[nullve]

I'll go first:

I played

1x/N-(2y)-?:

(...)
X = mildly penalty-oriented (partner assumes 3 y)
...P = usually 3+ y
...step 1 = Fishbein-like "takeout" (NF)
...(...)
step 1 = Fishbein-like "takeout" (NF)
(...)

for a while as a junior. It may have been in the context of a 4c major system, if that makes it any better.

I also once presented a partner with a 60+ page generalisation of this that was meant to cover most(!) competitive situations. He responded politely by lending me his copy of Robson/Segal's Partnership Bidding.

[pilowsky]

Outstanding idea.
Coming from a research background I immediately wanted to tinker with the system.
Combined with my inability to understand bidding, remember the cards and my terrible declarer play this pretty much guaranteed lousy results.
OTOH, I really liked the concept of preempts: telling your partner EXACTLY what you have with your first bid and at the same time annoying the opposition, "Approach" systems are based on this idea. Precision improves it by sharpening the ranges. So, I present "Clarity" where everything has a defined range.

Pass: either "5-10 HCP; 44/xx may contain a 5 card minor; no 5 card major" or as an overcaller "0-5HCP 4xxx"
2♣: "0-5 HCP; any shape" OR "15-19 HCP extreme shape not GF"


All 1 level bids are 11-13 exactly
All 1NT are 14-18 4xxx
All 2NT are 19-22 any shape
All 3NT and above are premptive
All 2 level bids are preemptive.


No transfers are needed in Clarity because you already know what each other has immediately. But you can still have them if you want.


This system works quite well with GIB.

[shyams]

A long long time ago, my friend and I had agreed to play 2♦ opening as 18-20 balanced.

I. A consequence of the agreement was that we had also agreed to open 3♣ and 3♦ preempts occasionally with six cards. We were young and pretty lousy at card play; this resulted far too many 800s and 1100s that it made us abandon the 2♦ structure.

II. If I recall, after a 2♦ opening, responder would act as follows:
.... 2♥ --- shows 4+ card ♠ any strength
.... 2♠ --- denies a spade suit any strength
etc. the rest of which I have forgotten.

III. The interesting part was not the 2♦ bid itself but the additional options it created for opener to show his distribution in 1m openings.

1m
...1M
......2NT showed exactly 5m and 4M, 18+ HCP. There were some follow-up structures.
......3m showed 6+m, no singletons or voids
......3M showed 6+m and 4+M, 16+ HCP

...etc (mostly forgotten). I used to find this portion of the bidding very useful because responder was much better placed to know what to do opposite a strong raise by opener. Responder knew that opener cannot be 4-4-3-2 18-20 when the jump "raises" were shown.

[FelicityR]

 nullve, on 2020-September-20, 17:47, said:

He responded politely by lending me his copy of Robson/Segal's Partnership Bidding.

That book is a good starting point before you try to improvise with your own methods

But it's not just invented bidding conventions that have ended up on the scrap heap. I'm sure many of us have thought we could do better than our predecessors and invent the ultimate bidding system, a hybrid of many other systems that are out there. Or come up with something anew, like Fantunes.

I have always dreamt about inventing one called 'Statistica' based solely around the probabilities of hand shapes and point counts. 70% of hands are covered in just 5 distributions, and 73% of hands are located in a 10 HCP range. But dreams are dreams...

And talking about dreams, I once read about a system I believe that was invented by the renowned British player John Collings. It was called the Walpurgis Club. And reputably, it was one of the worse systems ever played in a major championship. Has anyone else heard of this?

[pescetom]

For one year (I review agreements every year and scrap anything unused or disliked) we played that the control-bid immediately above a splinter checked back for singleton/void status. You can guess how useful that was compared to the damage caused.

But my favourites are the handful of famous but eminently scrappable inventions that somehow stand the test of time, at least on a national basis. Albarran 2♣ "Crodo" has been delighting opponents for over 70 years now

[FelicityR]

 pescetom, on 2020-September-21, 06:37, said:

For one year (I review agreements every year and scrap anything unused or disliked) we played that the control-bid immediately above a splinter checked back for singleton/void status. You can guess how useful that was compared to the damage caused.

