[mike777]

With summer in full spring and Ben and others mentioning Zar points it seems a good time to have someone discuss what Zar points are, how to count full zar adjuted points and what level to bid with them.

Reading the site seems almost impossible. I am more interested in how full Zar points are bid and to what level. I am not interested in yet another debate on whether they work or why they work or do not.

I appreciate Ben mentioning them but it seems many times he Counts Full Zar adjusted points and then downgrades them or upgrades them which make them seem even more mysterious.

I understand opening, fit or nonfit, opener or responder, sac etc may all count them differently, I am only asking how not the why.

[inquiry]

Well, I seem to be the only real advocate for ZAR points here... so let me give this a try.

Basic ZAR points are made up of three elements...

1) HCP (4,3,2,1 scale)
2) Controls (Ace = 2, K = 1)
3) Distribution (twice your longest suit plus your second longest minus your shortest).

Then add those up. One example will do....

AKxx Qxxxx Ax xx

Here, you have
1) 13 hcp
2) 5 controls (two for each ace, one for the king)
3) 12 distributional points (10 for five hearts times 2, 2 for 4 spades - 2 clubs).

The total is 30 ZAR points.

Zar claims you need 26 ZAR points to open, 25 if you hold 4 spades or more.
52 Zar points for game, 62 Zar points for slam, 67 Zar points for slam. Some distributional hands can add up to incrediable ZAR totals, be sure to check for ACES to avoid disasters.

As for adjustments to ZAR, there are several types. Of course, singleton or doubleton Q or J in suits partner didn't bid are dicounted. Zar removes one, I remove the jack always, and the queen, if singleton, I disount two, if doubleton, I discount 1. I also tend to discount a singleton King two if partner does not bid NT or that suit.

Then there are "fit" evaluations. If you have a fit, you get one bonus point for each honor (AKQJT) in partner's suit, up to max of 2 bonus points. You get 321 points for void, singleton, doubleton as well with a fit, and can get bonus those points if you hold extra cards in partner's suit (partner opens 1S and you have five spades, a void in some side suit is worth 6 points, not 3).. but with 3 card support, only worth 3.

Then there is ZAR MISFIT points which can be added or subtracted. Bascially what ZAR misfit point do when you lack a fit is tell you to bid cautiously. You don't really need ZAR misfit points, bridge judgement tells you exactly the same. But if you want to know what ZAR misfit points are (and you need them for some "fits"), they are difference in the suit legnths between the two hands...

When you have a misfit (7 card fit), you SUBTRACT the misfit points, when you have a normal fit (8 cards) you ignore the misfit points (use the normal fit points above), when you have a "super fit" you add either the MISFIT points to the normal ZAR total or the ZAR fit points mentioned ealier (which ever is greater, but not both).

Zar give this hand as an example.....

AKQJT98765432
-
-
-

-
xxxxx
xxxx
xxxx

Here north has 10 HCP, 3 control points and 26 distributional points, total = 39 ZAR points, south has 14 Zar points, and the ZAR misfit points is 26, since there is a super fit (13 card fit). Add it up, 39+14+26 = 70 Zar points, more than enough for grand slam. Ok, sort of artificial, but that is the theory. He also gives...

AKQJxxx
-
xxxxxx
-

Txxxxx
xxxxxxx
-
-

North has 33 ZARS, NS have a large misfit points (1+7+6+0) for 14 misfit points and south has 20 zars. 20+33+14 = 67 ZAR, Grand slam "values"

You can read about his MisFit and fit points in http://www.zarpoints...intsBidding.zip

That link contains his "bidding book" which explains his ZAR POINTS based bidding system. I ignore his bidding system, but the evaluation method is in there too.

[hrothgar]

Hi Mike

As you note, its possible to have a couple radically different conversations about Zar points.

Conversation 1: Are Zar points and accurate way to evaluate hand strength

Conversation 2: Assume that you believe that Zar Points are nice and accurate... How would you best integrate Zar's evaluation system into a bidding system.

