[shevek]

Perusing Ron Klinger's convention card for the current Australian playoffs, I saw

"Use 5-4-3-2-1 count for opener's balanced hands. 1NT = 22-25, 1x : 1y : 1NT = 17-21. etc"

Does anyone or any organisation feel uneasy about this 60 point deck?

[nigel_k]

IMO they should describe using the 4-3-2-1 count regardless of how they actually evaluate.

Many people cannot multiply by two thirds quickly enough to be able to understand what these descriptions actually mean if they just quickly browse the card.

[shevek]

nigel_k, on 2012-March-29, 20:13, said:

IMO they should describe using the 4-3-2-1 count regardless of how they actually evaluate.

Many people cannot multiply by two thirds quickly enough to be able to understand what these descriptions actually mean if they just quickly browse the card.

They also list 1NT as (14)15-17(18)

[mgoetze]

shevek, on 2012-March-29, 19:47, said:

Does anyone or any organisation feel uneasy about this 60 point deck?

Personally I am very much in favor of full disclosure.

[shevek]

The point being that

♠Axxx ♥Axxx ♦Ax ♣Axx

may be considered not strong enough for 1NT (5 x 4 = 20)
while

♠AKJT ♥QJT ♦JTx ♣T9x

could be fine. (12+6+3+1 = 22)

Now I doubt whether they would follow their evaluation method on these extreme hands but say they do.
What then?

Not so easy for defenders construct likely hands for declarer based on evaluation that is alien to them.

[mgoetze]

shevek, on 2012-March-30, 01:33, said:

Now I doubt whether they would follow their evaluation method on these extreme hands but say they do.
What then?

Not so easy for defenders construct likely hands for declarer based on evaluation that is alien to them.

So are you suggesting that they should lie about their methods instead? Or that the methods should be banned? I'm sorry but frankly I can't agree with either of these sentiments and I don't see what else you might be suggesting.

[gnasher]

I dislike the idea that you should be obliged to use a particular evaluation method in describing your system.

Kilnger and his partner seem to have done a good job of disclosure: they have described their actual method for those opponents who want complete disclosure, and they have provided an approximation for anyone who only understands Milton-Work. Perhaps they should have made it clearer that "(14)15-17(18) " is very approximate.

shevek, on 2012-March-30, 01:33, said:

♠Axxx ♥Axxx ♦Ax ♣Axx

may be considered not strong enough for 1NT (5 x 4 = 20)
while

♠AKJT ♥QJT ♦JTx ♣T9x

could be fine. (12+6+3+1 = 22)

Now I doubt whether they would follow their evaluation method on these extreme hands but say they do.

It appears that their point-count intentionally treats an ace as being worth much less than in the Milton-Work count, and intentionally treats minor honours as worth more. Whilst you might think this is barking mad, I don't see why you should assume that they don't actually use it. It would be odd to agree to play the 5-4-3-2-1 count, then make adjustments so as to turn it back into something approximating the 4-3-2-1 count.

Quote

What then?

Not so easy for defenders construct likely hands for declarer based on evaluation that is alien to them.

You could simply ask whether they ever upgrade or downgrade, and if so how - just as you can when playing against users of the 4-3-2-1 count.

[pran]

gnasher, on 2012-March-30, 04:44, said:

[...]
It appears that their point-count intentionally treats an ace as being worth much less than in the Milton-Work count, and intentionally treats minor honours as worth more. Whilst you might think this is barking mad, I don't see why you should assume that they don't actually use it. It would be odd to agree to play the 5-4-3-2-1 count, then make adjustments so as to turn it back into something approximating the 4-3-2-1 count.
[...]

I have a reference dating back to 1951 stating that careful mathematic calculations has led to the relative ratio between honours for a total of 10 points in a suit being:
Ace = 3,85 (instead of Milton-Works 4)
King = 2,95 (instead of Milton-Works 3)
Queen = 2,05 (instead of Milton-Works 2)
Jack = 1,15 (instead of Milton-Works 1)

For a total of 14 points in a suit (excluding the ten) the same calculation should give something like:

Ace = 5,4 (instead of 5)
King = 4,1 (instead of 4)
Queen = 2,9 (instead of 3)
Jack = 1,6 (instead of 2)

So Milton-Work apparently overrates the Ace and King while the 5-4-3-2-1 scale overrates the Queen and Jack. This over/underrate is most significant on the Ace and Jack and is greater with the 5-4-3-2-1 scale than with Milton-Work.

I believe this is a good justification for WBF (and others) to specify Milton-Work for permissible strength-ranges in their regulations?

[sfi]

shevek, on 2012-March-29, 19:47, said:

Perusing Ron Klinger's convention card for the current Australian playoffs, I saw

"Use 5-4-3-2-1 count for opener's balanced hands. 1NT = 22-25, 1x : 1y : 1NT = 17-21. etc"

Does anyone or any organisation feel uneasy about this 60 point deck?

I've never had any problems with this when playing against them. It's easy enough to work through when it becomes an issue and I ignore it the rest of the time. They are very ethical about it, providing details both before the round and when asked, and translate it into 4-3-2-1 count equivalents with caveats.

Whether it works or not is a different issue, but they seem happy enough with it. And I don't think you can or should be able to legislate hand evaluation at this level of detail.

