[Vampyr]

barmar, on 2015-May-15, 09:24, said:

Is there a missing "not" somewhere in the first sentence? Either before "make" (instead of "it") or after "does".

Yeah, instead of "it". I always have to be careful with my iPad because it makes up words.

[Trinidad]

pran, on 2015-May-15, 04:19, said:

I cannot believe that you are serious!

Maybe you need to know the results at the other tables in order to calculate respectively 40%, 50% and 60% of a top score on the Board - I do not.

Ok, NS make 4♥, vul with an overtrick. How many percent do they get for that?

Rik

[gordontd]

Trinidad, on 2015-May-15, 14:47, said:

Ok, NS make 4♥, vul with an overtrick. How many percent do they get for that?

Rik

They don't get an artificial adjusted score when an actual score was obtained at the table. Artificial adjusted scores don't depend on what the other tables did; they depend on the degree of fault of the pairs at the table.

[gordontd]

Trinidad, on 2015-May-15, 02:06, said:

Have you ever tried to calculate a percentage without using scores from other tables?

Sure. I've never used the scores from another table when awarding an artificial adjusted score. All we need to know to calculate a percentage is the number of results per board.

[Trinidad]

The point behind artificial adjusted scores and non artificial adjusted scores:

Before there is any significant action on the board, all scores are possible (or more correctly, just as likely as at any other table). NS might obtain 100% or 33% or 24%. If some irregularity occurs that makes any further play impossible, it is perfectly fine to come up with an artificial adjusted score.

If there has been significant action on the board, e.g. because the auction is over and a contract has been reached, the number of possible table results is limited. After the auction is over, there are at most 14 possible outcomes (and usually much less realistic outcomes). It is entirely possible that NS will never be able to score 60%, or even 30%, e.g. because they missed an obvious game.

The object of the laws is to restore equity after an irregularity: We try to assign a score that would have been obtained if the irregularity had not occurred. That means that you don't give NS 60% if without the irregularity they would have been able to achieve a score between 0 and 30%. Similarly, you don't give EW 60% if they were headed for a score between 70 and 100%. That is not restoring equity.

That is why you should not give artificial adjusted scores when there has been significant action on the board and the number of possible results are limited. In the case of this thread, there are only two results that are realistically possible: 2♠c and 2♠-1. Then you should base your AS on those two possible outcomes and not give an AAS of 60-60 which has nothing to do with restoring equity. 60% might correspond to a score of 4♠ E -3 or 1NT S c. Is that the AS you want to give?

Rik

[Trinidad]

You are a chemistry teacher and you have three students. It is your job to test them and give them a grade for their report card at the end of the term. You have come up with 8 tests that they need to take.

Case 1
Unfortunately, the first one is a lab test and one of the three students blunders and blows up the building. You cannot give any tests anymore. What grades are you going to hand out?

Case 2
Student A has aced the first 7 tests with flying colors. He is headed for a straight A, as long as he doesn't blow up anything. Student B has had major trouble getting sufficient grades, he might pass or he might not. Student C is hopeless. He has no chance whatsoever. The last test is a lab test and student C (who else) blows up the building. What grades are you going to hand out?

Answers:
Case 1: the student who blew up the building won't pass. The other two get the benefit of the doubt and will get a passing grade: a B.

Case 2: Student C will not pass. Student B gets the benefit of the doubt and gets a B. Student A also gets the benefit of the doubt and will get an A. Except if the chemistry teacher is Pran, then student A and B both will get a B, because the possible outcomes of the final grade are "numerous".

Rik

[gordontd]

Rik,

I think you earlier confused "when to give an artificial adjusted score" with "how to give an artificial adjusted score".

[Trinidad]

gordontd, on 2015-May-17, 03:17, said:

Rik,

I think you earlier confused "when to give an artificial adjusted score" with "how to give an artificial adjusted score".

Ok, suppose in this case that there would be 5 different possible outcomes:
2♠-2 (+100)
2♠-1 (+50)
2♠= (-110)
2♠+1 (-140)
2♠+2 (-170)

Then I don't have any problem with a TD who:

• looks at the score sheet
  
• sees that the possible results for NS vary from 10 (2♠+2, -170) to 50% (2♠-2, +100) and for EW from 50 to 90%
  
• decides to give NS 40% and EW 80%, giving them both some benefit of the doubt (since "on average" they would get 30% and 70%).

I am certainly not advocating that the TD starts to calculate (e.g.):
   MP for +100 x 1/16
   MP for +50 x 4/16
   MP for -110 x 6/16
   MP for -140 x 4/16
   MP for -170 x 1/16
   Total: MP for AS

So, I am not against an artificial AS. I am against the automatic 60%-60% since these might be results that would have been impossible without the irregularity.