But my favourites are the handful of famous but eminently scrappable inventions that somehow stand the test of time, at least on a national basis. Albarran 2♣ "Crodo" has been delighting opponents for over 70 years now

A good convention until the 2♣ opener has all four aces in their hand

[Vampyr]

 FelicityR, on 2020-September-21, 07:37, said:

A good convention until the 2♣ opener has all four aces in their hand

I don’t remember precisely, but I’m sure that sometime in my youth I briefly played control responses after a 2♣opening. Hopeless.

[pescetom]

 Vampyr, on 2020-September-21, 07:42, said:

I don’t remember precisely, but I’m sure that sometime in my youth I briefly played control responses after a 2♣opening. Hopeless.

It's even worse than that. Opener has 23+ HCP balanced or 3 losing tricks. The initial response is to show a specific ace or even two or three specific aces. More times than not the opener with strong hand would have known which ace(s) they are anyway, and just as well as remembering how to decrypt the 2 ace responses is a feat. After that opener either jumps to the wrong slam or it sputters out in an asymmetric fit seeking sequence where the responder should show honour fit (Kx or similar) but more often desperately bids the number of kings. Can hardly fail.

[awm]

I spent quite a bit of time trying to get a version of Polish club with transfer responses to work. The big problem is that the 1♣-1♠ auction is just ridiculously overloaded; it's like playing a 1♠ negative to a strong club (already quite bad) except that opener can also have a weak notrump.

I did also learn quite a bit about the weaknesses of Polish club as a system; most notably that it's really quite bad when opponents bid over your 1♣ opening. It's true they can't go as crazy as some opponents do over a strong club but you still get a lot of ugly auctions after sequences like 1♣-(2x natural preempt).

[pescetom]

 awm, on 2020-September-21, 12:11, said:

I did also learn quite a bit about the weaknesses of Polish club as a system; most notably that it's really quite bad when opponents bid over your 1♣ opening. It's true they can't go as crazy as some opponents do over a strong club but you still get a lot of ugly auctions after sequences like 1♣-(2x natural preempt).

Thanks for the hint

[mycroft]

I've seen suggestions that you should bid over 1 Polish Club using your weak NT defence, and figure out something sensible for the 3 1 level suit calls they've given you over an actual weak 1NT opener.

I haven't played against PC enough to try this, but it would make life really quite uncomfortable I would think.

[helene_t]

When Shogi and I started playing together we were both beginners, and since we are both computer geeks we obviously invented huge amount of nonsense. A few of our inventions actually made sense (T-Walsh, Raptor) but those had obviously been invented before.

Some of our agreements were:
- dbl shows any opening except for those that would make a natural 1NT overcall. That was just how everyone seemed to play it at our club. Fortunately we got rid of that after a few weeks.
- negative dbl shows 4+ cards in the highest unbid suit (as a consequence, 1c-(1d)-1S promises 5). This is maybe not so bad compared to so many other over-simplified definitions, and arguably over-simplified is better than unclear.
- A lot of emphasis on showing stoppers, sometimes before having investigated a minor suit fit. For example, our ghestem defense was to show stoppers straight away.
- Strong 3s and weak 2s or v/v depending on vulnerabilty. Arghhhhh.
- A double of a 1NT opening is take-out, i.e. shows a (4333) shape.
- Against artificial 1♣ openings we played Fishbein: dbl shows clubs while 1♦ is take-out. Maybe not so bad against really artificial 1♣ openings, but very few play that in the Netherlands, while it's not clear whether our defense should apply against a semi-natural 2+ 1♣ opening. Probably not.

[sfi]

 mycroft, on 2020-September-21, 17:07, said:

I've seen suggestions that you should bid over 1 Polish Club using your weak NT defence, and figure out something sensible for the 3 1 level suit calls they've given you over an actual weak 1NT opener.

I haven't played against PC enough to try this, but it would make life really quite uncomfortable I would think.

I've played quite a bit of Polish Club - something like 12 years - and it's pretty resilient to interference. I'm not sure what issues awm had with 2-level overcalls, but I can't see how you're worse off than if you're playing a 3+ 1C. Many opponents have tried strange things against us, and we're definitely better off than if they took my advice and just played whatever they do against a short (2+) club. You need a way to bid clubs (either 2C or 3C will work fine), you want a way to show both majors (2C or 2D are common), and your double should be major-oriented.

