The best system

[Flame]

Its time to ask the simple question, which system is the best ?

I cosider ability to play offline turnaments and simplicity important factors for a good sysem, but what is considered "fun system" isnt not importent (the fun part is temporary and a subjective factor, most of us have fun when we win).

This question is a practicle question for me and im sure for others who consider partnerships for serious bridge , so pls give practicle responses.

[Free]

First of all, this question has been asked before, but I can't find it anymore...

 

My answer:

From my experience, the science approach has a lot of advantages over the natural approach. This said and done, the natural approach has other advantages over the scientific one as well, but in general some science is needed. That's why in general, you should have a system which has natural constructive responses as well as a scientific part. I really like Moscito, because it combines these two very good. BUT it has a disadvantage: the strong ♣ opening. Why is it such a disadvantage? It's a big target for preemptive disruptive methods, and the entire system gets destroyed lots of times.

 

It's always a problem with strong hands. You can easily go high on brute power, but you want to stay as low as possible to make the slam bidding better. All by all, one of the best approaches for strong hands is to play them Polish-wise: strong any, balanced with a range or ♣s. That way, opps can't disrupt as agressive as they would on a strong ♣. If they would try it, they'd get lots of penalty doubles because opener didn't have the strong version, and the pair didn't even have a game to play. But I have to admit that I don't know enough about Polish Club to analyse it's scientific part.

 

I think the best system isn't invented yet, but if I had to choose between the systems I know (moscito, petit carreau, logsta, 2/1, acol, sayc) and some systems I've seen in action and I know some stuff about, I'd still consider Moscito as the best this far. The disadvantage of the strong ♣ can't compete with the light limited openings imo.

 

But you have to consider one more thing (I think Claus said it in the past): you need a system which doesn't care if your opps start bidding as well. If you can manage a system that has science and natural approach, which opens strong hands low, and which doesn't care about opps bidding, you have yourself the best system immaginable.

 

To end a quote from my f2f partner: the best system is partnership trust.

[flytoox]

the best system is partnership trust.

Thought I often disagree with you, free, I am with you this time:)

[flytoox]

the best system is partnership trust.

Thought I often disagree with you, free, I am with you this time:)

Or it is better to say partnership understanding?

[Free]

No no, it's TRUST! If you don't trust your partner's bids, you won't have any understanding as well...

[helene_t]

Mike Lawrence provides an excellent discussion in his 2-o-1 workbook. According to ML, a natural 4-card major system is easy to learn, and it's more fun because it forces you to make decisions more based on judgement than on system knowledge. 5-card major systems are more accurate, which may turn out to your disadvantage (the opponents are listening, too), but it is more likely to be to your advantage. That 5-card major systems are more acurate follows from the useful space principle: information theory predicts that a system in which the space-consuming openings (in casu, the majors) are used less frequently and are more specific than less space-consuming openings (in casu, the minors), conveys information more efficiently, at least as long as the opponents don't interfere.

 

I disagree on the assertion that 4-card major is easier to learn, but it may be a question of mentality, like students disagreeing of whether English or Spanish is easier to learn. The case for natural systems (and 4-card major in particular) is probably that they are more intuitive. However, in natural systems the forcing character of a bid is a very complicated matter. This issue may also be related to perfectionism: if a beginner recognizes that a freebid or rebid of, say, 2♥ shows hearts, but has no idea whether it is forcing or not, it could be argued that he is better of than somebody who doesn't even know if it is a natural bid or not.

[flytoox]

I am sure there is no answer to this kind question. This is not a simple question either.Bridge is not science. What is the definition of the best? Is it defined by the performance in the top level competition, or sth else? Perhaps we should say the system fits me best.

 

Having said that, I would say precision plus 2/1 GF will be an excellent system:)

 

Hongjun

[luis]

There's no such thing as a "best system".

In order to choose a system you have to analize the stregnths and weaknesses of each system, the requirements of each system and what you and your pd like to play, can remember and feel confortable with.