From my perspective, the best place to start for "Conversation 2" would be to focus on a fairly narrowly defined bidding problem and try to develop a good intuitive feel for applying Zar points. Use simulations to identify your break points. Once you feel comfortable with the basics, start to branch out.

Potentially the best place to start would be your 1NT opening structure. The 1NT opener's hand shapes are fairly well constrained, so you don't need to work about some 5-5 pattern distorting the hand strength. You also have a number of bidding situations where range asks are very important. If Zar points improve your judgement in auctions like

1N - 2♣
2M - 3M

you'll see some real benefits...

[inquiry]

hrothgar, on Jul 10 2006, 02:28 PM, said:

Hi Mike

As you note, its possible to have a couple radically different conversations about Zar points. 

Conversation 1:  Are Zar points and accurate way to evaluate hand strength

Conversation 2:  Assume that you believe that Zar Points are nice an accurate...  How would you best integrate Zar's evaluation system into a bidding system.

From my perspective, the best place to start for "Conversation 2" would be to focus on a fairly narrowly defined bidding problem and try to develop a good intuitive feel for apply Zar points.  Use simulations to identify your break points.  Once you feel comfortable with the basics, start to branch out.

Potentially the best place to start would be your 1NT opening structure.  The 1NT opener's hand shapes are fairly well constrained, so you don't need to work about some 5-5 pattern distorting the hand strength.  You also have a number of bidding situations where range asks are very important.  If Zar points improve your judgement in auctions like

1N - 2♣
2M - 3M

you'll see some real benefits...

I think it is safe to ignore the example auction Richard gives here...

For one thing, superfit is off the table, for another, opener tends to go to game with 4432 and tends to not with 4333. This is hardly a test for opener, especially with 15-17 hcp. With 4432 he will get 10 distrubtional points (8 with 4333), so he is at 23 to 27 ZARS before adding in his controls. The average 15=17 point hand has something like 4 to 5 controls (I have not made a study of that), so opener will have aound 27-32 ZAR points. A better test is if responder should bid 3M or 4M. The biggest decision maker here fpr opener is 4432 which gives him two more then the minimum. Bidding game with good shape and control rich is not exactly rocket science. I virtually never use ZAR on balanced hands. Responder with shape can use ZAR. Also not, most of ZAR points come into play with UNBALANCED hands anyway. (also not, opener does not get "fit" point for honors in his major and iwith 4432 does not get ruffing values for the doubleton, if opener has a five card major he would get a point for the doubleton).

For the record, I ran a quick BridgeBrowser simulation using the auction 1N-2C-2M-3M to see what people "do" at the table with 15 hcp. 51% of them bid game in the major (another 4% in NT). Of those bidding game on 15, less than 10% had 4333. Of those passing 3M, 27% with 15 hcp had 4333 distribution. So by almost 3 to 1, the shape factor (HCP being constant) seemed to drive the decision to bid on.

Best thing to do is actually quite different. Continue bidding and evaluating your normal way, and then look and see what "ZAR points" would have suggested. If you consistently do better than ZAR would have, ignore zar. If you find hands you "screw up on" would have worked well under ZAR evalutation, then you can decide to look deeper. And this is painless with myhands or bridgebrowser (homebase bridgebrowser is free), end evaluating the hands you played after the fact is a great way to improve your game, rather it is with ZAR points or any other metric... learning for our own mistakes is the best teacher.

[luke warm]

i don't use zar, not because i don't think it's worth using but because i can't seem to find the energy... it's hard to change methods unless the benefits are just too substantial to ignore, and i so far haven't seen that they are

so that means ben's last post makes more sense to me, just kib and evaluate the hands the way you normally would, then use zar, and see which way looks better

right now i simply use hcp and long suit points, then add up or down depending on bidding (working honors, etc) and distribution..

[hrothgar]

inquiry, on Jul 10 2006, 10:57 PM, said:

I think it is safe to ignore the example auction Richard gives here...

