[inquiry]

Mike has been pushing ZAR points for sometime, and I ignored them http://public.aci.on...eNeverMiss.html. Misho, however pointed me to them again earlier this week, and I have been looking though problem bidding hands in my past, as well as ones that crop up in the bbo... and well, there maybe something to it. And yes, I realize that hand evaluation is more than counting points, and that like any tool, ZAR points can be overly heavily relied upon. Here are back to back hands from a team match today involving stars versus yellows with lots of kibitizers....were the wrong decision was reached each time, i wonder if ZAR's were used if this would have been the case?

Scoring: BAM

East ZAR points... 32. West ZAR points start at 19. This totals 51... accord to the ZAR point page, 52 needed for game. However, when EAST bids 3♥ as a game try, WEST gets to add quite a few points to his hand according to the ZAR page. Two points for the heart honors, and two more points for the fifth ♥ and the singleton... for a total now not of 19, but of 23. This is more than enough to accept the game try. Bidding (1♦)-1♠-(2♦)-2♠; (P)-3♥-all pass

---

Scoring: BAM

East ZAR points... 32. West ZAR points start at 19. This totals 51... accord to the ZAR point page, 52 needed for game. However, when EAST bids 3♥ as a game try, WEST gets to add quite a few points to his hand according to the ZAR page. Two points for the heart honors, and two more points for the fifth ♥ and the singleton... for a total now not of 19, but of 23. This is more than enough to accept the game try. Bidding (1♦)-1♠-(2♦)-2♠; (P)-3♥-all pass

---

Scoring: BAM

East ZAR points... 32. West ZAR points start at 19. This totals 51... accord to the ZAR point page, 52 needed for game. However, when EAST bids 3♥ as a game try, WEST gets to add quite a few points to his hand according to the ZAR page. Two points for the heart honors, and two more points for the fifth ♥ and the singleton... for a total now not of 19, but of 23. This is more than enough to accept the game try. Bidding (1♦)-1♠-(2♦)-2♠; (P)-3♥-all pass

---

On this last hand, both teams rested peacefully in 4♠, due in part to the opening 2♠ rather than 1♠. However, even opposite a 2♠ bid, which should be 22-24 or so ZAR points, the ZAR total rapidly reaches the min 62 (you can count extra points for the singleton now as you know you have a ten card instead of nine card fit). But even 39 in west opposite 23 in EAST is enough for slam.

[hrothgar]

I had the chance to look over Zar's article. while I found the article interesting, however, as of yet I am still unable to make an informed decision regarding its technical merit.

From my perspective, the true “value” of any system of hand evaluation is the ability to create boundary conditions between different types of events. Ideally, I would want to see hand evaluation systems “judged” using the following type of approach:

Use a hand generator to create 1 million hands.

Use a double dummy engine to identify all hands that “make” precisely “X” tricks in a major suit contract. [10 tricks in a major suit contract or 9 tricks in NT are likely to be the most interest points of comparison]

Sum the Zar points for each hand. Plot a frequency distribution table, along with the mean and standard deviation.

Perform the same calculation for Goren Points, Bergen points, the Kaplan Rubens hand generator, or whatever other system that you want to compare.

Arguably, the system with the lowest standard deviation is the most accurate judge of trick taking potential.

[inquiry]

hrothgar, on Mar 17 2004, 12:28 PM, said:

I had the chance to look over Zar's article. while I found the article interesting, however, as of yet I am still unable to make an informed decision regarding its technical merit.

From my perspective, the true “value” of any system of hand evaluation is the ability to create boundary conditions between different types of events. Ideally, I would want to see hand evaluation systems “judged” using the following type of approach:

Use a hand generator to create 1 million hands.

Use a double dummy engine to identify all hands that “make” precisely “X” tricks in a major suit contract. [10 tricks in a major suit contract or 9 tricks in NT are likely to be the most interest points of comparison]

Sum the Zar points for each hand. Plot a frequency distribution table, along with the mean and standard deviation.

Perform the same calculation for Goren Points, Bergen points, the Kaplan Rubens hand generator, or whatever other system that you want to compare.

Arguably, the system with the lowest standard deviation is the most accurate judge of trick taking potential.

Hi Ron, like you, I found the article intersting, and like you I am uncertain how much I like the evaluation system and rather or not I think it "works".

Let me address some of your issues...

1) Technical merit.

Well this is a huge area, and sort of what I was driving at with the question of total waste of time or worth the effort. Let me put the first ZAR point question into persepective...is a 26 point ZAR hand with 7 or 8 hcp worth an opening bid? Here I think from my study, the answer is an unqualified yes. I like opening light (by hcp standards), I have mentioned that many times well before I ever heard of ZAR points, and I have altered my bidding system to allow for these light opening bids. Going back over my light opening bids, the vast majority of hands where I did open light would also qualify under the ZAR point schedule. The ones that don't are some balanced 11/12 point hands that I use to like to open 1NT, but now, playing 14-16 NT, I can open and rebid 1NT without too much extra risk.

