[msheald]

Hello! I'm curious. Why does BBO calculate final standings in matchpoint games based on an average of the percent for each of the hands rather than an average of match points for each hand?

By that I mean that BBO calculates the percent standing of each hand (say beating 80% of other pairs or 20% of other pairs) and then averages all percentages to achieve a final percent standing for the game. Based on the percent standing, the final matchpoints are calculated.

In the club games and ACBL tournaments which I've played, a match point is assigned to each hand and then the average matchpoints are calculated for the round. As a result, if a person has a low board, be that 0% or 40% compared to other pairs, the result would still be 0 match points. Additionally, if a person has a high board, be that 100% or 55%, the team would get the same number of matchpoints.

For me, it is much easier to get a 0% than a 100%. Because a low board is penalized the same amount in in-person club matches and tournaments, whether it was a 0% or say a 40%, it encouraged risk taking. The teaching is - "it is only one board." With BBO, that is not the case since there is a huge difference in final standings based on a 0% vs. a 40% even though both might have been low board for that hand.

So, it appears that BBO tournaments do not follow ACBL in how hand standings/matchpoints are calculated and how final matchpoints for an event are calculated.

Just curious as to why that is. Thank you and best regards.

Mike

[smerriman]

I don't understand your post. Are you saying that at your club, the bottom half of the field on each hand scores the same 0, and the top half scores the same 1 - even if some did considerably better than others? This certainly isn't matchpoints, or how normal ACBL scoring works.

[pescetom]

wikipedia said:

Matchpoint scoring
One common form of pairs scoring is by matchpoints. On each board, a partnership scores two matchpoints for each other partnership that scored fewer points with the same cards, and one point for each other partnership that scored the same number of points. Thus, every board is weighted equally, with the best result earning 100 percent of the matchpoints available, and the worst earning no matchpoints; the opponents receive the complement score, e.g. an 80% score for a N–S pair implies a 20% score for their E–W opponents. Colloquially, a maximum matchpoints score on a board is known as a "top", and a zero score is a "bottom". The terms "high board" and "low board" are also used.

Note 1: Using American Contract Bridge League (ACBL) methods, scoring is one point for each pair beaten, and one-half point for each pair tied.
Note 2: The rule of two matchpoints for each pair beaten is easy to apply in practice: if the board is played n times, the top result achieves 2n−2 matchpoints, the next 2n−4, down to zero. When there are several identical results, they receive the average. However, complications occur if not every board is played the same number of times, or when an "adjusted" (director-awarded) score occurs. These cases can result in non-integer matchpoint scores – see Neuberg formula.
These matchpoints are added across all the hands that a pair plays to determine the winner. Scores are usually given as percentages of a theoretical maximum: 100% would mean that the partnership achieved the best score on every single hand. In practice, a result of 60% or 65% is likely to win the tournament or come close. In a Mitchell movement (see above) the overall scores are usually compared separately for North–South pairs and for East–West pairs, so that there is one winner in each group (unless arrow-switching has been applied - see above).

[Gerardo]

Disclaimer: While I work for BBO, I was not involved in any of the descisions about this, nor I asked about them. This is completely unofficial.

For one, the number of matchpoints in itself is incomplete without the number of tables (or comparisons, usually tables - 1).

5 matchpoints in 6 tables is a shared top, 5 matchpoints in 16 tables is a quite poor result.

A percentage is just normalizing the result, so 5 in 6 tables is 5/(6-1)= 1, or 100% while 5 in 16 tables is 5/(16-1)=1/3 or 33.33%.

Without percentages, you need to stipulate another parameter (number of comparisons would do, in older times I saw the average instead, as in X matchpoints on a Y average).

With percentages, only one number is needed, and gives an easy read about the quality of your game.

Note that, while BBO across all sections (at the end; scores shown while tourney is running are different for each section), if you don't do this, an inconvenient number of tables, say, a prime number like 61, leaves you with sections of different sizes, then you need to normalize the results somehow, and using percentages does that, too.

[helene_t]

Matchpoints is also a confusing concept because some countries count 1 MP per comparison (half an MP for a tie) and some count 2 MP (one point for a tie). Percents are more international.

[Vampyr]

helene_t, on 2021-July-04, 19:42, said:

Matchpoints is also a confusing concept because some countries count 1 MP per comparison (half an MP for a tie) and some count 2 MP (one point for a tier). Percents are more international.

That is a good point, and is probably the answer in full to the OP’s question. Percentages are the same no matter how the matchpoints are counted; if you are very interested in the latter, you can easily work out how many matchpoints you achieved on a board, using whichever of the methods Helene mentioned. Just look in your history or later at the traveller, and count the pairs you beat. Give yourself one (or two, if you prefer) point for each of them. Then count the pairs you tied and give yourself one half (or one, if you prefer) point for each of them. Hey presto, now you know how many matchpoints you had on a board.

