WW Game "league"?

Isgrimnur · Post by **Isgrimnur** » Thu Aug 04, 2011 5:22 pm

So the Pittsburgh Pirates should get 2 points per win every year?

stessier · Post by **stessier** » Thu Aug 04, 2011 5:30 pm

triggercut wrote:when I originally conceived of the scoring, one of the things that made sense to me was that since WW games are essentially asymmetrical, that there should be asymmetrical scoring.

In other words, I think the Wolves have a tougher bit of sledding than do the Villagers. If the Villagers play a merely competent game, the Wolves must be outstanding to beat them. If the wolves play competently, the villagers have a decent shot of winning even if they have a few dramatic misplays. As such, my thoughts are that the scoring to offer a bit less risk to the wolves, so that if they play competently and are defeated, they may still score positive points.

My gut feeling is that the Wolves have the harder row to hoe and I feel like they should get bonus points when they win. I'm not sure if there are stats that actually back that up though - especially in the base game we will be playing.

The nice part about playing the exact same game over and over is that whatever scoring we decide on in the future can be retroactively applied without any bias.

Honestly I view the whole thing as an experiment, especially Season 1, and will agree to anything that isn't too off the wall.

stessier · Post by **stessier** » Thu Aug 04, 2011 5:38 pm

Isgrimnur wrote:So the Pittsburgh Pirates should get 2 points per win every year?

I think 2.10 is more appropriate.

bb2112 · Post by **bb2112** » Thu Aug 04, 2011 6:16 pm

stessier wrote: Honestly I view the whole thing as an experiment, especially Season 1, and will agree to anything that isn't too off the wall.

I agree. Since Triggercut and Stessier have generously decided to run this for us, why don't you two decide on a scoring system and get this thing rolling. If we find it too lopsided, we can tweak it for future rounds. Too many cooks are killing the soup at the moment.

Remus West · Post by **Remus West** » Thu Aug 04, 2011 7:52 pm

You guys figure out the scoring and give me the green light and I'll mod the first one.

triggercut · Post by **triggercut** » Tue Aug 09, 2011 7:24 pm

Scoring:

If stessier wants to use that 1500 points system with up/downs, I'm fine with it...but there's actual smartperson math involved there, so he's keeping score.

If not, we'll use the modified triggercut/bb2112 system as described upthread.

stessier · Post by **stessier** » Wed Aug 10, 2011 2:10 pm

triggercut wrote:Scoring:

If stessier wants to use that 1500 points system with up/downs, I'm fine with it...but there's actual smartperson math involved there, so he's keeping score.

If not, we'll use the modified triggercut/bb2112 system as described upthread.

Just use the modified system. I'll do the calculations this game for the ELO system and see if it is something we want to keep.

We should also have a request/rule that the Mod post either the actual voting results or a link to the actual voting results from each day in a post near the start of the thread. Will make double checking the Math Teacher's math a little easier.

purge · Post by **purge** » Thu Aug 11, 2011 12:14 am

I've now got to worry about scoring while playing fantasy games with a bunch of geeky guys?

Sounds like high school.

triggercut · Post by **triggercut** » Tue Sep 13, 2011 12:13 pm

triggercut wrote:
bb2112 wrote:
stessier wrote:
Remus West wrote:The way you have the Seer set up he could get killed the first night and still earn 3 points. I don't think he should get points for wolves killed when he is dead.
It's been a while, but I think the reason I put it in there is because even finding Villagers is useful. Not as useful, of course, but useful. So he should get some credit if he reveals 2 Villagers that lead to all the wolves being caught.

I'm not married to any of this or anything. This was trig's idea and all - don't want to step on any toes if people want the simpler scoring instead.
Yeah, I'm not married to that particular scoring system either, but I like the idea of correct actions/votes being rewarded, while incorrect actions/votes are penalized.

Maybe it is as simple as:

Wolf
+1 for every villager lynched while alive
+1 for a wolf win
-3 for a wolf loss

Max 4pts Min -3pts

Villager
+1 for every wolf lynched which has your vote
-1 for every villager lynch which has your vote
-2 if your vote is on a seer lynch
+2 if your team wins
+1 if you survive to the end
-1 if your team loses

Max 5 pts Min -5 pts

Seer
+1 for every wolf scanned
+1 for every wolf lynched which has your vote
-1 for every villager lynch which has your vote
+1 if your team wins
-2 if your team loses

Max 5 pts Min -5pts

This is just an idea. Not married to this either.
I like this scoring plan somewhat, but let's do away with "seer specific" scoring and just make all the villager scoring conform to that +5 to -5 score range.

