[Salih Pasa]

(“TL;DR” at the end)

Hello. I study Computer Engineering in Middle East Technical University. I am interested in algorithms and information theory. I started playing bridge in 2019 with the bridge club in my university. I played for 6 months until I decided that memorizing bidding systems that have been improved for years by experts is not for me. I have interests other than memorizing some systems and trying to learn card play to expertise in a complex game. I did not play the game afterwards…

However, some idea have been on my mind since then: what if one aims something other than being an expert or at least an intermediate player in this game? I will be having a lot of responsibilities in my life other than playing bridge. So I will never be able to play the game well. So having fun is my main aim -if any- in this game. With these, I had this idea: creating a fully original bidding and carding system for contract bridge.

I know that this sounds crazy considering I am a beginner and my card play is not even good. I know that even if we (since the game is played with a partner…) could create this system, it will be very hard to play with others since everytime we would need to explain the meanings. However, I believe that this problem can be solved in many ways that I won’t explain here in this post. I want to see how much of improvement can be done starting from scratch.

If you fancy the idea of “creating and constantly improving a totally original bidding system for this game just for fun, not for easy winning” (or if you know anyone who’d like this idea), show up and let us begin the journey!

“TL,DR”: I seek a parner or multiple people (preferably beginners) to create and constantly improve a bidding and carding system from stratch. My aim is to have fun creating something and watching it develop as I (and those who would join me) improve it…

[LBengtsson]

I like your idea but there are many players who have experience and have tried to create a new bidding system.

Perhaps it is easier to use a simple system as a start and change some ideas in it. EHAA (Every Hand An Adventure) could be a option.

https://en.wikipedia.org/wiki/EHAA

I do not like 10-12 no trump bid in it - my opinion - but you could change it to 12-14, 13-15, 14-16, 15-17.

[helene_t]

I don't think you should seek beginners for such a project. Beginners need something that is already developed and well documented. Developing your own bidding system (even a very primitive one) requires a theoretical understanding of bidding beyond what most bridge player will ever acquire, let alone possess as beginners.

You should rather seek someone with a theoretical interest in system design - they may well like the challenge of designing something very simple, even if their main interest is in high-tech systems.

Then again, a background in CS may be as relevant as bridge skills, so maybe CS students who are beginners as bridge players may be your audience.

Anyway, what about this system:

In every situation, you ask yourself if game is
- certain (your points plus partner's minimum is 25 or more)
- possible (your points plus partner's maximum is 25 or more)
- impossible (your points plus partner's maximum is less than 25)

Now you bid:
- game impossible (or 0-12 points in opening/overcall position): pass
- game possible: (or 13-18 points in opening/overcall position): make some bid without jump
- game certain: (or 19+ in opening/overcall position) make a jump bid (or bid game; or pass partner's game bid; or make a non-jump bid if we are already forced to game)

Which strain to bid:
- raise partner's suit if you know of an 8-card fit
- introduce a new suit as your first bid if you have 5+ cards
- introduce a new suit as your second bid if you have 4+ cards
- rebid (one of) your suit(s) if you have extra length
- default: bid notrump

I think this will land you on your feet most of the time. You can always add some exceptions (false preference, reverse bidding, offshape 1NT rebids, 1m opening rather than 1NT with 4441, take-out doubles, unassuming cuebid, preempts etc) once you have the basis under control and have appetite for improvements.

This is probably not going to evolve into a standard system like SA or Acol, it will rather evolve into something like Vienna or a relay system.

[hrothgar]

FWIW, I think that you might do better practicing with a fairly small / narrowly defined problem before moving over to a wide open search space.

My own recommendation would be that you create a system that is able to compare the efficiency of different auction termination mechanisms after shape is known in a relay system. This would be of interest to (at least some) system designers and give you some useful experiece.

[helene_t]

hrothgar, on 2022-February-16, 09:26, said:

FWIW, I think that you might do better practicing with a fairly small / narrowly defined problem before moving over to a wide open search space.

My own recommendation would be that you create a system that is able to compare the efficiency of different auction termination mechanisms after shape is known in a relay system. This would be of interest to (at least some) system designers and give you some useful experiece.

Wrong thread, Richard This is about designing something ultra-simple for people who find Acol too difficult to learn.

[Salih Pasa]

Thank you for all responses! I would like to answer your responses one by one.

[Salih Pasa]

LBengtsson, on 2022-February-16, 06:12, said:

I like your idea but there are many players who have experience and have tried to create a new bidding system.

Perhaps it is easier to use a simple system as a start and change some ideas in it. EHAA (Every Hand An Adventure) could be a option.

https://en.wikipedia.org/wiki/EHAA

I do not like 10-12 no trump bid in it - my opinion - but you could change it to 12-14, 13-15, 14-16, 15-17.

helene_t, on 2022-February-16, 09:56, said:

Wrong thread, Richard This is about designing something ultra-simple for people who find Acol too difficult to learn.

