For All the Bergen Haters out there

[whereagles]

Its job is to get to the LAW

 

can we stop referring to it as the "LAW"..? it's nothing but a statistical tendency, and not even a very concrete one (it's only correct on about 40% of the cases)

 

if something, use "law" :P

[rhm]

Interesting read :) I'm not at all shocked that Mike Lawrence hates them - they basically seem to turn on what you think of the LoTT.

 

Has anyone reviewed Lawrence's anti-Law books on these forums, btw? I read it recently and it became one of my favourite bridge books, so if not, I should write something.

I found the book interesting but not convincing.

The book is a one sided attack on the law, which is okay since there are so many proponents of the "law"

 

To the credit of the authors they have established a site

 

http://www.newbridgelaw.com/

 

where you get some critical comments, which complements the book well and is recommended. .

 

In the book they substitute the law by a SST+WP theory, which is much more complex and difficult to apply single handed at the table. In my mind this evaluation method may in theory be more precise (not clear though) but it is impractical.

I use SST and the lack of it to overrule the law. For example if I am 4♠333 and my partner opens or overcalls 1♠, I will refrain from a "lawful" raise to 3♠.

It seems to me that the expert community is divided on this subject, but I suspect Lawrence / Windgren are a minority and their book did not change many minds in the expert community.

There are many more fans of the law than of Bergen raises, for example Kit Woolsey.

 

Rainer Herrmann

[whereagles]

Well, I find Lawrence/Wirgren's objections quite convincing. SST/WP is not that cumbersome to use after you get used to it, but I agree it is more practical to use the law + corrections. The difference shouldn't be too big.

[Jinksy]

Likewise, I think it takes a bit of getting used to, but not so much as to make it wildly impractical. If you think their theory is poor that's another question, but it seemed much more reliable than LoTT to me.

 

That said, they don't seem to overlap as much as much as Mike Lawrence claims. LoTT seems like a decent approximation when you don't have much info, and don't have strength/room to convey it. Lawrence's approach seems to work better when you have the relevant info (and is more relevant for constructive bidding).

 

I suspect a decent part of the reason for the law's relative success is better marketing by Cohen and Bergen, who never seem to shut up about it, and are incapable of writing it in lower case (to my consternation as well as whereagles'), not to mention came up with a much catchier name.

 

Also, while they're overly ebullient (at least to us restrained British folks), Lawrence comes across as being unnecessarily spiteful in his criticism, which probably didn't do him any favours.

[rhm]

Likewise, I think it takes a bit of getting used to, but not so much as to make it wildly impractical. If you think their theory is poor that's another question, but it seemed much more reliable than LoTT to me.

 

That said, they don't seem to overlap as much as much as Mike Lawrence claims. LoTT seems like a decent approximation when you don't have much info, and don't have strength/room to convey it. Lawrence's approach seems to work better when you have the relevant info (and is more relevant for constructive bidding).

 

I suspect a decent part of the reason for the law's relative success is better marketing by Cohen and Bergen, who never seem to shut up about it, and are incapable of writing it in lower case (to my consternation as well as whereagles'), not to mention came up with a much catchier name.

 

Also, while they're overly ebullient (at least to us restrained British folks), Lawrence comes across as being unnecessarily spiteful in his criticism, which probably didn't do him any favours.

When I said I consider Lawrence alternative impractical and difficult to apply single handedness at the table, I was hinting at the very fact that I often lack the relevant information for applying Lawrence method with any degree of certainty and you seem to confirm that.

Sometimes it is possible to exchange some of the required information but this often helps opponents and the cost is high.

Lawrence evaluation relies on the combination of two hands and the law in principle as well.

However, different to lawrence method exchanging information about suit length is needed to select a trump suit and the law works almost exclusively on that information.

Evaluation methods of what a hand is worth (e.g. "working points") can be much more precise once the opposite hand is known. But this renders those methods useless.

In a nutshell once I know the opposite hand I do not need any evaluation method to tell me what I can make.

I need good evaluation to categorize my hand properly before I make my first or second bid when I have little or at best very limited information about the opposite hand.

 

Rainer Herrmann

[helene_t]

Lawrence/Virgren are addressing a completely different issue than LOTT addresses. LOTT is used in situations in which one doesn't know (or doesn't care) how much strength one's partnership has - when making a LOTT based competitive bid, one does not acess the likelyhood that the contract makes. Just the likelihood that the total number of tricks is high enough that it either makes OR is a good sac.

 

Lawrence/Virgren's method serves a purpose identical to that of MLTC or fit points, and should be compared to those. It would be meaningful to discuss whether fit points, MLTC or Lawrence/Virgren is more accurate. Comparing those methods to LOTT is like comparing either to sliced bread.

[rhm]

Lawrence/Virgren are addressing a completely different issue than LOTT addresses. LOTT is used in situations in which one doesn't know (or doesn't care) how much strength one's partnership has - when making a LOTT based competitive bid, one does not acess the likelyhood that the contract makes. Just the likelihood that the total number of tricks is high enough that it either makes OR is a good sac.