Also, I would say that making a wolf team win a +2 for all team members would be a good adjustment.

This is our current scoring system. Absolutely I am not married to it at all.

Some adjustments I think are necessary based on game 1:

1. +2 points to every villager on a successful Wolf lynch vote. (Changed from +1)
2. +1 (additional) point to the FIRST TWO villagers on a successful wolf lynch vote
3. Consecutive +1 kickers to both teams for votes leading to the deaths of members of the other team. In other words, if the Wolves get two villagers in a row killed, they get an additional +1 for as a kicker for the second villager they take down in a row, and a +2 kicker for voting to successfully lynch a third villager in a row.. Similarly, villagers get an extra +1 if they're in on the vote for a wolf as the second consecutive wolf lynch.
4. Elimination of "surviving to the end" points for villagers and wolves.

Thoughts:

On 1.: Villagers have trouble scoring points, since an essential tool of good villaging is usually killing one of your own and then figuring out how that happened. Seems to put the villagers at a -1 disadvantage from the start.
On 2.: Mindful and understanding of Grundbegriff's points that surviving isn't necessarily a good end by itself, rewarding with an extra point would seem to favor the provocateurs AND a seer in the early game, and favor village survivors in the late game. Thoughts?
On 3.: Definitely like the idea of the "perfect game", and I think consecutive kickers rewards that.

Thoughts?

triggercut · Post by **triggercut** » Tue Sep 13, 2011 12:15 pm

Give me a sec to reconfigure scoring based on that adjustment, just to see what it looks like. I'll post it to see how it looks.

Remus West · Post by **Remus West** » Tue Sep 13, 2011 12:19 pm

purge wrote:I've now got to worry about scoring while playing fantasy games with a bunch of geeky guys?

Sounds like high school.

Nah, we rarely break down into a discussion of how hot the senior girl that just walked by is and what we would like to do with her but never will because she doesn't know we are alive.

triggercut · Post by **triggercut** » Tue Sep 13, 2011 12:28 pm

TOTALLY HYPOTHETICAL SCORING BASED ON GAME 1 WITH SOME OF THE CHANGES SUGGESTED.

Newcastle lynched, villager:

trig -1
Qant +1
pr0n +1
Isg -1
Grund -1

(trig killed overnight)

pr0ner lynched, wolf

theo +3
Grund +3
bb +2
Isg +2

(successful villagers receiving an additional point, with theo and grund receiving a bonus for being the first two.)

(Grund killed overnight)

RMC lynched, villager

theo -1
bb -1
Qant +1
Isg -1

(theo killed by wolves overnight)

(bb2112 lynched, villager)

Qant +2 (1 + 1 consec kicker)
Isg -1

Wolves win!

pr0ner +2 (ADJUSTED SCORE. WOLF WIN WORTH 2 POINTS)
Qant +2 (ADJUSTED SCORE. WOLF WIN WORTH 2 POINTS)

Newc -2
trig -2
grund -2
RMC -2
theo -2
bb2112 -2
Isgrim -2

Qantaga: 6 pts
pr0ner: 3 pts
Grundb: 0 pts
theo: 0 pts
bb2112: -1 pts
RMC -2 pts
Newc: -2 pts
Isgrim: -3 pts
trig: -3 pts

Thoughts on this?

bb2112 · Post by **bb2112** » Tue Sep 13, 2011 1:06 pm

That does look more fair. Let's try it for the next game and see if it holds up over time.

Grundbegriff · Post by **Grundbegriff** » Tue Sep 13, 2011 1:26 pm

That does look better.

I especially like your mechanism of rewarding those who are instigators of, or early on, a bandwagon rather than those who hop on later. This achieves three important goals:
(1) rewards leadership
(2) spreads bonuses more evenly across those who lead early and those who survive (and therefore lead) later.
(3) avoids rewarding wolves merely for jumping on a bandwagon at N-1 or N -- low hanging fruit for them rather than a significant risk.

Sure, there's always a lot of heat on the person who places a killing vote, but the heat hardly ever lasts, and the person is seldom an evildoer anyhow. Usually, "I just got sick of waiting" or "let's learn something" or "I genuinely thought he was evil" dissipates the heat.

