Total Posts:25|Showing Posts:1-25
Jump to topic:

Mafia Scoring System

tvellalott
Posts: 10,864
Add as Friend
Challenge to a Debate
Send a Message
12/23/2010 11:14:40 PM
Posted: 5 years ago
I'm hoping that people don't look on me trying to impose more work on mods, but I have an idea and I need community input for it to work.
Simply scoring people by win/loss ratio is not an accurate way of ranking players. First of all, terrible players may be on the winning team. Excellent players may be killed by rogue vigilantes.

I have two points:
1) I am hoping that people will keep a record of key events of their games. I DON'T mean you have to even read the game threads, which I suppose some mods don't. However, things like who was involved in the final lynching of each day phase, who the cop investigated on what nights, who the mafia killed and so on can easily noted down, since knowing the information is part of modding the games anyway. I suggest a simple spreadsheet.
2) I'm also hoping people will contribute how much each action is worth. For example, a vanilla townie who is killed NP1 won't receive any role points, but a vanilla townie who survives to the end of the game would receive a good amount of points, because VT claim's are often lynched because they aren't much of a loss if it turns out the they were telling the truth.

So, lets start with the basics:
Cop surviving to the end of the game:
Cop making a guilty investigation:
Doctor surviving to the end of the game:
Doctor protecting an innocent from death:
Cop surviving to the end of the game:
Vigilante kiling a non-town player:
Vigilante killing a pro-town non-role:
Vigilante killing a pro-town with role:
Miller surviving to the end of the game:
"Caitlyn Jenner is an incredibly brave and stunningly beautiful woman."

Muh threads
Using mafia tactics in real-life: http://www.debate.org...
6 years of DDO: http://www.debate.org...
Ore_Ele
Posts: 25,980
Add as Friend
Challenge to a Debate
Send a Message
12/23/2010 11:53:38 PM
Posted: 5 years ago
At 12/23/2010 11:14:40 PM, tvellalott wrote:
I'm hoping that people don't look on me trying to impose more work on mods, but I have an idea and I need community input for it to work.
Simply scoring people by win/loss ratio is not an accurate way of ranking players. First of all, terrible players may be on the winning team. Excellent players may be killed by rogue vigilantes.

I have two points:
1) I am hoping that people will keep a record of key events of their games. I DON'T mean you have to even read the game threads, which I suppose some mods don't. However, things like who was involved in the final lynching of each day phase, who the cop investigated on what nights, who the mafia killed and so on can easily noted down, since knowing the information is part of modding the games anyway. I suggest a simple spreadsheet.
2) I'm also hoping people will contribute how much each action is worth. For example, a vanilla townie who is killed NP1 won't receive any role points, but a vanilla townie who survives to the end of the game would receive a good amount of points, because VT claim's are often lynched because they aren't much of a loss if it turns out the they were telling the truth.

So, lets start with the basics:
Cop surviving to the end of the game:
Cop making a guilty investigation:
Doctor surviving to the end of the game:
Doctor protecting an innocent from death:
Cop surviving to the end of the game:
Vigilante kiling a non-town player:
Vigilante killing a pro-town non-role:
Vigilante killing a pro-town with role:
Miller surviving to the end of the game:

I think what is more important then how people use their roles is how they act during the day phase. Obviously, if a vanilla townie is doing all the logic and reasoning and the vigilante is just following orders and making all the good kills, the logic driven townie should get the points.
"Wanting Red Rhino Pill to have gender"
Danielle
Posts: 21,330
Add as Friend
Challenge to a Debate
Send a Message
12/24/2010 12:04:58 AM
Posted: 5 years ago
At 12/23/2010 11:14:40 PM, tvellalott wrote:
For example, a vanilla townie who is killed NP1 won't receive any role points, but a vanilla townie who survives to the end of the game would receive a good amount of points, because VT claim's are often lynched because they aren't much of a loss if it turns out the they were telling the truth.

But doesn't this sort of give way to the luck thing again, which you already noted as problematic? If I say I'm a vanilla townie and get lynched, but you say you're a vanilla townie in the next game and people keep you, why should you get more points? It would have nothing to do with skill and everything to do with the other players. I personally think that since all methods seem to present particular conflicts, you might as well just go with the easiest (win-loss ratio). However if you're interested in doing something like this, I see no reason to say you shouldn't :) I'm sure people's main concern is the amount of work it will be, in which case they'll have a more reasonable opinion once they see how it actually works (completed spread sheet and whatnot).
President of DDO
tvellalott
Posts: 10,864
Add as Friend
Challenge to a Debate
Send a Message
12/24/2010 4:55:34 AM
Posted: 5 years ago
At 12/24/2010 12:04:58 AM, theLwerd wrote:
At 12/23/2010 11:14:40 PM, tvellalott wrote:
For example, a vanilla townie who is killed NP1 won't receive any role points, but a vanilla townie who survives to the end of the game would receive a good amount of points, because VT claim's are often lynched because they aren't much of a loss if it turns out the they were telling the truth.