 

It depends on you and your pd, what you like, what's your style, etc. This is like choosing a pen to write. Each pen has its reputation probably coming from a) techonolgy/science behind or the public opinion about it. The challenge is that you have to test and test and choose the pen that you like the most.

 

With my mother I've always played Moscito with great success. It's a wonderful system but you must practice because when you forget a bid you are terminated. I'd say Moscito has the best "technology" behind.

 

With my father I played Colonial Acol and won some tournaments with a system everybody defined as "that ancient piece of #####".

 

With pd#1 I played a very very bad system that he invented and had great results because he loved the system and knew every single bid. Since I have good memory we knew exactly what we were doing even when our methods were not state of the art.

 

With pd#2 I started playing precision and after some frustrations switched to 2/1 and now we switched to what we call "Unregulated Sayc", and now we are having the best results. We went from science to natural!

 

Finally with pd#3 I started with Sayc and then 2/1 and the results were so horrible that we switched to a strong club artificial system, results improved a lot and we are now happy. This time we went from natural to science.

 

And with pd#4 2/1 and some simple conventions have always worked excellent for us.

 

So I play and played in the last 2 years:

Moscito,

An artificial strong club system

Sayc,

Colonial Acol,

2/1

A system invented by my pd

 

And in each case that system was the best option for my patrneship and other systems failed.

[mishovnbg]

Best system can be found by comparison between systems. To do this you need a tool - method of comparation. Using different methods can lead to different best system. I tried to invent universal method of comparison between systems and posted it here in forum. I was so disappointed that nobody in forum didn't help me to tune it, that ask moderator (Dwayne) to erase it from forum!!! :) .

 

For best system look:

http://bridgebase.lunarpages.com/~bridge2/...indpost&p=20769

 

Misho

[Flame]

I was waiting for misho's and luis's responses because i wanted to see what you will say about systems which are related to you, this was NTC and the new kind of PC with misho and german moscito with luis.

I also wish to see the hog response to know if he really believe strefa is the best.

Yes i know you cant have an exact definition for the best system, but for me its really importent, for example i cant play syac or 2/1 because i believe they are inferior to other systems, i am gelous of people that can play for fun, i never could have done that on any game i played, my fun is to find the best system, just like i was sitting hours with the chess encyclopedies, or looking for the best strategies in age of empires game.

[mikestar]

In Journalist Leads Jeff Rubens writes (discussing lead conventions) that all methods have losing cases and a partnership must choose which mistakes it prefers to make.

 

I believe that this applies at least equally well to bidding systems and that the quest for one best system is futile. Every system will shine at something and do something else badly--which mistakes do you prefer to make?

 

Certainly the quest for a partnership's preferred system is fruitful--it is possible to find a bidding system that works better than others for two particular players: there will be something that fits their skills and temperaments better than some other system does.

[Shrike]

I believe that it is possible to answer this question definitively. But not now.

 

Unlike Misho (sorry, Misho, if I've misinterpreted), I believe that no first-principles analysis of systems will tell which is better -- they must be tested against each other, in play, for many thousands of hands. We must correct for player experience and judgement, and control (no, not eliminate) such variables as deviations and outright psyches. Then, we should compare the results achieved at the table (we don't care about the par contract or double-dummy defense).

 

I think it is clear that this test cannot rely on humans -- too many extraneous, uncontrollable variables. When (no, not if) computers can be taught to bid as well as humans in all respects, and when they can be taught to do so in any system we can come up with, THEN we'll know which system is best in theory. Yes, I agree this ability is quite a ways beyond our current grasp.

 

After that, we program in a certain realistic probaility of misunderstandings and memory lapses for each call, as a function of the complexity of the system, frequency of the situation, similarity to other situations, intuitivenesss of the meaning. . . Now rerun the test, and we find out which system ought to be best in practice.