The ranking generated by your recalculation looks pretty good, with one exception: it seems somehow wrong that you, triggercut, would end up at -3, in the same position as a figure who lasted a long time but had a bad run, when you were nuked almost right out of the gate.

What that means is that someone who puts forth a bad vote and belongs to a team that loses is penalized pretty severely for not having had a chance, by staying alive, to earn some offsetting positive points. Since (all else equal) wolves tend to kill the apparently good or apparently perceptive and to tolerate the ambiguous, suspected, or silent, the scoring system favors silents and suspects. Maybe that's not a bug; maybe it's a feature. But it seems somehow harsh that you're punished for dying early when the reason you died early is that they feared you or wanted you out of the way.

Thoughts on that last point, anyone?

stessier · Post by **stessier** » Tue Sep 13, 2011 1:28 pm

Grundbegriff wrote:The ranking generated by your recalculation looks pretty good, with one exception: it seems somehow wrong that you, triggercut, would end up at -3, in the same position as a figure who lasted a long time but had a bad run, when you were nuked almost right out of the gate.

Sucking up to the Scorer is beneath you.

Grundbegriff · Post by **Grundbegriff** » Tue Sep 13, 2011 1:31 pm

stessier wrote:
Grundbegriff wrote:The ranking generated by your recalculation looks pretty good, with one exception: it seems somehow wrong that you, triggercut, would end up at -3, in the same position as a figure who lasted a long time but had a bad run, when you were nuked almost right out of the gate.
Sucking up to the Scorer is beneath you.

I'm trying to signal him, with as much subtlety as possible, that I'm the actual Scorer and that he can safely continue his spoof because I'm behind him all the way.

stessier · Post by **stessier** » Tue Sep 13, 2011 1:49 pm

Grundbegriff wrote:
stessier wrote:
Grundbegriff wrote:The ranking generated by your recalculation looks pretty good, with one exception: it seems somehow wrong that you, triggercut, would end up at -3, in the same position as a figure who lasted a long time but had a bad run, when you were nuked almost right out of the gate.
Sucking up to the Scorer is beneath you.
I'm trying to signal him, with as much subtlety as possible, that I'm the actual Scorer and that he can safely continue his spoof because I'm behind him all the way.

Well played.

triggercut · Post by **triggercut** » Tue Sep 13, 2011 2:07 pm

Regarding my -3...

Village lost. I managed to not only place a death vote on a villager, but got myself killed in the process.

Look at it this way, though: had the village won, I'd be at +1 for the game.

One way of remedying this: no one--wolf or villager--receives a + or - point on the day's first lynch vote, because it is so blind, and because it allows wolves to potentially "game" the points system.

Kickers still affect subsequent votes, so if the wolves get a villager on day 1 they get no points, but if they get a villager lynched on day 2, every wolf vote in that train is rewarded with a +2 instead of just a +1, because they get the additional consistency kicker related to the (no point awarded) first day lynch of an innocent.

Thoughts?

Isgrimnur · Post by **Isgrimnur** » Wed Sep 14, 2011 11:34 am

Suggestions for player precedence in signups. Award one point for powered role, one point for each round number present for. If you die in Day/Night 2, you get two points.

So a Seer getting offed in Night 3 would receive 4 points, a village getting lynched on Day 2 would receive 2 points.

Signups for the next round would be given priority in a low number highest fashion to allow those who have not played or were eliminated early over those that have routinely played and survived to the endgame.

Example

Newcastle 1
triggercut 1
pr0ner 3
Grund 3
RMC 3
theohall 3
bb2112 4
Isgrimnur 4
Qantaga 5

So for Round 2, non-players would get precedence, but if there were, say, 11 signups, Qantaga would be the first bumped, bb and myself would be next, etc.

Thoughts?

Qantaga · Post by **Qantaga** » Wed Sep 14, 2011 1:27 pm

That looks like a very effective way to prioritize playing spots Isgrimnur.

stessier · Post by **stessier** » Wed Sep 14, 2011 2:37 pm

Looks like a good system to me. Very nice!

Grundbegriff · Post by **Grundbegriff** » Wed Sep 14, 2011 4:59 pm

I don't see why Specialness should figure into it.

