[DavidKok]

This will be a very long post. If you are scared of the C-word ('canapé'), it is not too late to close the tab.

For a bit over a year now I've been very interested in strong club systems, and written a number of posts on them. I thought I had included my motivation for systems like this in https://www.bridgeba...lub-with-a-4cm/ and in my own documentation of Kok Canapé, which can be found in an external google documents link here as well as a pdf (available on request), but this is only partially true. So this post will be a motivation for this approach to systems in general, and my steps to creating Kok Canapé from Cottontail Club in particular. This will be a long story with multiple sections. I will go over the design choices and tradeoffs of the opening system, and some brief tangents into the continuations. This is not a full overview of my bidding system, though the aforementioned google document does a much better job of that. Instead I hope to show the strengths and weaknesses of this system, and how I went about discovering and designing it.

This post is split into the following sections:

• Strong club.
  
• The notrump ladder.
  
• Four card majors.
  
• Canapé.
  
• Weaknesses and problem hands.
  
• Competitive auctions.
  
• Conclusions.

As a general remark, I will be talking about HCP ranges and shape requirements a lot. Please keep in mind that I always encourage deviating from written norms if you feel the hand is suitable. This includes, but is not limited to upgrades, downgrades or describing the hand as a different pattern than the one you have if you feel it is suitable. I use HCP frequently but loosely as a communication tool for sharing what I expect a typical hand in that range to do, e.g. "11-15 HCP". It would be more accurate but, in my opinion, less clear, to instead refer to this as "an aggressive minimum or slightly above minimum opening". Please exercise your own bridge judgement in deciding which way to go with any particular hand on close decisions, using the written descriptions as a guide at best.

[DavidKok]

1. Strong club.
I like strong club systems. I've previously mentioned this, for example here, here and here, though this particular comment gives a better insight in my opinion. The reason I like strong club systems is quite simple: we embrace the weakness of the 1♣ opening, but in return get to benefit (at least in theory) on the limited openings.
The limited openings are narrowly defined 'for free' on constructive auctions - we can play relatively standard gadgets here, with the bonus that the limited nature of the hands means that all ranges and descriptions automatically become narrower. But it is the competitive auctions where the limited openings really get their edge. Responder can bid more freely in competition both with and without support, knowing that opener will rebid only with great shape and not jump to some random hopeless game based on having a few extra kings and a misfit. Opener can rebid more freely in competition as well with shapely minima, knowing partner won't play them for a big hand. And both opener and responder can also pass in competition more freely, knowing that they are unlikely to be missing good games or partscores when partner cannot act despite all this newfound freedom.
In addition to the above, limited opening systems famously have room to open more aggressively than standard systems, and getting the first blow in can be half the battle. The current frontier seems to be opening all (or 'all', please do not let high card point count be a substitute for judgement ever) 11-counts and a good fraction of unbalanced 10-counts, something that is risky but probably percentage in standard systems and simply percentage in strong club systems.
The numbers on how many auctions are competitive versus constructive differ. Going through vugraph records, about 60-70% of all auctions seem to be competitive at top level events. At my club it seems to be closer to 80-85%. In preliminary rounds of strong events (again, going by vugraph) it seems closer to 50%.
A lot of players agree (and I certainly believe) that the game is getting more competitive, so I expect these numbers to go up rather than down. In my view a limited opening system is a great choice for dealing with this competitive landscape, provided the tradeoffs are chosen carefully. We want the generic strong bid to be infrequent, but not so infrequent that we lose the upsides of the limited openings. We'll be behind almost every time we open 1♣, but not by an insurmountable degree. As long as we use the other openings much more often we will probably be ahead on balance. The frequencies are very important here, see also the particular link above. If 1♣ is "16+ unbal or 17+ bal", as I prefer, that covers approximately 7.9% of all hands. The "11-15 any or 16 balanced" category covers approximately 34.3%. And "preempt or pass" covers the remaining 57.8%. I think those are good ratios, though we may wish to reduce the pass frequency by including e.g. a Kamikaze (10-13) NT, aggressive weak 2's, or reducing our opening requirements to include 10-counts (my figures above exclude 10-counts, under the hope that infrequent nice 10-counts are compensated for by the lousy 11-counts worth a downgrade). To summarise: if we keep the 1♣ opening rare and the limited openings common, I think we come out ahead with this system design, as we are well prepared for competition on the most frequent opening hands.

Lastly, getting ahead of ourselves a bit, several other frequent styles of limited opening systems exist - i.e. Polish/Swedish club (including a weak option, traditionally the weak notrump, in the 1♣ opening), strong diamond systems, strong pass systems and even wilder proposals. I won't go into detail here but I think these systems take on significant weaknesses in competition, ironically exposing themselves to the very thing they set out to combat.