For one thing, superfit is off the table, for another, opener tends to go to game with 4432 and tends to not with 4333. This is hardly a test for opener, especially with 15-17 hcp. With 4432 he will get 10 distrubtional points (8 with 4333), so he is at 23 to 27 ZARS before adding in his controls. The average 15=17 point hand has something like 4 to 5 controls (I have not made a study of that), so opener will have aound 27-32 ZAR points. A better test is if responder should bid 3M or 4M. The biggest decision maker here fpr opener is 4432 which gives him two more then the minimum. Bidding game with good shape and control rich is not exactly rocket science. I virtually never use ZAR on balanced hands. Responder with shape can use ZAR. Also not, most of ZAR points come into play with UNBALANCED hands anyway. (also not, opener does not get "fit" point for honors in his major and iwith 4432 does not get ruffing values for the doubleton, if opener has a five card major he would get a point for the doubleton).

I conciously chose the auction

1N - 2♣
2M - 3M

for all the reasons that Ben notes. Of course, the big difference is that those areas that Ben condemns as bugs are, in fact, useful features.

I'm a firm believer in start small, establishing a firm foundation, and building from there. Accordingly, I like starting with auctions that ignore Zar-Super fits and all the other crap that will get dragged in later on.

Me, I'd want to be practicing a couple very basic points:

1. Mastering the mechanics and ensuring that all the calculations become second nature

2. Identifying "break points" How many Zar points are necessary for a major suit game. What hands are worth an initial Stayman type invite. What hands are worth accepting a game invite once you've identified a fit.

Of course, you always have the option to jump in willy-nilly, don't focus at all, try to absorb everything at once, and pray that what you think is is controlling the system actually is controlling the system.

[inquiry]

Well the problem with your auction, Richard, is opener in tha auction can not upgrad his hand at all. No points for fitting honors, no extra points for a doubleton, so again, no fit or misfit points. What kind of way is that to leann. Perhaps if you had ever used ZAR points you would know how CRAPPY a suggested auction you chose, so the only conclusion is you are clueless in their application.

Chose almost any other auction, you would have been on sounder ground... say,

1m-1H-2H-then invite.... Opener gets to evaluat a lot of stuff.. did responder make short suit game try? Help suit game try? etc. Even on 2NT as a relay, opener can evaluate his points in term of ZAR to decide how to continue... The action you gave (1N-2C-2M-3M) opener has no up-evalaution, he simple bids game or not on his starting evalaution. HArdly a test of ZAR or any other method. But certainly not a bug... in the method, you can look and see if you have a max or min... again, not rocket science.

[mikeh]

I confess that my comments should be read in the context that I have not tried to learn the use of zar points.

However, in my view, those who focus on these issues are diverting their time and energy away from learning to play the game. Zar may well represent a significant advance in numerical hand evaluation, but:

1) there are limits to the degree that any valuation system, playable by an unaugmented human, can develop

2) there are limits to the degree to which any bidding system can incorporate the values ascribed to the hands by the valuation system

3) there are limits to the mental energy available to any player, and the amount of energy available for 'being at the table' and the actual play of the cards will diminish as the energy required for valuing the hand in an arithmetical fashion increases

4) ditto the time element: if we are taking a long time to do the arithmetic, we will be pressed in the play in most situations... even long team matches have time limits and mp pairs has strict limits. Even if we internalize zar in the manner most have internalized 4321, every hand will take a little longer simply because there are more variables to compute.

My own experience has been that the better one becomes as a player, the less one relies upon any arithmetical evaluation.

I used to, years ago, look at adding points to my hand value due to shortness once we have a trump fit, or adding points because of extra cards in a suit, and so on. I doubt that I have done this for at least 15 years....

If asked to verbalize how I value a hand, sure it starts with adding the 4321 points, but that is a very small part of the process. In complex auctions, the idea of saying, for example, that AKxxx AJxx xxx x has become worth 15 points (or whatever) after partner's 2♥ response to my 1♠ opener would not occur to me at the table.

I would, rather, say... my hand looked ok to start with and has now become huge... I really like my hand and can see that slam is possible...and I'd either splinter or raise to 3♥, depending on the style my partner preferred re splinters.