[gnasher]

pran, on 2012-March-30, 06:03, said:

I have a reference dating back to 1951 stating that careful mathematic calculations has led to the relative ratio between honours for a total of 10 points in a suit being:

There have been a number of studies that attempt to do this, based on double-dummy analysis, analysis of single-dummy results, evaluation of specific combinations of high cards, inspection of tea-leaves, and who knows what else. They all produced different answers.

Quote

So Milton-Work apparently overrates the Ace and King while the 5-4-3-2-1 scale overrates the Queen and Jack. This over/underrate is most significant on the Ace and Jack and is greater with the 5-4-3-2-1 scale than with Milton-Work.

I believe this is a good justification for WBF (and others) to specify Milton-Work for permissible strength-ranges in their regulations?

That's a separate question. I can understand that a regulator doesn't want to specify its rules using half a dozen different metrics. We don't want rules on the lines of "You may not agree to open 1♠ with less than 10 4-3-2-1 points, 15 5-4-3-2-1 points, 7 losers, 2.5 honour tricks, or 11.2 Binkies."

What we're talking about here is a pair who define their agreements using their own point-count system, and therefore explain them using the same method.

(But, since you mention it, I don't see how you reached this conclusion. All you have told us is that you know of two different methods of evaluation, and both are inaccurate. That doesn't seem a reason to choose one over the other. A better reason to use 4-3-2-1 points in writing regulations is that everybody understands them.)

[RMB1]

gnasher, on 2012-March-30, 07:27, said:

A better reason to use 4-3-2-1 points in writing regulations is that everybody understands them.

Indeed, I think this is the only reason.

[blackshoe]

pran, on 2012-March-30, 06:03, said:

I believe this is a good justification for WBF (and others) to specify Milton-Work for permissible strength-ranges in their regulations?

gnasher, on 2012-March-30, 07:27, said:

(But, since you mention it, I don't see how you reached this conclusion. All you have told us is that you know of two different methods of evaluation, and both are inaccurate. That doesn't seem a reason to choose one over the other. A better reason to use 4-3-2-1 points in writing regulations is that everybody understands them.)

RMB1, on 2012-March-30, 07:44, said:

Indeed, I think this is the only reason.

Afraid I have to agree with Andy and Robin on this one. Even the notorious ACBL has not gone quite so far — you are permitted to use your own evaluation methods, and to specify the range of hand types in terms of your methods, but you must be able to describe that range in terms of Work count when asked. Which is precisely what Klinger and company do.

[pran]

gnasher, on 2012-March-30, 07:27, said:

[...]
(But, since you mention it, I don't see how you reached this conclusion. All you have told us is that you know of two different methods of evaluation, and both are inaccurate. That doesn't seem a reason to choose one over the other. A better reason to use 4-3-2-1 points in writing regulations is that everybody understands them.)

Oh, I did give the comparison between each method and the quoted weights, didn't you notice?
And the agreement with the quoted weights is evidently far better with 4-3-2-1 than with 5-4-3-2-(1).

(I know of at least two more scales that have been used: 3-2-1 (for Ace - King - Queen) during the early thirties, and of course the Vienna system during the forties with its 7-5-3-1 scale.)

I don't think it is accidental that the 4-3-2-1 scale prevails, it is older than Culbertson's Honour trick scale and there is a strong suspicion that Culbertson introduced his honour trick scale primarily as a means for competition with Milton Works. The return to and survival of 4-3-2-1 (rather than any of the other scales) is probably the best evidence that this is generally considered the "best" scale for comparing strengths.

[blackshoe]

"Everybody knows" that the Work scale is the "best". OTOH, "If 'everybody knows' such-and-such, then it ain't so" (Robert A. Heinlein)

[RMB1]

I think it is inevitable that regulatory authorities will use Miltons to define permitted methods: minimum agreements for 1 bid, strong bids, etc.

But if players are using different evaluation methods in deciding what to bid it will not achieve full disclosure if these players are instructed to describe their agreements solely in terms of Miltons. Regulatory authorities need to require disclosure in a way which is sufficiently accurate and sufficiently understandable to the opponents.

[Cthulhu D]

pran, on 2012-March-30, 15:57, said:

(I know of at least two more scales that have been used: 3-2-1 (for Ace - King - Queen) during the early thirties, and of course the Vienna system during the forties with its 7-5-3-1 scale.)

People use the 3-2-1 scale (calling it Queen Points for reasons I don't understand), when playing MOSCITO relays around here.

This is one area where the ABF regulations are poorly written. The complete list of system categories are 'Green, Blue, Red and Yellow' Green, Blue and Red define the minimum for an opening hand as:

All one-level opening bids, must by agreement, promise at least 8 HCP.

but yellow sticker (HUMs) are defined as:

An opening bid at the one level that may be made on high card strength a king or more below that of an average hand (i.e., ~ 0-7 HCP and insufficient compensating distributional values).

So what about if our agreement promises 8 HCP or compensating distributional values? I have no idea.

[mgoetze]

Cthulhu D, on 2012-April-01, 22:47, said:

People use the 3-2-1 scale (calling it Queen Points for reasons I don't understand), when playing MOSCITO relays around here.

This is one area where the ABF regulations are poorly written. The complete list of system categories are 'Green, Blue, Red and Yellow' Green, Blue and Red define the minimum for an opening hand as:

Sounds like a poor copy of the WBF Systems Policy, which define what a Yellow system is, and then say that anything which is not a yellow system is Green, Blue or Red.

In Germany by the way a clear dividing line has been drawn at the rule of 18. Some think this is a bit high and the line should be lower (especially considering 3rd seat openings).