But doesn't this sort of give way to the luck thing again, which you already noted as problematic?
Indeed it does, but we both know that some players are indeed better than others and I'm trying to find the best way to capture that.

If I say I'm a vanilla townie and get lynched, but you say you're a vanilla townie in the next game and people keep you, why should you get more points?
Perhaps you're suggesting we have some kind of handicap for players who are regularly killed early in the game?

It would have nothing to do with skill and everything to do with the other players. I personally think that since all methods seem to present particular conflicts, you might as well just go with the easiest (win-loss ratio). However if you're interested in doing something like this, I see no reason to say you shouldn't :) I'm sure people's main concern is the amount of work it will be, in which case they'll have a more reasonable opinion once they see how it actually works (completed spread sheet and whatnot).

Indeed. Lets see what I come up with and we'll work from there.
"Caitlyn Jenner is an incredibly brave and stunningly beautiful woman."

Muh threads
Using mafia tactics in real-life: http://www.debate.org...
6 years of DDO: http://www.debate.org...
mongeese
Posts: 5,387
Add as Friend
Challenge to a Debate
Send a Message
12/24/2010 10:03:49 AM
Posted: 5 years ago
I don't think a scoring system is a very good idea. It would overcomplicate the game and make things too confusing. Luck would be extremely influential. The rare willingful deaths would't be properly factored. And I can almost guarantee that if the scoring system becomes good enough, somebody sometime is going to cast some lynch vote not for the sake of the game, but for the sake of the scoring system.
wamba
Posts: 688
Add as Friend
Challenge to a Debate
Send a Message
12/24/2010 1:45:42 PM
Posted: 5 years ago
we just need to go on a wins/loss system. Simple.

Luck happens, but to be consistently lucky is a skill indeed, unless you're philo.
wamba
Posts: 688
Add as Friend
Challenge to a Debate
Send a Message
12/24/2010 1:50:31 PM
Posted: 5 years ago
Suggestion for games forum:

When posting in games forum users have the ability to change their username/profile picture when they are not active in a game. Post count will be hid.

This way:

Before - TheLwerd 132432 posts
After - Don
Game 1 - Don (secretly thelwerd)
Break - Fred (secretly theLwerd)
Game 2 - Fred

Game posts - user names will not change when they have been posted.

This way if TheLwerd as above posts under the alias of Don
then when she changes her username i between game 1 and 2, all of her Game 1 posts will say "Don" even after she changes her name to Fred.

This way users can be anonymous when playing games if they wish.

Users are not allowed to change their usernames in mid-game.

This is my suggestion.
ccstate4peat
Posts: 2,022
Add as Friend
Challenge to a Debate
Send a Message
12/24/2010 5:29:42 PM
Posted: 5 years ago
At 12/24/2010 2:29:13 PM, mongeese wrote:
And how many days do you think would pass until we realize who's who based on post style/grammar?

I'd presume people would think ahead and change their writing style beforehand.
wamba
Posts: 688
Add as Friend
Challenge to a Debate
Send a Message
12/25/2010 12:58:22 AM
Posted: 5 years ago
At 12/24/2010 2:29:13 PM, mongeese wrote:
And how many days do you think would pass until we realize who's who based on post style/grammar?

would be a lot harder to do this vs. the current system.
OberHerr
Posts: 13,062
Add as Friend
Challenge to a Debate
Send a Message
3/25/2012 11:35:13 PM
Posted: 4 years ago
Bumpity bump.
-~-~-~-~-~-~-~-OBERHERR'S SIGNATURE-~-~-~-~-~-~-~-

Official Enforcer for the DDO Elite(if they existed).

"Cases are anti-town." - FourTrouble

-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-
F-16_Fighting_Falcon
Posts: 18,324
Add as Friend
Challenge to a Debate
Send a Message
3/25/2012 11:54:56 PM
Posted: 4 years ago
I think there are a lot of things that could be quantified and many that cannot be quantified.

