Programming Arbitrary Scores
I am writing a Bridge Scoring program for small clubs.
I have reached the stage where I am programming corrections to scores.
Having read the EBU’s White Book on this topic I am aware that for Match Point Scoring, the Arbitrary score 'Ave+' translates to 60%, 'Ave' to 50% and 'Ave'- to 40%.
My question is 60% of what? Is it:-
a) 60% of the Average Score on the Board
b) 60% of Average MPs awarded to the Board
c) 60% of the Pair’s Average Session MPs
I could find just one reference in the White Book in paragraph 8.12.13 which indicated that option (c) is the correct definition. Is this a correct interpretation?
Also am I correct in assuming that a Neuberg Correction will need to be applied to the other Results on the Board, since the Board will have been 'played' one less time than other the other Boards?
Many thanks for your help in advance.
Comments
Sorry if I failed to reply to an earlier.
The starting point is not the White Book but Law 12C2.
The 'available matchpoints' is the top which all boards are scored: 2 x number_of_possible_results - 2.
If there is an artificial score then the remaining real scores have to be neuberged.
Thank you Robin for your reply.
For clarification the 60%/50%/40% is of the Maximum MPs that can be obtained on the Board, but before or after the Neuberg Correction is applied?
If a top on a board is 100 MPs, then Ave+ is 60 MPs (or session average if higher), regardless of what else happens on the board. i.e. this applies, even if the actual top achieved by another pair drops to 99 MPs (say) due to Neuberg.
The artificial scores are not included in the neuberg calculations.
On the subject of arbitrary scores, I had a pair that left early from a club session, so were awarded 40/60 for the 3 boards remaining. Their average at the time was less than 40% (or more accurately less than -2 IMPs per board as it was a X-IMPed event). Bizarrely, the boards were awarded as -2.39, -2.26 and -2.13 IMPs. Does anyone know why EBUScore does this? (Their average for the session up until that point was -2.39 IMPs).
More specifically, 18 boards had been played and their total was -43, so their average at that point was -2.39. -2.26 corresponds to -43/19 and -2.13 corresponds to -43/20, which seems to be how EBUScore is programmed. However, this doesn't really feel like the correct interpretation of the White Book?
Thanks again Robin but your answer does not clear up the question I was trying to pose.
Let us say that a board is played 6 times, but for whatever reason the TD decides to give an Arbitrary Score to one of the results.
So it is played 5 time plus an Arbitrary Score, see attached Excel file
The Arbitrarily Scored hand has not been included in the Neuberg correction and I assume that the Arbitrary Score is awarded NS-6, EW-4, and not NS-4.8, EW-3.2.
Is this correct?
Yes, 6-4 not 4.8-3.2
The point that you might have missed from the start is that all boards in the session should be factored to a common top, regardless of how many times they are played. & this applies whether a board is played fewer times because of an 'arbitrary' score, or because they were scheduled to be played fewer times (e.g. because of a sit-out in an incomplete movement).
The common top should be taken from whichever board has been played the most times. So, if most boards (or even just one!) have been played six times, then the theoretical top on every board should be 10 MPs.
Theoretical top may be 10, but the actually top, after Neuberging, may be slightly less. But what does not change is the 'average', which should be 5 MPs regardless. Hence 50% = 5 MPs; "average plus" = 6MPs.
Thanks for confirming that Mitch.
It's the conclusion I reached after working up the spreadsheet.