Quick Fix Game 25: A Sundered Sky

[Elbereth]

5 hours ago, Sart said:

Changing the number of elims is risky. Too many, and they overwhelm the village in the voting. Too little, and the village might overwhelm them. The reason why we have a normal number of eliminators is for balance purposes. Meta did a mathematical proof that stated games are balanced when the number of eliminators equals the square root of the number of players. That's why we have that tradition. That being said, you can tip the scales provided you balance it well enough. You could run a game with 8 elims, but all the villagers are Seekers, or something silly like that. You could have a conversion role, and have that reek havoc with the town. Or you could run a game with a Serial Killer, as shown in this game. That's not a bad thing, and it can be a good change from the norm.

The problem with this game is that the Serial Killer was more powerful than a regular eliminator team. First, having a smaller number of eliminators is an advantage, not a disadvantage. Less eliminators means less chances to make mistakes in thread, and less chances to be caught defending one another. My favorite part of the game is seeing that an eliminator died, and then deducing who else must be evil because of it. To balance out this inherent strength, a smaller team has less power to vote. That can be especially dangerous if one of them dies early on. Thus, serial killers usually have some sort of extra power to balance it out. There are 3 common ones.

• Having some sort of protection
• Having some sort of vote manipulation
• Having some sort of extra kill power.

Having one or two of those can make for a balanced game. The problem was, the Skybreakers had all three!

I'm not mad; I'm just disappointed. I enjoyed this game while I was playing it. I loved seeing the debate, the back and forth of conversation, all the crazy conjectures people were making. I hated the ending of this game. I've been writing this post for a while now, and I finally figured out why I don't like this game.

Voting didn't matter. No, scratch that. Voting was harmful to everyone. Let me explain from each player's perspective.

• The villagers:
  • On another villager: This is normal, and drives discussion. However, it still hurts the village in the long run. Games should try to encourage this type of behavior.
  • On Stink: They lose their lives. It benefits the person being lynched more than the person killing him. If it's a neutral role we can avoid, that's fine. However, we had to lynch Stink. There was no other way to kill him. We had to act against our own self-interest to win the game. Yuck.
  • On Orlok: Hooray! We lynched the elim we're supposed to lynch. What's that? He had an extra life? Oh well, I guess he might be an Elsecaller. Guess we learned nothing from that lynch. What's that? We lynched him again, and he still didn't die? Huh, guess he might be some type of Survival role. Better lynch him again to make sure. Yay! We finally killed him! And it only took 2 cycles without any relevant discussion. Hooray!
• The eliminators:
  • On another villager: This will just make me look suspicious. I guess I will if I start looking suspicious, but there's no rush. I can just hide among the crowd anyways. Games should try to encourage this type of behavior, to make mafia stand out.
  • On Stink: Haha. Suck it losers, get wrecked son! I guess we still lose a teammate... and Orlok loses his extra life... why do we have to do this again?
  • On Orlok: I really don't want to vote on myself. Eh, worse case scenario I just lose an extra life.

Do you see the problem here? This isn't healthy. We want players to vote and engage with this discussion. I want to see the benefits of me voting, and see "So and So has died, they were Evil" I especially want to see that text when the only thing I can literally do in this game is vote on people! Seriously, when 90% of your players can do nothing besides vote or influence the vote, I want my vote to actually be meaningful!

I could forgive all that for just being bad game balance. It's fine, it's just an experiment that didn't go well. We need those every once in a while. They're healthy for the community. But you crossed a line. You made this game broken on purpose. You gave your boyfriend the most powerful role in the game. I think your romance is one of the coolest things that has come out of this forum, but you have to understand how this looks. You gave the troll the trolliest role in the game. You didn't do this randomly! I play with the expectation that everyone has a fair shot of being evil, and a fair shot of being good. That didn't happen this game. If you needed to tell Orlok the role distribution twist, Orlok shouldn't have played.

In the past 2 games I've played (this and the last MR), GMs have given out roles to players. On the Question and Answer page, both Aman and Elb have admitted to tweaking distributions after seeing the player list. This can't continue! Just look at the discussion above me. Head-hunting has been occurring because "there must be one experienced eliminator on a team". That type of logic is cancerous, and just leads to experienced players having less fun, while other players never get a chance to be bad guys. This can't hold.