 

Notice that the test needs to pit each system against each other system. It is possible that we'll discover that system power is not an easily characterizable function (e.g., suppose 2/1 beats Regres, and Regres beats Moscito, but Moscito beats 2/1).

 

What then? Doesn't matter. By then WBF will have mandated 2/1 with whatever gadgets are currently in vogue among bunnies, and no psyches.

[hrothgar]

Notice that the test needs to pit each system against each other system.  It is possible that we'll discover that system power is not an easily characterizable function (e.g., suppose 2/1 beats Regres, and Regres beats Moscito, but Moscito beats 2/1).

Shrike introduces the interesting notion the the relationship between bidding systems does not preserve transitivity. [Transitivity means that if A >B and B >C, the A > C]

 

There is a fairly simple way to model this type of dynamic:

 

Start by creating a random population population of bidding system.

 

During each period, the bidding systems are mixed and pairs of systems compete against one another in a long match.

 

The results of the match are used to determine the extent to which the bidding systems reproduce. Systems that score "well" produce more offspring than systems thatscore poorly.

 

Next period, spin the whell, create new matchings and repeat.

 

Over time the system should converge to an equilbirum in which you have an optimal ratio of systems.

 

I'll note in passing that most "stable" equilibia are contain a mixture of many different independ critters that have settled into some sort of balance. Consider the human stomach as an example. THE G/I track is full of all sorts of rather nasty bacteria strains. When a new strain of E-Coli hits the G/I track, it has a very difficult time establishing itself.

 

In contrast, mono-cultures are very easy to disturb/destroy.

 

For whats its worth, I used to do acadmic research on precisely this type of topic [sure, we were using programs that played tick-tac-toe rather than bridge, but the same logic prevails]

[Free]

Sorry for saying Shrike, but that test is garbage! If you want to know what SYSTEM is the best, you don't have to consider mistakes, memory lapses, play,... of the players who use the system. Human mistakes have NOTHING to do with a certain system. Just for a second say that a relay system is the best. Then you let some human play it and he makes a mistake. Is it the system's fault that it didn't work? No, it's the human's fault. And ok, you can say that the complexity of the system is too high, but that doesn't change anything about the capabilities of it.

[Flame]

Sorry for saying Shrike, but that test is garbage! If you want to know what SYSTEM is the best, you don't have to consider mistakes, memory lapses, play,... of the players who use the system. Human mistakes have NOTHING to do with a certain system. Just for a second say that a relay system is the best. Then you let some human play it and he makes a mistake. Is it the system's fault that it didn't work? No, it's the human's fault. And ok, you can say that the complexity of the system is too high, but that doesn't change anything about the capabilities of it.

I dont agree with you, like i said at my first post, simplicity is a factor in how good a system is, if a system is harder to play it is less effective.

[Free]

Opinions differ :(

[mishovnbg]

Opinions differ :(

Hehe, Frederick!

Are your actual partner really share your opinion?

Misho

[1eyedjack]

Perhaps another relevant question is: how do you measure the best system?

 

Suppose system A gets you into the top 10 places consistently, but system B gets you bottom place 80% of the time but top place 20% of the time? Which is better? Most would say that system A is better, but some might be content to shine on 20% of occasions and let the other 80% disappear into obscurity, when there is a trophy at stake for 1st place and 2nd place just gets you a pile of monster points.

[mikestar]

. . . If you want to know what SYSTEM is the best, you don't have to consider mistakes, memory lapses, play,... of the players who use the system.  Human mistakes have NOTHING to do with a certain system.  . . .

Free,

 

Your point is well taken if you are considering what bidding system is best for computers--but for play by humans, given two systems of approxiamtely equal technical merit, the more easily remembered one is clearly superior--that is, your partnership will make fewer mistakes with it and win more often. And isn't winning the primary reason for the quest for the best system?