For instance, we could have a system like this for townies:

- Voting on a scum lynch = +10 points
- Voting on a mislynch = -10 points
- Getting mislynched = -10 points
- Not voting by the end of the DP = 0 points

But then there are the unquantifiable things like who busted a mafia member and who just followed in their wake etc.
Zaradi
Posts: 14,125
Add as Friend
Challenge to a Debate
Send a Message
3/26/2012 1:08:39 AM
Posted: 4 years ago
Ah, bumping old threads. Oh well, at least now I can add my two worthless cents to the discussion.

I think the entire premise of a scoring system is entirely subjective as to what actions merit what points and what actions don't merit what points. How can one reasonably decide between two people who both claimed to have outted a mafioso and lead the inevitable lynch who deserve the credit for the lynch and who doesn't? It would all be arbitrary.

To make matters worse, a win/loss record bites into the same harms, as mentioned by a few other people. What if I happen to really fvcking suck at the game, but I always end up on the winning team somehow? My win/loss record would be absolutely incredible, but my skills would not be on par with what my w/l record showed. How do we fix this?

I say instead of trying to avoid the subjectivity, we embrace it and allow everyone to form opinions on who is the better players. Everyone does it anyway, so why not just use that? For example, it's mostly agreed upon by all the normal players that Royalpaladin, bluesteel, F-16, and a few others whose names slip from my tired mind at 1 in the morning are fairly experienced and fairly good players of mafia. Why do we have to come up with some arbitrary scoring system to try and justify it instead of just leaving it as it is? There really isn't a reason not to.

Anyway, that's my two cents. Had to spend them somewhere, vending machines don't take pennies.
Want to debate? Pick a topic and hit me up! - http://www.debate.org...
Buddamoose
Posts: 19,448
Add as Friend
Challenge to a Debate
Send a Message
3/26/2012 8:51:52 AM
Posted: 4 years ago
At 3/25/2012 11:54:56 PM, F-16_Fighting_Falcon wrote:
I think there are a lot of things that could be quantified and many that cannot be quantified.

For instance, we could have a system like this for townies:

- Voting on a scum lynch = +10 points
- Voting on a mislynch = -10 points
- Getting mislynched = -10 points
- Not voting by the end of the DP = 0 points

But then there are the unquantifiable things like who busted a mafia member and who just followed in their wake etc.

+10 ;)
"Reality is an illusion created due to a lack of alcohol"
-Airmax1227

"You were the moon all this time, and he was always there to make you shine."

"Was he the sun?"

"No honey, he was the darkness"

-Kazekirion
Buddamoose
Posts: 19,448
Add as Friend
Challenge to a Debate
Send a Message
3/26/2012 8:52:16 AM
Posted: 4 years ago
At 3/26/2012 1:08:39 AM, Zaradi wrote:
Ah, bumping old threads. Oh well, at least now I can add my two worthless cents to the discussion.

I think the entire premise of a scoring system is entirely subjective as to what actions merit what points and what actions don't merit what points. How can one reasonably decide between two people who both claimed to have outted a mafioso and lead the inevitable lynch who deserve the credit for the lynch and who doesn't? It would all be arbitrary.

To make matters worse, a win/loss record bites into the same harms, as mentioned by a few other people. What if I happen to really fvcking suck at the game, but I always end up on the winning team somehow? My win/loss record would be absolutely incredible, but my skills would not be on par with what my w/l record showed. How do we fix this?

I say instead of trying to avoid the subjectivity, we embrace it and allow everyone to form opinions on who is the better players. Everyone does it anyway, so why not just use that? For example, it's mostly agreed upon by all the normal players that Royalpaladin, bluesteel, F-16, and a few others whose names slip from my tired mind at 1 in the morning are fairly experienced and fairly good players of mafia. Why do we have to come up with some arbitrary scoring system to try and justify it instead of just leaving it as it is? There really isn't a reason not to.

Anyway, that's my two cents. Had to spend them somewhere, vending machines don't take pennies.

+10
"Reality is an illusion created due to a lack of alcohol"
-Airmax1227

"You were the moon all this time, and he was always there to make you shine."

"Was he the sun?"

"No honey, he was the darkness"

-Kazekirion
Buddamoose
Posts: 19,448
Add as Friend
Challenge to a Debate
Send a Message
3/26/2012 1:38:30 PM
Posted: 4 years ago
At 3/26/2012 1:08:39 AM, Zaradi wrote:
Ah, bumping old threads. Oh well, at least now I can add my two worthless cents to the discussion.

