Can someone confirm this?

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
SimplyShane
Posts: 7
Joined: 14 February 2016, 21:54

Can someone confirm this?

Post by SimplyShane »

Long story short: I was playing a game of Puerto Rico. Someone disconnected and ran out of time. Everyone else decided to just keep playing, so we finished the game. I actually came back from a huge deficit to win on the very last move, but when I look at how the points were divided among all of us, I only earned +2 ELO while everybody else got quite a bit more. Is this because I was last place when the disconnect happened?
User avatar
Jest Phulin
Posts: 1856
Joined: 08 July 2013, 21:50

Re: Can someone confirm this?

Post by Jest Phulin »

First, when discussing something that happened at a table and it doesn't seem right, please include the table number so it is easier to track. :D

But, I'm assuming you are talking about table 19838602, since that looks like what you described.

This is how ELO works. You get more ELO by winning against people who are ranked better than you, and less against those ranked below you.

Your ELO is 1611, just barely less than the person listed as zombied (1670)
The other players at the table had significantly less ELO (1540, 1423, 1516), so they gained more than you.

Additionally, since only 63% of the game was actually played (player was zombied at 63% progression), only 63% of the ELO was awarded.

Long story short -- it's working the way it should.

(Lengthening the story) It isn't because of the scores at the zombification. It's because your high ELO was matched with an almost equal ELO at a partial game, and everyone else's low ELO was matched with a higher ELO.
User avatar
SimplyShane
Posts: 7
Joined: 14 February 2016, 21:54

Re: Can someone confirm this?

Post by SimplyShane »

Thank you for explaining it so clearly. I fully understand now why it happened that way.
User avatar
sprockitz
Posts: 666
Joined: 23 October 2014, 02:22

Re: Can someone confirm this?

Post by sprockitz »

the problem is that the system counts your game as a tie with the other non abandoning players. So you are actually losing points for tying against the other players that don't abandon while gaining points against the player who abandoned. This is what results in a negligible elo gain.

I don't think this is how it should work, the 'ties' should really be ignored in this case...but that wasn't the question :-)
User avatar
Oso
Posts: 701
Joined: 29 September 2011, 04:04

Re: Can someone confirm this?

Post by Oso »

Exactly, it's the problem with the "all are winners", because in ties... who has more ELO will loss some points against the opponent.

In ties, people who has more ELO loss around 1 point per each 20 ELO difference (1.5 points if you've 1 to 29 games played). Of course, this number is around 100% of game progress.
Locked

Return to “Discussions”