[kenrexford]

I have had a number of discussions with a friend of mine who is very MP skilllful. I consider myself wildly tilted to IMP in my thinking and skills. This, of course, causes some problems.

In any event, it seems to me that a large number of situational issues (what does such-and-such mean) in bidding have a "conventional wisdom" meaning, perhaps a very advanced or even expert CW meaning, that has a MP bias in the thinking that resulted in the conclusion.

It then occurred to me that one could in theory develop an entirely different set of GP defaults depending upon whether the form of scoring in MP or IMP, but that perhaps no one would ever want to learn and most would not be competent anyway in the entirely different ways of thinking as to bidding.

I think of concepts like "improving the partial," ideas that make little sense in the context of IMP's, and yet are presumptive base principles that govern the interpretation of some sequences. The whole concept of when to bid 3NT and when to play the major at game is equally affected, as at MP the goal is bidding 3NT when an equal likelihood of total tricks is expected, whereas the usual concern at IMPs is bidding 3NT when you expect either game to be set one trick most of the time, in a manner of speaking.

It seems to me, then, that systems, conventions, treatments, GP defaults, and CW understandings typically have a pro-MP bias that is often not recognized. The "cure" is often to use the MP-biased agreements with an eye to IMP principles, which is strained, rather than to rework the entire set toward IMP thinking.

I feel that I am rambling, but I'm curious if others see this as well and have thoughts to share.

[dake50]

I too think in IMP, so adjust when playing MP.
Another aspect: the best try to set (esp. games) is seldom chosen in MP for fear of losing Overtrick. Which cyclical reasoning feeds itself: defenders can't afford 'best' lead to set so offense can push another trick.
The IMP aspect I hate is Vul games needn't be likely makes only 40%(10 v 6 IMP), so non-V obstruction doesn't force even a narrow decision. Obstruction must keep them out or don't bother. Obstructing equal-Vul at least forces a narrow (=near 50+%) decision, so much more leeway.

Think of MP as experiment lab. See what worked/failed. Then SHOULD that be IMP incorporated ?? With what constraints?

Barry Crane DID develop MP strategy failing in IMP. So the divergence in theory by MP/IMP does exist.

I, too like IMP based system write-ups. So jaundiced look at MP systems.

Where are the IT guys with IMP expectation analyses? Systems analyzed, not conventions?

[hrothgar]

dake50, on Mar 16 2008, 05:51 PM, said:

Where are the IT guys with IMP expectation analyses? Systems analyzed, not conventions?

Most of us got disgusted by system regulations and turned our attention to more practical issues

[hrothgar]

Hi Ken

Few quick comments:

1. I agree with your basic conjecture. IMPs and MP are very different games. I expect that an IMP oriented bidding system would probably look different than a MP oriented bidding system. There are also a fair number of pairs that are known as IMP players or MP players. I'm not sure whether this is a question of bidding system/ temperament /whatever. However, it has been commented about on many occasions.

2. Its far from clear whether the difference in system design are enough to overcome inertia/friction/whatever. There's a cost associated with mastering a bidding system. I'm not sure whether the gains associated in specializing your bidding system are sufficient to compensate for the costs in switching back and forth.

[pclayton]

I think there's some truth to this.

I can think of many treatments I wish I was (or wasn't playing) in MP's. I like the idea of 3N as a choice of games even when we have a major fit. Playing 3N as friv or serious makes this impossible.

Sometimes 2N makes a tasty mp spot. Its usually artificial.

Puppet stayman over 1N is useful.

[kenrexford]

hrothgar, on Mar 16 2008, 10:14 AM, said:

Hi Ken

Few quick comments:

1. I agree with your basic conjecture. IMPs and MP are very different games. I expect that an IMP oriented bidding system would probably look different than a MP oriented bidding system. There are also a fair number of pairs that are known as IMP players or MP players. I'm not sure whether this is a question of bidding system/ temperament /whatever. However, it has been commented about on many occasions.

2. Its far from clear whether the difference in system design are enough to overcome inertia/friction/whatever. There's a cost associated with mastering a bidding system. I'm not sure whether the gains associated in specializing your bidding system are sufficient to compensate for the costs in switching back and forth.