It would never occur to me, or to any of my serious partners, to later analyze the auction on the basis of an arithmetical assessment of my hand or his. We'd talk about degree of fit, controls held or not held, possession or lack of working cards.. but not once would we say, for example, that his holding the Kx♠ (obviously a valuable holding) made his hand worth 'x' points more than the 4321 count would have it... we would say that the Kx was a powerful holding: a great working card.. and so on.

Holding Qxx Ax AKxxx Axx (a hand that came up on Saturday night) both players who held the hand opened 1N (15-17). I am sure that zar would say this hand was too good for that (as would adding a length point in 4321 or adding a point due to the extra controls), but neither I nor any good player needs zar to tell us that.

A hand with this many controls and a decent 5 card suit has to be upgraded and we do not need to use an adding machine to tell us that.

This holistic approach to valuation ( I hope I am not misusing the word holistic) takes longer to learn than an arithmetical approach but it is, I believe, far more rewarding.

That's why I pay little attention to formulae, such as the rule of 20 or FTL... I do use the LOTT on occasion, but not slavishly... and it is really a guide to competitive auctions, not to constructive game or slam decisions.

So while I recognize that the developer(s) of zar have put a lot of work into it, and what little I have seen suggests that it is indeed an improvement over other arithmetical formulae, I believe that it is a step in the WRONG direction in terms of a human learning to play bridge. It might well be a useful step for computers, but the programming of computers to play bridge is a topic about which I know nothing.

I recognize that I may be akin to those who predicted that home computers would never catch on

[Walddk]

I agree with all MikeH is saying and would like to add:

1. I am truly impressed that Ben is able to remember all this, along with all his special gadgets (I am sure they are fine).

2. Since many top class players have a memory as good as Ben's, it strikes me as odd that you don't see any (or at least very few) who actually use ZAR-points as their guidelines.

Why is that? Haven't they seen the light?

Roland

[inquiry]

Well for sure ZAR evaluation is not as good as true expert "holistic" approach. But while it takes years and years to become as skilled as an expert in evalauting plus and minus features of a hand, it takes very little time at all to learn a simple mathematical method. HAving said that, you must pay attention to the ZAR potential on in partners hand as well as yours. Zar inviented a new bidding system based totally on zar points. I wouldn't play that I don't think.

And I don't use ZAR on every hand, nor do I slavishly follow ZAR (nor LOTT). Hannie had a habit or asking me "how many ZAR points on the last one?" usually because I bid with too few or passed with too many "zar points", in part because I didn't count them and in part because even if I did, I choose not to follow them.

But the suggestion here is in response to a question about ZAR and then Richards how to "study" it. I will make a futher suggestion. If anyone agrees, I will use BridgeBrowser to pull up the first 25 boards they played (they have to have played the hand as declarer) since Jan 1, and we will see how ZAR would have done versus what they did for good or bad. All we need is a guinea pig, and it will provide 25 hands to see how the ZAR evaluation might have worked using exactly the rules laid out above (or we can use ZAR rules which are slighlty more liberal giving 2 points for QJ doubleton, I give one, and giving two points for two doubletons with a fit, I give only one).

[inquiry]

Walddk, on Jul 10 2006, 04:09 PM, said:

I agree with all MikeH is saying and would like to add:

1. I am truly impressed that Ben is able to remember all this, along with all his special gadgets (I am sure they are fine).

2. Since many top class players have a memory as good as Ben's, it strikes me as odd that you don't see any (or at least very few) who actually use ZAR-points as their guidelines.

Why is that? Haven't they seen the light?

Roland

Experts don't need ZAR... don't be silly. And how much harder is it to count 6 for an ace, 4 for a king, 2 for a queen and 1 for jack, plus the distibution calculation given rather than 3 points for a void, and 2 points for singleton and 1 point for doubleton?

And compared to most "experts" my bidding gadgets are few....