Here's an alternate system:

To everyone who expresses interest in a particular game, a point value is assigned according to a simple algorithm:

(A) ∞: If you managed the immediately preceding game, you're automatically admitted.
(B) 1 per skip: For every game that has passed in which you didn't play, you are assigned a point.
(C) order: Ties go to the person who has been waiting longer, if any, or to random.org or courtesy.

Here's an example using a cast of 13 players competing for nine slots in game 05:

p0 (managed the immediately preceding game, has infinite points)
Sam (played 0 games, has 4 points)
Tony (played 0 games, has 4 points)
Violet (played 1 game, has 3 points)
Ursula (played 2 games but not game 04, has 2 points)
Wilhelm (played 2 games including game 04, has 2 points)
p1 (played all 4 games, has 0 points)
p2 (played all 4 games, has 0 points)
p3 (played all 4 games, has 0 points)
p4 (played all 4 games, has 0 points)
p5 (played all 4 games, has 0 points)
p6 (played all 4 games, has 0 points)
p7 (played all 4 games, has 0 points)

It's game 05 of season 1. Let's say that all 9 of the people who played in game 04 ask to play in game 05, and that the manager of game 04 asks, and that three additional people ask: Sam and Tony, who have never played, and Ursula, who has played twice and skipped twice.

* The manager of game 04 automatically enters game 05.
* Since they have never played, Sam and Tony have a point for each game missed: 4 points each. Nobody else will have so many, so they enter game 05.
* Ursula has 2 points. Whether she enters the game depends on whether any of the 9 who just played have missed more than 2 games. Let's say that 7 of the players had played all games and have 0 points apiece. But let's also say that for one of the players, Violet, game 04 was her first game. Since she missed games 01-03, she has three points. She enters game 05, even though she just played, because she has played fewer times than Ursula.
* Let's say that the last of the players in game 04, Wilhelm, had played twice including game 04; having missed two games, Wilhelm has two points. He's tied with Ursula. In the case of a tie, the player who played most recently yields to the one who has waited longer, so Ursula enters game 05.

I think Isgrimnur's system of points for longevity could be used to settle the 7-way tie among those who have played the most. Clearly, someone who hasn't played at all should trump someone who has played but was killed early, so it would make sense to use Igrimnur's algorithm, or something like it, to settle the issue after newbs and the previous manager have been accommodated.

Make sense?

Isgrimnur · Post by **Isgrimnur** » Wed Sep 14, 2011 5:44 pm

My system already gives n00bs an instant advantage, as they're already at zero points, and it's lowest number gets priority. I have no issues giving last manager a free pass into the game.

And I give a point for specialness because they get to "play" more per night than most people, thus meaning that they got to have more interactions every evening. That leads me to suggest that a special killed in a Day 1 lynch probably shouldn't get a point for specialness, as they never got their "extra play".

Remus West · Post by **Remus West** » Wed Sep 14, 2011 5:49 pm

Specials only get to "play" more if the villagers are lazy.

Isgrimnur · Post by **Isgrimnur** » Wed Sep 14, 2011 5:58 pm

I'm speaking to the night time actions, not their analytical tasks.

Remus West · Post by **Remus West** » Wed Sep 14, 2011 7:20 pm

Just because you don't get any "night time action".....

this is where you would say something about a pot and a kettle unfortunately

Isgrimnur · Post by **Isgrimnur** » Wed Sep 14, 2011 9:01 pm

There may be a thread around EBG about chocolate that makes your tiny statement invalid.

Remus West · Post by **Remus West** » Wed Sep 14, 2011 9:57 pm

What ever it is isn't going to invalidate it for me, that's for sure.

triggercut · Post by **triggercut** » Fri Sep 30, 2011 8:17 am

Game 2 scores

Day 1:

Triggercut lynched (Wolf)

Theohall +2 (wolf vote)
Theohall +1 (1st villager on wolf vote)
Remus West +2 (Wolf vote)
Remus West +1 (1st villager on wolf vote)
Grundbegriff +2 (Wolf vote)
bb2112 +2 (Wolf vote)
Lassr +2 (Wolf vote)

Night 1:

Grundbegriff killed.

Day 2:

Scoop lynched (villager)

+1 Isgrimnur

Night 2:

bb2112 killed

Day 3:

RMC lynched (villager)

+1 Isgrimnur
+1 Isgrimnur (consistency kicker, 2 villagers in a row)

Night 3:

Remus West killed

Day 4:

Lassr lynched

+1 Isgrimnur
+2 Isgrimnur (consistency kicker, 3 villagers in a row).