Rik

[blackshoe]

Trinidad, on 2015-May-17, 05:26, said:

I am certainly not advocating that the TD starts to calculate (e.g.):
   MP for +100 x 1/16
   MP for +50 x 4/16
   MP for -110 x 6/16
   MP for -140 x 4/16
   MP for -170 x 1/16
   Total: MP for AS

Isn't this exactly how one would calculate a weighted assigned adjusted score?

Trinidad, on 2015-May-17, 05:46, said:

Deleted. (I am starting to suffer from double posting syndrome.)

I deleted two of these. Generally, I will delete any post consisting of the word "delete[d]" simply to avoid cluttering up the thread. If the poster makes additional comments, I have to decide whether to keep that post. Usually, I won't.

[Trinidad]

blackshoe, on 2015-May-17, 09:13, said:

Trinidad, on 2015-May-17, 05:26, said:

I am certainly not advocating that the TD starts to calculate (e.g.):
   MP for +100 x 1/16
   MP for +50 x 4/16
   MP for -110 x 6/16
   MP for -140 x 4/16
   MP for -170 x 1/16
   Total: MP for AS

Isn't this exactly how one would calculate a weighted assigned adjusted score?

Pretty much, except that I made an assumption of the weighting (1:4:6:4:1 /16). I think that if the outcomes start to be "too numerous" then it doesn't make much sense to assign a weighted score. The weighting factors are getting pretty arbitrary. The factors that I came up with at least had some (also arbitrary) reasoning behind them, but often one sees weighting factors that seem to simply be random guesses. These guesses can still be reasonably accurate when there are two or three possible outcomes, but when you get to five it starts to be silly.

Rik

[RMB1]

Under the previous law (Law 12C3:1997 "vary an adjusted score to do equity"), there were a variety of ways of calculating the adjusted score. My local experience was weighting the scores for various results on the board, as now envisaged (if not explicit) in Law 12C1(c):2007. But we did see European appeals committees just give a match-point or IMP score - sometimes the scribe would "reverse engineer" the weightings of two possible outcomes from the adjusted score.

What Rik is doing appears to be the second approach above: assigning a score which does equity without worrying about the precise weightings. I think this a still a ruling under Law 12C1(c) (not Law 12C1(d) or Law 12C2) and is therefore an assigned adjusted score (not an artificial adjusted score).

[pran]

RMB1, on 2015-May-17, 11:28, said:

Under the previous law (Law 12C3:1997 "vary an adjusted score to do equity"), there were a variety of ways of calculating the adjusted score. My local experience was weighting the scores for various results on the board, as now envisaged (if not explicit) in Law 12C1(c):2007. But we did see European appeals committees just give a match-point or IMP score - sometimes the scribe would "reverse engineer" the weightings of two possible outcomes from the adjusted score.

What Rik is doing appears to be the second approach above: assigning a score which does equity without worrying about the precise weightings. I think this a still a ruling under Law 12C1(c) (not Law 12C1(d) or Law 12C2) and is therefore an assigned adjusted score (not an artificial adjusted score).

Precisely.

[Trinidad]

RMB1, on 2015-May-17, 11:28, said:

What Rik is doing appears to be the second approach above: assigning a score which does equity without worrying about the precise weightings. I think this a still a ruling under Law 12C1(c) (not Law 12C1(d) or Law 12C2) and is therefore an assigned adjusted score (not an artificial adjusted score).

True, but what reason should there be to invoke Law 12C2 when the condition for it is not fulfilled: "When owing to an irregularity no result can be obtained"? (Note that this condition comes after listing all the possibilities of assigning an AS in law 12C1, obviously meaning that if all the other ways to assign an AS won't work then -and only then- you apply Law 12C2.)

In this case there are 2 possible outcomes: 2♠-1 and 2♠c. You simply apply Law 12C1c or, if necessary, 12C1d and you have your AS. There is no need to use the "If all else fails" method of law 12C2.

Rik

[VixTD]

Trinidad, on 2015-May-17, 03:04, said:

You are a chemistry teacher and you have three students. It is your job to test them and give them a grade for their report card at the end of the term. You have come up with 8 tests that they need to take.

Case 1
Unfortunately, the first one is a lab test and one of the three students blunders and blows up the building. You cannot give any tests anymore. What grades are you going to hand out?

Case 2
Student A has aced the first 7 tests with flying colors. He is headed for a straight A, as long as he doesn't blow up anything. Student B has had major trouble getting sufficient grades, he might pass or he might not. Student C is hopeless. He has no chance whatsoever. The last test is a lab test and student C (who else) blows up the building. What grades are you going to hand out?

Answers:
Case 1: the student who blew up the building won't pass. The other two get the benefit of the doubt and will get a passing grade: a B.

Case 2: Student C will not pass. Student B gets the benefit of the doubt and gets a B. Student A also gets the benefit of the doubt and will get an A. Except if the chemistry teacher is Pran, then student A and B both will get a B, because the possible outcomes of the final grade are "numerous".