Do you want to know something? I've run two games recently, and I think they've been good games. I created a role list before the game began, detailing roles and alignments. Then I assigned players randomly, and hit start. I didn't look if the team would be good, if the team had experienced players, or if the team was composed entirely of new players. It shouldn't matter. Yes it sucks if Ruin suddenly goes inactive, and he's the only evil kill role. However, that's an outcome we may face.

I propose the following rule: All games should be made name-blind. That is to say, role lists should not be based off of particular people, and roles should be assigned randomly to all players. It's the only way to keep the process fair and balanced. It's how we've done things in the past, but we've just strayed away from it. We need to be better than this.

I have a few responses to this. 

I agree about the game being difficult lynch-wise. I hadn't considered that aspect, and apologise that it makes the game less fun. (I disagree that eliminators are disadvantaged in voting for villagers - they want villagers dead, after all. Using the lynch to do so is a perfectly valid eliminator strategy.) Voting was one of my main worries about this game, though for a different reason - the Truthless discourages it. I don't like that, but accepted it as the cost for the other things I wanted out of the game. 

These are the particular things I disagree with: 

1. As I've already explained, I gave the role to Orlok because he knew the twist. As I recall, he came up with the twist, or at least a good part of the twists in this game. (There were initially more - I was playing with a Lover role that didn't end up working at all.) He's the one who pointed out that my current QF that I was planning on running was horrendously broken, and helped me come up with a new one. And he wanted to play. I didn't give it to him because he's my boyfriend, and am frankly offended that you would accuse me of doing so - I did it for the same reason Meta made Cessie the eliminator in LG2. Orlok knew, and he wanted to play, and he's the one who helped me make the game interesting and not horribly broken in the first place. I don't think it would be fair to ban him from playing a game which he helped to create when he wants to, and there is a way to allow him to play. 

2. I disagree about the square root rule - could you link me to Meta's proof, since I don't think I've ever actually seen it? It works, but only in non-extreme cases. A game of 9 players shouldn't have 3 eliminators. That's too many. They should have two. (Or even more extreme, 2 players out of 4 obviously makes no sense.) On the other end, 36 players should have more eliminators than 6. I think something like a 15-25% ratio makes more sense, personally.

3. Smaller number of eliminators have an advantage in less connections, yes. At the same time, it's just them. They can't afford to lose anything - a single mislynch could be devastating. In this game that's particularly so - even one lynch on Orlok would've essentially ended the game for him, leaving it up to Stink. And to win, Stink would've had to survive to endgame and get all but one player to vote on him (because he could kill them). That's why a couple extra cycles for the eliminators to get closer to a win seemed important to include. 

4. Could you please clarify what you mean by "you made this game broken on purpose"? I feel that I have made it very clear that I did my absolute best to make this game as balanced as I could. I didn't succeed in a few ways, as I've set out, but I did not set out to make a broken game, and I feel that you're disrespecting the work I've put into this game by saying that. 

5. The difference between a Serial Killer in a normal game, I'll note, and in this one, is that in this one the Serial Killer has to have a decent chance to win. Most don't have much chance at all - the only one I recall ever succeeding was Aman in QF14, and he gained the role late in the game. So comparing this game to a normal Serial Killer who can hide because the eliminators are the main threat isn't quite accurate, in my opinion. 

 

Lastly, regarding distributions. I agree in a standard game that roles shouldn't be chosen for player-specific reasons. It only becomes necessary in games in which certain roles enormously impact the outcome of the game, like this one. I would argue that that's an argument for more standard games, though, rather than no choosing players whatsoever. Even had Orlok not played, I would not have given Nale to Jebus, for instance. He notified me halfway through the game that he was going inactive, so he wouldn't have ever died to the filter, but the game would essentially never have ended until the village tore itself apart after who-knows-how-many cycles. That does not sound fun, to me. So it had to be someone who I felt reasonably sure would be active. In such a case (similar to LG21, which I referenced in Game Creation), I would not be comfortable with a game being completely random. Semi-random, yes - I'd probably pick a shortlist of players I thought would do and pick from that, rather than handpick. But I'm firmly of the opinion that totally name-blind games will lead to an increase in broken or less balanced games. I don't think that's a sacrifice I could support in order for games to be entirely random. 