Wolves win:

+2 Triggercut
+2 Isgrimnur

Village loses:

-1 Grundbegriff
-1 Scoop
-1 bb2112
-1 RMC
-1 Remus West
-1 Lassr
-1 theohall

Totals:

Triggercut 2
Grundbegriff 1
Scoop -1
bb2112 1
RMC -1
Remus West 2
Lassr 1
theohall 2
Isgrimnur 8

Isgrimnur · Post by **Isgrimnur** » Sat Oct 01, 2011 1:56 pm

Here's my version of player precedence after Game 2. Exclusion would go from the top down.

Name	Total	Rd 2
Isgrimnur	9	5
RMC	7	3
theohall	7	4
bb2112	7	3
Qantaga	5	-
Grundbegriff	4	1
Lassr	4	4
triggercut	3	2
pr0ner	3	-
Remus West	3	3
Scoop20906	2	2
Newcastle	1	-

A winning wolf or seer gets 4 points for the rounds and 1 for the powered status, giving me 5 for last game. Trig gets one point for being a wolf, one for being a first round player. (I'm still pondering the powered point for a first-round lynch...) bb2112 gets 1 point for being the seer, two for lasting two days. Everything else is based on round survival.

theohall · Post by **theohall** » Sun Oct 02, 2011 11:00 am

Isgrimnur wrote:Here's my version of player precedence after Game 2. Exclusion would go from the top down.

Name Total Rd 2
Isgrimnur 9 5
RMC 7 3
theohall 7 4
bb2112 7 3
Qantaga 5 -
Grundbegriff 4 1
Lassr 4 4
triggercut 3 2
pr0ner 3 -
Remus West 3 3
Scoop20906 2 2
Newcastle 1 -
A winning wolf or seer gets 4 points for the rounds and 1 for the powered status, giving me 5 for last game. Trig gets one point for being a wolf, one for being a first round player. (I'm still pondering the powered point for a first-round lynch...) bb2112 gets 1 point for being the seer, two for lasting two days. Everything else is based on round survival.

That looks like an excellent way to determine priority. Thanks!

purge · Post by **purge** » Fri Oct 28, 2011 11:09 am

So do we have some sort of readable scoring system?

Also, where's the OOWWL Round4 signups at?

Lassr · Post by **Lassr** » Fri Oct 28, 2011 11:13 am

purge wrote:So do we have some sort of readable scoring system?

Also, where's the OOWWL Round4 signups at?

I'll let Triggercut do the scoring since it's his baby.

Isgrimnur · Post by **Isgrimnur** » Fri Oct 28, 2011 11:46 am

I'll try to get my precedence chart updated this weekend. I will still be at the bottom of the list.

theohall · Post by **theohall** » Fri Oct 28, 2011 2:36 pm

I won't be ready to moderate the next one, but will happily do the one after the next one. Need to get write-ups prepped and it will be my first time moderating.

stessier · Post by **stessier** » Fri Oct 28, 2011 3:41 pm

If no one else wants to do it, I can moderate a game.

Edit: I should mention I am in a creative trough so the write-ups will likely be of the form "Hey look - he's dead!"

LordMortis · Post by **LordMortis** » Fri Oct 28, 2011 3:46 pm

stessier wrote:If no one else wants to do it, I can moderate a game.

Edit: I should mention I am in a creative trough so the write-ups will likely be of the form "Hey look - he's dead!"

That's better than I'd do. I totally admire the AARs the collective write around here and the story teller aspect of the games is why I poke my head in here and read.

So don't disappoint me.

purge · Post by **purge** » Fri Oct 28, 2011 3:46 pm

So it's going to look just like a Unagi game then?

Remus West · Post by **Remus West** » Fri Oct 28, 2011 3:53 pm

stessier wrote:If no one else wants to do it, I can moderate a game.

Edit: I should mention I am in a creative trough so the write-ups will likely be of the form "Hey look - he's dead!"

Thats pretty much exactly what I did.

Qantaga · Post by **Qantaga** » Fri Oct 28, 2011 4:02 pm

stessier wrote:If no one else wants to do it, I can moderate a game.

Edit: I should mention I am in a creative trough so the write-ups will likely be of the form "Hey look - he's dead!"

stess, I can run this one, if you want to play.