I don't like your choice of analogy. Students should only pass assessments if they can demonstrate competence and ability. If they are unable to demonstrate such competence (whether through their own fault or not) they should not be given the benefit of the doubt. I work as a university lecturer and I have to be constantly on my guard against this sort of thinking, that if a student has been unable to take a test because, for instance, a university administrator told them the wrong exam date, they should be paid compensation in the form of academic credit.

You wouldn't dream of doing this for a driving test or pilot's licence, or a licence to practise medicine.

[blackshoe]

It occurred to me when I first read Rik's analogy that the question what grades to award is probably moot, because the students are probably dead.

[barmar]

VixTD, on 2015-May-18, 06:38, said:

You wouldn't dream of doing this for a driving test or pilot's licence, or a licence to practise medicine.

Because those aren't games. We don't give them the benefit of the doubt because it could kill someone.

[VixTD]

barmar, on 2015-May-18, 09:53, said:

Because those aren't games. We don't give them the benefit of the doubt because it could kill someone.

Do you mean you would give the benefit of the doubt to an arts student is unlikely to be in a position to kill someone through incompetence (except perhaps by mistranslating a crucial text), but you wouldn't to a lawyer, an engineer or a medical student?

[barmar]

VixTD, on 2015-May-18, 12:08, said:

Do you mean you would give the benefit of the doubt to an arts student is unlikely to be in a position to kill someone through incompetence (except perhaps by mistranslating a crucial text), but you wouldn't to a lawyer, an engineer or a medical student?

I wouldn't consider it a travesty if someone did that. We don't even require people to pass a test to make art.

[Trinidad]

VixTD, on 2015-May-18, 12:08, said:

Do you mean you would give the benefit of the doubt to an arts student is unlikely to be in a position to kill someone through incompetence (except perhaps by mistranslating a crucial text), but you wouldn't to a lawyer, an engineer or a medical student?

It happens all the time. I don't know what you are lecturing, but I am a chemical engineer. Throughout my life, as a high school student, a university student, as a grad student and when working at the university, I have seen situations where due to unforeseen circumstances tests could not be completed.

If you would have to fail each university student who couldn't complete a lab, then only very few, very lucky ones (as in: "not necessarily good, just lucky") would graduate with an engineering degree.

Take a typical engineering lab. It is full of equipment kindly donated by industry (read: "old junk that might still be used for teaching purposes"). The lab course is given once a year or once every two years. (These labs are expensive to run.) The students are divided into groups. Each group will run each piece of equipment in a rotation, similar to a bridge movement... or so is the plan. Because on day three, invariably one of these things will break. That happens with old equipment. This means that only two groups could run the equipment according to plan. The other 13 groups can't.

Do you mean that now only those first two groups get to pass the lab course because only they have been working the complete planned program? And the others have to wait two years when the lab runs the next time? In the next term, these same students go through the analytical chemistry lab. On day four, there is a short-circuit in the IR spectrometer. 12 out of the 15 groups cannot do their IR testing. "Sorry guys, see you next year"?

That would not be the engineering approach. The standard solution is that the students who can't use the equipment copy the data from one of the groups that could use it. And the teacher will make sure that the data is good. Then every group produces their own report on all the pieces of equipment, including the one that they never saw.

That means that those students will never turn any valves or pull any levers on the broken piece of equipment. They do not need to think which gauge to read at what particular point in time. And they do not need to make any decisions on what to do (which valve to turn or what lever to pull) with those readings. They will get good data for free.

In addition, when grading the report the lecturer will typically keep in mind that the students haven't seen the equipment when it was running. Mistakes in the report that are caused by this, will be corrected (so that the students will learn that it doesn't work like that) but also forgiven.

I would call that "getting the benefit of the doubt".

Rik

P.S. Engineers don't get actual specific industrial equipment training at a university. Operators who run equipment will have learned that on the job (i.e. in industry). They will be trained (and certified) for that specific piece of equipment by someone who knows that specific piece of equipment. University engineers typically do not operate or handle industrial equipment. They do not turn valves or pull levers in a plant. It would be one of the faster ways to blow up the place.

[VixTD]

In the modern university every course and module has documented "learning outcomes" which the students have to meet in order to pass. If your learning outcomes are "analyse adequately data generated from a practical exercise using a certain piece of equipment" then it's fine to allow the students to pass if they are given someone else's data and show they can analyse it because the machine has broken down, even though you would have preferred them to have had a go at working the machine themselves as well. Working the machine wasn't part of the learning outcomes, but analysing and presenting the data was. What I would object to is saying "the machine's broken down, you can't generate data, but we'll pretend you've passed".

I recall years ago a student was unable to take a particular module which he needed for the degree he'd started because of an accident that put him out of action for several months. By the time he was well enough to have another go the course had changed and the module no longer existed. The powers that be wanted to just give him the academic credit for the work he was unable to do. I wanted us to construct some sort of personalised work plan that the student would have to carry out (a research dissertation or something) that would as far as possible meet the learning outcomes, so that we weren't giving away degrees that other students have to work for. (I lost.)