And remember the original poster said "I am more interested in how full Zar points are bid and to what level. I am not interested in yet another debate on whether they work or why they work or do not." If we want to discuss teh merits of ZAR, that probably should be for a different thread (and we have done that before, to great lengths). This thread really should stay on point of how to calculate them... as the original poster asked.. that is "how" not "why" (his words).

[mr1303]

A decent pair in the world junior pairs had a neatly filled in convention card involving a 33 ZAR + 1C opening, and 26-32 ZAR 1D-1NT openings, so maybe there is some support. But I guess most people just haven't heard of it.

[awm]

We'd all like to be better bridge players. The best way to accomplish this is to play a very large number of hands, preferably with a compatible partner against the best possible opposition. However, most of us don't have the time/money/opportunity to play tens of thousands of hands against world class opposition. So is there any other way to improve?

In terms of improving our declarer play, the answers are easy. There are plenty of books filled with "problem hands." Fred Gitelman provides excellent software (the bridgemaster series) to practice your play. You can even declare hands against a computer defending double-dummy.

Improving defense is harder, since it's so much a partnership thing. But there are still books of defensive problems, and you can still defend hands against a computer playing double-dummy and see how you do.

So how do you improve your bidding judgement? There are bidding "problem hands" of course, but these are often extremely esoteric (the bridge world polls) or depend a lot on partnership style (partner opens a weak three, do you bid game? depends on pd...) Computers are notoriously bad bidders so playing against them doesn't help much.

ZAR points are an attempt to formulate some rules that can help us become better bidders without resorting to playing thousands of hands. There have, of course, been other such attempts. A classic example is the 4-3-2-1 point count. Obviously the experts of the time didn't need it, but the bidding of relative beginners improved immensely with the introduction of this simple method. Assume for the moment we have good memories and are good at math, but haven't played all that many bridge hands (yes I know many of us are exactly the opposite of this...) Is there a more complex method we can use that will approximate "expert judgement" and allow us to obtain better results?

ZAR points are not a bad idea, and there are plenty of simulation to back them up. I think there are several questions we need to ask about them:

(1) Are they too complex to figure out at the table? Obviously this will depend on the person.

(2) How closely do they approximate expert judgement? Are they accurate?

(3) How do they interact with the rest of the bidding system?

It's point (3) where I think there are really some major issues. The problem is that ZAR seems to sometimes mandate that you subtract points from your hand after discovering a misfit. Obviously this means that a hand which was once worth bidding game might be no longer so. If you're going to play this way, you need to have methods that allow you to "back off" with a formerly promising hand once this is discovered. Modern 2/1 GF methods do not often allow this. Curiously, old fashioned "one-round forcing" methods do.

[P_Marlowe]

Hi,

I want to second awn post:
If you use ZAR points, you need to
adapt your complete system.

An example would be forcing pass
agreements, because ZAR favors
distributionand light openers, you will
get stuck, if you use forcing pass
agreements passed on the old HCP count.

With kind regards
Marlowe

[hrothgar]

inquiry, on Jul 10 2006, 11:39 PM, said:

Well the problem with your auction, Richard, is opener in tha auction can not upgrad  his hand at all. No points for fitting honors, no extra points for a doubleton, so again, no fit or misfit points. What kind of way is that to leann. Perhaps if  you had ever used ZAR points you would know how CRAPPY a suggested auction you chose, so the only conclusion is you are clueless in their application.

As I noted before, I think that there is some real value to starting with the simple and working towards the complex. It might not be interesting or exciting, but it provides a strong foundation for further study.

I was hoping that it might be possible to actually have a reasonably civil discussion about Zar points for a change...

(edited by rain)

[Walddk]

hrothgar, on Jul 10 2006, 11:50 PM, said:

I was hoping that it might be possible to actually have a reasonably civil discussion about Zar points for a change...   

Why must this develop into a personal attack? I simply don't understand. Richard, you used "civil" yourself, so why don't you keep it "civil"?