[DavidKok]

2. The notrump ladder.
If you open all 11-counts and no 10-counts, approximately 48% of all opening hands are balanced (i.e. 4333, 4432 or 5332). If you include semibalanced hands with suitable values in the short suits this number goes up all the way to 59%, though of course this is subject to personal style and preferences. Either way any bidding system relies a great deal on its ability to handle balanced opening hands, simply because of their overwhelming frequency.
There is a second reason to place special emphasis on balanced hands - these hands struggle most in competition. If you have to pass or make a nebulous bid with a balanced hand (even if it's a strong nebulous bid) it can be difficult to enter the auction on later rounds as you don't have a long suit to bid. In accordance with my previously stated desire to prepare for the competitive auction being ready with balanced hands is one of the key design criteria.
In particular, combining the wish to open balanced 11's, to keep 1♣ frequent and to do well with balanced hands, we are going to need to have two notrump ranges below 1♣. The ladder I prefer and that matches my criteria is 11-13, 14-16, 17+ (putting 17+ in 1♣), though others exist (in particular 10-13 Kamikaze 1NT, 14-16 in 1-suit, 17+ in 1♣, or adjust the strength of the above by a point if desired). The options that I wish to avoid are systems along the lines of '12-15 opens 1NT, 16+ opens 1♣' which

• Makes the fragile 1♣ more frequent and vulnerable.
  
• Passes balanced 11-counts.
  
• Has an uncomfortably wide range for the 1NT opening.

Kamikaze and variable NT structures exist and do very well, but for now I'd like to go with the 'simple' option of putting 11-13 balanced in the 1-suit openings and 14-16 balanced in the 1NT opening. This choice is not arbitrary - playing a strong notrump has three marked benefits over the converse.

• We get to play a 2/1 structure with a semiforcing notrump over the 1-suit openings, where the weak notrump hands can pass the 1NT response and the non-minimum hands are not balanced and therefore have a natural suit rebid available. Strong notrump hands can be very awkward on the 1X-1NT start to the auction, and are generally paired with a weak (6-9 ish) notrump response to avoid this problem, putting more strain on the rest of the system.
  
• Playing a strong notrump finds more major suit fits total than a weak notrump. Facing a strong notrump responder more often has the strength required to make a Stayman asking bid, finding the 4-4 major fit if one exists, while we will locate a 4-4 major suit fit 'always' on the slower 1-suit start of the auction.
  
• In most weak notrump systems responder wishes to play opener for some extras (in the form of strength or shape) on competitive auctions, which discourages opening unbalanced light hands with the playing strength of a weak notrump. In a limited opening systems these ~11 HCP unbalanced hands are ideally opened regularly while failing to meet this criterium, creating competitive problems for responder.

Combining the above we have a rudimentary notrump ladder:

• 11-13: Open 1-suit, bid (or pass) 1NT.
  
• 14-16: Open 1NT.
  
• 17-19: Open 1♣, bid 1NT.
  
• 20+: Open 1♣, jump rebid something.

This brings me to the structure over the 1♣ opening. I won't go into detail - many great options exist, and I think it is not a very important part of the system. Some options can handle a continuous NT range in 1♣, others struggle with it and demand a 20-21 2NT opening to introduce (strong) split ranges. I think either is fine and we can decide if we want this 2NT opening for something else later, though on general principles I'd prefer to put big balanced hands in the 1♣ opening.

[DavidKok]

3. Four card majors.
Way back when, this started as an experiment. Precision systems, i.e. strong club with five card majors, incur a second weakness alongside the strong 1♣. I've gone into this in more detail before in this post, but the summary is as follows: if we want 1♣ to be strong, 1M to show five, and 2♣ to not contain Raptor (in particular, 4M5♣) hands, we end up with a nebulous 1♦ opening (or worse). I am particularly fond of the Auby Ebenius Club approach to Precision, handling the nebulous 1♦ better (in my opinion) than most other Precision systems. In fact, I like it so much I based my simplified Precision system on it.
That being said, the nebulous diamond is still a weakness. As mentioned in the links above, it is simultaneously the most frequent opening bid, the least descriptive opening bid and contains the weakest hands (11-13 balanced no 5cM), making it a competitive liability. This may well be the least of all evils - Precision systems get their powerhouse 1M openings and keep 2♣ pure. But, just as an experiment, I was wondering how much we could 'clean up' the nebulous 1♦. Auby Ebenius had one approach, IMPrecision and SCUM/SCRUM a second, and in the spirit of exploring all options I looked to a different one: four card majors.

