Probably need a bit more active of a userbase before something like this would work out. If a single 15-20 person game cant fill, finding 8-11 people to run an 8 week tournament is gonna be tough.
I don’t see majority of players for this coming from our current userbase.
I’d invite/advertise elsewhere - the “competitive tournament” demo is a pretty different one.
That is very true.
Maybe make it best 4 of 6? Like play 6 of 8 games and take your four best scores
or something similar. Whatever makes the most mathematical sense. Should be able to drop at least your worst score imo so it doesnt tank your overall
1 Like
Eh, I don’t really agree with this. It’s not like liars club where you can get super fucked by rand.
I think 6-0 should always beat 4-2.
You can get fucked by a bad teammate tho
If I go 6-0 and you go 5-1 because your town teammate fucked up in a 3v1, does that make me necessarily a better player than you?
Dropping correctly measured “outliers” is a bad thing in general - https://stats.stackexchange.com/questions/200534/is-it-ok-to-remove-outliers-from-data/200545#200545
Going 6-0 doesn’t mean you’re a better player than 0-6. It just means you won the tournament. Kasparov and Carlsen didn’t win every tournament they played but they’re still the “better player”.
You can easily argue that both no drop vs. dropped aren’t perfect measures of “betterness” but what’s harder to back up is that dropped is a better measure of betterness.
I’m happy to use Liars club’s algo instead of raw score to rank players but I’m pretty emphatically against dropping games. Definitely wouldn’t mind playing with a dataset after the fact to see how it changes rankings though.
And a nice thing about nightless is that if you’re not around to effect something it means you got lynched. 
I agree with liars club in that if there is any sort of round-based scoring people on the same team should get the same score on a given round when applicable.
I also wouldn’t mind it being straight up 8 people who play 6 games against each other.
Yeah I guess since its a tournament you are strictly scoring by wins and losses with the round scores as like potential tiebreakers then?
That’s the system we have in the op right now.
Sorry I guess I had gotten the round score part of it confused
So let me drop this question in the thread then:
Why not just have a site wide leaderboard and have mods designate if they want their game to be considered for the leaderboard total or not? Make it a seasonal like “League” thing instead of an actual tournament
main reason was I thought tournament better for marketting
we’re 100% doing that for Espionage since it’s the same setup over and over but it’s harder for different setups.
Has to probably be opens where the probability tree is fairly clear so outcome and round scores are easy to get. (It only turned out to be plusminus 0.5 this time because the EV’s were 0.5).
My suggestion would be to have both- the ultimately competitive and very strictly regulated setups for the tournament and then an overall leaderboard for any games that mods want to submit for leaderboard recognition. Gives people a way to track their stats as well as being like a more friendly competive bragging rights sort of deal. Having a few unbalanced games won’t really matter because there will be more of them and over a longer period of time.
I’m worried this will be a turn off for a lot of people. I want to keep the casual atmosphere for most games here and have any leaderboard stuff be for a special subset. Probably just a straight up separate category.
I’ve done that before in the past, you do it like
+3 points for a town win
+5 points for a scum win
+1 point for loss
+1 point for MVP
+0.5 for Losing team MVP
-1 point for LVP
or something similar in scope - I’m bad at math things
Well I was thinking of it as like a casual thing not a competitive one
Getting off topic at this point, but you can put me in for the tournament LOL