Scummies Ideas, Suggestions and Comments Thread

[Thestatusquo]

I think it might be a great idea to make the scummies mean something more than a mockery of a popularity contest. Does anyone have any ideas for how this could be accomplished?

(note, this is serious. I am not trying to start a fight, I just think that the scummies have become a little bit of a joke.)

[Thestatusquo]

^^^

I think that's one of the crucial things about mafia; it's not baseball, where you can go all SABERMETRICS and figure out a player's effectiveness based purely on stats and it's pretty much going to be correct. Certain playstyles are effective in certain games with certain other players with certain roles and mechanics and etc. There are simply too many variables, and the game is based too heavily around human interaction to put together any sort of truly meaningful statistics. Hell, even win/loss ratios are unreliable; you could've gotten vigged N1 for being a VI in every town game you play, and still have a 7/2 town record just because the rest of the town was good.

I think judging pretty much has to be subjective; I honestly don't feel that there are any truly meaningful individual statistics in mafia.

I almost completely disagree. I 100% think that there are almost surely possible metrics that could be used to objectively rate mafia skill. Just because we haven't created them yet does not mean they are impossible or even that difficult.

To continue your analogy, sabermetric analysis of baseball is a relatively new development. Old metrics, like wins for a pitcher, RBIs, etc, are useless statistics, just like game wins are useless statistics in mafia. There is almost surely, however, metrics that can be used, and developed.

Furthermore, there are statistics in baseball that are capable of powerfully synthesizing different things to get overall pictures. Stats like wOBA generally paint a portrait of offense, and is useful even when comparing players with completely different playstyles (i.e. Power hitters who strike out a lot vs. Fast guys with high OB%) Hell, in baseball we've even invented metrics which allow us to compare players across eras and accounting for park factors. (WAR).

Basically, your statement doesn't make any sense at all.

Although, I would say all of this is besides the point. Less popularity contest != we need to create objective statistical measures. It is a strawman you are attacking, sir. There just needs to not be a system where basically winning a scummy means you happen to be friends/in a game with one of the 10 people who care about the scummies.

[Thestatusquo]

I would prefer a system where ballots are given to players who either play or mod a lot of games. Perhaps ex scummy winners could be given ballots as well, as well as any site mods. Basically create an academy of the scummies. You get a wide diverse group of people who have votes, which nullifies the popularity contest aspect of a) the nomination process (which is really the worst part of the whole thing) and b) the small sample and biases of the tiny group of scummies judges.

The main disadvantage of this system, as I see it, would be that with a large amount of people, less of them would read every game. But I think that this is answered back soundly by aggregation. By giving a large number of ballots, you get more eyes reading more games which means you're more likely to reach the correct outcome. And also, this is answered back by the fact that we are giving ballots to a group which has seen the most games and has the most experience with good mafia.

Shrug. I just hate the current system. I think it gives the wrong people awards a large % of the time, and rewards flashy attention getting play and mafiascum politics rather than actual playskill.

[Thestatusquo]

I think that would be good in that it would get rid of the nomination process that we currently have now, but I think there are some issues. For one, who is doing the nomination? The players in the game? the mod of the game? Etc. For another, I don't think it solves the issues in terms of team performances. It doesn't work with best team scum and town performance. It also doesn't work with body of work awards such as paragon.

Lastly, and perhaps most importantly, I think it puts an unreasonable burden on whomever you have judging the ceremony. Are you really asking judges to read every game that is played in a given year? That's just not possible for people who have lives outside of reading mafia games.

[Thestatusquo]

I remember one time I nominated a popular scummer for a complete joke reason that was clearly not scummy worthy, and like 2-3 people 2nded it.

I dont think the problem is necessarily with the nomination in the first place, though that is sometimes a problem, but I think that people generally just second nominations for their friends. I also agree with what you've said about not all deserving people getting nominated.

No, they all just get one vote for each catagory. I think if maybe you had a group of 30-50 people you will get enough ballots and it would be more accurate and less biased than the current system. It's the same way that MVP voting and such work in various sports.

[Thestatusquo]

In my system the academy DOES choose the winner outright, reck.

Zora, that is an interesting point, but I think its an implimentation problem. If you choose a good group of people, you'll be more likely to get a good result than the current system.

[Thestatusquo]

Ok. New system. I just pick all the awards.

[Thestatusquo]

its ok. the scummies are just a popularity contest anyway.

[Thestatusquo]

k.

you realize I've judged the scummies too, right?