Without going through the arguments at the start of the aforementioned thread again, four card majors can solve most of the problems with nebulous bids. We can start with an all natural base of 1♦ promising 4, 1M promising 4 and 1♣ promising 5 (but denying a 4cM) and then we only have to put the 3=3=3=4 11-13 hands somewhere - I suggest 1♦. Easy, right? This 1♦ can be raised on four, and we open our longest suit or four card suits up the line except for 4M5♣ hands (open 1M rather than 2♣), 3=3=3=4 11-13 hands (open 1♦) and we bypass four clubs if balanced.
While one could certainly play an opening system like this, it almost definitely wouldn't be that good. We'll explore the reasons for this in more detail in the next two sections, but for now let's begin by observing that 4cM systems are very unpopular at the highest level, and I don't think it's a coincidence. The problem is that this style of four card major opening contains three hand types opposite which responder has to evaluate their hand very differently: 4cM balanced (notably 4M333, but also e.g. 2=4=3=4 11-13 in 1♥), the unbalanced Raptor type (e.g. 4=1=3=5 in 1♠) and the unbalanced 5(+)M. The first should rarely be raised, and is often happy defending. The second hand type is aggressive but often needs four card support to play well in the major. The third wants to be raised aggressively, even on three card support. And poor responder has no idea which hand to cater for, especially in competitive auctions. Oops.

[DavidKok]

4. Canapé.
The problem highlighted at the end of last section can be made more specific - what should responder do with 3-card support?
At this point a little digression is helpful: in standard 4cM systems, for example Acol, usually you would open four card suits up the line. As a result the 1♠ opening is only four when exactly 4=3=3=3, and in particular is very similar to a 5(+) card 1♠ opening. Similarly, 1♥ is only four on 3=4=3=3 and 4=4=(32). And some modern adaptations of these systems even open a minor suit, typically 1♣, with the 4M333 hands outside of the 1NT range. As a result the 1M openings are very similar to 5cM openings, and responder does well to raise on 3 always.
I've been told some 4cM players go a different route, stressing opening the major suits, or at least granting the flexibility to exercise judgement and open 1M on a four card suit more regularly. This means responder is facing more of a guess with three card support, but this style might well prove beneficial on balance - after all, the opponents are also guessing. Either way we don't have a choice. The whole point of the 4cM approach to strong club is to get the Raptor hands out of 2♣ and out of the nebulous bids by putting them in 1M, which means the frequency of 1M being exactly four will always be significant. So instead of shying away from this, let's lean into it. Open the 4cM ahead of an equally long or longer minor suit, applying more pressure. Responder was always going to need systemic solutions with 3-card support, so it's fine to increase the frequency of 1M being four if we accept that. I think of these hands as 'one bid hands' - if we have a minimum (or slight extras) opening with a 4cM, probably that is most of the information partner is interested in. This is similar to the Walsh principle over partner's 1♣ opening - if we aren't strong enough to take control of the auction, which almost tautologically means 'strong enough to open 1♣' in this system, show our 4(+) card major suit(s) and leave the rest to partner. The analogy is far from perfect, but I do think we frequently have revealing and convoluted auctions showing balanced or semibalanced minimum opening hands in standard when really all I'm looking to find out is a specific 4cM and then I want to sign off in 3NT versus 4M.

So responder needs a systemic solution for bidding with three card support. Roughly speaking, there are four approaches. Obviously I don't like 1 through 3 and I put canapé last. Here we go.

• Frequently raise on 3, accepting possible Moysians (especially at the partscore level or in competition). Moscito does this, as a notable example. Typically they remove balanced hands from the (openings showing) 1M, so that if we do play a partscore in a Moysian at least there will be ruffing values and a good chance the opponents have a fit and a making contract themselves. This is a nice idea - put 11-13 balanced in 1♦ even with a 4cM, and make 1M unbal 4(+)M, somewhat similar to SCUM/SCRUM though far more extreme in its application. In this style 1M is very frequently 4 though, and we are exposing ourselves to a great risk with the frequent Moysians. Also we are stuck for an action if the opponents raise the stakes past 2M, as now we are the ones guessing the total trumps on the deal.
  
• Play two raise systems, one with 3-card raises and one with 4-card raises. Standard bidders already do this both in and out of competition. Auken - von Arnim played a style like this for a while, as did some other pairs. A big downside is that we often have no clue about our safety level with 3-card support. E.g. the standard 1M-(2♣)-3♣ as an invitational+ 3-card raise does not work if our best score is defending 2♣, or playing some other suit at the 2-level, both of which are realistic scenarios if opener can be 11-13 balanced with four cards in the major. Even the constructive auctions, e.g. 1♥-2♦* (good 3-card raise), aren't completely comfortable (we might well belong in 1NT).
  
