Sanderson Elimination: Questions & Answers and Game Meta Discussion

[Elbereth she/her]

Here’s a few quotes (from various posts in this discussion) from Yitzi about game balance, which caught my attention: 

Quote

2. The strategy/battle of wits.  For this, I think our primary goal needs to be that the game should be balanced.

iii) People who enjoy the strategy (and therefore need a balanced game) feel that the game is not balanced because one side has significantly more non-optimal players (weighting for non-optimality and proportionate to the side's size) than the other.  The only way I can see to resolve this (though I'd like to hear if anyone else has ideas) is for those players who play non-optimally to be known, and GMs make sure to consider that when balancing sides.

d) How should we make sure that games are balanced (and thus fun for the strategizers) in light of the fact that some players are significantly more likely to play non-optimally due to finding it fun than others are?

3. People who play in a more relaxed manner making the game less fun for those who prefer a more optimal playstyle.  This gets tricky, and in extreme situations may require the use of publicly known handicaps.  (For instance, if a player likes to do crazy things sometimes, they might ask for a public bonus role requiring them to take an interesting but suboptimal action every so often, and then when figuring out team numbers they might only count half.)  After all, playing around suboptimal players is a lot more fun for the optimal-style players when the suboptimal play is an official part of the game.

I am looking primarily for a game in which I can attempt to win, and feel confident that the game gives me a fair chance of achieving this.  When I say a "fair chance", the distribution of extremely skilled players and newbies may or may not be taken into account (i.e. if Aman and Wilson are both elims and the elim team is smaller as a result, or Aman and Wilson are both elims and the elim team is not smaller as a result, either way is ok), since those can be compensated for by helping newbies and trying to improve to match the top players, but players who choose not to try to win need to be accounted for in alignment distribution in order for me to feel that I have a fair chance.
I am also interested in RP, and would be interested in an SE "game" that is so RP-heavy that there's no possibility to try to win, but as long as it's a game where "trying to win" is a meaningful concept, I cannot have fun unless I can make a fair attempt to win.

 

One thing that hasn’t been addressed, which I consider important, is the GMing aspect laid out in the points above. 

The main assumption being made here is that GMs don’t already balance games based on playstyle. I’m… not sure where you got that impression. I’ve balanced/distributed 8 games by my count (soon to be 9), which I believe is equal to the most of any other GM on this site. I’ve done a lot of balancing. 

You’re underestimating how difficult balancing is. It depends on the game (QF11 with three roles total was a lot easier than LG26 with each player having three win conditions all of which interact), but it’s not exactly easy. The most obvious place that playstyle is balanced for, naturally, is people who are typically inactive. Eliminator teams full of those players are never a good idea, which is why they never happen (except in cases where someone is unexpectedly inactive - QF16, for instance, had Joe, Aman, Cloudjumper, and Winter Cloud. Aman was lynched before he was able to get online D1, and Cloud and Winter were basically inactive, leaving Joe the only active elim. If Aman hadn’t died, there wouldn’t have been an issue there at all.). 

We also balance for other types of playstyle. (Side note: you’re also assuming that GMs need a list of players who play suboptimally. Almost always, they’ve played a number of games (and often have also co-GMed) before they GM - simply because it takes so long to get up one of the GM lists and run a game. They know the playerbase already. They don’t need to be told that kind of thing.) People who are likely to go wild with them are less likely to get kill roles, or if they do that’s accounted for. People who don’t have issues with dictatorships are less likely to be seekers (at least in my games). Et cetera. Eliminator teams always have at least one experienced player on the team, or enough good middling players to make up for that (except in AG2, which was a larger team of almost entirely the newest players in the game, because Meta is a troll). When you say, then, that we don’t balance for suboptimal playstyle, we most certainly do. And that’s in addition to balancing roles (which is annoying in itself), experienced players, and so on. 

Furthermore, looking through the SE Stats Spreadsheet, you’ve never played a game broken through distribution. We have had those, although they’re unusual (13 of 87 games, though about 5 games aren’t yet marked). In fact, of the completed games you’ve played, only one has been broken at all - LG35, through mechanics, and that only affected the neutrals, not you. So it makes sense that you would take the good balance of recent games for granted. Individual games can have balancing problems, yes, but I don’t see that as such an issue that it needs special attention to be solved. 