Bidding games. So far the vast majority of hands with 52 ZAR points and a fit have at least a play for game (needing a finessee or a friendly split). Similar for 62 points and slam. The first problem for slam, sort of glossed over in the article, is the need to pay attention to controls... you can get some very high ZAR values and yet be off two aces or a side suit AK. So it is not like bidding a 35 hcp small slam in NT.

2) Judging a million hands.. .way overkill, especially for me. The vast majority of hands I do not even count my hcp during the bidding. I just decide if my hand is good, bad or average for the bidding. On defense, I do count mine and dummy and try to guess about what is in the other hand. But the point is, I doubt I am going to become a zealot at counting ZAR points either. But who knows, if it works as well in real life as it should have in the hands in this thread, maybe counting them is not a bad idea. I will let you know that misho who pointed at the website (but maybe without interest in ZAR specifically), is not a believer in ZAR points. One thing for sure... if bridge devolves into accurate bidding with just "counting" some point scale that would be a sad day indeed. Much better to use your own judgement, enhanced perhaps, by your favorite evaluation tool...but table feel, opponent and partner's tendencies all should play a role in the decision process.

Ben

[luke warm]

ben, i believe you said that you looked at some of your 'light' openings of the past and saw that they did indeed conform to zar points... now if those openings were based upon your judgement, and if your judgement and zar point evaluation coincide, isn't that a good testimony for the efficacy of zar points (or of your judgement, or both)?

misho prefers to rely on his own method of evaluation, coupled with his experience and judgement... maybe he's checked, i don't know, but i'd like to see how his evaluation/judgement conform to zar points... i'd almost be willing to bet that it comes out pretty much the same as yours

[JonnyB]

Talking about hand evaluation......Iread a book a couple of years ago, by Laurence Drabble. Unfortunately, cant remember the title, but it dealt with a different way to evaluate combined hand strength - "Complete Hand Evaluation" he called it.
Has anyone heard of this..it seems to be totaly unknown here in NZ, but is possibly the most consistently accurate way to judge hand stength, that I have ever come across.
The second half of the book dealt with his bidding system.. "MidMac" if I recall. Well worth a read, if you can find a copy!

[mikestar]

The Zarpoints article briefly describes the Drabble method for comparison.

[mikestar]

I've been doing some study and experimentation with Zarpoints and the idea seems to have fairly considerable merit.

I don't think I will adopt the method, more likely I will use its main features as adjustments to my evaluation technique, giving greater weight to controls and extreme shape.

[cwiggins]

inquiry, on Mar 17 2004, 05:59 PM, said:

[Let me put the first ZAR point question into persepective...is a 26 point ZAR hand with 7 or 8 hcp worth an opening bid? Here I think from my study, the answer is an unqualified yes. I like opening light (by hcp standards), I have mentioned that many times well before I ever heard of ZAR points, and I have altered my bidding system to allow for these light opening bids.

1) What bidding system did you start with?
2) What alterations did you make?

For example, if you are opening 8 HCP with 5-5, then 17-18 balanced is just enough to force to game.

Thanks.

[Free]

You don't need 25HCP all the time to make games my friend, my record is 4♠ with 12HCP combined (doubled and made ), but there are even slams with 11HCP which were made in the past... Unbalanced hands have more potential than balanced hands with more HCP, that's proven i think.

[EricK]

JonnyB, on Mar 18 2004, 08:08 AM, said:

Talking about hand evaluation......Iread a book a couple of years ago, by Laurence Drabble. Unfortunately, cant remember the title, but it dealt with a different way to evaluate combined hand strength - "Complete Hand Evaluation" he called it.
Has anyone heard of this..it seems to be totaly unknown here in NZ, but is possibly the most consistently accurate way to judge hand stength, that I have ever come across.
The second half of the book dealt with his bidding system.. "MidMac" if I recall. Well worth a read, if you can find a copy!

"A New Approach to Bidding" by Jon Drabble.

His method of hand evaluation is surprisingly accurate. it is slightly less aggressive than Zar points, but not much!

One of the differences is that Zar points explicitly add points for controls (2 for an Ace, 1 for a King) whereas Drabble's CV compares the controls actually held with the expected number (CV/3) and upgrades or downgrades the hand accordingly

CV is explained briefly at

http://www.maths.uwa...itzpatr/cv.html

Midmac is quite a nice bidding system, but I think the minor openings are rather too open to pre-emption

Eric

[Trpltrbl]

Free, on Mar 19 2004, 09:46 AM, said:

You don't need 25HCP all the time to make games my friend, my record is 4♠ with 12HCP combined (doubled and made ), but there are even slams with 11HCP which were made in the past... Unbalanced hands have more potential than balanced hands with more HCP, that's proven i think.

I can even make a grandslam with 10 HCP hehehe

Mike

[Trpltrbl]

Zar system is easy to explain, you can play it with any natural system you want. Maybe even with the relays systems, but not sure, and to be honest not even interested in it.
The whole idea is that long suits and Quick tricks play a lot better then having a hand that is balanced with with Queens and Jacks. I like it and it seems to work for me. But I am sure that there are other out there that find something else better or are more comfortable with. As long as I keep winning I don't care hehehe

Mike

[JonnyB]

where can I find a copy of this "Zar Points"? The hands given as an example, at the start of this topic.....they could have been bid correctly, using the losing trick count, and total trumps surely. I would be interested to see how this system deals with mirror hands.

