The Law of Total Tricks

[PhilG007]

Although Larry Cohen wrote about this in his best selling book "To Bid or Not to Bid",it actually goes much further back than that. The system(if one can call it that) first appeared in an article by Jules-Rene Vernes in 1969 in the June issue of the "Bridge World" magazine. At the time,Vernes espousal caused barely a ripple. However,when Cohen's book was published in 1993,it was hailed as a major advancement in bidding theory. However,several top experts began to doubt the

accuracy of the system,claiming it had several serious flaws. These were highlighted by top U.S. expert Mike Lawrence in his

book "I fought the Law" (1998) Has anyone here ever used this device? Can they testify as to whether it brought them improved results.

or was Lawrence correct in saying it's more of a liability than an asset?

[manudude03]

I would use it often on partscore battles where each side has close to 20 HCP. That's when it's most accurate. At game or higher, I am more likely to use ODR (offense-defense ratio). The logic behind the LoTT is sound enough, and generally speaking, my results are better when using it. I would say that the vast majority of non-novices use the LoTT to at least some extent. An auction like (1C)-1H-(X)-3H showing a weak hand with 4 hearts is very common, and sound in LoTT terms (since you have a 9 card fit, the opponents must have at least an 8 card fit somewhere, therefore there are at least 17 total tricks which warrants a 3-level bid).

[diana_eva]

You may find these useful:

 

 

http://bridgebase.com/forums/topic/69995-the-law-of-total-tricks-lott

http://bridgebase.com/forums/topic/69009-law-of-total-tricks-question

http://bridgebase.com/forums/topic/54809-the-law

 

I'm sure there are more out there, that's just what came up with a simple "total tricks" search. You can make your search more specific if you are interested in something in particular other than "do you use it"

[Zelandakh]

It's a very useful tool for the typical club player. I would certainly recommend understanding it for someone at your level of ability.

[PhilG007]

It's a very useful tool for the typical club player. I would certainly recommend understanding it for someone at your level of ability.

I don't need it...I use the Losing Trick Count :)

[Zelandakh]

I don't need it...I use the Losing Trick Count :)

Then you have already decided that LTC is better than LoTT in all situations? Why did you ask the question about LoTT then?

 

As an side, do you use the original LTC or the MLTC? For the latter, you might find some previous BBF threads on the subject enlightening.

[nige1]

IMO, both the Losing Trick Count and the Law of Total Tricks aid judgement. Some experts probably think so too.

 

F. Dudley Courtenay: The System the Experts Play (1934) : LTC.

Jean-René Vernes: Bridge moderne de la défense (1966): LoTT from '50s WC analysis.

Dick Payne & Joe Amsbury: Bridge: TNT and Competitive Bidding (1981).

Larry Cohen: To Bid or Not to Bid: the Law of Total Tricks (1992).

Mike Lawrence & Anders Wirgren: I Fought the Law of Total Tricks (1998).

[barmar]

Although Larry Cohen wrote about this in his best selling book "To Bid or Not to Bid",it actually goes much further back than that. The system(if one can call it that) first appeared in an article by Jules-Rene Vernes in 1969 in the June issue of the "Bridge World" magazine.

Is this supposed to be a major revelation? Vernes's article is acknowledged in the 3rd paragraph of Marty Bergen's Foreword to Cohen's book. I think every well-read bridge player knows that the LOTT is based on this article.

[PhilG007]

Then you have already decided that LTC is better than LoTT in all situations? Why did you ask the question about LoTT then?

 

As an side, do you use the original LTC or the MLTC? For the latter, you might find some previous BBF threads on the subject enlightening.

 

To answer your first question....curiosity about the opinions of others.

To answer your second question...the LTC because it's best by test...

[ArtK78]

To answer your second question...the LTC because it's best by test...

I guess you are consistent. MLTC is so much better than the original LTC that there is no real comparison.

[PhilG007]

I guess you are consistent. MLTC is so much better than the original LTC that there is no real comparison.

That's as maybe but I guess I'm set in my ways :)

[Zelandakh]

To answer your first question....curiosity about the opinions of others.

