[hrothgar]

fred, on May 10 2009, 05:14 PM, said:

Some types of artificial systems and conventions are primarily designed to increase volatility thus giving lesser pairs a chance to perform much better than expectation if they happen to get lucky. Such methods often turn "normalish" deals deals into deals in which either side, more or less randomly, will win or lose 100s of IMPs.

Hi Fred:

For what its worth, I have some sympathy for this line of argument. (This probably doesn't come as much of a surprise; I often frame arguments regarding system regulation in terms of variance, expected value, and the like)

Personally, I hope that the big "fights" over system regulation might be able to more forward to a more productive place if more people adopted this same lens.

I've always felt that the whole "destructive" versus "constructive" lens prevents the different sides from being able to advance the discussion. We spend lots of time dancing around various side issues:

Can you define an appropriate suggested defense?
Is the phrase "destructive" a pejorative?
Yada, yada, yada

If the core issue is actually variance, lets have a discussion about variance...

[fred]

hrothgar, on May 10 2009, 02:29 PM, said:

If the core issue is actually variance, lets have a discussion about variance...

OK, but let's wait for another day - I still have 27 boards to play today

Fred Gitelman
Bridge Base Inc.
www.bridgebase.com

[hrothgar]

fred, on May 10 2009, 05:49 PM, said:

hrothgar, on May 10 2009, 02:29 PM, said:

If the core issue is actually variance, lets have a discussion about variance...

OK, but let's wait for another day - I still have 27 boards to play today

Fred Gitelman
Bridge Base Inc.
www.bridgebase.com

Fair enough

Break a leg

[hanp]

I think it is pretty nuts that Fred is having this same old discussion again while playing in the Cavendish. Good luck Fred!

[kfay]

hanp, on May 10 2009, 11:59 AM, said:

I think it is pretty nuts that Fred is having this same old discussion again while playing in the Cavendish. Good luck Fred!

Same thoughts but I didn't feel much like commenting on the same old stuff

[helene_t]

Ditto Kevin, I won't contribute to this thread either

[awm]

Apparently Zia and partner are playing a 2NT opening as 12-14 balanced.

This seems far more randomizing than any of the commonly played methods that people complain against. But it's "natural" so certainly legal under the regulations (or ACBL's regulations, etc).

[Trinidad]

fred, on May 10 2009, 09:14 AM, said:

Trinidad, on May 10 2009, 09:58 AM, said:

- Artificial systems are randomizing the field, making it hard to recognize the players with the best 'skill' in bridge. (In this sence 'skill' is defined as skill in card play, you know: the type of skills that you use in dozens of other trick based card games like whist or hearts. Designing a good bidding system is not perceived as a bridge skill.)

I can't speak for anyone else's perceptions, but I certainly perceive that designing a good bidding system is an important bridge skill. It seems fairly clear to me that you did not understand my use of the word "random"...

Hi Fred,

I think that I understand perfectly well what you mean by "random", i.e. pretty much what you describe below. (BTW, I wasn't aware that you were using the term "random" in this thread.)

fred, on May 10 2009, 09:14 AM, said:

Some types of artificial systems and conventions are primarily designed to increase volatility thus giving lesser pairs a chance to perform much better than expectation if they happen to get lucky. Such methods often turn "normalish" deals deals into deals in which either side, more or less randomly, will win or lose 100s of IMPs.

I fully agree. One simple example of such a convention are the upside down suit preference signals that Meckstroth-Rodwell (used to?) play. No theoretical advantage (other than confusion), no drawback either, only adding randomness.

Another example is the 12-14 2NT opening that you played against. I will not go into the theoretical advantages and drawbacks. But it surely adds randomness.

fred, on May 10 2009, 09:14 AM, said:

Suppose an "average Cavendish pair" that would normally not have much of a chance to do well in the event is allowed to play a convention that 50% of the time will win 300 IMPs and 50% will lose 300 IMPs. Suppose this convention comes up once in each of the 5 sessions. If the pair gets lucky and wins 300 IMPs on all 5 occasions (a 1 in 32 chance) they rate to come in the money. But regardless of how much luck that pair gets, the impact of their random convention on the rest of the field is extreme - 5 random pairs will get either a 300 IMP gift or a 300 IMP penalty, purely by luck of the draw.