[inquiry]

JonnyB, on Mar 20 2004, 01:59 AM, said:

where can I find a copy of this "Zar Points"? The hands given as an example, at the start of this topic.....they could have been bid correctly, using the losing trick count, and total trumps surely. I would be interested to see how this system deals with mirror hands.

http://public.aci.on...eNeverMiss.html

[tysen2k]

Waste of energy. BUM-RAP (A=4.5, K=3, Q=1.5, J=0.75, T=0.25) plus a simple 321 distribution scheme is more accurate. Even dropping the T completely is still better. Zar's method of distribution counting just doesn't reflect real trick-taking ability accurately.

I wrote an article about this a few weeks ago on rgb comparing different evaluation schemes:

http://tinyurl.com/32tgc


For those intereseted in hand evaluation, that article also references two other articles I wrote on the subject:

[Improving Hand Evaluation Part 1]
http://tinyurl.com/25huc

[Improving Hand Evaluation Part 2]
http://tinyurl.com/383e6

Tysen

[luke warm]

tysen2k, on Mar 26 2004, 01:23 AM, said:

Waste of energy. BUM-RAP (A=4.5, K=3, Q=1.5, J=0.75, T=0.25) plus a simple 321 distribution scheme is more accurate. Even dropping the T completely is still better. Zar's method of distribution counting just doesn't reflect real trick-taking ability accurately.

I wrote an article about this a few weeks ago on rgb comparing different evaluation schemes:

http://tinyurl.com/32tgc


For those intereseted in hand evaluation, that article also references two other articles I wrote on the subject:

[Improving Hand Evaluation Part 1]
http://tinyurl.com/25huc

[Improving Hand Evaluation Part 2]
http://tinyurl.com/383e6

Tysen

you think you could post a few random hands for us? with examples of this 'counting'?

[1eyedjack]

I vaguely remember reading a book about 20 years ago called Cobra by E T Lindeloef. Didn't think much of the system. And the book cost a small fortune, but it had some interesting points about hand evaluation. Never hear about it any more. Anyone else with any experience of this to compare with modern evaluation methods?

[Trpltrbl]

Everybody has there own way of evaluating. I like ZAR points and it seems to work nicely for me, and it is simple. I am not trying to push it on anybody, I bet ya most of you are old enough to make your own choices, I made mine...

Mike

[tysen2k]

luke warm, on Mar 25 2004, 06:51 PM, said:

you think you could post a few random hands for us? with examples of this 'counting'?

Since the HCP portion of Zar is essentially equal to BUM-RAP x1.5, the hands that Zar goes off from reality depend only on distribution and not high-card strength. Zar points undervalue some distributions and overvalue others. When you do a statistical analysis of all hands, you can pick out the distributions that mess up the most. When you weight them by frequency, the biggest errors are:

5332 – Zar gives this 11 distribution points, or 3 more than a 4333 pattern. Therefore Zar says it should take 0.6 tricks more than 4333, when in reality it only takes 0.339 more tricks.

6322 – Similar story. Zar says it’s 1 trick better than 4333 when it’s only 0.660 tricks better.

5422 – Zar says 0.8 tricks when it’s 0.595

4441 – This time Zar undervalues the holding. Zar says it’s 0.6 tricks better when it’s really 0.810.

There are many other errors all over the place and many have bigger magnitude, but their frequency doesn’t put them at the top of the list.

[inquiry]

Soap box time.... but first, let me say that I am not an advocate of any particular hand evaluation style, i make my decisions mostly on, for lack of a better phrase, past experience.

Ok.. standing firmly on the soap box. Goren's point count method (4321) took off and became the defacto standard for decades not because of its accuracy, but because of its simplicity. Any one can count on their fingers and toes up to however many points. And it was easy... 13 points to open.. two 13 point hands = game... This was easy to count, easy to understand.

Alright there are problems with Goren point count. But somehow, adding 0.25 points for a Ten and 0.75 points for a J, and so forth is hard on the cognative powers of those bridge players among us who are feeble minded, like me. I frequently never count my points at all, much less try to add decimal levels. Then to use decimal for distributions too... gack... forget it.

ZAR shares many of the simplicity of Goren style. Add hcp his way, add 2 more for each ACE, and one for each king (control points), do the distribution (again in whole numbers)... opening bid =26, game needs 52, each level is roughly 5 points. Say what you will, this is simple.

I read tysen2k's articles on hand evalaution. I read about Blinky points where you basically have to use a calculator to figure them out, I read about Bum Rap points and Bum Rap + 321, I read about Bergen (already knew that one). All in all, I am not convinced any of these are all that good... but they are all better than just counting HCP (how many times have we heard..."but partner, I had 17 hcp???" , or "but partner, I only had 10 hcp?, when overbidding or underbidding based upon what can make. But if I was going to use one of these, I would us ZAR simply because I can add whole numbers a lot easier in my head and with less stress than fraction of numbers...

Ben