To answer your second question...the LTC because it's best by test...

The general opinion here is that LoTT is a helpful guideline in certain circumstances but is no replacement for expert judgement.

The opinion of MLTC is mixed - some posters like it a lot and others don't. Mathematically it is a hcp count by a different name so not as revolutionary as some would have you believe.

Most here would say that the original LTC is close to useless. Statistically it is amongst the worst evaluation methods available. If you seriously believe this to be optimal then you probably need to work on your basics before incorporating LoTT into your game.

 

 

PhilG, you started out like a normal poster and now you're becoming more and more aggressive.

And I have to ask why you allowed this Diana. If hrothgar or the hog had made such a post questioning whether PhilG had any friends, etc, then I suspect you would have edited that out of the message. Yet here you have chosen to leave it. Admittedly the message says nothing about Mike and everything about the OP and we can all see that but there is a consistency to be had. When one of our "pricklier" posters next gets upset and launches a personal attack against one of our more sensitive posters or a newbie, they can rightfully feel aggrieved to be singled out. Personal attacks are against the forum rules. This post contained such an attack without any veneer of prentence and I think you are wrong to leave it uncensured.

[rhm]

I don't need it...I use the Losing Trick Count :)

This is a misunderstanding

LTC tries to estimate how many tricks your side can win when you play a trump contract.

LOTT tries to answer whether it might be profitable to bid on in a competitive situation or in anticipation of a competitive situation.

LOTT does not answer the question how many tricks your side can make.

Some have concluded that LOTT is useless unless HCP are somewhat evenly distributed between both sides of a deal.

 

Mathematically it(LTC) is a hcp count by a different name so not as revolutionary as some would have you believe.

I disagree, though I agree with the latter that there is nothing revolutionary any more about a basic concept (later substantially refined), which was first introduced 80 years ago.

HCP usually work best for notrump contracts, LTC was designed for trump fits, where HCP does not work so well since distribution gets very important.

LTC tends to be conservative on balanced hands and is aggressive on unbalanced ones, which is sensible given the different ODR ratio of balanced and unbalanced hands.

LTC is essentially not a "high card point" evaluation method even though some have tried unconvincingly to equalize it to one.

 

Rainer Herrmann

[Zelandakh]

I have explained the conversion here often enough Rainer. MLTC uses a combination of hcp (A = 1.5; Kx = 1; Qxx = 0.5) and dp (void = 3; sgl = 1.5; dbl = 0.5). It is a trivial matter to normalise this to hcp of 4.5/3/1.5 which then gives an associated dp of 9/4.5/1.5. It is thus trivial to see that MLTC overvalues shortages in comparison with more common hcp + dp schemes. Of course you can choose to use even higher dp values; the question is simply what the optimal levels are. And my belief is strongly that the dp values used in MLTC are not optimal - I understand your opinion is different. But I think both of us can agree that the original LTC is not good.

[ArtK78]

I have explained the conversion here often enough Rainer. MLTC uses a combination of hcp (A = 1.5; Kx = 1; Qxx = 0.5) and dp (void = 3; sgl = 1.5; dbl = 0.5). It is a trivial matter to normalise this to hcp of 4.5/3/1.5 which then gives an associated dp of 9/4.5/1.5. It is thus trivial to see that MLTC overvalues shortages in comparison with more common hcp + dp schemes. Of course you can choose to use even higher dp values; the question is simply what the optimal levels are. And my belief is strongly that the dp values used in MLTC are not optimal - I understand your opinion is different. But I think both of us can agree that the original LTC is not good.

Z, we may be talking about two entirely different concepts.

 

MLTC has nothing to do with HCP. It has to do with losers and cover cards. I first ran into the concept in the first Romex book by George Rosenkranz. I have used it since, and found it to be very effective. In the first Romex book, MLTC is used as the basis for the major suit raise structure and many other bids, such as the requirements for opening a Dynamic 1NT, the original Mexican 2♦ and the very strong 2♣ opening, as well as for determining which bids are minimum, middle and maximum.

