Slow play - board not played
We had a Pairs League running last night with 5 divisions, each of 3 tables, on BBO - all tables playing the same boards.
When a table is slow, BBO will not allow the starting of a board if there is less than 3 (?) minutes to go in the round.
In one division, two of the tables didn't get to play board 8, while the third table did play it and got - when compared with the other divisions - a very favourable result.
With only 3 tables in each division, the system, naturally, produced no comparisons for that board in that division and the pair with the favourable result got no reward.
Is that how it should be, or should there be some compensation?
Comments
The TD can over-ride the result - I assume that you mean that one pair got a good result (and weren't rewarded) and one pair got a poor result (and didn't suffer).
You could just make up some scores e.g. if the very favourable result was +630 then the scores at the other two tables could be 3NT making for only +600 - the pair with the very good result get their top, their opponents get a bottom. The matchpoints would be split by BBO as
2-0
1/2 - 1 1/2
1/2 - 1 1/2
First, let's see what happens under the Laws' definition of "matchpoints":
The letter of Law 86B1 doesn't apply here, because it only applies to team games. However, I think the spirit of the Law applies somewhat, here: it effectively says that if there are no comparisons for a given board, the Director should assign a score to the tables that didn't receive a result (thus allowing the table which did receive a result to have a comparison). Although it's limited to team games, the same principle should presumably apply at pairs (especially because your three-table pair event has fewer comparisons than a teams-of-8 event!).
Law 87B attempts to cover a comparable situation at pairs (in which a board is fouled and produces insufficient comparisons as a result), and basically leaves the correct adjustment up to the CoC, or the Director if it doesn't specify one. Again, though, the lack of comparisons here were the consequence of an irregularity at every table but one, rather than specifically due to a board being fouled.
It would probably make sense to update the Laws covering lack-of-comparison situations to simply say "if no comparisons are available for a result, the Director assigns a score to compare against" in general, rather than only applying to specific circumstances (such as teams play and fouled boards). As of now, though, it doesn't, so Law 78A is the overriding factor.
I do have some serious concerns over Law 78A, though, now that I read it – it seems to reduce the top on boards for which there are artificial adjusted scores at other tables. Every Director and scoring program knows that if, say, 4 of 9 tables end up with artificial adjusted scores, you have to double the number of matchpoints available at the other tables in order to compensate for there being only half as many comparisons available (a primitive method would be to just double everyone's matchpoint award individually, but it's better to do this via a more sophisticated algorithm like Neuberg). I can't see any permission to actually do that in the Laws, though, when playing an event scored under "matchpoints" as the Laws define it. So this is a good reason to not play the Laws-defined "matchpoints" method, and in fact BBO doesn't use it.
However, we don't need to use matchpoint scoring as the Laws define it; Law 78D allows Tournament Organizers to define their own scoring mechanisms. The EBU has effectively done this using the White Book, inventing what we can think of as an "EBU matchpoints" scoring system, which defines different algorithms for what to do when a board is unplayable at some tables. For example, WB 4.2.3 gives a scoring system, different from matchpoints, for handling boards which have fewer comparisons than they should have (but at least one comparison).
WB 3.3.2 handles the situation you're asking about – if a board is played only once in a particular form, it should be given an artificial score. Unfortunately, the artificial score it suggests in this case is average+ to both sides (because neither pair at the table were responsible for there being no comparisons for the result they gained), which doesn't seem like enough compensation for the successful pair if the result is clearly good for one side.
This is a situation where we might plausibly want to update the White Book, specifying what to do in this situation; this would effectively change the EBU's definition of "EBU matchpoint scoring", and therefore allow for the comparisons to be done differently. (It is a pretty rare situation for a board to have an artificial score at every table but one, though, so perhaps it's unsurprising that neither the Laws of Bridge nor the White Book suggest a "fair" way to recover from it.)
Are those divisions scored independently of each other? If so, it seems to me that at the table that played the board, both pairs should get Av+. Those at the other tables should get some sorts of artificial scores depending on degree of fault for the boards not being played - most likely Av- for all four pairs, unless the log makes clear that one side was not directly to blame.
The divisions are scores independently of each other - and it is only the fact that they are playing the same boards which makes it so clear that a very favourable result was obtained by one pair.
Giving both pairs at the one table which played the board AV+ seems like a quite unfair reward to the pair who did badly on the board.
Well they did nothing to make the board unscoreable - it was the pairs at the other table who did that. As you say, if they hadn't happened to play the same boards as the other divisions, it wouldn't have been know how well or badly they had done.