I think the entire premise of a scoring system is entirely subjective as to what actions merit what points and what actions don't merit what points. How can one reasonably decide between two people who both claimed to have outted a mafioso and lead the inevitable lynch who deserve the credit for the lynch and who doesn't? It would all be arbitrary.

To make matters worse, a win/loss record bites into the same harms, as mentioned by a few other people. What if I happen to really fvcking suck at the game, but I always end up on the winning team somehow? My win/loss record would be absolutely incredible, but my skills would not be on par with what my w/l record showed. How do we fix this?

I say instead of trying to avoid the subjectivity, we embrace it and allow everyone to form opinions on who is the better players. Everyone does it anyway, so why not just use that? For example, it's mostly agreed upon by all the normal players that Royalpaladin, bluesteel, F-16, and a few others whose names slip from my tired mind at 1 in the morning are fairly experienced and fairly good players of mafia. Why do we have to come up with some arbitrary scoring system to try and justify it instead of just leaving it as it is? There really isn't a reason not to.

Anyway, that's my two cents. Had to spend them somewhere, vending machines don't take pennies.

I agree, however a scoring system can still be formed on a subjective basis. Lets say at the end of a game each player is given a vote for who they believe played the best. Of course votes for oneself not being allowed. When the votes are tallies:

1st in votes: 5 points
2nd: 3 points
1st: 1 point

I dont belueve much abuse would occur on this system, it would happen(people voting on bias) but overall i think it would give an accurate portrayal of the best players.
"Reality is an illusion created due to a lack of alcohol"
-Airmax1227

"You were the moon all this time, and he was always there to make you shine."

"Was he the sun?"

"No honey, he was the darkness"

-Kazekirion
F-16_Fighting_Falcon
Posts: 18,324
Add as Friend
Challenge to a Debate
Send a Message
3/27/2012 12:46:58 AM
Posted: 4 years ago
Another interesting thing I wanted to point out:

1) Good townies are unlikely to have high win rates because they are often killed early by the mafia.

2) Good mafiosos on the other hand can have very high win rates sometimes even achieving 100% win rates (like Drafterman for instance).

So, there should be a different scoring system for town play and mafia play given that some players are better as town and some are better as mafia.
Buddamoose
Posts: 19,448
Add as Friend
Challenge to a Debate
Send a Message
3/27/2012 12:52:05 AM
Posted: 4 years ago
At 3/27/2012 12:46:58 AM, F-16_Fighting_Falcon wrote:
Another interesting thing I wanted to point out:

1) Good townies are unlikely to have high win rates because they are often killed early by the mafia.

2) Good mafiosos on the other hand can have very high win rates sometimes even achieving 100% win rates (like Drafterman for instance).

So, there should be a different scoring system for town play and mafia play given that some players are better as town and some are better as mafia.

Drafter does not lose 0_o ?
"Reality is an illusion created due to a lack of alcohol"
-Airmax1227

"You were the moon all this time, and he was always there to make you shine."

"Was he the sun?"

"No honey, he was the darkness"

-Kazekirion
airmax1227
Posts: 13,241
Add as Friend
Challenge to a Debate
Send a Message
3/27/2012 12:52:17 AM
Posted: 4 years ago
At 3/27/2012 12:46:58 AM, F-16_Fighting_Falcon wrote:
Another interesting thing I wanted to point out:

1) Good townies are unlikely to have high win rates because they are often killed early by the mafia.

2) Good mafiosos on the other hand can have very high win rates sometimes even achieving 100% win rates (like Drafterman for instance).

So, there should be a different scoring system for town play and mafia play given that some players are better as town and some are better as mafia.

yeah these are good points and I don't know how one can fairly, statistically, assign value to a mafia players ability... I'm 5-0 in my games for example.. but that can't really be attributed to me..

But separating qualifiers between mafia and town seems like a good start..
Debate.org Moderator
Buddamoose
Posts: 19,448
Add as Friend
Challenge to a Debate
Send a Message
3/27/2012 12:54:32 AM
Posted: 4 years ago
At 3/27/2012 12:46:58 AM, F-16_Fighting_Falcon wrote:
Another interesting thing I wanted to point out:

1) Good townies are unlikely to have high win rates because they are often killed early by the mafia.

2) Good mafiosos on the other hand can have very high win rates sometimes even achieving 100% win rates (like Drafterman for instance).

So, there should be a different scoring system for town play and mafia play given that some players are better as town and some are better as mafia.

+1
"Reality is an illusion created due to a lack of alcohol"
-Airmax1227

"You were the moon all this time, and he was always there to make you shine."

"Was he the sun?"

"No honey, he was the darkness"

-Kazekirion