• Give up, it was a nice try but this is the (unacceptable) cost of 4cM.
  
• Take the five card hands out of 1M - canapé.

The last option is radical and uncanny, but immediately solves the problem of what to do with 3-card support (do not raise). We wish to suppress the frequency with which a 1M opening contains exactly five cards in the suit bid, making it "4 or 6(+)". This can be achieved to a reasonable degree by playing a canapé style, where with an unbalanced hand with a five card suit and a four card suit we open the four card suit. This way the only hands that have to open their five card major are 55 majors hands and 5M332 11-13 - the former is very rare (and we actually gain here compared to standard) while the latter plays well in 1NT anyway, so losing the 5-3 fit on competitive auctions is not disastrous.
Traditionally strong club 4cM canapé bidding systems insist on their 1-suit openings being unbalanced, keeping the inference of "6+ in the suit opened or an undisclosed outside 5(+) card suit" pure, aiding responder in competition. This is unfortunately incompatible with our notrump ladder, so we won't be doing that. But we can try to salvage the "4 or 6+" principle as much as possible. I've tried to summarise the opening rules here in the hopes of providing clarity while keeping it brief:

• With an 11-13 balanced hand, open the 4(+)cM (if any, and 1♥ with both), or 2♣ with 5♣332, or 1♦ with the remainder.
  
• With a single-suited hand, open that suit.
  
• With a two-suiter (at least 5-4 in length) not including clubs, open the shorter suit.
  
• With a two-suiter with clubs, open the other suit unless it's 4♦5(+)♣, then open 2♣.

This way opener can raise 1M on four and relatively safely not raise 1M on three.

[DavidKok]

5. Weaknesses and problem hands.
I mentioned attempting to salvage the "4 or 6(+)" idea earlier. We won't quite get there, and this style of opening structure incurs losses on four particular hand types. In this section I want to briefly go over the four predictable problem hands for this system, and the next section will explain how to deal with these issues on competitive auctions.

• 11-15 hands with 5(+)M4(+)♣. In an earlier version of this system I took on the Roman 2M openings, showing exactly these hands. I have since abandoned that - the openings are lousy, and I'd much rather have a preempt. So these hands regrettably have to open 1M, which comes with a double cost: not only might we now miss a 5-3 fit, also the clubs rebid is no longer specific about relative lengths. Both of these problems can be solved to a decent degree - using 1M-1X; 2♣ as non-specific relative lengths is surprisingly fine (there is plenty of room for a 2♦ asking bid). Furthermore, in competition we can use 2NT as a gadget to disambiguate relative lengths. This is probably one of the biggest system losses, and while the above allow for reasonable recoveries they don't cover the full problem.
  
• 11-15 hands with 5♥4(+)♦ especially can struggle, opening 1♦ and finding 2♠, 3♠ or 4♠ come back to them without having shown the major. So far I've never actually missed a valuable fit here in competition (see also next section), but inevitably it will happen. We can recover somewhat with a good application of takeout doubles by opener, that way we are not committed to the 3-level to show the suit. Also using 2NT as a gadget can again play a role. But sometimes I expect to lose here.
  
• 11-13 5♣332 hands. I don't really get why this works, so far opening 2♣ with them has been a big winner. But theoretically I expect it to be a loser on balance, especially if the hand has poor texture for a suit contract. Currently I am allowing opening 1♦ on three with (32)=3=5 11-13 to remedy this somewhat, despite the fact that results so far indicate it's not strictly required, but this certainly is one of the scarier aspects of the opening system to me.
  
• 16+ hands with short spades and long hearts, especially two-suiters. These systemically open 1♣ in strong club systems, and may well have to find a bid over 4♠ when the auction gets back to us. This is unpleasant but can be resolved somewhat by allowing responder to compete aggressively, as detailed in the next section.

That being said, this system 'naturally', i.e. we haven't even defined the responses and continuations yet, has a lot of strengths. In no particular order:

• We find our 4-4 major fits immediately, which often play better than (even a longer) side suit fit.
  
• The negative inferences about opener not having (exactly) four cards in unopened majors often allows us to identify a fit for the opponents before even they can, and we can apply maximum pressure in competition.
  
• We can stop lower on misfit deals, allocating more sequences to unbalanced hands (standard and also our reverses) and clarifying shape and strength faster.
  
• We find 3-5 fits faster.
  