Please don’t. Balancing is hard and annoying. Until you’ve balanced one yourself or played through a game that wasn’t balanced properly, try not to presume that you know how balancing works, because it feels to GMs (at least, to me) like you’re not appreciating a lot of the work we put into making games fun for everyone. Kolo? 

Something I haven't seen brought up in any discussion yet is action restrictions: I like games with no action limit, so you don't have to give up good actions just to use necessary actions. If you have a role and two items I think you should be able to use all three, not just one.

Now we're getting into the really small preferences, but I think GMs should experiment more with non-standard role distributions. There are some notable examples of where GMs have broken the role distribution norms, but for the most part elim teams have been very predictable role-wise.

Interesting. You’re conflating role distributions overall and those for eliminator teams - there have been plenty of interesting overall role distributions over time, though those are limited by types of roles. (For instance, take the AGs - a Rioter/Soother/Smoker heavy game has to have all of those three. You can do that, you can do Lurcher/Coinshot heavy, or you can do normal, or mostly vanilla. Those are kind of the main four distributions, and you can’t vary much further than that in such a setting, though obviously different roles can be changed slightly more.) For eliminator teams, I agree that more variety could be done. At the same time, there is a normal eliminator distribution for a reason - mainly, no more kill roles than the natural one, and usually a protect when there are other kill roles in the game because that’s really nice for elims to have (particularly since vigkill success rate is significantly higher than the lynch rate). 
Also, what are the ‘notable exceptions’ you’re thinking of? I’m not sure I can recall any particularly notable role distributions except AG2/3 with the amount of vote manipulation on the eliminator team. Unless you mean player distribution instead? 

That’s the entire point of action restrictions. Forcing you to choose which to use, giving you more options so that you can figure out which is best to use each turn. The optimal action, as it were. No action restrictions can be more fun, yes, in games where there are lots of roles, but I disagree that action restrictions are bad. They force you to make sacrifices. They give you choices so that each player when given a set of actions could play the game differently because they prioritise differently. (Also, if someone has a number of powers and others only have one, it makes the first much less OP compared to everyone else.)

[Elenion he/him]

@Elbereth AG2 was the game I had in mind when I mentioned games with unexpected distributions, and I just generalized that because I've only seen or heard about half of all SE games.

[Yitzi2]

8 hours ago, Elbereth said:

Please don’t. Balancing is hard and annoying. Until you’ve balanced one yourself or played through a game that wasn’t balanced properly, try not to presume that you know how balancing works, because it feels to GMs (at least, to me) like you’re not appreciating a lot of the work we put into making games fun for everyone. Kolo?

Got it.  I'm wondering if you know of an example where the elim team was larger than usual due to having a large number of likely-to-be-inactive or otherwise highly suboptimal players?

Edited July 26, 2017 by Yitzi2

[Elbereth she/her]

Just now, Elenion said:

@Elbereth AG2 was the game I had in mind when I mentioned games with unexpected distributions, and I just generalized that because I've only seen or heard about half of all SE games.

Sure. There have been others - LG14 had a fun role distribution with the Lifeless, as I recall - but you're correct that role distribution isn't played with as much as player distribution. That's partly because when people want to play with roles it tends to be through new ones or an interesting set in the rules themselves, but we definitely shouldn't leave out that aspect. Thank you for pointing it out. 

[King Cole he/him]

How long do games last on average?

[Elbereth she/her]

Depends on the format. Those labeled as Quick Fixes usually last about a week to 10 days. Mid-Range games last roughly two weeks, and Long Games are a month or longer, generally. 

[little wilson she/her]

2 hours ago, Yitzi2 said:

Got it.  I'm wondering if you know of an example where the elim team was larger than usual due to having a large number of likely-to-be-inactive or otherwise highly suboptimal players?