I agree that switching back-and-forth between IMP-fine-tuned systems and MP-fine-tuned systems is impractical.

What I am noticing, however, is that the "expert standard" thinking process is very MP-tiuned, in my opinion, and IMP-deficient. The general meanings of calls are governed by MP analysis, with experts using judgment to switch tactics within the MP-dominant theory. That seems to mean an inability to make certain calls at IMPs that might be useful if meaning whatever the IMP-preference meaning would be an instead zooming or "making the practical IMP call" too often.

Which is fine, of course.

However, I personally cannot even think MP right, as to theory. I think IMP. This causes me to assess an undiscussed auction as "obvious" toward one conclusion that in retrospect, in thinking MP with great effort, is suddenly not obvious but actually counter-intuitive for the MP scenario.

So, do I start to develop my MP-dominant thinking, that I may better mesh with a MP-dominant bridge world, or do I stay IMP-dominant and find a partner of equal IMP-dominant theory to discuss and develop IMP-dominant system and defaults?

I mean, here's the problem. I firmly believe that I could spend years re-thinking MP-dominant theory to end up a once-in-a-while MP success, which frankly bores me anyway. Or, I could try to find someone IMP-theory based to develop a partnership that utterly sucks at MP play but is a serious threat at IMP play. That would maximize my personal talents, but the problem is that most available "better players" think MP, live in a MP-dominant world, and don't even recognize this. I'm not even sure how to find someone who actually thinks IMP, and well.

Of course, maybe "psychotic idiot fringe" is what I am interpreting as "IMP dominant thinking" because I just don't get it.

[awm]

As someone who plays a lot of MP, it seems to me that the expert community has an awful lot of IMP-oriented thinking. This is backed up in part by a cadre of people who think MP is "not real bridge" and by the fact that a lot of conventions are geared toward better slam bidding which is low frequency and high (IMP) reward.

For example, take the basic 2/1 GF method. You lose on invitational hands, sometimes playing the wrong partscore, sometimes even having trouble in game auctions, in exchange for better slam bidding all the times you start with a 2/1 bid.

There are many more examples. Opening 1NT with five-card major. Might help you get to game when you have it. Doesn't help you get to 4M, which is often a better MP spot (one more trick) or the major suit partscore which is very often better than the 1NT partscore (esp. at MP).

Lebensohl and good/bad 2NT. Help you decide whether you have game or not, but make it more difficult to compete to four-over-three because partner doesn't know in which suit you are competing. At IMPs maybe this doesn't matter, because a lot of times one of the two partials is going down and even if they both make it's just a small minus to not compete... more important to make the right decision about bidding 3NT. But at MP the difference between +50 and +130 can be huge.

Maximal doubles -- obvious IMP convention. At MP you want to be able to whack these partscores when opponents overcompete the hand.

To take things to an extreme, have a look at Viking Club. A zillion relay sequences to look for slam on every hand. Very few methods to look for the right partial -- the argument seems to be "if it's a partscore deal, just stop low, who cares if you're in the wrong strain." This kind of method will lose matchpoints left and right, but it could be a winner at IMPs if it helps you make one slam decision per session...

There are a lot of first-time partnerships doing well in MP events -- it seems to happen quite frequently. Presumably first-time partnerships are playing something simple without all the latest gadgets/agreements. This seems to suggest that people playing complex methods are not obtaining a huge advantage from this at MP even though it seems likely they are at IMPs. A good reason to conclude that most of the complex methods/conventions floating around are IMP-oriented?

Of course it's true that slam-bidding methods are not as big a deal at MP as at IMPs, and if your system is designed to facilitate slam bidding (as many systems are) more than anything else you will benefit a lot more from this system at IMPs. But I'd say that most complex systems are IMP oriented and that no one has really done a good job of designing a MP-oriented system yet (if such is even possible) rather than vice-versa.

[Winstonm]

I believe the answer obvious and have long thought seperate systems depending on scoring was the ideal. After all, the goals of the forms of scoring are totally different - frequency versus gain.

[kenrexford]

Interesting.