• The openings 2♦ and up are completely free for preemptive uses, not requiring plugs for constructive system gaps.
  
• We gain all the benefits of a limited opening system - i.e. we can jump to game more frequently than any other system (4-4 fits are more common than 5-3 fits), bid with weak shapely hands on competitive auctions or pass even with quite some strength anticipating a misfit.
  
• The system has no nebulous openings, even 1♦ can (and should) be raised on four both in and out of competition.
  
• Standard tools for competitive auctions, notably takeout doubles, are less effective against canapé openings (the doubler often has a takeout double of our long suit, rather than the suit we opened).

[DavidKok]

6. Competitive auctions.
Competitive auctions after a 1-suit opening in Kok Canapé are fascinating. Wear a seatbelt.
There are several main ideas that help decide what to do in competition. A lot of these auctions rely to a huge degree on negative inferences, and it takes time to get used to these. Personally I get a kind of soft whiplash switching between this system and standard over how much the meaning of every bid changes on competitive auctions in particular, even though (almost) everything is natural. First, some ground observations and rules:

• New bids by responder are forcing and natural, double is takeout, raises are competitive, the cue is a good raise (but keep in mind that in a limited opening system responder may also choose to simply blast game).
  
• New bids by opener are natural, though we reserve the cue and 2NT for possible gadgets.
  
• If opener can show their canapé at the 2-level they will always do so.
  
• Voluntarily bidding a new suit at the 3-level as opener shows extra playing strength, usually a 6-4 hand (maximum 5-4 hands can make a takeout double instead).
  
• If responder makes a takeout double at the 1- or 2-level this means 'I want to hear about your canapé suit even if you have to go to the 3-level to tell me'. It denies a fit and promises 3(+) cards support in every unbid suit as well as the opening suit (or sufficient values to control the hand despite a possible misfit). If fourth hand passes and opener has the 11-13 balanced hand they will make the cheapest available rebid of 1NT, 2M (having opened 1M) or 2NT - the single-suited major hands have to jump rebid.

So far everything is natural, though that takeout double might look scary. Actually it's just a negative inference from playing canapé - if partner opens 1♥ (say), RHO overcalls 1♠ and we wish to compete, we already have the options of raising hearts with four, bidding a five card minor suit or bidding 1NT with a defensive hand. As opener will rebid their 2m canapé if they have one always, double is used to say "hey, if they raise to 2♠, I still want to hear about your suit". This is the percentage action with hands as skewed as 2=3=(53), as we now wish to play the (possible) 5-3 minor fit rather than let the opponents rest in their (likely) 8(+)-card spade fit.
A slight word of caution though - this double does promise some values, approximately 7(+) HCP or so in a suitable hand. Opener is allowed to jump rebid with extras (based on shape and playing strength and the assumed fit opposite your double), so if you really have a dreadful 6-8 count just pass and prepare to defend. Facing 11-15 this is a very low risk action, and one example of taking more liberties with passing I alluded to previously.

The takeout double tool is particularly relevant on the 1♦-(1♠) start. We have the following negative inferences: partner does not have (exactly) four spades, and likely does not have a spade canapé (and if so, we'll gobble them right up). That means partner has at most 3 spades, so if we have 2 we know they have a fit. Do we want to play 3m or 3♥ rather than defend 1♠? If so, we should bid our long suit now or double. If not, partner will compete to 2♥ but not past 2♠, unless they hold significant extra playing strength. The same principle is at work on many other auctions: 1♦-(2♣)-? and we have a 3=3=4=3 10-count - bid 2♦! Opener will show their major if they have one, and 2♦ is likely the limit facing 11-13 balanced with 4(+)♦ (and it is highly unlikely that opener has got the 3=3=3=4 or (32)=3=5 on this auction). 1♦-(X)-? with a 2=4=4=3 5-count, bid 3♦! Unless partner has the spade canapé they've got a spade fit and we're taking LAWful action (again, braving the 3=3=3=4 shape), let them sort it out at the 3-level. For all I know game their way might be cold.

I promised to say something about interference over the strong 1♣ opening, in particular with regards to recovering a possible heart fit when opener has short spades and long hearts. My opinion is as follows: since opener promises strength, responder should be free to act with shape even on relatively weak hands. Facing a 16+ (not 16 balanced) 1♣ opening the relevant ranges for responder are approximately:

• Weak: 0-4 HCP (not an ace).
  
• Invitational: 5-7 HCP or an ace.
  
• Game forcing: 8+ HCP.