 

As an example comparing two systems of the same type, Meckwell Precision is clearly superior to Goren-Wei precision if you are a computer or a Bridge god like Meckstroth or Rodwell, but two average players on BBO or in a club will have a much higher win percentage with Goren-Wei: Meckwell will confuse them so badly that it will cause them to make hundreds of mistakes.

 

In math terms, If System A is only 50% as good as system B on pure technical merit but a given partnership makes 10 times as many mistakes playing system B, which system should it play?

[Free]

now we're back at the beginning: the best system is the one which suits YOU the most... if you can bid every slam with pure natural bids and common sense, why play something else? :(

 

just play something where you can rely on partner's bids (partnership trust) and you'll achieve the best results. yesterday my partner once didn't trust my bid and we had a bottom, where as if he had trusted me we'd have a top! Luckily he trusted me for the rest of the day :(

[the hog]

The best system, in fact the only system possible, is clearly sayc. Nothing else should even remotely be allowed.

 

A day or so ago someone was thrown out of a bbo tournament for playing a totally natural system. The TD whom I will not name made the following comment:

"natural systems do not include weak nt"

The pair in question was playing everything natural, 5 card ms, weak nt, all 2 bids weak and natural.

 

rgb has the thread, called "The Punch Line".

 

For once I am speechless!

[mikestar]

. . . A day or so ago someone was thrown out of a bbo tournament for playing a totally natural system. The TD whom I will not name made the following comment:

"natural systems do not include weak nt"

The pair in question was playing everything natural, 5 card ms, weak nt, all 2 bids weak and natural.

 

. . . For once I am speechless!

Someone thrown out for playing something ACBL Limited Convention Chart legal? Good grief!

[hrothgar]

The best system, in fact the only system possible, is clearly sayc. Nothing else should even remotely be allowed.

 

A day or so ago someone was thrown out of a bbo tournament for playing a totally natural system. The TD whom I will not name made the following comment:

"natural systems do not include weak nt"

The pair in question was playing everything natural, 5 card ms, weak nt, all 2 bids weak and natural.

 

rgb has the thread, called "The Punch Line".

 

For once I am speechless!

I saw the RGB posting as well.

 

Sadly, I don't find this overly surprising. When most players use the expression "Natural", they - in fact - mean "What I Play".

 

The expression "Natural" is mainly used for rhetorical effect in an attempt to frame the conversation.

 

[Much like naming your party the "Bolsheviks" despite your minory position or naming a major encroachment on civil liberties the "Patriot" act]

[DrTodd13]

I want to clear up a point that Luis made. There has been some

work that has shown that there is such a thing as a perfect system.

The system is not static as are all systems today. The system is

dynamic and changes constantly depending on the context. At

every opportunity to bid, both partners would perform a complex

computation to determine what information is needed and how to

encode it. It is possible to encode this information optimally. Then

they match their hand against that list and make the bid into which

their hand type falls. Of course, only computers could do such a

thing so all human systems are a pathetic attempt to get closest

to the optimal system while at the same time allowing humans to

remember the system. The perfect system would also have to

have an understanding of psychology and it's opponents. Against

humans the computer might preempt more knowing that human

judgement is flawed in those circumstances. Against computers a

computer may behave differently. So part of this depends on

knowing your opponent but I would still say that a perfect system

exists even though for all intents and purposes it is unattainable

to humans forever and currently unattainable to computers as well.

 

Todd

[Flame]

Good point Tod.

I think even for computer there is alot of payoff for this prefect system and therefore i dont think will see anything like that soon, but the point that it exist is true.

I actually ment this post to be practicle, i am going for a new partnership and was considering either a polish club version that open 1M when it has it (like strefa or the system described by misho which i really like but will be harder to get a good full implementationmany) or a major system like midmac or maf or one of my own systems which are more majorish then those two.

I took out of the picture 2/1 because i dont like their 1c/1d duplication, and also precision because i dont like opening 1c with a 5 card major, atleast not that often.