I did that with MR10. 36 players with 8 eliminators, 2 of which I was near certain would go inactive for at least half the game. I was right too. Not that it helped the village because even though they had 3 kill roles, they never killed the inactive eliminators and one of them came back on the last turn. (Though the village made a number of mistakes that game, like letting a publicly known elim kill role survive for a full two cycles after he'd claimed, which conveniently was just long enough for the elims to take control and beat the village >> )

[Steeldancer he/him]

2 minutes ago, little wilson said:

I did that with MR10. 36 players with 8 eliminators, 2 of which I was near certain would go inactive for at least half the game. I was right too. Not that it helped the village because even though they had 3 kill roles, they never killed the inactive eliminators and one of them came back on the last turn. (Though the village made a number of mistakes that game, like letting a publicly known elim kill role survive for a full two cycles after he'd claimed, which conveniently was just long enough for the elims to take control and beat the village >> )

8 elims?! Holy cow! That's a lot of elims! I'm totally reading that one later. 

[little wilson she/her]

It was mostly because in one of the games that had just barely ended, someone made a comment about no game ever having more than 6-7 eliminators, and I thought "oh? We'll see about that..." So when I got a 36 player game....well. I certainly had a bit of fun.

[Jondesu he/him]

2 hours ago, little wilson said:

It was mostly because in one of the games that had just barely ended, someone made a comment about no game ever having more than 6-7 eliminators, and I thought "oh? We'll see about that..." So when I got a 36 player game....well. I certainly had a bit of fun.

Now you're making me want to troll and have a game where the majority is "elims", but the village roles are the ones with real power, and maybe even give the village a doc and not the Elims…oh the fun.  Just completely reverse their expectations.

[StrikerEZ he/him]

6 minutes ago, Jondesu said:

Now you're making me want to troll and have a game where the majority is "elims", but the village roles are the ones with real power, and maybe even give the village a doc and not the Elims…oh the fun.  Just completely reverse their expectations.

Oh, I'd love to play this.

[Elbereth she/her]

8 minutes ago, Jondesu said:

Now you're making me want to troll and have a game where the majority is "elims", but the village roles are the ones with real power, and maybe even give the village a doc and not the Elims…oh the fun.  Just completely reverse their expectations.

While that would be amusing, that'd quickly and obviously turn into a normal game where everyone realises what's going on. 

Also, eliminators not having a doc is a really tricky thing. Be very careful with that kind of mechanic.

[Yitzi2]

1 hour ago, Elbereth said:

While that would be amusing, that'd quickly and obviously turn into a normal game where everyone realises what's going on. 

Also, eliminators not having a doc is a really tricky thing. Be very careful with that kind of mechanic.

Though it might be interesting in a game with limited PMs, especially if elims have the ability to alignment-scan people they have PMs with.

But yeah, if the elims have no way to identify each other, then they won't really function as an elim faction.

[Alvron]

3 minutes ago, Yitzi2 said:

Though it might be interesting in a game with limited PMs, especially if elims have the ability to alignment-scan people they have PMs with.

But yeah, if the elims have no way to identify each other, then they won't really function as an elim faction.

They can have PMs.  There have been a couple of games where they knew each other but had no Doc.  It was a fun challenge.

[Yitzi2]

19 minutes ago, Alvron said:

They can have PMs.  There have been a couple of games where they knew each other but had no Doc.  It was a fun challenge.

If there are unlimited PMs, then you could easily end up with group PMs that might as well be a doc.

There's also fun to be had by having most, but not all, of the people with access to the doc be elims, and most but not all of the people without access to the doc be village.  (This also lends itself well to games with a conversion mechanic.)  Of course, there would need to be some mechanism to discourage just lynching everybody in the doc.

[Elbereth she/her]

Uh. Yeah. That's... eesh. Essentially, that removes any use out of the doc whatsoever - members would either not name themselves at all or not talk. The point of eliminators having a doc is unlimited conversation without the village seeing. It essentially removes all the use of that doc for them. 

[Metacognition he/him]

This is getting into a conversation that would fit better in AoGC. At this point, it's less about the meta and more about how such a mechanic would work. And I'd hate to lose the discussion we were having, unless everyone feels like they've said their peace. 

[Megasif he/him]

Btw, have we not got a discord channel?

[Jondesu he/him]

12 hours ago, Elbereth said:

While that would be amusing, that'd quickly and obviously turn into a normal game where everyone realises what's going on. 

Also, eliminators not having a doc is a really tricky thing. Be very careful with that kind of mechanic.

