[manudude03]

Woah, I completely forgot about this thread.

Gwnn: Most forms of scoring (MPs, IMPs, Butler etc) don't take into account of who made it so the table didn't get the par result. It can just as easily be that declarer has done something brilliant, or a combination of both. I'll just post 1 (made up) hand for argument sake:

---

Let's say everyone leads a heart, doesn't give anything away.

Hope I haven't messed this up too bad. But lets see 3 lines, 1 normal, 1 "double dummy" and 1 "silly"

Normal: Win the lead in hand and run the ♦9 (catering for Kx(x) or JTx onside), it loses to the stiff K, declarer wins the heart return in dummy, cashes the ♦A finding the 4-1 break. He then finesses clubs out of necessity and cashes the ♣A, spade to the Ace and cashes the ♣A throwing a diamond. Declarer now needs 3-3 spades or a ♦+♠ squeeze and when neither materialises, down one.

Double Dummy: Again, declarer wins the ♥ lead in hand, and plays a small diamond to the ace dropping the stiff K. He then plays a club to the jack and ducks a diamond. He wins the ♥ return in hand, cashes the ♣A and finesses diamonds (West only has 1 diamond honour left). When the diamond is won in dummy, cash the ♣A throwing the losing ♠. Contract made.

Silly: Again, the lead is won in hand, but declarer plays a small diamond to the Queen. This loses to the stiff K and a heart returned. Declarer wins in dummy and finesses clubs which wins, cashes the ace and plays a diamond to the A. He cashes the ♣K throwing a diamond and plays 3 rounds of spades. When they don't break 3-3, declarer is down 2.

[gwnn]

Yes I know that there are a lot of situations, and I believe I worded my first post accordingly (in the footnote).

All I'm asking is why is it a problem that Butler gives a different value than ximps? You seem to be quite sure that the people with -50 "should get" -1 imp. I don't know why.

[manudude03]

Well, all the declarers "failed" to find the winning line, so should be punished for it, right? And the only reason it would only be like -1 and not -14 or whatever is that there were a lot of them who had the same result implying that it may have been a difficult problem. Often the only difference between butler and XIMPs is relativity. XIMPs just stretches the margins. I don't really consider it that much of a problem, but in general in XIMPs, it is more often the case you have to do something good to gain imps which imo is how it should be.

[nige1]

Duplicate

[nige1]

Siegmund, on 2011-January-07, 03:46, said:

Gotta define "better."
If you mean in the statistical sense of estimating as accurately as possible what the distribution of results on the board should be if played many times, no, throwing out results is a Bad Thing. If you mean some very specialized sense like trying to estimate what par should be on a board, you can make a case for it, but you can make even better cases for all of means, medians, and modes than you can for trimmed means, except in a few really odd situations.

FrancesHinden, on 2011-January-10, 16:01, said:

Really? I've always found that when I win, I deserved it because I played well. When I don't win, I was either unlucky (pairs), or my teammates had a bad day (teams). Seems to be true all the time.

Agree with Siegmund. And Frances (except that a bad pairs-result is often partner's fault).

FWIW, IMO, cross-imps is best because it retains all relevant data and makes good use of it.

In Butler scoring you discard one of two scores from each extreme and imp against the average of the remaining scores. Instead, you could discard novice and beginner scores. Or you could introduce a new loony category and discard the scores of such players. All this is superficially plausible. The basic argument is "Why should I be penalised because somebody else, my way, lost 1700 in a part-score deal?" The problem is that extreme scores are not always stupidities. For example, most pairs reach the excellent 6N contract. Most declarers give up when they hit a bad break. A couple of resilient declarers manage to recover with a squeeze-strip. Now, Butler discards the par scores, which makes little sense.

When choosing an average, to imp against, the normal choice is the mean. Although arguably the median -- or even the mode -- would be better. Or you could imp against an artificial average e.g. a Gib DD result. The objection to all these is that you lose data. You replace masses of perfectly valid data with one arbitrary "representative" score.

Among the attractions of Bridge are the the human dimension and the role of chance. Eliminate the rub of the green and you detract from the game.

Luckily, it is rare that choice of scoring method materially affects the overall tournament results.

[hotShot]

Let us assume a board where the declaring side playing 4M (nonvul) is missing 3 aces and there is no way they can avoid to lose these 3 tricks.
At one table the contract is doubled making.

Score method 1:
Both sides at tables where 4M= was played receive 0 IMPs.
The pair declaring 4MX= wins 5 IMPs over 4M=.
The pair losing 4MX= loses 5 IMPs againt 4M=.

No result is lost.

Score method 2:
Assuming that the board was played n-times.
Every declarer making 4M= gets -5/(n-1) IMPs.
Every defender gets +5/(n-1) IMPs.
The pair declaring 4MX= wins 5 IMPs over 4M=.
The pair losing 4MX= loses 5 IMPs againt 4M=.


The question is, do we want that the pairs playing 4M=, where both declarer and defender made only right choices, separate in their score by 2*5/(n-1) IMPs?

Obviously for large n, the methods produce equal results.

Playing in the MBC the "Highscore" could be 7MXX-3 producing a even larger distortion.

[mycroft]

mycroft, on 2011-January-11, 15:18, said:

And why should defenders be punished because we are the only ones in the room to have a blackwood error and bid grand off a finessible trump king? Because that's the way it works. Swings and roundabouts.

Heh - serendipity strikes (otherwise known as Mycroft should watch what he says).

1) Both partners invert DEPO, but one remembers in time to pull 6Cx to 6H. Makes on two of two finessable kings.
2) 1NT-2H; 2S-4H; 4NT-5S; 6S. Partner forgot 3H was forcing, so 4H is splinter, so we're keycarding spades, not hearts. Makes when SAQx is onside.

Embarassed? Sure. Gonna give back the matchpoints? No way.