 

Please enlighten me as to your idea of what MLTC is.

[PhilKing]

Z, we may be talking about two entirely different concepts.

 

MLTC has nothing to do with HCP. It has to do with losers and cover cards. I first ran into the concept in the first Romex book by George Rosenkranz. I have used it since, and found it to be very effective. In the first Romex book, MLTC is used as the basis for the major suit raise structure and many other bids, such as the requirements for opening a Dynamic 1NT, the original Mexican 2♦ and the very strong 2♣ opening, as well as for determining which bids are minimum, middle and maximum.

 

Please enlighten me as to your idea of what MLTC is.

 

Well I understand what he means, but it is not for me to explain ...

[Zelandakh]

Please enlighten me as to your idea of what MLTC is.

You are still blinded by the marketing Art. Let me explain it a different way. Take a 12 loser hand, that is a 4333 hand without any aces, kings or queens, and call that zero points. To improve that to an 11.5 loser hand we need to add either a queen (call that 1.5 hcp) or change the distribution to 4432/5332 (call that 1.5 dp). Similarly, to go from 12 to 11 losers we add either a king (3 hcp) or 2 queens (1.5 x 2 = 3 hcp). And so on - each 0.5 losers equates to 1.5 points on this scale. Of course we could normalise it to a different scale if we wanted such as Zar points (6/4/2) or QP (3/2/1). I chose to use 3 points for a trick to correspond to the regular point count most people use.

 

Do you see the trick now? If you have understood, it should now be trivial for you to produce a point count system equivalent to the original LTC. Once you see what it really means it becomes painfully obvious just how bad that method is. And yet many club players use it and swear by it. And that is why it is my opinion that the LTC has on average lowered the standard of bridge. Even if there are a few who understand what it does, most users do not and are too blinded by the marketing. It is just another point count method, no more, no less.

[akwoo]

LoTT is not perfect, but I think most intermediates are better off using it than not using it.

 

One of the main reasons for mediocre players staying mediocre is that they don't realize that half the time your job is to keep the opponents from playing in their best contract, not to get to your best contract. If nothing else, using LoTT helps you realize this, because to apply LoTT properly you need to compare various projected scores on a regular basis.

 

Actually, (M)LTC is very good replacement for LoTT, IF you remember to calculate the opponents' LTC as well as yours. That's more work (and requires more guessing) than LoTT though.

 

MLTC does differ from normal point count in how it evaluates singleton and doubleton honors. Having Qxx in one suit and xx in another is better than xxx and Qx. (This is even more true for QJx/xx vs Jxx/Qx or QJx/xx vs xxx/QJ.) It's much easier to see this difference with LTC based methods than with normal point count methods.

[Vampyr]

LoTT is not perfect, but I think most intermediates are better off using it than not using it.

 

One of the main reasons for mediocre players staying mediocre is that they don't realize that half the time your job is to keep the opponents from playing in their best contract, not to get to your best contract. If nothing else, using LoTT helps you realize this, because to apply LoTT properly you need to compare various projected scores on a regular basis.

 

It seems to me that the most common application of the LoTT is simply "the three level belongs to the opponents". So bid 3/2 if it seems like each side has an 8-card fit, and bid 3/3 if you have a 9-card fit or have a good reason to suspect that they do.

 

Any more than this is a) putting too much faith in the Law and b) pretty difficult, since at higher levels there are loads of adjustments and much of the information you need to make these adjustments will not always be available to you,

[ArtK78]

You are still blinded by the marketing Art. Let me explain it a different way. Take a 12 loser hand, that is a 4333 hand without any aces, kings or queens, and call that zero points. To improve that to an 11.5 loser hand we need to add either a queen (call that 1.5 hcp) or change the distribution to 4432/5332 (call that 1.5 dp). Similarly, to go from 12 to 11 losers we add either a king (3 hcp) or 2 queens (1.5 x 2 = 3 hcp). And so on - each 0.5 losers equates to 1.5 points on this scale. Of course we could normalise it to a different scale if we wanted such as Zar points (6/4/2) or QP (3/2/1). I chose to use 3 points for a trick to correspond to the regular point count most people use.