 

Lawrence/Virgren's method serves a purpose identical to that of MLTC or fit points, and should be compared to those. It would be meaningful to discuss whether fit points, MLTC or Lawrence/Virgren is more accurate. Comparing those methods to LOTT is like comparing either to sliced bread.

I agree.

Law evaluation is unique in the sense that it is a competitive tool and does not look at par but at absolute par.

MLTC is somewhat in between and that is why I do like it. (My own version of course, but close to the one published in the BW)

Many have attacked MLTC because it "overvalues" shortages. (One is Zelandakh). A misunderstanding.

They do not understand that a good evaluation method should point you to the right bid and a little optimistic "overbidding" with a distributional raise rather than one based on HCP power is a good thing because your objective needs to shift from par to absolute par.

MLTC does a good job there, being aggressive with distributional hands and conservative with balanced hands, sometimes without their proponents and critics understanding that.

My critic on Lawrence methods (lack of information to apply it properly) is not affected by the above differentiation.

 

Rainer Herrmann

[Zelandakh]

Yes I do criticise the MLTC for this. Thank you for telling me it is from a lack of understanding - as always I bow to your superior intellect.

 

In an unopposed auction it is my view that it is better to show partner where our shortage is along with our general strength and let them work out for themselves how much it is worth. That is why the schedule of raises posted above contains 3 separate splinter ranges. If we want to "overbid" with a shapely hand then it seems like a good idea to put some system in place for this rather than just assign some arbitrarily inflated figure and lump this together with other hands that have more trick-taking capability.

 

In a competitive auction we may not have enough space for such refinements. Then we have to look at the auction and try to guess what the shortage is worth. Again, using the MLTC blindly in such circumstances is highly questionable. It is fine if the shortage is in a suit where we expect no wastage but in those circumstances we are upgrading whichever the method. The problems come from making this assumption without any indications of it being the case.

[fromageGB]

I can't really see the point in having a theory that tells you what the right contract is when you can see all the hands, but can't be applied at the table. What I want is a method I can use looking at my hand and the bidding. In an unopposed auction, Bergen does a fair job, and is easy for average players. Where it particularly falls down is on hands with shortages, and like others I prefer a method that as well as showing length of support and strength ranges for balanced hands, also shows shortage support hands in various strengths. As Zel says, showing the latter and letting partner make a judgement is better than a unilateral decision.

[rhm]

In an unopposed auction it is my view that it is better to show partner where our shortage is along with our general strength and let them work out for themselves how much it is worth. That is why the schedule of raises posted above contains 3 separate splinter ranges. If we want to "overbid" with a shapely hand then it seems like a good idea to put some system in place for this rather than just assign some arbitrarily inflated figure and lump this together with other hands that have more trick-taking capability.

If you could show it only to partner that would be great. Unfortunately there are two opponents who also get this information.

For slam investigation that could still show a profit.

For game bidding the cost is prohibitive, even more so at matchpoints.

 

Rainer Herrmann

[Zelandakh]

For game bidding the cost is prohibitive, even more so at matchpoints.

That a quick search at ecats for mini-splinter produces quite a few hits, for example Berg-Svendsen of Norway, suggests that an alternative opinion is possible on the comparative cost.

[Jinksy]

Lawrence/Virgren are addressing a completely different issue than LOTT addresses. LOTT is used in situations in which one doesn't know (or doesn't care) how much strength one's partnership has - when making a LOTT based competitive bid, one does not acess the likelyhood that the contract makes. Just the likelihood that the total number of tricks is high enough that it either makes OR is a good sac.

 

Lawrence/Virgren's method serves a purpose identical to that of MLTC or fit points, and should be compared to those. It would be meaningful to discuss whether fit points, MLTC or Lawrence/Virgren is more accurate. Comparing those methods to LOTT is like comparing either to sliced bread.

 

I don't agree with this.

 

Their method/s(better marketing needed) comprise a two hand evaluation approaches that combine to give a near-perfect description of how many tricks the hands can take, if given full information.

 

Looking at the two:

 

1) working points are often hard to gauge at the table - but methods like splinters, game tries etc can give you some part of the picture

2) SST is often easy to gauge at the table, and similar methods can clarify it even further

 

1) is mainly of use in constructive bidding (but can help eg when thinking of marginal decisions when the vul is such that whether you'll make your contract is a more important question than total tricks)

2) is useful in both constructive and competitive auctions

 

Obviously we don't have full information on these issues when bidding, but we don't have zero info either, and each approach's value is proportionate to the info we have about it. I prefer that an approach that, even with full information, can't reliably tell you how many tricks either side (or both sides) will take.

 

Neither subsumes the other, but whichever you prefer, I do think they overlap heavily (as do Lawrence and Virgren - and apparently Larry Cohen, come to that).