[Double !]

P_Marlowe, on Jul 8 2006, 10:32 AM, said:

Hi,

just a couple of comments.
The Law of Total Tricks was formulated by Verne
around 1966. (+- 3 years), i.e. it was around a couple
of years before it was mentioned by Kit Woolsey or
by Larry Cohen.
http://www.bridgeworld.com/default.asp?d=a...r&f=samltt.html

With kind regards
Marlowe

I just wanted to thank you for this link.

DHL

[luke warm]

Quote

JTxx
Qxx
xxxxx
x

2S in the first one because I have a decent hand in support of spades

ok.. the bidding goes

1S (p) 2S (p) p (x) p (3H) p (p) your bid... do you pass, double, or bid 3S? i'm simply saying that most who now know of LOTT, mainly due to cohen's book, would bid 3S (assuming they didn't the first time) and that they'd be right to do so *most* of the time... it might seem intuitive to do so, but LOTT has replaced intuition for a lot of people, for better or worse

Quote

AKxx
Kx
ATxx
QJxx

3S in the second one because I have a hand that might produce game opposite a hand that would pass 2S

yes you do... but suppose you held

AKX
Kx
ATxxx
QJx

you're still max but now you have (as far as you know) an 8 card vs. 9 card fit.. would you still super accept? if not, why not? all i'm saying is, the 9th trump plays a big role, and we can base that on LOTT or bridge logic or on whatever we choose... LOTT has simply given a greater number of average (and better) players another weapon to use

[mike777]

helene_t, on Jul 8 2006, 04:52 AM, said:

I suppose this "FTL" method is what Lawrence and Wirgren call SST/WP (Short Suit Totals/Working Points). It's the same as LTC except that it's downwards adjusted in case of duplication.

Not very original, thus, and besides the information you need is not available in practice. In particular, it doesn't make sense to compare it to the LOTT since it uses different covariates to predict a different outcome.

The SST/WP method is just postulated without any systematic evidense. The problems they present with the LOTT is what Cohen also admits in his own books. "I Fought the Law" is just a cruscade against the LOTT, and a very poor one that is.

It's the worst bridge book I ever read. The owner of the local bridge bookshop says the same, which is unusual - they usually talk about the books they sell in a more diplomatic (if not positive) tone.

I do not understand how a book that basically says if you estimate you have more working hcp and you estimate you have more short suits without duplication, you will make more tricks, with less you will make fewer can be the worst book ever written. The book just seems to tell you to focus on 2 main issues to simplify things for nonexperts.

I would think this would be a very good thing to encourage newer players to do.

If you are an expert and focus on other issues that is great, but we got to start somewhere. Working hcp and shape/shortness seems like a good place to start. It gets you thinking on visualizing the other hands, that must be a good thingy....

[luke warm]

mike777, on Jul 8 2006, 08:07 PM, said:

Working hcp and shape/shortness seems like a good place to start. It gets you thinking of visualizing the other hands, that must be a good thingy....

i agree, which is why (when i play 2/1) i much prefer treatments that show different ranges/shapes.. over and under j/s, swiss raises, etc

however, i still think LOTT is more than adequate for competitive decisions

[awm]

I'm not really sure what the point was for the FTL book. Especially since the next line is "... and the law won."

The authors seem to really miss several points:

(1) LOTT was never claimed to be more than "right on average." It would be unbelievable if something that simple was "always right" or even "right on a substantial majority of hands." Producing hands where LOTT is inaccurate doesn't prove anything, and their own statistics support that it's right on average, at least when the number of trumps is moderate (like 8-10 to a side).

(2) The primary purpose of LOTT is in helping with quick competitive decisions. For example, LHO opens 1♥, partner bids 1♠, RHO bids 3♥. Do you bid 3♠? If LHO then competes to 4♥, do you bid 4♠? The kind of information that FTL requires about relative suit lengths in your hand and partners, or wasted values opposite shortness in side suits, is not typically available in these auctions. Similarly ZAR fit/misfit points are hard to evaluate.