[Arrae]

At the very least this game is creating a fair bit of dicussion.

I personally enjoyed it, though I feel like it might have been better if we'd had more hints that something was off. Something more than the placement of Truthless in the roles list or the auto-PAFO to questions about secret roles/mechanics.

I'd like to thank everyone who stayed civil when discussing this game and its distribution. To those who have feedback about the game: talk about the game all you want -- it's a fair target and you can feel however you want about it -- but let's keep some of the more personal accusations out of here unless they are relevant to the game, shall we? We're giving GMs a huge insight on how players view the game here, and we might get some useful ideas out of this if we stay level-headed.

[Elenion]

10 hours ago, little wilson said:

You're about to join the ranks of those players, Len. You're over 20 games now. We've had 220 players and you're in the top 30. That's pretty high up. Do you want people saying that about you? That you have to be evil, and then targeting you for that logic? Pushing your death for it? Do you want to die merely because of how many games you've played and not because of how  you're playing a particular game?

1. I've already seen something kinda like experience votes: being the votes coming on me from players who always suspect me. I'm not being targeted for my activity per se, but for my maneuvering in previous games, which comes from being an active player. While I can't say that I enjoy those votes, I take them as a sort of compliment: if I'm village and being lynched as evil, that means that the village thinks I'm a crafty enough eliminator to be playing village as well as I am as when I am village.

2. In a game where people are saying "shshshsh, xxxxxx, and Len are the most experienced players, so one is likely an elim", what that's going to cause me to do is take a look at the posts of shshshsh and xxxxxx and see which of them is more likely an eliminator, and push their lynch. If I can get an evil, that makes me soft-cleared and in a good position, and if I can't, then some of the blame's on me for not analyzing the eliminator out of the bunch. The only scenario where that's a problem is if in reality none of the three are evil.

3. I never said that we should lynch Orlok because he played a lot of games. I said that we should lynch him because Orlok and Lopen are both tacticians, and the elim team needs a tactician. If, say, Randuir were in this game, I would have put him on that list as well, despite him having many fewer games under his belt than Lopen or Orlok. So maybe it's a minor nit-pick, but it's not necessarily Orlok's experience that got him on my list, it's his track record of correct analysis.

4. I don't bargain on surviving games. In fact, according to your spreadsheet I'm on a six-game and counting death streak. The only active SE players that show as having a longer current streak are Rae and Winter. If I die to an elim fear-kill, that just means I took a bullet away from someone else who IMO could have done a lot more damage than me.

[little wilson]

9 hours ago, Elenion said:

3. I never said that we should lynch Orlok because he played a lot of games. I said that we should lynch him because Orlok and Lopen are both tacticians, and the elim team needs a tactician. If, say, Randuir were in this game, I would have put him on that list as well, despite him having many fewer games under his belt than Lopen or Orlok. So maybe it's a minor nit-pick, but it's not necessarily Orlok's experience that got him on my list, it's his track record of correct analysis.

Really? Link. Relevant bit:

Quote

Also, from my reads post I noted Jon, Sart, Orlok, and Elith as acting suspiciously. I'm going to support Lopen's vote on Orlok because every elim team needs an experienced player, and Orlok fits the bill, hasn't been targeted yet, and hasn't been openly suspected until now.

I'm fairly sure that's a vote on someone specifically because they are experienced and you think someone experienced must be in the team. Not tactician. Experienced.

[Sart]

20 hours ago, Elbereth said:

1. As I've already explained, I gave the role to Orlok because he knew the twist. As I recall, he came up with the twist, or at least a good part of the twists in this game. (There were initially more - I was playing with a Lover role that didn't end up working at all.) He's the one who pointed out that my current QF that I was planning on running was horrendously broken, and helped me come up with a new one. And he wanted to play. I didn't give it to him because he's my boyfriend, and am frankly offended that you would accuse me of doing so - I did it for the same reason Meta made Cessie the eliminator in LG2. Orlok knew, and he wanted to play, and he's the one who helped me make the game interesting and not horribly broken in the first place. I don't think it would be fair to ban him from playing a game which he helped to create when he wants to, and there is a way to allow him to play. 