So leverage this. Start bidding with 5-counts. Get your suits in. Bids should show shape and an invitational+ hand. 1♣-(2♠)-?, time to make a takeout double with 4 hearts and 5 points. Transfer approaches are particularly effective here, combining a bid below the safety level with rightsiding. There has been detailed and extensive discussion on this topic previously in this thread and this thread, and I stand by most of what is written in there.

I also wanted to include the hand from this post the other day: assuming a P-(P)-1♠*-(P) start - nothing out of the ordinary so far - responder should reason as follows: "I have 9 facing 11-15, let me look at my outside shortages. Partner has either got at most 3 hearts or 5(+), because with exactly 4 they would have opened 1♥. If it really is 3(-) then the opponents are concealing their 9(+)-card heart fit while having approximately half the deck, and we likely belong in 3m. There is a good chance we can get there if they compete in passout over 1♠, and the chance that opener has the perfect maximum in spades while the opponents are concealing their hearts is remote. Much more likely is that partner has a spade-heart canapé, in which case our combined 20-24 HCP plus misfit in both majors means game is very unlikely. On misfit deals stay low, so pass in the 4-3 fit at the 1-level (or, if more optimistic, respond 1NT intending to pull 2♥ to 2♠)".

[DavidKok]

7. Conclusions.
The development of Kok Canapé, based on Cottontail Club with inspiration taken from MICS, Larry Lowell and others, is still ongoing. The system has notable weaknesses, as listed in an earlier section. At the same time it has exceeded all my expectations. I listed some (though far from all) upsides previously - the system does many of the things I want, and almost none of the things I don't want in a bidding system. I will continue to develop it and hope to be able to play this seriously for an extended period of time. Personally I expect the system to do better at the table than it does on paper - as mentioned takeout doubles are of limited value against a canapé system, and many opponents will be caught unprepared. The high frequency of 1M-2M auctions will also apply considerable pressure, and I haven't even introduced the ways to quickly bid games while leaking minimum information yet in this post. While figuring out this system from the ground up I kept finding new situations where everything slotted in place just right without any effort on my part.

There are aspects that I left out in the discussion above, as well as points that I am still trying to figure out. Should I try to include a Kamikaze NT or variable NT? Should the opening requirements be different in third and fourth seat? Those 5♣332 11-13 hands, shouldn't they just go to 1♦? What's a good system to play over the 1♣ opening, every third week of the month when LHO is silent? How many gadgets do we want to have in competition, and which ones are most important? Does the canapé style help with defending through inferences, or does it do more harm than good? I don't have the answers, but I'm looking forward to exploring all these options and improving the system as I go.

[PrecisionL]

Well done, David.
We (and David) have used Roy Hughes suggestions for dealing with opponents' interference over strong ♣ opening with judicious use of transfers.

Building A Bidding System, 2005, Roy Hughes, pgs. 58-59.

We (and not David) have also been experimenting with a 2/1 response of 2♣ to be GI or better. This allows 11-13 openers to rebid 2♦ showing that range and balanced without a canape.

[awm]

While I agree with many of the things you say, I don't agree with the conclusions. An 11-13 notrump hand (especially without a five-card suit) will usually be a disappointment to partner in competition. There are lots of sequences where you want to compete if partner has an unbalanced hand but not opposite the weak notrump. For example, 1♥-(2♠) and you have 1345, how much do you need to double? If partner has six hearts or a five-card minor or 5♥/4♣ you really don't want to pass here and force him to guess whether to balance (not to mention the opponents might raise to 3♠ or 4♠). But if partner has some 44(23) hand your double could easily get you too high. Even if you have four-card heart support on the same auction, you need to decide how much strength is necessary for a 3♥ raise, and the answer will be quite different opposite 11-13 balanced than the other options.

Since 11-13 notrump hands are really common, I'd prefer to lump them all into one bid. This means the other openings guarantee that they are not this hand, giving much better results in competition. After the "usually a weak notrump" 1♦ opening, partner knows to go slow/passive in competition. This also allows us to play things like competitive transfers or negative free bids opposite the "usually a weak notrump" opening.

Of course there are not enough calls to play 1♦=11-13 balanced and 1NT=14-16 balanced (you have to put the raptor hands somewhere, three-suiters somewhere, minor one-suiters somewhere, etc). So for me the question is "what other hands do I lump in with 11-13 balanced and hope to survive"?

By having 11-13 balanced as a common option for all three natural one-suit openings, you seem to create more problems for yourself.