(3) Obviously if you tell me partner's exact distribution and high cards, I will do a better job of gauging the final contract than LOTT. This is not particularly mysterious. It's also not particularly helpful when partner has made one call and the bidding is at the four-level, and I have to decide what to do.

(4) Even the proponents of LOTT admit that there are adjustments for various situations. Having very flat hands (i.e. suits break) tends to reduce the number of total tricks. Having wasted high cards in the opponents suits tends to reduce the number of total tricks. If you look at Ben's example hands, the clubs are 2-2 and a substantial amount of N/S high card strength is tied up in the AK tight of ♦ (in a suit where opponents hold the length). In the E/W hands, the spades are 1-1 and the hearts also divide relatively evenly. Yes, this effects the FTL and ZAR evaluation, but that type of issue was well-known before those methods were popularized. I think even Marty Bergen tends not to make a "Bergen 4-card raise" with 4333 shape.

As to ZAR points, I'll even give you that they're more accurate than many other evaluation methods especially once fit/misfit is added in. But that doesn't mean I want to use them. Counting straight high card points (plus mild adjustments for spots and suit lengths) is quite accurate in the presence of a misfit. Especially in the context of modern bidding where responder often has to decide whether to game force at first turn, it's best to start off conservative. This way you get decent results when you have no fit. If you find a fit, you can upgrade later. The ZAR method adds a lot of points for shape at the beginning. Then if you discover you have no fit you have to downgrade later, but the structure of modern bidding means you are often in a game forcing auction when you discover that partner has two cards in each of your five card suits, and your beautiful 5-5 ten count suddenly isn't worth half what the partnership needs for game anymore.

[kenberg]

"Even the proponents of LOTT admit that there are adjustments for various situations."

Not only do they admit it, they make heavy use of it in support of LOTT. In the hand I present, suppose 3CX is off 1. LOTT supporters say "Of course you should bid 3C, just follow the Law". If 3CX is off two, they say "In order to use LOTT properly, you have to make adjustements. This is clearly spelled out in the Book".

With this versatility, the Law is never wrong.


I have little quarrel with the Law, only a quarrel with those who want to elevate it to a position that I don't think is justified. Coupled with some judgment, I'm sure it is useful.

[kenberg]

luke warm, on Jul 8 2006, 06:23 PM, said:

Quote

JTxx
Qxx
xxxxx
x

2S in the first one because I have a decent hand in support of spades

ok.. the bidding goes

1S (p) 2S (p) p (x) p (3H) p (p) your bid... do you pass, double, or bid 3S? i'm simply saying that most who now know of LOTT, mainly due to cohen's book, would bid 3S (assuming they didn't the first time) and that they'd be right to do so *most* of the time... it might seem intuitive to do so, but LOTT has replaced intuition for a lot of people, for better or worse

Quote

AKxx
Kx
ATxx
QJxx

3S in the second one because I have a hand that might produce game opposite a hand that would pass 2S

yes you do... but suppose you held

AKX
Kx
ATxxx
QJx

you're still max but now you have (as far as you know) an 8 card vs. 9 card fit.. would you still super accept? if not, why not? all i'm saying is, the 9th trump plays a big role, and we can base that on LOTT or bridge logic or on whatever we choose... LOTT has simply given a greater number of average (and better) players another weapon to use

Second hand first: You have really given me a great hand, one that I might well have opened 1D intending to rebid 2NT over a major response. But given that I open this 1NT, I think I would bid 3S. Partner will expect four spades in my hand, but AKx may do.


Second hand: I am not sure what is right here. My concern is that 3S may push them into 4H which I am not so sure I can beat. Perhaps 3S on the first round, if I were playing Bergen, would have worked better, or perhaps not. I might pass and hope that they have an untouchable ten tricks. LOTT plays a role here, but I wouldn't say: 17 total trump, therefore 3S. There is a lot of shape here, and their trick total in hearts may be ten, whatever the combined total is.