2. I disagree about the square root rule - could you link me to Meta's proof, since I don't think I've ever actually seen it? It works, but only in non-extreme cases. A game of 9 players shouldn't have 3 eliminators. That's too many. They should have two. (Or even more extreme, 2 players out of 4 obviously makes no sense.) On the other end, 36 players should have more eliminators than 6. I think something like a 15-25% ratio makes more sense, personally.

3. Smaller number of eliminators have an advantage in less connections, yes. At the same time, it's just them. They can't afford to lose anything - a single mislynch could be devastating. In this game that's particularly so - even one lynch on Orlok would've essentially ended the game for him, leaving it up to Stink. And to win, Stink would've had to survive to endgame and get all but one player to vote on him (because he could kill them). That's why a couple extra cycles for the eliminators to get closer to a win seemed important to include. 

4. Could you please clarify what you mean by "you made this game broken on purpose"? I feel that I have made it very clear that I did my absolute best to make this game as balanced as I could. I didn't succeed in a few ways, as I've set out, but I did not set out to make a broken game, and I feel that you're disrespecting the work I've put into this game by saying that. 

5. The difference between a Serial Killer in a normal game, I'll note, and in this one, is that in this one the Serial Killer has to have a decent chance to win. Most don't have much chance at all - the only one I recall ever succeeding was Aman in QF14, and he gained the role late in the game. So comparing this game to a normal Serial Killer who can hide because the eliminators are the main threat isn't quite accurate, in my opinion. 

 

Lastly, regarding distributions. I agree in a standard game that roles shouldn't be chosen for player-specific reasons. It only becomes necessary in games in which certain roles enormously impact the outcome of the game, like this one. I would argue that that's an argument for more standard games, though, rather than no choosing players whatsoever. Even had Orlok not played, I would not have given Nale to Jebus, for instance. He notified me halfway through the game that he was going inactive, so he wouldn't have ever died to the filter, but the game would essentially never have ended until the village tore itself apart after who-knows-how-many cycles. That does not sound fun, to me. So it had to be someone who I felt reasonably sure would be active. In such a case (similar to LG21, which I referenced in Game Creation), I would not be comfortable with a game being completely random. Semi-random, yes - I'd probably pick a shortlist of players I thought would do and pick from that, rather than handpick. But I'm firmly of the opinion that totally name-blind games will lead to an increase in broken or less balanced games. I don't think that's a sacrifice I could support in order for games to be entirely random. 

Sorry about getting so heated.

1. I didn't realize Cessie knew about the twist. I thought she was just a random player, but I could be wrong about that. Still, I dislike giving out roles on principle, but agree to disagree.
2. Don't actually have a link. I think it was also assuming random voting and no scanners, which isn't accurate.
3. Agreed. They need some protection.
4. I was just mad that you gave a powerful role out. I don't like that on principle.
5. Very good point.
6. I don't think nameblind games are in-balanced. The main source of imbalance comes from people going inactive, right? No matter if the person is on the village or the eliminator team, it's going to be bad for balance. It might not be as bad for the village, but it still hurts their chances of winning.

[Elenion]

On 08/12/2017 at 10:47 AM, little wilson said:

Really? Link. Relevant bit:

I'm fairly sure that's a vote on someone specifically because they are experienced and you think someone experienced must be in the team. Not tactician. Experienced.

Huh. Looks like you're right on that regard, so I guess I should have clarified in that post that I correlated Orlok's experience with his accuracy and planning. I still hold that were Rand in this game he would have made the list.

[Elbereth]

On 8/12/2017 at 1:03 AM, Elenion said:

1. I've already seen something kinda like experience votes: being the votes coming on me from players who always suspect me. I'm not being targeted for my activity per se, but for my maneuvering in previous games, which comes from being an active player. While I can't say that I enjoy those votes, I take them as a sort of compliment: if I'm village and being lynched as evil, that means that the village thinks I'm a crafty enough eliminator to be playing village as well as I am as when I am village.