[DavidKok]

Thank you for your comment. I mentioned AEC (1♦ = 11-13 bal, 11-15 Raptor or 4441's), Moscito (the 1M-1's are 5(+) or Raptor but not balanced), SCUM (1♦ = 11-13 bal, 1M = 11-15 5(+) or 4441, Raptor hands open 2M) and traditional strong club 4cM canapé (1NT = 12-15, 1M is 4(+) unbal) in the text, among others. All different approaches to cordoning off the weak notrump. My experience has been the opposite: these systems struggle with 3-card (non-)raises more than my approach on the 1M openings, and have a high frequency nebulous bid that is difficult to leverage in competition. Conversely, showing additional shape immediately on the weak notrump hand is a frequent winner, as partner can now get a lot of 2-over-2 and 3-over-2 decisions right facing the common WNT. If you do wish to go this route I think AEC is very strong and sensible.

In Kok Canapé the playing strength of a lot of the unbal hands in 1M is similar to that of 11-13 balanced. In particular, the minimum Raptor hands aren't all that different from balanced, and the most important piece of information to share in these 'one bid hands' is their major suit length.
Conversely, the upper end of the range (13-15 unbal) is quite different in playing strength, but these hands can frequently take another bid in competition. This is a benefit of the strong NT in this system - all maxima in 1-suit openings are not balanced, and therefore have a natural rebid.

I want to stress that I'm not claiming that there aren't any competitive issues. Of course there are, all systems have some weaknesses. But I think that this style is ahead of the alternatives on most of these competitive auctions, averaged over all opening hands. If we know that LHO is about to (jump) overcall 2♥, how would you prefer to distribute the hands? I think that having openings that help partner with what to do with exactly 3 or 4 spades and non-GF values is a good start.

[nullve]

What are Opener's rebids with

a) 11-13, 3433
b) 11-13, 3424
c) MIN, 3415
d) MAX, 3415
e) MIN, 3514
f) MAX, 3514
g) MIN, 3613
h) MAX, 3613

over 1♥-(2♠)-X-(P) ?

[PrecisionL]

nullve, on 2024-February-23, 10:35, said:

What are Opener's rebids with

a) 11-13, 3433
b) 11-13, 3424
c) MIN, 3415
d) MAX, 3415
e) MIN, 3514
f) MAX, 3514
g) MIN, 3613
h) MAX, 3613

over 1♥-(2♠)-X-(P) ?

Our canape system limits 1♦ openings to 4 or more cards unless 3-3 in the majors. (1M response = 5-cd suit).

a) Open 1♥ if Qxxx or better otherwise pass, pass 1NT response, over 2♣ rebid 2♦ = 11-13
b) Same as a)
c) Open 1♥ and rebid 2♣ (canape)
d) Same as c) unless 5-losers, then jump rebid
e) Open 1♥ and pass 1NT response (don't canape with weak hand unless suits are pure)
f) Open 1♥ and rebid in NT to show max and 4♣ (Weiss design)
g) Open 1♥ and rebid in ♥.
h) Same as g) unless 5-losers then jump rebid.

HAPPY CANAPING!

[DavidKok]

nullve, on 2024-February-23, 10:35, said:

What are Opener's rebids with

a) 11-13, 3433
b) 11-13, 3424
c) MIN, 3415
d) MAX, 3415
e) MIN, 3514
f) MAX, 3514
g) MIN, 3613
h) MAX, 3613

over 1♥-(2♠)-X-(P) ?

With so much bidding space gone I would bid naturally. So a+b bid 2NT, c+d bid 3♣, e+f bid 3♥ though f might upgrade to 4♥ with little spade wastage, g has to decide between 3♥ and 4♥ while h bids 4♥ always. There is room for gadgets here, e.g. good/bad 2NT or something more fancy, but I'm not sure it's necessary. Partner's double denies four hearts but does show a desire to compete over 2♠, so partner will be approxmately 2=3=4=4 with competitive values or slightly offshape with game forcing values. With the entire 2-level gone it's difficult to invite, and responder should more frequently blast or pass without clear direction (keeping in mind that pass need not end the auction).

[awm]

DavidKok, on 2024-February-23, 11:51, said:

With so much bidding space gone I would bid naturally. So a+b bid 2NT, c+d bid 3♣, e+f bid 3♥ though f might upgrade to 4♥ with little spade wastage, g has to decide between 3♥ and 4♥ while h bids 4♥ always. There is room for gadgets here, e.g. good/bad 2NT or something more fancy, but I'm not sure it's necessary. Partner's double denies four hearts but does show a desire to compete over 2♠, so partner will be approxmately 2=3=4=4 with competitive values or slightly offshape with game forcing values. With the entire 2-level gone it's difficult to invite, and responder should more frequently blast or pass without clear direction (keeping in mind that pass need not end the auction).