Maybe there are some IMP intrusions into system, as well as MP intrusions, which at times makes the entire approach to expert bidding schizophrenic and inconsistent. You have theoreticians with their own biases tinkering with structure in such a way as to wiggle back-and-forth between approaches in a way that makes little sense as far as consistency is concerned. Strangely, however, you rarely seem to see intelligent commentary as to the form of scoring being critical as to a meaning in its development.

Take one awm example.

2/1 GF seems oriented toward IMPs. However, most people who play 2/1 GF are miserable at using it, because they think and play MP. So, you end up with MP-biased people playing an IMP-biased structure and yielding terrible results, hampered by system more than helped. The cures become contortions of 2/1 GF with tweakings that sometimes makes sense and sometimes do not, based on whether the originator of the idea is IMP-biased or MP-biased.

I'm not so sure that this bias, however, is purely a slam phenomenon. There seem to me to be a lot of biases in systemic approaches against techniques useful to find marginal games. Fit-bids, sound overcalls, and the like facilitate right-fitting game exploration, but these are rejected with mantras like "if you make 10 tricks, you probably score above average anyway, but a set is terrible." These mantras are true for MP and govern meanings of bids in strange sequences. Instead, bids that could be extremely useful to marginal game exploration are instead used to improve partscore contracts, such as scoring -50 instead of -100.

This also happens in partscore battles. Scoring +90 for 2♦ is horrible at MP when compared to +110 or +120 for NT or a major, but at IMP scoring this is relatively irrelevant. However, bidding is pro-major and pro-1NT too much for IMP thinkers, not only as to decisions but as to structure.

[Cascade]

I am certainly aware of this problem and we have made a conscious decision to bias our style and methods towards IMPs.

This is a little odd as there are many more MPs events for us to play. However we have decided it is the IMP events that have more reward for us. When we play MPs we adapt our style a little but we play our IMP-designed system and just accept that sometimes this will cause problems.

[helene_t]

I don't think it would be worthwhile for me to agree on two different systems with the same p. In one partnership I agreed to play Multi-Landy at IMPs and DONT at matchpoints. Now that is an example of counter-productive complexity.

OTOH the emphasis of partnership discussions may depend on the form of scoring we typically do. In an IMPs partnership I would spend a lot of time on slam bidding technique. In one partnership that was MP oriented, we spend a lot of time of discussing criteria for converting various low-level doubles, and in my perception this payed off (obviously this perception is biased).

Although there are clearly situations in which one should adopt a slightly different style (and it is worthwhile to discuss this with p) I think the differences between the two games is often exaggerated. Whenever I hear someone saying "I would do this at matchpoints but that at IMPs" I cannot help thinking that such a comment has more to do with a desire to sound smart than with rational considerations. OK, I know I am insulting players far better than me.

After all, if 3N and 4M both make, either could be better at MP. And if no-one else bids and makes game, making a trick more in 4M than in 3N may earn an IMP while it can't earn any matchpoints. And 3m+1 is better than 2NT= at matchpoints. A frivolous overcall may be a matchpoint disaster (-200 against a partscore) while not making the overcall may be an IMP disaster (we get stolen from when we could make a vulnerable game). Etc etc.

I do find it puzzling that the Jack software makes its analysis on a total points basis. I wonder if anyone has analyzed the extent of the bias introduced by that technique. Would Jack really bid a 40% vulnerable game at matchpoints? Maybe it's irrelevant since the World championships for computers are IMPs.

[Winstonm]

Barry Crane was the undisputed master of matchpoints. Someone may want to correct me on this, but I remember from the alerts and explanations that his system used a Drury-like response in all seats. By being able to play virtually all major-suit partscores at the 2-level instead of the 3-level gains a huge advantage over thousands of hands.

[MickyB]

Agreed Helene - I think people overrate the difference between MP and IMP bidding, many think you should be much more aggressive in part-score battles at MPs which is a massive over-simplification.

The vul matters much more for part-score battles at MPs. I'd consider playing a more aggressive defence to 1NT at love-all and a less aggressive one at game-all. At IMPs, you are fine using a middle-of-the-road defence throughout.

[han]

Agree with the over-simplication Mike, and I think you just made one.

I also agree that judgement, style and technique should be adjusted rather more than system (although it is undoubtedly true that some treatments are more suitable for MP or more for IMPs).