The advantage is the three basic hand types are well defined and split. A weak NT will generally only bid if forced to, the 15+ with clubs will bid clubs (we often use 2NT to show the weaker club range), and stronger hands bid their suit. Double is available for hands with no clear bid.

[Zelandakh]

The first convention I invented (about 13 or 14 years old I guess) was 2-way jump shifts, where a JS is either a normal, natural SJS or a mini-splinter with shortage. I also "invented" 1-level Lebensohl after a takeout double and using a 1NT response to show diamonds (and nothing about spades) after 1♣ - (1♠) along with several other (long forgotten) conventions before ever meeting a real life bridge partner.

When I did meet a bridge partner he mentioned this idea about using both 2♣ and 2♦ to show strong hands but he could not remember how it worked. I had never heard of Benjamin at this point so I thought about it for 5 minutes and came up with a workable structure that more or less resembles what I now know as Reverse Benji, which we ended up playing the entire duration of our partnership.

The same partner had some other ideas too. He insisted on playing something he called the "2♣ Asking Range" convention (Opener's rebid of 2♣ is an artificial relay), which unfortunately made several hand types effectively unbiddable. I did manage to talk him out of that one after a couple of months though.

Some more recent ideas that I have abandoned include:-
Bambi (BAlanced Majors, Big 1NT) - {1♣ = 10-14unbal; 1♦ = 15-22unbal or 18-20bal; 1♥ = 12-14bal or 21+bal; 1♠ = 15-17bal or 23+bal; 1NT = any unbal GF};
Tropic (I think it was TRansfer OPenings Incorporating Canapé) - {1♣ = 13+bal or 16+unbal or 11-15 4414; 1♦ = 10-15 4+♥; 1♥ = 10-15 4+♠; 1♠ = 10-15 4+♦ no 4M; 1NT = 10-12bal; 3♣ = 10-15 6+♣};
Big Heart - {1♣ = 10+ 4+♥; 1♦ = 10+ 5+♣ or 10+ 5+♦ or 17-19bal; 1♥ = 11-13bal or 20+bal or 20+ with ♣, ♦ or ♠}; 1♠ = 10-19 5+♠; 1NT = 14-16bal};
FP - {P = 0-7 6+♣ or 12+ OB in ♦/♥/♠ or 15-17bal or 19+ 3-suited or 20+ any; 1♣ = 12-14bal or 18-19bal or 12+ 5+♣ or 12-18 4414/4144; 1♦ = 8-11 any; 1♥+ = 0-7};
2 Under (SNT) - {1♣ = 10+unbal 4+♥ or 18-19bal 4+♥; 1♦ = 10+unbal 4+♠; 1♥ = 12-14bal or 15+unbal 5+♣ no 4M or 18-19bal <4♥ or 20+bal; 1♠ = 10+unbal 5+♦ no 4M; 1NT = 15-17bal; 2♣ = 10-14unbal 5+♣ no 4M; 2♦ = W2M or Acol 2♦}; and
2 Under 2 (WNT) - {1♣ = 10+unbal 4+♥ or 18-19bal 4+♥; 1♦ = 10+unbal 4+♠; 1♥ = 15-17bal or 15-19unbal 5+♣ or 18-19bal <4♥ or 20+bal or 20+ 5+♣ no 4M; 1♠ = 10-14unbal 5+♦ no 4M or 15-19unbal 5+♦ or GF 5+♦ no 4M; 1NT = 12-14bal; 2♣ = 10-14unbal 5+♣ no 4M; 2♦ = W2M or Acol 2♦}

I could probably fill a couple of full length bridge books with the ideas I have had and abandoned realising I could never persuade anyone to play it. Maybe I will go back and develop a couple of them properly now I have a little time after taking a furlough/voluntary redundancy package. Probably not though - so few bridge players want to try anything different out ever. It is kind of sad really.