Roland

[hotShot]

ZarPointsBook_page_10 said:

NOTE, that counting Zar Points with a balanced hand will NOT help you – with these 15 HCP you collect
only 25 Zar Points which “formally” means you should pass.
Zar Points are geared towards aggressive bidding with distributional power rather than hands with brute
HCP force and balanced hands – every pair would bid and make 3 NT on this first board with a natural and
simple sequence of 1NT – 3NT (not even a Stayman used :-).

So any sequence starting with NT is not a good base to evaluate ZAR-Points.

There is at least one local organisation that order the use of 4321 HCP for filling the convention card (it's in the national condition of contest). Since it is not possible to give reasonable HCP descriptions of ZAR-Points, you can't use them there.

[hrothgar]

Walddk, on Jul 11 2006, 01:12 AM, said:

hrothgar, on Jul 10 2006, 11:50 PM, said:

I was hoping that it might be possible to actually have a reasonably civil discussion about Zar points for a change...   
Pity that you're acting like a defensive little piece of *****.

Why must this develop into a personal attack? I simply don't understand. Richard, you used "civil" yourself, so why don't you keep it "civil"?

Roland

I could certainly have allowed the conversation to progress along its predetermined path. Ben and I would direct more and more insulting comments at one another until the whole thread collapsed into ugliness.

Sadly, I have a lot of stuff on my plate right now and really don't have as much time as I'd like. I decided to skip over the long slow dance of insinuation and jump directly to the insults...

[Walddk]

hrothgar, on Jul 11 2006, 12:46 AM, said:

I decided to skip over the long slow dance of insinuation and jump directly to the insults...

Heaven knows that Ben and I agree on very little bridge related stuff, but we don't call each other names. Feel free to disagree, but there is no need to be rude and insult fellow members deliberately!

Roland

[inquiry]

Hotshot and I are both correct.. ZAR's are not needed by 1NT opener. Oh you can use it, but anyone knows a "maximum" 1NT opener when invited versus a "minimum" 1NT opener when invited. For someone to argue that using 1NT opener hand as the metric to study ZAR evalution is a horrible and totally useless waste of time.

Peope will look at their 1NT and say "hey, thihs is a quacky hand" (lots of Queens and jacks) or this is dead distribution (4333) and soft pedal. or they will say, I got quick tricks glore and good shape (4432 or wonderful 5332 or very rarely 5422 or 1444). This really is not the place to study ZAR. Seriously. It is a VERY CRAPPY idea. I wonder where crappy ideas come from? Who knows, oh I have an idea.

But here is what happened in the thread.. Richard suggest a "way to study ZAR". He doensn't use ZAR, I do. I explained why his was a horrible idea, and offered the two better ways. That is as a Zar user, I offered a counter suggestion to the broke one Richard suggested.

Richard countered attacked, with ignore superfits and the like.. do it his way (and in fact ignore ALL THE RE-EVALUTON as opener of 1NT will never RE-EVALUATE on the given auction). [b]AND PLEASE REMEMBER, this is exactly what the original poster asked for, not how well lthey worked, not where the break points were, or anyohter issue Richard raised... he asked how

Quote

"many times he Counts Full Zar adjusted points and then downgrades them or upgrades them which make them seem even more mysterious.

I understand opening, fit or nonfit, opener or responder, sac etc may all count them differently, I am only asking how not the why,

For pointing out the WHY RICHARD SUGGESTION is the worse possible one for this suggestion (yet again, above), I get called a "*****". Lucky no board adminstrator is reading your post Richard. BUT I WILL REPEAT FOR A LAST TIME, the auction you gave is CRAPPY ONE for ZAR.. I will spell it out in detail why.

1) No short suit upgrade for your own suit unless you have unusaual legnth. So operen with a 4 card major gets no upgrade for a side suit doubleton (unless he has a five card major). Responder DOES get upgrades for fit with opener
2) No upgrade for honors in your own suit. But here, RESPONDER would get upgrades for spade honors.
3) No superfit is deteminable, so forget Superfit/Misfit points.

In other words, your example RICHARD IS ALWAYS GOING TO FAIL 1OO% TO ANSWER WHAT THE POST ASKED FOR.

(edited by rain)
Ben