Does double here promise 3♥? My impression was not necessarily, for example with 22(45) and a good hand you really want to find your nine-ten card minor fit. Even with a shape like 3154, while I suppose you could bid 3♦ (forcing) you might want to double (accepting a 6-1 heart fit on occasion), but if opener is blasting 4♥ on a max with five hearts this all seems too dangerous.

[DavidKok]

awm, on 2024-February-23, 12:11, said:

Does double here promise 3♥?

DavidKok, on 2024-February-23, 11:51, said:

Partner's double denies four hearts but does show a desire to compete over 2♠, so partner will be approxmately 2=3=4=4 with competitive values or slightly offshape with game forcing values.

awm, on 2024-February-23, 12:11, said:

Even with a shape like 3154, while I suppose you could bid 3♦ (forcing) you might want to double (accepting a 6-1 heart fit on occasion), but if opener is blasting 4♥ on a max with five hearts this all seems too dangerous.

DavidKok, on 2024-February-23, 11:51, said:

e+f bid 3♥ though f might upgrade to 4♥ with little spade wastage

I don't really understand your question, sorry. Opener's hand will most likely have decreased in value with three spades sitting under the 2♠ jump overcall, though a hand that sees game opposite a 2=3=4=4 8-count or so should make a strong rebid (4m or 4♥). I would not double with 3=1=(54) and so-so values, rather pass and wait for partner to keep it open with short spades, defend knowing there's a misfit, or upgrade into 3m GF (or 3NT to play). Opener will rebid hearts every time they have six, and the 6-1 fit at the 3(+)-level does not seem good to me.

[awm]

---

---

Where did NS go wrong?

[DavidKok]

On the first hand I would bid 3♠ or 3♦ with South, on the second hand I would rebid 3♠ or 3NT as North rather than 4♥. As mentioned before I could attempt to introduce more gadgets here to address some of these problem hands, but I think I'm gaining enough with all natural that I wish to give that a try first. It also helps keep the system simple.

But I think that doesn't actually answer the question you haven't dared to ask, which is "if I construct a sufficiently perfect example, does that discredit the method?". I think the answer is "no". I'm happy you are helping come up with possible problem auctions and hands to help sharpen my judgement, but to me your attempts feel hostile rather than helpful, almost like a "gotcha". Could you perhaps embed it in a statistical analysis, or list advantages and disadvantages compared to other approaches? I've mentioned this twice before: I think any system has problem hands and weaknesses, and I think this system will do well on average (which is obviously different from "do well in every scenario").

Or were you only looking for clarification on the meaning of the bids?

[awm]

DavidKok, on 2024-February-23, 16:15, said:

On the first hand I would bid 3♠ or 3♦ with South, on the second hand I would rebid 3♠ or 3NT as North rather than 4♥. As mentioned before I could attempt to introduce more gadgets here to address some of these problem hands, but I think I'm gaining enough with all natural that I wish to give that a try first. It also helps keep the system simple.

But I think that doesn't actually answer the question you haven't dared to ask, which is "if I construct a sufficiently perfect example, does that discredit the method?". I think the answer is "no". I'm happy you are helping come up with possible problem auctions and hands to help sharpen my judgement, but to me your attempts feel hostile rather than helpful, almost like a "gotcha". Could you perhaps embed it in a statistical analysis, or list advantages and disadvantages compared to other approaches? I've mentioned this twice before: I think any system has problem hands and weaknesses, and I think this system will do well on average (which is obviously different from "do well in every scenario").

Or were you only looking for clarification on the meaning of the bids?

The first hand is illustrative of the problems of the weak notrump hand -- opposite 11-13 balanced it's quite common that 3m will be the limit on a hand like this and selecting a game-forcing sequence with this 10-point hand is dangerous. Of course, if partner has a canape into a minor (or really any hand in the 14-15 range) you have very good game chances. Obviously you can't solve all problems, but the 11-13 notrump is surely common enough that this sort of issue will arise.

It seemed extremely weird to me that opener with 5♥-4♣ and max would jump to game in hearts after 1♥-(2♠)-X, since the double does not at all promise three hearts. You suggested this action earlier, but it sounds like you would actually not do that (preferring to bid 3♠) which makes more sense to me. Of course you've already indicated that the 5M-4+♣ hands are problematic for the system, which is fair but also seems rather high frequency.

[nullve]

PrecisionL, on 2024-February-23, 11:35, said:

a) Open 1♥ if Qxxx or better otherwise pass, pass 1NT response, over 2♣ rebid 2♦ = 11-13
b) Same as a)

So you're already passing some balanced hands in the 11-13 range.