 

11 hours ago, Yitzi2 said:

Though it might be interesting in a game with limited PMs, especially if elims have the ability to alignment-scan people they have PMs with.

But yeah, if the elims have no way to identify each other, then they won't really function as an elim faction.

 

11 hours ago, Alvron said:

They can have PMs.  There have been a couple of games where they knew each other but had no Doc.  It was a fun challenge.

Definitely would have to work on that awhile. I know in MR17, we started out with the name of one other Elim and a PM with them (sort of round robin, so we knew two others right away actually), but group PMs were allowed so we set one up quickly (then ended up with others when we included Lopen, who worked with us, and then excluded Len, who was betraying us). I preferred the group PM over a doc, personally.  If I didn't allow group PMs, though, inactive Elims might cause big issues by breaking a chain of communication.

[StrikerEZ he/him]

1 minute ago, Megasif said:

Btw, have we not got a discord channel?

We do, in fact.

https://discord.gg/xuMqRD

[Megasif he/him]

16 minutes ago, StrikerEZ said:

We do, in fact.

https://discord.gg/xuMqRD

Oh thanks. I was expecting a channel on the 17th shard server. All this time. Sigh.

[StrikerEZ he/him]

2 minutes ago, Megasif said:

Oh thanks. I was expecting a channel on the 17th shard server. All this time. Sigh.

You're welcome! And it's fine lol

[little wilson she/her]

Quote

The question at hand is: How do we ensure a forum that is fun for everyone?

What are you trying to get out of an SE game, to the point that it's no fun if it's not absent?

I think the question at the heart of this discussion is rather “To what extent should games be competitive?” You also suggest that we should have rules in place to control the behavior of players, which I disagree with vehemently.

I’m pretty sure I’ve answered the question, but just in case that’s unclear, let me reiterate. I do not think the focus of these games should be on competition and winning. It should be on fun, camaraderie, and the community.
 

As for the second part, I already mentioned that the moderator team generally prefers a hands-off policy for most things. We believe that it’s better for the players to influence and change the metagame than for rules to force a change in the metagame. When the players bring about the change, it’s more likely to last, but rules not only limit player choices, but they also tend to be irritating from a player point of view, in having restrictions. Additionally, it’s difficult to enforce in grey areas, and even in the black/white, it can take a lot of work.

The moderator team has three people on it, and all three of us are busy. We are also volunteers. While moderating the games is important for us, there are many other things that are just as important or more important in our lives that we have to deal with first. Like paying jobs or our mental health.

As Meta already mentioned, the General Rules and Etiquette Policy should be more than enough to direct players in what is acceptable. The fewer additional rules we have to enforce to make the games run smoothly, the better things will be.

 

Also, the question you say we should be answering implies that we’re not already making the games fun. But why would people sign up and continue to play at all if they didn’t find the games fun? All of our regular players are ones that don’t prioritize competition, so clearly competition isn’t the driving motivator to play SE.

Yitzi, why do you think controlling players’ playstyles, and therefore limiting their options, will make the games more enjoyable?
 

Speaking of control, let’s move onto mayors/dictators. There is a difference between a leader of a village and a dictator. Not all leaders are dictators. A leader who listens to other ideas, giving people information they need to make decisions, while giving suggestions for role actions, but ultimately letting each individual player play how they want to play is not a dictator. That is not a mayor. A dictator is someone who controls the village. Someone who expects everyone to roleclaim to them, listens to no one except their small cabal of trusted players, and threatens other players with death if they don’t follow them exactly.

LG4 had a dictatorship. In one cycle, one of the players crucial to the village died, but there was a role, the Returned, who could sacrifice themselves to bring a very recently deceased player back (they learned the list of the dead 12 hours before the end of the night). The Returned in the dictator’s cabal threatened another Returned with death if he didn’t sacrifice himself for that crucial villager. So he did. Because either way, he was going to die.

 

Moving on. Rand’s post reminded me of something that I think bears mentioning here, since it answers more about competitively-geared playstyles.