My judgment is only so good, but with practice I can hope it gets better. Just bidding by the Law keeps me from exercising, and hopefully improving, my judgment.

This may all come down to personal preference. I have resistance, perhaps an excessive resistance, to formulas.

[mike777]

Is not judgment just a set of rules or formulas, consciously or unconsciously?

It seems even if you use a random number generator to make your decision, that is some kind of rule making device.

[luke warm]

mike777, on Jul 9 2006, 11:17 AM, said:

Is not judgment just a set of rules or formulas, consciously or unconsciously?

It seems even if you use a random number generator to make your decision, that is some kind of rule making device.

of course you're right, mike... another thing i found interesting about ken's post was when he said, when affirming a super accept, "Partner will expect four spades in my hand, but AKx may do."

the question i kept trying to get him to address is, *why* would partner expect 4 spades? 9 trumps = 9 tricks, conciously or subconciously, enters into it even when responder xfers with xxxxx xx xxx xxx ... i know i'd xfer with that hand if p opened 1nt

[mike777]

It does seem using judgment is using a very complicated, excessive amount of rules or formulas, so complicated we may not be able to fully express it to others or ourselves.

[kenberg]

In Jimmy's NT hand, partner will expect four spades when I super accept because I rarely would super accept with three. This hand is sort of an exception, but not really because with a 17 count and a decent five card minor I would have opened it 1D. Turn a long diamond into a club and I would open 1NT and not super accept.


As to whether more or less any judgment call is some sort of formula, perhpas in some sense it is true but I think it also scuttles a worthwhile distinction.

Go back to the club hand, and suppose a reasonable auction of
P P 1H P
2H P P ?

Balancer doesn't have much but their side is non-vul, he expects the heart ace to be favorably located, partner may have spades, the scoring is matchpoints. All in all, it's probably a balancing double. I wouldn't quarrel with a partner who passed it out, I wouldn't quarrel if he doubled and it worked badly. He has to place a bet, I think he should take more things than LOTT into consideration, and then partner should accept his best effort regardless of the result. To say this reasoning is "by formula" may be correct in some philosophical sense (and who am I to dis philosphy) but in practical terms I think it is different from saying "Follow the Law".

Doing a little result merchanting, I think in this case 3C works badly, since the poster of the hand was annoyed that 2H was passed out. I accept that one hand proves nothing, of course.

Ken

[han]

mike777, on Jul 8 2006, 08:07 PM, said:

helene_t, on Jul 8 2006, 04:52 AM, said:

I suppose this "FTL" method is what Lawrence and Wirgren call SST/WP (Short Suit Totals/Working Points). It's the same as LTC except that it's downwards adjusted in case of duplication.

Not very original, thus, and besides the information you need is not available in practice. In particular, it doesn't make sense to compare it to the LOTT since it uses different covariates to predict a different outcome.

The SST/WP method is just postulated without any systematic evidense. The problems they present with the LOTT is what Cohen also admits in his own books. "I Fought the Law" is just a cruscade against the LOTT, and a very poor one that is.

It's the worst bridge book I ever read. The owner of the local bridge bookshop says the same, which is unusual - they usually talk about the books they sell in a more diplomatic (if not positive) tone.

I do not understand how a book that basically says if you estimate you have more working hcp and you estimate you have more short suits without duplication, you will make more tricks, with less you will make fewer can be the worst book ever written.

I haven't read the book, so I don't know if the book is good or bad.

But Mike, if that's basically what the book says, then it should be able to fit on one page. Heck, you say it so well in less than two lines!

[cherdano]

Hannie, on Jul 9 2006, 07:49 PM, said:

mike777, on Jul 8 2006, 08:07 PM, said:

helene_t, on Jul 8 2006, 04:52 AM, said:

(...)"I Fought the Law" is just a cruscade against the LOTT, and a very poor one that is.

It's the worst bridge book I ever read. The owner of the local bridge bookshop says the same, which is unusual - they usually talk about the books they sell in a more diplomatic (if not positive) tone.

