[ljwobker]

When playing bridge-4 or other similar games, it would be nice to see the results at the end as a percentile rather than just a rank. For instance, if I finish a quick Bridge 4 game and ranked “1034”, I have no idea or not whether that’s actually any good. If 1200 people have played that set of hands, I played like crap. But if 15,000 people played that set of hands, I played really well.

If I was also told “your rank of 1034 is the 83rd percentile” then I’d have a pretty good idea that I played well (and how many people played...)

[smerriman]

ljwobker, on 2021-May-15, 19:03, said:

If 1200 people have played that set of hands, I played like crap. But if 15,000 people played that set of hands, I played really well.

1 person played that set of hands; you.

The leaderboard is entirely meaningless since everyone else is playing different boards from you. The only way to rank highly is to be dealt 4 high-scoring boards in a row; there's no correlation between this and playing well.

[armantt2k]

Perhaps then BBO could provide a comparison to GIB double dummy prediction of best play? Again, DDA is not always realistic, but ... this would be better than current "result"

[smerriman]

If you want to see how well you've done why not just play one of the duplicate games?

[pescetom]

I think the OP proposal that rankings should indicate percentile is very logical, however silly the actual ranking may be.

It certainly would be nice to also have an indication of how significant the given ranking is in mathematical terms... that however is another can of worms and probably not beneficial to the BBO business model either