How is the win percentage calculated?

Discussions about BGA (all languages)
Forum rules
Warning: challenging a moderation in Forum = 10 days ban
More info & details about how to challenge a moderation: viewtopic.php?p=119756
Locked
User avatar
Schachus
Posts: 6
Joined: 18 April 2020, 14:35

How is the win percentage calculated?

Post by Schachus »

On the XP page of a player, for each game it says something like x Games y Victories z% wins. Now you could thingk that z was the quotient of y/x as a percentage, and it seems to be the case for 2 player games (i.e. "35 Games • 29 Victories • 83% wins") for me with Hex.

for multi-player games, the number seems quite strange though, sometimes lower, sometimes higher than expected, some examples from my page : "9 Games • 7 Victories • 39% wins" (Hanabi), "25 Games • 8 Victories • 72% wins" (6 nimmt), and the percantage can even surpass 100% ("2 Games • 2 Victories • 150% wins" at Puerto Rico).

Does anybody know how it works?
User avatar
Steve N
Posts: 13
Joined: 03 November 2011, 18:35

Re: How is the win percentage calculated?

Post by Steve N »

I think it is calculated on a “per player” basis, rather than “per game” basis.

So if you play one 4-player game and come second, you have beaten two players and lost to one. That would give you a 67% win ratio.

I think.
angelslight
Posts: 2
Joined: 10 May 2020, 14:48

Re: How is the win percentage calculated?

Post by angelslight »

No. No matter the player count, only wins (1st places) earn you winning percentage points.

The basis is 2 players = 100 % and the following are proportionate: 3 players 150 % (3/2 * 100 %), 4 players 200 % and so on.
As far as I know, when there is a shared first place, that value is divided again by the number of players that got first. That's why highest in Hanabi is 50 % because one can only win together, or as another example in 5p a tied win with one other player would earn you 125 %.
User avatar
Schachus
Posts: 6
Joined: 18 April 2020, 14:35

Re: How is the win percentage calculated?

Post by Schachus »

you seem to be right, at least that would fit the data. What Steve N suggested, would make much more sense to me as a system though. 150% win percentage is just not a real thing...
Locked

Return to “Discussions”