[smerriman]

helene_t, on 2021-July-04, 19:42, said:

Matchpoints is also a confusing concept because some countries count 1 MP per comparison (half an MP for a tie) and some count 2 MP (one point for a tier). Percents are more international.

That's true, but the OP stated that in their club a score of 40% could receive 0 matchpoints. That's not possible regardless of whether you're using half points or percents, unless they're using a different scoring system altogether.

[helene_t]

smerriman, on 2021-July-04, 21:02, said:

the OP stated that in their club a score of 40% could receive 0 matchpoints. That's not possible regardless of whether you're using half points or percents, unless they're using a different scoring system altogether.

Yeah.
There may be some confusion caused by scores that are calculated across the whole field versus per-section results, and scores that don't auto-refresh when more tables complete the board, and tallys that are partially grandfathered from other players who played in the same direction before one subbed in. Also, BBO averages the percentages while most scoring software AFAIK calculates total matchpoints across the whole tally and calculates percentages at the end, iow BBO doesn't weight the boards by the number of comparisons.

[Vampyr]

msheald, on 2021-July-04, 06:52, said:

Hello! I'm curious. Why does BBO calculate final standings in matchpoint games based on an average of the percent for each of the hands rather than an average of match points for each hand?

……..

Mike

So as has been explained, the percentage is just a way of repenting the number of matchpoints; for most people, seeing the percentages is more readily understandable. Say there are 11 tables. How are the matchpoints being counted — is 10 a top or is 20 a top? Do I have a top or an average board? So if you were just given the matchpoints, you would have to know both how many tables there were and how the matchpoints are being counted. It I beat 70% of the pairs in this game, I would rather see my score expressed as 70% than 7 matchpoints, although they are the same thing.

[dsLawsd]

Some time ago the ACBL went to percentage instead of match points so that reports made more sense to non-duplicate players when reporting overall results where the matchpoints on a board might vary. It took a bit to make the adjustment I suppose. Unlike bowling a '200' game could be anywhere from winning to being a dreadful score.

[mycroft]

And with more games scoring across-the-board, and with web movements (all this IRL, of course), the numbers were no longer memorable. 156 (26 6s, 13, 2-board rounds) average is fine to work off of; as is 108 (27 4s). But when top could be 19, or 15, or 65, the numbers stop making sense. Also, when there are assigned or artificial scores, and it is not hand-scored, we end up with "fractional matchpoints" (that aren't 1/2). So there's really little difference between percentages and whole number matchpoints now, from the POV of a player; and for players who weren't playing in 1995, percentages are just easier than 185 on a 156 average.

But I'm very confused as well as to where the "0 or 40 scores the same, 55 or 100 scores the same" - because at least in the ACBL, that's never been the case, except for BAM scoring. I know England does MPs-to-VPs for short-match "Swiss Pairs" events, but that's over the entire round, not each board. Please do explain, OP!

[barmar]

I think the OP may be misunderstanding something in the result reports.

[armantt2k]

Vampyr, on 2021-July-04, 20:23, said:

That is a good point, and is probably the answer in full to the OP’s question. Percentages are the same no matter how the matchpoints are counted; if you are very interested in the latter, you can easily work out how many matchpoints you achieved on a board, using whichever of the methods Helene mentioned. Just look in your history or later at the traveller, and count the pairs you beat. Give yourself one (or two, if you prefer) point for each of them. Then count the pairs you tied and give yourself one half (or one, if you prefer) point for each of them. Hey presto, now you know how many matchpoints you had on a board.

Doing what you say will result in final total matchpoints scores for each pair that are either whole or half (e.g. 63.00 or 62.50), no decimals other than x.00 or x.50. This is what matchpoint game scoring (using the ACBLscore software approved by the ACBL to run tournaments) is and is the basis for MP awards (*). This often results in pairs tied for a position in the final standings and sharing awarded MPs.

Duplicate Bridge pairs are familiar with these ties and comfortable with sharing the MPs. BBO's scoring results in pairs tied with same matchpoints, but only one pair getting ALL the MPs, while the other pair feels cheated because even though they tied for the award position based on established, well-known, and time-honored ACBL scoring rules, they receive no MPs.

(*) Yes, I am aware that TD rulings (giving Average +,=,- on disputed boards) can result in some odd decimal scores in ACBLscore, but those are relatively rare exceptions, and unrelated to the question of the validity of BBO's scoring methodology.