I would not want to play in a pairs event in which the field was littered with such pairs playing such conventions and I suspect that neither would most (all?) of the "premium pairs" in the Cavendish.

Luck is important at bridge regardless and even some system design issues involving purely natural methods (weak notrump versus strong notrump for example) will result in considerable randomness. That being said, disallowing highly volatile methods decreases the premium on luck thereby increasing the premium on skill.

Fred Gitelman
Bridge Base Inc.
www.bridgebase.com

Again, I fully agree. But what if the numbers are changed slightly? A pair has devised a bidding system where they will win 300 IMPs 55% of the time and lose 300 IMPs 45% of the time? There are various examples of pairs who were forced to modify their system to an inferior version because they weren't allowed to play the better, original system.

One such example is the Swedish group of players around Anders 'Carrot' Morath, who had to modify their Carrotti system (a forcing pass system) to Magic Diamond (a strong diamond system) or Carrot club (a two way club system) to be able to participate in international tournaments. (And they were fairly successful.) Ask any of these players and they will say that Carrotti was theoretically sounder than Magic Diamond or Carrot club. If we can believe them, one can conclude that they would have been even more successful if they had been allowed to play Carotti. I am sure that you will find similar examples in Poland and currently in New Zealand.

I used to live in Sweden, so I know a little about the situation there. At the top of the list of requirements for participating in the team trials in Sweden you could find that you have to play a system that can be played in international competition. From the point of view of the Swedish Bridge League, that obviously made a lot of sense. Why would one select a team with players that is not allowed to play their system? But what did it do to the ambitious Swedish players who play these, at least in their views, superior systems? In Swedish competitions, they regularly beat the internationalists, but they had no chance to show the world how good they (and their systems) were.

And (OK, I am putting it on thick now) all of that is caused by the strange perception that it is normal to open strong hands with 2♣ (?!?, SA, Acol) or if necessary with 1♣ (?!?, Precision, Polish type Club), 1♦ (?!?, Strong Diamond systems) or 1NT (?!?, Vienna), but that opening them with 'Pass' is 'Highly Unusual'. 'Highly unusual' to whom? The American Bridge Teachers Association?

Rik

[slothy]

awm, on May 10 2009, 01:13 PM, said:

Apparently Zia and partner are playing a 2NT opening as 12-14 balanced.
....

I was playing in a tournament recently on BBO and our opponents played this too

What was nice was that one of the partnership wasnt aware of it and they ended up in a 6NT with 3 aces missing.

I think he cottoned on the 2nd time he did it

[qwery_hi]

hanp, on May 10 2009, 08:59 AM, said:

I think it is pretty nuts that Fred is having this same old discussion again while playing in the Cavendish. Good luck Fred!

I think its even more nuts that Fred didn't consult you before participating in this discussion

[barmar]

qwery_hi, on May 10 2009, 08:29 PM, said:

hanp, on May 10 2009, 08:59 AM, said:

I think it is pretty nuts that Fred is having this same old discussion again while playing in the Cavendish. Good luck Fred!

I think its even more nuts that Fred didn't consult you before participating in this discussion

I found it surprising, too. Fred usually doesn't participate in the forums while he's at a big tournament. Is it because the tourney is local for him?

[the hog]

I am sure that Fred is old enough to know what he is doing. I am sure he wants to win enough that he would not participate if it would affect his results.

[barmar]

The_Hog, on May 10 2009, 09:04 PM, said:

I am sure that Fred is old enough to know what he is doing. I am sure he wants to win enough that he would not participate if it would affect his results.

It just seemed out of character. I can't recall seeing him participate in the past while he's been playing in a national tournament.

[the hog]

barmar, on May 11 2009, 09:10 AM, said:

The_Hog, on May 10 2009, 09:04 PM, said:

I am sure that Fred is old enough to know what he is doing. I am sure he wants to win enough that he would not participate if it would affect his results.

It just seemed out of character. I can't recall seeing him participate in the past while he's been playing in a national tournament.