 

Do you see the trick now? If you have understood, it should now be trivial for you to produce a point count system equivalent to the original LTC. Once you see what it really means it becomes painfully obvious just how bad that method is. And yet many club players use it and swear by it. And that is why it is my opinion that the LTC has on average lowered the standard of bridge. Even if there are a few who understand what it does, most users do not and are too blinded by the marketing. It is just another point count method, no more, no less.

I read what you wrote. I just don't agree with you.

[StevenG]

Do you see the trick now? If you have understood, it should now be trivial for you to produce a point count system equivalent to the original LTC. Once you see what it really means it becomes painfully obvious just how bad that method is. And yet many club players use it and swear by it. And that is why it is my opinion that the LTC has on average lowered the standard of bridge. Even if there are a few who understand what it does, most users do not and are too blinded by the marketing. It is just another point count method, no more, no less.

I think you are completely wrong on this, Zel. Learning the LTC massively improved my bridge, because it gave me an insight into how to evaluate unbalanced hands. Now, I'm not stupid and I don't use it blindly. After the initial revelation, I slowly worked out what was good and what was bad. But I still do a HCP count and a raw LTC, and if the answers suggest different things, then I stop and think about WHY they are different, and that gives me clues as to how to reevaluate my hand in the light of an ongoing auction.

 

I have a longstanding partnership with an essentially social player who uses the (raw) LTC without understanding what's going on. I had a longstanding club/tournament partnership with another player who generally bids well, but doesn't declare well and doesn't use LTC. On distributional hands, we find good contracts in the former partnership, but miss them in the latter. So, my belief is that LTC raises the standard of club-level bridge, even for those who do not make sensible adjustments.

[gwnn]

I read what you wrote. I just don't agree with you.

Do you disagree with the "made club bridge worse" statement, which is obviously controversial, or his maths that shows that MLTC considers a void to be worth 9 points, or do you disagree with the concept that 9 points is too much for a void? Or do you disagree with all of the above, or perhaps, on a more metaphysical level, you disagree with the concept of Zelandakh as a person?

[rhm]

Do you disagree with the "made club bridge worse" statement, which is obviously controversial, or his maths that shows that MLTC considers a void to be worth 9 points, or do you disagree with the concept that 9 points is too much for a void? Or do you disagree with all of the above, or perhaps, on a more metaphysical level, you disagree with the concept of Zelandakh as a person?

Nobody disagrees with Zelandakh as a person. We all like him.

However, there is a good reason to raise more aggressively with a side suit void rather than holding AKQ in a side suit.

An evaluation method which does appreciate this difference does not equalize a void with AKQ.

But there might be Bridge wisdom to raise in both cases to the same level.

If your chances making your contract are poorer with the void you might still show a profit.

High cards are a zero sum game. The aces you hold can not be held by your opponents.

Distribution is not a zero sum game.

In other words some deals contain a total of 40 HCP and some much more.

I like an evaluation method, which appreciates this tactical differences, which Zelandakh mathematics does not.

 

When I show a limit raise over partner's 1♥ opening I might hold ♠x ♥Qxxx ♦Axxx ♣xxxx or ♠Kxx ♥Qxxx ♦Kxx ♣Kxx

I could not care less whether Zelandakh believes the singleton spade in the first case is not worth 5 HCP.

I believe it is good Bridge to raise with both hands to the same level and invite game.

 

Rainer Herrmann

[cherdano]

So, LTC is meant not to measure how likely we are to make a given contract, but how likely it is right to bid a given contract (given that opponents might be making something on their own)? That's a good defense of LTC. It's not, however, how most proponents seem to use it.

[rhm]

So, LTC is meant not to measure how likely we are to make a given contract, but how likely it is right to bid a given contract (given that opponents might be making something on their own)? That's a good defense of LTC. It's not, however, how most proponents seem to use it.

I am not sure, but the distinction is crucial.

It is my impression that those who like LTC like it, because it improved their game, which means they get better results using LTC.

 

Rainer Herrmann