2. In a game where people are saying "shshshsh, xxxxxx, and Len are the most experienced players, so one is likely an elim", what that's going to cause me to do is take a look at the posts of shshshsh and xxxxxx and see which of them is more likely an eliminator, and push their lynch. If I can get an evil, that makes me soft-cleared and in a good position, and if I can't, then some of the blame's on me for not analyzing the eliminator out of the bunch. The only scenario where that's a problem is if in reality none of the three are evil.

3. I never said that we should lynch Orlok because he played a lot of games. I said that we should lynch him because Orlok and Lopen are both tacticians, and the elim team needs a tactician. If, say, Randuir were in this game, I would have put him on that list as well, despite him having many fewer games under his belt than Lopen or Orlok. So maybe it's a minor nit-pick, but it's not necessarily Orlok's experience that got him on my list, it's his track record of correct analysis.

4. I don't bargain on surviving games. In fact, according to your spreadsheet I'm on a six-game and counting death streak. The only active SE players that show as having a longer current streak are Rae and Winter. If I die to an elim fear-kill, that just means I took a bullet away from someone else who IMO could have done a lot more damage than me.

1. It's a compliment for a while - just like early deaths are something of a compliment. Eventually, however, they also just get irritating. And also, a difference between being voted for because of past game fear, and voted for because you're experienced and thus one of those players 'must' be evil. 

2. The important thing to note here is that there's a difference between lynching someone only because they're experienced, and lynching them using that as one factor in your suspicion of them. As far as I recall (and correct me if I'm recalling it incorrectly), you didn't have any other reasons for suspecting Orlok than that one of he and Lopen must be evil. While I don't think it's good to use that argument in any circumstance, it would be better had you had other points which can actually be argued against. Because there's no response or defence to that. So I don't have an issue at picking which players to analyse based on experience, but I do have an issue with that being the only basis with which to lynch them. 

3. A couple things here - one, you're implying that analysis makes someone a good tactician. Lopen and Orlok are both good at analysis, yes. Definitely. That's one of the main things that makes them feared players. But analysis and strategy are separate skills, and one doesn't necessarily lead to the other. Orlok is also a very good strategist, yes. Lopen is decent. As far as I can recall, I've never seen Lopen do any particularly brilliant strategising (though correct me if I'm wrong, Lopen, since I'm doing this off the top of my head). He's not bad at it, certainly, but I wouldn't rank him higher than someone like you, for instance. Lopen has a wonderful skillset, which makes him feared, but being a 'tactician' isn't part of that for him, at least to me. So I'm not certain how them both being tacticians is any better an argument than their experience. (Also, I don't think the elims necessarily need a good tactician. And even if you do argue that, there's a number of people I'd put into that category - Orlok and Lopen, yes, but I'd also put in you, Jondesu, Drake, Yitzi, and more that I don't want to go back and check in the player list. There are a few very good strategisers - Wilson and Orlok, for instance - but I don't think an elim team needs more than a decent one if any, which is most mid-level players.)

4. The issue isn't really surviving games. Most experienced players tend not to care anymore, because they can't. They always die. The point is that either they die because they're experienced rather than because they're actually suspected. Which is really irritating reasoning, because there's nothing they can do about it. Why would they join if they know that it's certain that even if they manage to survive to C3-4, they'll then get lynched simply because they're not dead yet and they're experienced? We've had players nearly leave the games multiple times because of the string of deaths they were experiencing and the knowledge that there wasn't any likelihood of that changing. A six-game streak is slightly different than spending two whole years without surviving a single game, for instance. 

On 8/12/2017 at 7:02 PM, Sart said:

Sorry about getting so heated.

1. I didn't realize Cessie knew about the twist. I thought she was just a random player, but I could be wrong about that. Still, I dislike giving out roles on principle, but agree to disagree.
2. Don't actually have a link. I think it was also assuming random voting and no scanners, which isn't accurate.
3. Agreed. They need some protection.
4. I was just mad that you gave a powerful role out. I don't like that on principle.
5. Very good point.
6. I don't think nameblind games are in-balanced. The main source of imbalance comes from people going inactive, right? No matter if the person is on the village or the eliminator team, it's going to be bad for balance. It might not be as bad for the village, but it still hurts their chances of winning.