[pilowsky]

 Zelandakh, on 2020-September-22, 01:37, said:

... Probably not though - so few bridge players want to try anything different out ever. It is kind of sad really.

yeah everyone just wants to win!

[Vampyr]

 Zelandakh, on 2020-September-22, 01:37, said:

The first convention I invented (about 13 or 14 years old I guess) was 2-way jump shifts, where a JS is either a normal, natural SJS or a mini-splinter with shortage. I also "invented" 1-level Lebensohl after a takeout double and using a 1NT response to show diamonds (and nothing about spades) after 1♣ - (1♠) along with several other (long forgotten) conventions before ever meeting a real life bridge partner.

LOL when I was 13-14 all of my bridge ideas came from Contract Bridge Complete by Charles Goren.

I am curious — with whom did you play before you met a real-life bridge partner?

[pescetom]

 sfi, on 2020-September-21, 18:21, said:

I've played quite a bit of Polish Club - something like 12 years - and it's pretty resilient to interference. I'm not sure what issues awm had with 2-level overcalls, but I can't see how you're worse off than if you're playing a 3+ 1C. Many opponents have tried strange things against us, and we're definitely better off than if they took my advice and just played whatever they do against a short (2+) club. You need a way to bid clubs (either 2C or 3C will work fine), you want a way to show both majors (2C or 2D are common), and your double should be major-oriented.

The advantage is the three basic hand types are well defined and split. A weak NT will generally only bid if forced to, the 15+ with clubs will bid clubs (we often use 2NT to show the weaker club range), and stronger hands bid their suit. Double is available for hands with no clear bid.

By an outrageous coincidence we will be playing against Polish Club for the first time tomorrow, so this will be very useful, thanks.

[Zelandakh]

 Vampyr, on 2020-September-22, 04:23, said:

I am curious — with whom did you play before you met a real-life bridge partner?

I taught myself how to play bridge from 2 library books - Culbertson's White Book and "Basic Acol", which was a 5cM 16-18NT system. I cobbled together the parts of these 2 systems I liked and added my own ideas by bidding all 4 hands alone and then playing them out. What I ended up with looked remarkably like Standard English Acol, something I found out a year or so later on when I got a hold of the Acol Blue Book. Playing bridge in this way did not seem odd to me at all at the time as this is also how I analysed chess openings. Back then, bridge was to me more of a theoretical puzzle on how to codify hands into bids rather than a social card game. In some ways it still is, though I enjoy the play as well providing the table is a pleasant one and have basically accepted that there is no single optimal solution but rather a range of solutions that offer pluses and minuses in different areas according to taste.

[mikestar13]

Not actually my system, the essential idea was my friend's, I was tasked with making it GCC legal and at least somewhat logically coherent given the principles he gave me. He called it the Crappy Club.

1♣ = in principle, 10-13 with any shape. A 1♦ was artificial invitation or better, again with any shape. All other reposes (including pass) were attempts to place the contract.
1♦ = 14+, four or more diamonds, might have longer clubs, no four+ card major.
1♥ = 14+, four or more hearts, might have longer minor, denied 4+ spades.
1♠ = 14+, four or more spades, might have longer second suit.
1NT/2NT/3NT = 4♣ -3-3-3 or 5♣-3-3-2. I forget the strength ranges.
2 or more ♣ = 14+ , six or more clubs, not forcing, bid what you think you can make.2 of higher in other suits = standard preemepts.
Sorta like Fantunes meets insanity, but it was invented long before Fantunes (EHAA was still new).

I only played this with him in practice matches. He played with his wife in novice games and actually won a few.

[rbforster]

For a while I tried to get a 2/1 non-forcing, non-constructive framework to work over a limited/precision 1M opener. The general idea was that 1N was forcing and included all GF hands without M support, while 1M-2x would be a 5-6+ suit with short M somewhat like a weak two. After 1M-1N, opener responded like a forcing NT in 2/1 making a natural rebid, after which responder could pass, preference to 2M, or bid a new suit to show a GF hand. Even with lots of articulate responses at that point, there just wasn’t space to reasonably sort out possible fits below 3N with many possible responding GF hands.

This was partially hobbled by trying to be GCC at the time, which precludes things like transfers over 1M and such, that seem more promising.