True. Maybe he feels strongly about this issue, or maybe he is just killing some time to relax.

[mike777]

Interesting,many more posts on Cavendish regulations in the forum than on actual Cavendish bridge deals.

[JoAnneM]

And even posts about why Fred is posting. I find that pretty incredible.

[qwery_hi]

JoAnneM, on May 10 2009, 08:02 PM, said:

And even posts about why Fred is posting.  I find that pretty incredible. 

Well, Fred is a celebrity - Its kinda like discussing what motivated Brad Pitt to change his hair style

[cherdanno]

Meanwhile, Gerben's friend from whom he asked about this is (I assume) Alex Smirnov, who came in 5th overall.
Congratulations, a nice success for a pair that's fairly new on the scene (compared to most other pairs participating)!

[Gerben42]

Quote

Meanwhile, Gerben's friend from whom he asked about this is (I assume) Alex Smirnov, who came in 5th overall.
Congratulations, a nice success for a pair that's fairly new on the scene (compared to most other pairs participating)!

That's right, I wrote already in the other thread that Alex & Josef are "hot" are in heat 1 recently

[fred]

Trinidad, on May 10 2009, 06:56 PM, said:

fred, on May 10 2009, 09:14 AM, said:

Suppose an "average Cavendish pair" that would normally not have much of a chance to do well in the event is allowed to play a convention that 50% of the time will win 300 IMPs and 50% will lose 300 IMPs. Suppose this convention comes up once in each of the 5 sessions. If the pair gets lucky and wins 300 IMPs on all 5 occasions (a 1 in 32 chance) they rate to come in the money. But regardless of how much luck that pair gets, the impact of their random convention on the rest of the field is extreme - 5 random pairs will get either a 300 IMP gift or a 300 IMP penalty, purely by luck of the draw.

I would not want to play in a pairs event in which the field was littered with such pairs playing such conventions and I suspect that neither would most (all?) of the "premium pairs" in the Cavendish.

Luck is important at bridge regardless and even some system design issues involving purely natural methods (weak notrump versus strong notrump for example) will result in considerable randomness. That being said, disallowing highly volatile methods decreases the premium on luck thereby increasing the premium on skill.

Fred Gitelman
Bridge Base Inc.
www.bridgebase.com

Again, I fully agree. But what if the numbers are changed slightly? A pair has devised a bidding system where they will win 300 IMPs 55% of the time and lose 300 IMPs 45% of the time?

Good question. Other questions arise by changing the frequency in my example from once per session to (say) 10 times per session, changing the expected IMP gain or loss from 300 IMPs to (say) 100 IMPs, and/or making the formula for expected gain and expected loss more complex (say 10% of the time you break even, 60% of the time you lose 100 IMPs, and 30% of the time you gain 200 IMPs thus breaking even overall - not that even this "more complex" formulation is anywhere close to being "complicated enough" to describe the expectation of any real convention).

The only way I can really answer such questions is to say that my original example was not intended to be realistic. In practice it is impossible to calculate the expected gain and loss of any convention. Even if it was possible, deciding what is allowed on such a basis does not make sense for a variety of reasons.

So tournament organizers are left with only 2 choices:

1) Allow anything
2) Do the best they can with the arbitrary line they draw by taking into account factors like (in no particular order):

- what do the players want?
- what conventions and systems are sufficiently "mainstream"?
- what will be best in terms of increasing interest in the tournament (and in bridge in general) among the general public?
- what families of systems and conventions are deemed "sufficiently volatile" that their randomizing effect is unwelcome?
- how much administration are we willing and able to do in order to make sure that all pairs know what all other pairs are playing in a reasonable time frame, to make sure that pairs don't change their methods mid-event, etc?

I know some will disagree, but I strongly believe that 1 above, "allow anything", just won't work in a pairs tournaments (in my view this would score very badly in all 5 of the considerations mentioned in 2 above).

So the organizers have to draw a line and, I must say, I do not envy them this task

My personal view is that the Cavendish organizers have done fairly well in this regard (though as I said earlier, if it were up to me, I suspect I would be slightly more liberal and allow multi for example).

Fred Gitelman
Bridge Base Inc.
www.bridgebase.com