[glen]

awm, on Mar 16 2008, 11:12 AM, said:

...But I'd say that most complex systems are IMP oriented and that no one has really done a good job of designing a MP-oriented system yet (if such is even possible) rather than vice-versa.

1) MP oriented systems should not be complex, and thus all complex systems are essentially IMP oriented or flawed for MP purposes;
2) Of the systems I've put on the web, the MP oriented stuff has received little interest relative to the other systems.

[MFA]

awm, on Mar 16 2008, 11:12 AM, said:

To take things to an extreme, have a look at Viking Club. A zillion relay sequences to look for slam on every hand. Very few methods to look for the right partial -- the argument seems to be "if it's a partscore deal, just stop low, who cares if you're in the wrong strain." This kind of method will lose matchpoints left and right, but it could be a winner at IMPs if it helps you make one slam decision per session...

Don't insult my regular system, please

The Viking club is fine for MPs, because it's an aggressive system with 10-15 openings, and the competitive and invitational sequences are very strong to cope with the light openings (not hanging partner, but getting in there).
It has some flaws in uncontested partscore auctions (1M a.p. or after 2♣ precision), but this is not really a MP problem.
I wouldn't dream of changing the system for MP events.

All in all I feel these discussions often result in big overthinks. The bidding situation in which I would like to play different methods at IMPs/MPs is a relatively rare bird. The difference is usually a matter of judgement instead, and even that is frequently overthought also IMO.

[the hog]

I very much agree with MFA's comments above. Playing 5 card Ms in a NT opening is a big winner at MPs. How less revealing is a sequence like 1NT Swish compared to 1H 1S 1NT or even 1M Swish.

The aggressive openings of high relay systems such as Moscito give you a big advantage. I would suggest that Adam's comments are those made by someone who has never played those methods and is therefore arguing in the dark, (am I right Adam?), and who has an innate and admitted bias against anything complicated.

[benlessard]

MP events usually bring a weaker field then imps events, so many thing out of the ordinary will do much better in MP then in Imps.

LOB and weak nt effectivness are strongly dependant on the skill of the opponents imo. We play 10-14 nt and the number of gift we get is anormal.

[awm]

The_Hog, on Mar 16 2008, 10:54 PM, said:

I would suggest that Adam's comments are those made by someone who has never played those methods and is therefore arguing in the dark, (am I right Adam?), and who has an innate and admitted bias against anything complicated.

Wow, we are quick to degenerate into attacks. Well if you're really interested...

I am sure I have not played the latest version of Moscito. The versions seem to come a mile a minute, and none are all that well-documented as best I can tell. The last document I saw from Marston (ten pages!) didn't describe any relay structure and was very heavy on philosophy and very light on follow-ups after the first round of bidding.

In any case, I have played a number of transfer opening systems, most often Josh Sher's "Transfer Oriented Symmetric Relay" but also one of Moscito's earlier variants. I have also played against these systems, as well as a few boards against Gnome's transfer-opening system and a few boards against Dr. Todd's strong pass system. And I have kibitzed top-level teams playing these methods on many many occasions.

One of my regular partnerships plays methods which are probably as complex as anyone's, and include light opening bids, invitational or better responses to strong club, and a wide variety of different relay sequences and structures. I would not say I have an innate or admitted bias against anything complicated.

In fact, my experience has been that transfer openings and MAFIA are lousy methods. And I don't see people winning things left and right by playing them... in fact I seem to recall a very unheralded group of Aussies upsetting the "Oz One" team to represent Australia in the Bermuda Bowl (even though this group was supposedly inferior players and playing supposedly inferior natural methods) and then having one of the best performances by an Australian Bermuda Bowl team in recent years. And time and again, I see "first time partnerships" of two good players using very vanilla methods winning pairs events, not folks playing canape-relay or meckwell precision.

I could enumerate a bunch of reasons I believe Moscito is a lousy system for matchpoints, but I don't really see the need. Just because bureaucrats have unfairly outlawed something doesn't mean that method is any good. I'm not allowed to play 3♣ mini-roman in ACBL-land either.

[han]

Uh oh, saying something bad about Moscito, you are in trouble Adam. Ron has already launched a completely ridiculous personal attack, others soon will follow. Probably all their anger about system regulations will now be aimed at you.