCUM 1M openings can easily be raised on three, both in and out of competition. Even if this might lose on some hand types it will be the percentage action, so I would systemically raise with that hand type.
  
• I think my approach gains when we have a 4-4 major fit as well as when we have a diamond fit (particularly 4-4) and can apply pressure. Auctions like 1♠-(X)-2♠ and 1♦-(1♥)-3♦ are difficult for the opponents to deal with, and are in great part facilitated by the high frequency of these openings and the negative inferences of opener not holding 4oM. I also obviously get to bid 1X-3NT and 1M-4M more often, which I think will be a noticeable single dummy advantage. Also when it is not our deal, e.g. 1♠-(2♥)-P-?, this applies more pressure than most other systems.

I have been having trouble with posting special characters in my BBO posts in particular, and think other people may be running into the same issue. I've been typing 'canape' instead of 'canapé' and then editing the é in after, which seems to be a workaround.

[pescetom]

I thank you for this interesting thread, useful to a wide range of skill levels and whether actually playing strong club or not.

Can I drop back for a moment to the NT ladder, and in particular the ranges above 1NT and the possible uses of 2NT.
I briefly tried 14-16 1NT in a standard system and while I was happy with how the subsystem still worked and the higher frequency of the opening, I was concerned that the 17-19 range below 2NT was rather wide. Did you consider addressing this by using 17-18, 19-20 and so on? It would seem that using Birthright over 1♣ you have enough NT ranges to drop all by 1 without problems.
I also must admit that if I was to play this, my instinct would be to retain 2NT natural as you mentioned being possible, as I imagine these are the most vulnerable of all hands going through 1♣.
It also seems like a low cost change, as all our strong club pairs seem to use 2NT for some very low frequency meaning (probably sound 5-5 minors). Is there a convincing artificial use for 2NT?

[ulven]

As the author of SCUM I enjoy following this thread and the comments. My only comment for now is that Raptor 2M openings have statistically been a big winner spanning over 20+ years (there has been periods where I for other reasons haven't played them). Sometimes what looks questionable in theory has to be evaluated in actual play. And my experience is against the very best. I have had partners who failed to get full mileage out of them.

I now play 2D as weak M, 2C as minors and 1D as weak NT or 6+m (no M).

I will leave my comments on 5332 for another day.

[DavidKok]

pescetom, on 2024-February-25, 10:47, said:

I thank you for this interesting thread, useful to a wide range of skill levels and whether actually playing strong club or not.

Can I drop back for a moment to the NT ladder, and in particular the ranges above 1NT and the possible uses of 2NT.
I briefly tried 14-16 1NT in a standard system and while I was happy with how the subsystem still worked and the higher frequency of the opening, I was concerned that the 17-19 range below 2NT was rather wide. Did you consider addressing this by using 17-18, 19-20 and so on? It would seem that using Birthright over 1♣ you have enough NT ranges to drop all by 1 without problems.
I also must admit that if I was to play this, my instinct would be to retain 2NT natural as you mentioned being possible, as I imagine these are the most vulnerable of all hands going through 1♣.
It also seems like a low cost change, as all our strong club pairs seem to use 2NT for some very low frequency meaning (probably sound 5-5 minors). Is there a convincing artificial use for 2NT?

I have three partial answers to your questions, which I hope combine into a full answer.

• On using 2NT as strong (say, 20-21) in a strong club system: I don't know what's better. The weak use, e.g. 4-10 HCP 5-5 or longer in the minors, is very rare and doesn't win some of the time when it comes up (you are often either in a misfit or telling the opponents that they have a major suit fit and can play the hand). I've looked into other preemptive uses, most of the treatments I found interesting are brown sticker, rather poor, or both. The problem with 2NT weak is in my opinion threefold:
  • We have bypassed 2M and 1NT, so we need to have enough shape to justify going past the most common partscores.
    
  • It is a practically forcing preempt, allowing LHO two bites at the apple. This makes it less useful than most natural preempts, regardless of meaning.
    
  • It wrongsides 3NT should we wish to play there later.
  These factors combine and limit the interesting uses. By contrast 20-21 balanced is fine, though it is 'the slam killer'. Personally I feel like I'd rather open those hands with 1♣, especially as the transfers I play in competition after interference can handle the 20+ balanced hands well. So by putting this hand type in 1♣ I'm hoping to gain on the balanced hands, in addition to getting some rare use for 2NT. I don't have enough experience to say whether opening 1♣ with this hand type is a winner on balance, but I suspect it is.
  
• The standard auction 1X-1Y; 2NT (showing 18-19 balanced) is uncomfortable, especially if responder can be quite light for their initial action. Being able to show this hand type at at the 1NT-level gains frequently and eliminates most of the issues of increasing the range to 17-19. Out of the standard systems I prefer Dutch Doubleton, but e.g. T-Walsh can also do this (and if you pair it with a form of Gazzilli over 1M-1Y the only hand type where you have this problem is 5♦332, and I prefer to open 1♣ with that in standard systems for a couple of reasons anyway). Also the 17-19 NT can be a liability in standard systems on competitive auctions, especially in combination with support doubles, but strong club does not have this problem (partner will assume 17-19 balanced in competition as the weakest hand) though it does have different ones.
  
• Strong clubbers get to show the 17-19 balanced hand 'for free' - 1♣-1♦; 1NT shows the bottom range at the 1-level. With this I feel that the 3-point range is fine, though you could decide to make this 17-18 and put the 19+ through a Birthright 1♥ continuation. To be honest I do in general not care that much for the uninterrupted responses to 1♣ - the numbers are just lousy. I've ranted a bit too much about this before but for an overview: the 1♣ opening is ~8% of all hands, LHO passing is maybe half the time so we're at 4%, and then we have to split those between our entire response structure, most of which is an easy game force at the 1-level. I'm still experimenting with options here, having played (among other things) KK Relay, Standard Modern Precision, TOSR, Transfer Superpositives, 1♣-1♦; 1♥ Birthright and C.C. Wei's 'all natural', and I didn't notice big differences (though I did like SMP, i.e. Mecklite, on 1♣-1♦). I'd love to have a more complete set of notes on how to use a Birthright 1♣-1♦; 1♥ - the versions I've tried were all a bit clunky.

[etha]

The latest 2NT I am trying is weak 5+ !h and 5+ minor. The idea being they can no longer bid 2!s