I recently drafted Magic: the Gathering with Brandon Sanderson and six other people. Magic can be a very competitive game, but Magic with Brandon is very casual. Now, I haven’t played Magic for a good year, and I’m novice anyway, so I was pretty fuzzy on some of the rules, like if you look at your hand, you’ve basically said that you’re going first. Usually, this wouldn’t matter at all at Magic with Brandon, but one of the people there was very rules-based. He was also very loud about the rules. This is not how things generally are there, even though that’s how most Magic players would be when playing.

Additionally, I have this irrational fear of playing Brandon in Magic, because I know he’s very good and like I said - I’m very novice. I worry that when he whoops me, I’ll have made an utter fool of myself, and he’ll think I’m an idiot - which would be the case if he were a generic Magic player. However, if Brandon plays and beats a novice, he takes a look at their deck, helps them fix it, and then gives them pointers to improve not only their general Magic play but also their game with that specific deck. He will never be like “You sucked,” and move onto the next person.

Just because a game is generally seen as competitive doesn’t mean it always has to be played competitively. I personally feel like that casual play Brandon does with Magic is kind of the perfect example of how we should play mafia.

 

Also, the problem isn’t one person playing over-competitively. It’s players in aggregate playing competitively. If it were only a single player, that one person is unlikely to have a large impact on the metagame, even if they’re being super competitive. But take a number of players all playing competitively - even if it’s not strongly so - and they will have a significantly bigger impact on the overall tone of the games. It’s like when I was trying to fix the metagame by myself a year ago and wasn’t having any kind of impact, but when 4-5 of us all joined efforts for the same goal, we made progress and things got better.

I do not remotely believe we should cater SE around people who wish to play competitively, or group “suboptimal” players differently. Also, I don’t like calling players who prioritize fun (however they define that) “suboptimal.” They may be playing suboptimally from a competitive standpoint, but that does not make them suboptimal. Fun would be a better term. Fun players vs Competitive players. Our goal here in SE isn’t to play competitively. It’s to play a fun game of mafia with our friends, because killing friends is fun.

I’d like to reiterate once again that if the games did become that competitive sanctuary some of you seem to want, they’d be shut down. You would no longer be able to play competitively here because there would be no more games to play. If you want ultra-competitive play, go to Mafia Universe, Epic Mafia, Town of Salem, or really almost any other mafia site. There are more than enough to choose from. This forum is a haven of civility and tolerance, where the members care about each other, and it would be a real shame if those of us who enjoy being Gentleman Killers were suddenly unable to play.

[DrakeMarshall he/him]

If by competitive play you mean putting winning over fun, I wholeheartedly agree that we should discourage such behavior.

The point of any competition should ultimately be to have fun. So if competition doesn't produce fun, it's kind of pointless.

[little wilson she/her]

2 hours ago, Drake Marshall said:

If by competitive play you mean putting winning over fun, I wholeheartedly agree that we should discourage such behavior.

The point of any competition should ultimately be to have fun. So if competition doesn't produce fun, it's kind of pointless.

Winning over fun and also encouraging optimal play from all players.

Magic: the Gathering has specific archetyes I think work well to show exactly what I mean. There are three basic types of players: Timmy, Johnny and Spike.

Timmy plays to have fun. He likes things to be interesting and the game is less about the win than it is about how the win came about (note: I actually like this description of Timmy better than the other one, but the other one is at least a start).

Johnny likes combos. He likes gambits and messing with the rules. He plays to get creative with the game.

Spike plays to win. He plays to compete. He cares very much about if he's winning or if he's losing and that factors into his ability to have fun.

My post isn't trying to say that we prefer Timmy and Johnny's and Spikes can just play elsewhere. On the contrary, we have a number of long-term Spike players. There's nothing wrong with Spike. The issue is when Spike players try to get non-Spike players to be a Spike. That's the whole bit with "optimal" play. Spike is really the only one that cares about optimal play in terms of winning. Timmy and Johnny have other ideas of what is optimal play, because it's different for them.

This is why we don't want to ban any specific play style because some people are able to play them in a way that's perfectly fine for this forum, because they're not trying to pressure everyone else to play like them. And that's the important bit: play how you want (within the forum/subforum rules) but don't try to push for your playstyle above others. Let everyone else play how they want to as well, even if that doesn't include as much of a focus on winning as you might like.