LBengtsson and helene, I actually did not aim to create or use a simpler system to have a ease in bidding. What I want is the joy of creating, testing and improving a system from stratch. (By the way, I am still open to any volunteers who'd enjoy doing such a work.)

[Salih Pasa]

helene_t, on 2022-February-16, 06:30, said:

I don't think you should seek beginners for such a project. Beginners need something that is already developed and well documented. Developing your own bidding system (even a very primitive one) requires a theoretical understanding of bidding beyond what most bridge player will ever acquire, let alone possess as beginners.

You should rather seek someone with a theoretical interest in system design - they may well like the challenge of designing something very simple, even if their main interest is in high-tech systems.

Then again, a background in CS may be as relevant as bridge skills, so maybe CS students who are beginners as bridge players may be your audience.

Anyway, what about this system:

In every situation, you ask yourself if game is
- certain (your points plus partner's minimum is 25 or more)
- possible (your points plus partner's maximum is 25 or more)
- impossible (your points plus partner's maximum is less than 25)

Now you bid:
- game impossible (or 0-12 points in opening/overcall position): pass
- game possible: (or 13-18 points in opening/overcall position): make some bid without jump
- game certain: (or 19+ in opening/overcall position) make a jump bid (or bid game; or pass partner's game bid; or make a non-jump bid if we are already forced to game)

Which strain to bid:
- raise partner's suit if you know of an 8-card fit
- introduce a new suit as your first bid if you have 5+ cards
- introduce a new suit as your second bid if you have 4+ cards
- rebid (one of) your suit(s) if you have extra length
- default: bid notrump

I think this will land you on your feet most of the time. You can always add some exceptions (false preference, reverse bidding, offshape 1NT rebids, 1m opening rather than 1NT with 4441, take-out doubles, unassuming cuebid, preempts etc) once you have the basis under control and have appetite for improvements.

This is probably not going to evolve into a standard system like SA or Acol, it will rather evolve into something like Vienna or a relay system.

Helene, it is a wise statement that people who has a theoretic interest in systems design (I am, probably.) would be the ones who'd enjoy my idea. I will consider this not just in this topic, but in any relevant situation.
Although that I really wanted to create a system from "stratch", the bidding system you proposed'd be very useful in dealing with the fundamental problems and starting to design a completely new system from there. Then the fundamental system you suggested would be very useful to test the additional rules applied to it... This is a option. Thank you for your contributions. I'd like to read any other contributions and comments from you and others.

[Salih Pasa]

hrothgar, on 2022-February-16, 09:26, said:

FWIW, I think that you might do better practicing with a fairly small / narrowly defined problem before moving over to a wide open search space.

My own recommendation would be that you create a system that is able to compare the efficiency of different auction termination mechanisms after shape is known in a relay system. This would be of interest to (at least some) system designers and give you some useful experiece.

Hrothgar, I will have your suggestion on my mind. It sounds enjoyable to design and test such a system. However, I'd still like to use and test my system in a game (bridge) which is played by tens of thousands of players everyday worldwide. This is like a hobby of me, if I had the opportunity...

[helene_t]

Salih Pasa, on 2022-February-22, 16:14, said:

What I want is the joy of creating, testing and improving a system from stratch.

That is great, then you are in good company here Keep an eye on posts by AWM, Straube, Foobar, nullve, Kungsgeten and various others who have posted interesting ideas for new bidding systems.

Quote

Although that I really wanted to create a system from "stratch", the bidding system you proposed'd be very useful in dealing with the fundamental problems and starting to design a completely new system from there. Then the fundamental system you suggested would be very useful to test the additional rules applied to it... This is a option.

I think that the ultra-simple system could serve as a fall-back if you want to implement a bidding system in software in order to test its performance with simulations. There will be millions of convoluted auctions which your new system doesn't cover initially, so you need some default system that provides definitions for all those situations that aren't covered.

In particular, I think that defining most non-jump bids as constructive but non-forcing may be a reasonable default for contested auctions where we are still in the game exploration phase.

As a basis for a more advanced system, however, I don't think the system is so great, you would need to shake it up pretty fundamentally to make a good system out of it. You will need some low-level forcing bids, the NT range is too wide, and you probably don't want the 1♣ opening to be natural. You also don't want to force to game with random 19-counts.

One improvement I thought of that would cover quite a lot of the issues with the system is the principle that there is always some "grobe" bid available which is ostensibly a hand without clear direction but could also be some strong hand. This would be
- double if available
- otherwise, cuebid if available
- otherwise, cheapest unbid suit (or maybe 2NT if cheaper, or maybe it should be cheaper minor?)

This would cover (some distorted version of) take-out doubles, Polish 1♣ opening, FSF, CBS and scrambling 2NT. It may not be very good to build the whole system on that principle, but again it could work as a default.