1. She did, yes - Cessie/firstRainbowRose helped Meta come up with the game, and I think even came up with the Inquisitor role. 
2. Sure. I know Meta's fond of the square root rule, but I hadn't ever heard of an actual proof before and was wondering if there actually was one. 
4. Which is fair of you, and entirely within your rights. 
6. I'm not saying they're always imbalanced, only that they have a higher potential to be. And I'm not just talking about the eliminator team, here: I'm also talking about roles like Devotion in LG29, which was an important village role that Seonid and I considered important enough to select from a shorter list of players that wouldn't go inactive. 

[Elenion]

 @Elbereth @Sart In my opinion, choosing roles should be done nameblind and then balanced if necessary, but not handpicked. In LG36, for instance, the item drops were assigned by me with the names hidden. I could ensure that player 17 got a balance of both weak and powerful items, but I couldn't see who player 17 on the item sheet was. That way the items were done roleblind and we deemed the balance was okay.

I think what Sart would be really against is like LG35's neutral roles. I was hoping to get one of those, but they all ended up being given to the social circle of experienced players. I can think of a few other players, such as BR, who were also hoping for one, but didn't get an RNG's chance at it. I'm not in Hael's head, so I don't know why he did it that way, but if I was the GM I would have randomized all roles and then re-chose the parole officer and ringleader if need be.

[little wilson]

18 hours ago, Elbereth said:

A six-game streak is slightly different than spending two whole years without surviving a single game, for instance. 

I could not agree with this enough. And the thing is, the reasons why I was killed in every game I played in those two years wouldn't have changed had it been fifteen or forty. My fifteen game death streak could've easily been 20 or 30. I just stopped playing as frequently as I might've liked because I knew I would die - and not that I play to survive, because I don't. My death streak amused me until I hit 10-11, and then it just got annoying, because in every game I played, I was attacked by cycle 4. Every. Single. One. And if I didn't die, I'd die by the end, because I'd keep getting attacked.

My survival percentage was about 20% (for reference, at the time, the survival rate on average was closer to 33%, and every player people would put me on par with in terms of skill was around 30%). So yeah. A six game streak when you play almost every game that comes along is vastly different than a two year long streak. That's only a couple months. It'll pass.

[Eternum]

Wow.

Just.. Wow.

Wilson you have my eternal respect. A 2 year long deathstreak and you still kept playing, still kept being a part of the community...

There's something to be said about dedication here, but the fact that you're a mod speaks for itself, I think

Also, regarding the discussion going on here: Can't we all just get along? Sure, Orlok was OP. Sure, an elim Truthless was GOD *BLEEP* HORRIFYING (Props to whoever gets the joke), but I can tell you with certainty that as my first SE game, this was fun and engaging. It was a great introduction for me, and honestly I didn't care that I lost, by the end. All I cared about was the experience. Knowing that I need to step up my game to compete with y'all makes this so much more exciting. Thing is, you'll lose. You'll make mistakes. We're human, obviously. Whatever happens (and now I think I'm talking as much about SE as real life ) you just have to live with it, because you can't change the past. What you can change, though, is the future, by simply learning from your mistakes and helping others learn from theirs. Heated discussions are not the best way of accomplishing that. (I realize the discussion kinda cooled down now, but still. This post should have gone up much sooner, but I wrote 3/4 of it, told myself I'd finish and post the next day, then I forgot, 3 days passed, I remembered it today and now I rewrote it completely )

I won't talk about the balance or meta or anything because I don't know nearly enough to do so, but these were my two cents on the matter.

Edited August 15, 2017 by Eternum

[Orlok Tsubodai]

It's been more than a month since the end of the game, so this is well overdue being closed. Whilst the aftermath was contentious, I think most people greatly enjoyed the game, and would like to express gratitude to El for running a game with such an interesting twist, and to all the players for contributing to such an exciting game.

As always, if anyone would like to try their hand at running a game, just get a hold of Wilson, Alvron, Seonid, or myself. Not only will we get you added to the list (and the GM PM group), but we'll be more than willing to help out in any way we can. 

You can also post game ideas, ask questions, and get feedback from everyone over in our Art of Game Creation thread too. With all the games that we've run so far, we have plenty of experienced GMs that can help you refine any game you're working on.

Thanks again to everyone who played, and we look forward to killing seeing you in future games!