I do not understand how a book that basically says if you estimate you have more working hcp and you estimate you have more short suits without duplication, you will make more tricks, with less you will make fewer can be the worst book ever written.

I haven't read the book, so I don't know if the book is good or bad.

Han, you have me confused, I distinctly remembered a post where you said you have read it (and didn't like it).

BBF's memory agrees:
http://forums.bridge...showtopic=11028

Was it so boring that you forgot it? Did I mix s.th. up? Did I spoil some subtle irony?

[han]

Huh, maybe maybe my old me is right. I don't have the book myself, but my former f2f partner does, I must have read it while visiting him. The book obviously didn't make a strong impression on me.

[whereagles]

FTL is an unbelievably boring, but good, book. Replacing LOTT with SST/WP does increase bidding precision considerably, regardless of it not being too well motivated in the book.

However, it will never be popular because it demands way too many calculations at table. People who have computers for brains (e.g. Alan Sontag or Michael Rosenberg) could use it on a regular basis, but most players will never have anything to do with it. They'll keep using the LOTT, which does fine in most cases and is far easier to use.

[helene_t]

Even if one adopts the WP/SST method (or LTC with a duplication corrrection with is just another formalism for the same), one should still use the LOTT because the two are used for different purposes. WP/SST predicts your own side's tricks given detailed information about your own side's assets, which is useful for constructive, uncontested bidding. The LOTT predicts the total number of tricks given very little information about both sides' assets, which is useful for fast, competitive bidding.

It might be interesting to compare WP/SST to some other predictor for one's own sides tricks, such as ZAR points. Comparing the accuracy of LOTT to that of WP/SST is like comparing the hight of the Eifel Tower to that of a musical node.

Personally, I use both LTC and LOTT and I don't feel that the two are conflicting since I use them in different situations.

[ArcLight]

>However, it will never be popular because it demands way too many calculations at table. People who have computers for brains (e.g. Alan Sontag or Michael Rosenberg) could use it on a regular basis, but most players will never have anything to do with it. They'll keep using the LOTT, which does fine in most cases and is far easier to use.

What it requires you to do is picture everyones distribution.
That may not be easy, but in my limited experience that is far more important to do (or try to do) than knowing 15 conventions like Namyats, Kaplan Inversion, etc.

Visualization is a vital skill in this game.


ZAR points are a hand evaluation technique, no so much a competitive bidding technique.

[whereagles]

ArcLight, on Jul 10 2006, 12:52 PM, said:

1.
What it requires you to do is picture everyones distribution. That may not be easy,

2.
but in my limited experience that is far more important to do (or try to do) than knowing 15 conventions like Namyats, Kaplan Inversion, etc.

1. Indeed. It is very stressful and time consuming. I've tried it. The good news is most hands actually don't require using SST/WP, so you can relax in between. But in the end it still is cumbersome and mentally demanding, so I'm not sure it will improve your bidding in the long run.

2. True. With practice visualization gets easier. Adding WPs doesn't get much easier though.

[inquiry]

ArcLight, on Jul 10 2006, 07:52 AM, said:

ZAR points are a hand evaluation technique, no so much a competitive bidding technique.

You are correct in a way. But FTL is also a hand evaluation method, not so much a competitive bidding technique as well. Lawrence suggest not using FTL until a fit is found, and many ZAR points (Fit points) don't come in until a fit is found, and if you have read Zar's newest online book, he subtracts ZAR MISFIT points if no fit is found (reducing the ZAR count). Zar and Lawrence (and Cohen) also devaluate for wasted (or not working) points, in Cohen's case is terms of "purity" but the issue is the same.

[han]

ArcLight, on Jul 10 2006, 07:52 AM, said:

What it requires you to do is picture everyones distribution.
That may not be easy, but in my limited experience that is far more important to do (or try to do) than knowing 15 conventions like Namyats, Kaplan Inversion, etc.

I agree that it is more important, but knowing 15 conventions like Namyats and Kaplan inversion is much easier.