Board Game Arena

Some funny mathematics from my "prestige page":

Hanabi: 3 615 Games • 3 243 Victories • 46% wins (I'd say it's closer to 90% wins...)

Saboteur: 7 Games • 3 Victories • 79% wins (3 is less than half of 7...)

Coup: 1 Games • 1 Victories • 200% wins (Oh yeah!)

I know some of it may be due to abandoned/practice games being counted weirdly, but really, 200%?

Jellby wrote: ↑06 September 2021, 21:11 Coup: 1 Games • 1 Victories • 200% wins (Oh yeah!)

I know some of it may be due to abandoned/practice games being counted weirdly, but really, 200%?

Yep, if you are first in a 3p games, it is considered as a win against 2nd and 3rd players.

Likewise, for Hanabi, a shared win only counts as 50%

The percentage of wins is basically the percentage vs a single opponent. So it does look strange if you are trying to apply that number to multiple opponents.

aesche wrote: ↑06 September 2021, 22:11 Likewise, for Hanabi, a shared win only counts as 50%

Should it be less if the win is shared between more than 2 players? Most of my Hanabi games are more than 2 players.

I‘d say so, but not sure. Your winrate suggests otherwise.

Jellby wrote: ↑06 September 2021, 21:11 Some funny mathematics from my "prestige page":

Hanabi: 3 615 Games • 3 243 Victories • 46% wins (I'd say it's closer to 90% wins...)

Saboteur: 7 Games • 3 Victories • 79% wins (3 is less than half of 7...)

Coup: 1 Games • 1 Victories • 200% wins (Oh yeah!)

I know some of it may be due to abandoned/practice games being counted weirdly, but really, 200%?

The maths are quite simple (though not intuitive): a victory is worth more or less depending on the number of players that scored less than you did. (In competitive games, only 1st place count as a victory)

• everyone ex-equo => Victory = 0.5 win
• one defeated player => Victory = 1 win
• two defeated players => Victory = 1.5 wins
• three defeated players => Victory = 2 wins
• etc

This mean that when playing Hanabi, since there is only one team, the best you can hope for is to get half a win per game, hence 50% is the maximal score physically possible. You had a few lost games, hence 46% wins.

For Coup, I'm assuming this was a 4 player game, so the 1st place is worth 2 wins, hence 200%.

For Saboteur, you had 3 victories out of 7 games:
First victory was in a 4p game, with two victorious players, hence two defeated players => victory is worth 1.5 wins
Second victory was in a 5p game, with three victorious players, hence two defeated players => victory is worth 1.5 wins
Third victory was in a 5p game, with one victorious player, hence four defeated players => victory is worth 2.5 wins
TOTAL: 5.5 wins for 7 games, hence 79% wins.

Since it's not intuitive, and it's using some arbitrary and not clearly documented formulas, I'd suggest giving it some other name rather than "wins", something like "BGA player score". And since a "%" sign usually indicates that 100 is the maximum, I'd drop it.