Boosting in-game activity: difficulties and solutions

[NerfedBuJ]

Low in-game activity is bad.

How can we encourage more discussion WITHOUT sacrificing post quality?

Increasing quantity is doable. Currently most mods don't, but it can be done. You just make stricter activity rules. Moe replacements, more modkillings.

However that will either just ruin games because of lack of players or too many modkills, or the thread is filled with nonsense and spam. Higher page counts, people unwilling to catch up, game suffers, and we're back to square one.

I was thinking what if we add a secret vote each night (or some other interval for nightless games), where 1 of the let's say 3 bottom posters have to be lynched. Each player send a PM to the mod with their choice (the idea is to choose who they feel is contributing less regardless of actual postcount), and whoever gets the most votes gets killed. (In case of a tie the lower post count would get killed).

Obviously this is too much power for scum as they will likely have enough votes to gain plurality in most cases. So I figured they just aren't included. Since the vote is secret the mod just contacts each townie individually and asks for their vote.

It's not ideal but we're a step closer I think. However we now have to tackle the problem of balancing, since this is now an extra kill each night. We can restrict it so it isn't every night by first having all players vote for if the activity level is satisfactory or not, but even if that would help, it doesn't solve the balance issue. You're still facing the potential for too many kills.

The other problem is post length differs from person to person. Someone who posts walls will always have less posts than someone who posts one liners. That is a fundamental difference in style and punishes one style for doing nothing wrong.

I know we can all just accepts it as a natural side effect of forum mafia, as we've all been doing anyway, but there's no harm in trying to come up with solutions. If the discussion doesn't lead anywhere we're no worse then when we started. So pitch in, all ideas are welcome.

[Ramcius]

You can't enforce higher activity with quality posts

If you want solve this, find a way to make people more interested in playing games they are in

[Lycanfire]

Spread a rumor that getting the high post # not only makes you lock town, but is also proof that you weren't born with a micropenis. If somebody threatens to take the top slot from you you're going to need to treat it like a personal attack. The meltdown you have in the thread should be worth a good 100 or more posts depending on how much you milk it.

Also because you are the towniest player with a magnum dong you are going to die on N1, so it's important to get your first thousand posts out of the way on day 1, because people are petty enough to kill you for that reason alone.

[MathBlade]

Low in-game activity is bad.

How can we encourage more discussion WITHOUT sacrificing post quality?

Increasing quantity is doable. Currently most mods don't, but it can be done. You just make stricter activity rules. Moe replacements, more modkillings.

However that will either just ruin games because of lack of players or too many modkills, or the thread is filled with nonsense and spam. Higher page counts, people unwilling to catch up, game suffers, and we're back to square one.

I was thinking what if we add a secret vote each night (or some other interval for nightless games), where 1 of the let's say 3 bottom posters have to be lynched. Each player send a PM to the mod with their choice (the idea is to choose who they feel is contributing less regardless of actual postcount), and whoever gets the most votes gets killed. (In case of a tie the lower post count would get killed).

Obviously this is too much power for scum as they will likely have enough votes to gain plurality in most cases. So I figured they just aren't included. Since the vote is secret the mod just contacts each townie individually and asks for their vote.

It's not ideal but we're a step closer I think. However we now have to tackle the problem of balancing, since this is now an extra kill each night. We can restrict it so it isn't every night by first having all players vote for if the activity level is satisfactory or not, but even if that would help, it doesn't solve the balance issue. You're still facing the potential for too many kills.

The other problem is post length differs from person to person. Someone who posts walls will always have less posts than someone who posts one liners. That is a fundamental difference in style and punishes one style for doing nothing wrong.

I know we can all just accepts it as a natural side effect of forum mafia, as we've all been doing anyway, but there's no harm in trying to come up with solutions. If the discussion doesn't lead anywhere we're no worse then when we started. So pitch in, all ideas are welcome.

Here’s the issue with that.

You won’t if ever lynch scum with that unless scum want the buddy dead.

What in general other sites have done is a slank vig which while not perfect can be town or scum aligned and can just kill the lurkers while giving information to go off of.

[MathBlade]

Spread a rumor that getting the high post # not only makes you lock town, but is also proof that you weren't born with a micropenis. If somebody threatens to take the top slot from you you're going to need to treat it like a personal attack. The meltdown you have in the thread should be worth a good 100 or more posts depending on how much you milk it.

Also because you are the towniest player with a magnum dong you are going to die on N1, so it's important to get your first thousand posts out of the way on day 1, because people are petty enough to kill you for that reason alone.

No penis jokes please.

[popsofctown]

I don't like the angle of adding a "vig the lurker mechanic" into games. To me it seems like you want to modkill people for low activity and you're shirking the responsibility for taking that drastic step onto the players. You can't really say "it's going to be baked into the balance of the game". Is it going to be a system that players can use to help their faction win? Then it won't improve activity levels, it will only be used to help their faction win, towns will shoot wallposting scums.
Is it going to be a system carefully designed so it's too hard to leverage it for a factional win? Well then, it's going to kill players, removing them from the game, independent of their alignment, voting record, or stances. It's essentially a modkill. The only difference is you're having the players manage it. Which, like asking for volunteers to post votecounts, is just not as good as modding your game yourself.

I am not taking the position that a modkill is too drastic a step to improve activity levels, I am just equating these two things. Prevailing popular sentiment is that a modkill is too drastic a step, though, with force-replace being more controversial.

I think it's better to focus on ways to incentivize low-activity posters to post, and ways to prevent low-activity players from entering your games in the first place. Focusing on mechanics the entire town uses together publically could be a good way to foster activity, as seen in a lot of open setups people kick around. Traditional PRs and VTs setups risk making a lazy player a VT and doubling their laziness. If powerful PRs that make them want to play like a survivor aren't even worse.. I can't decide.

Preventing low-activity players from entering games in the first place is controversial and high risk high reward and has been discussed before and I can appreciate by talking about baking things into the setup you're trying to get an end run around that angle. But I really think it would end up being a reflavored modkill, and the ranking seems pretty obviously "I'm not gonna let you /in >>>> I'm going to force replace you >>>>> modkilling your slot"

There's also the concept of instituting a maximum number of games a player can play, to discourage people from spreading themselves too thin. People don't want it queue-wide but an individual mod could adopt it if they really wanted to. "You can't /in to my game if you are already in N other games. If I notice you join other games before you die in the game and reach N, I will modkill you." There's definitely reasons to be against that, some players play lots of games and handle them all.

And to reiterate, game activity is important, and if you wanted to just modkill people for being low activity in your games, like, you can do it. I would recommend giving the PRs backups in such a game. It's heavyhanded and draconian and some people might not want to join that game, but some people would.

[Psyche]

make your games interesting
design your setups to facilitate engagement and even drama
feel too many focus on balance or thematic fit or novelty without giving much thought to player experience beyond "will they riot"
it's attractive to put all the burden to make a game fun on your audience, some mafia equivalent of giving kids a cardboard box and telling them to use their imagination but the community is so broad and diffuse that this is a sort of russian roulette
how do you design games to be reliably exhilarating? that's the big q, not this

[Lycanfire]

I'd say the problem isn't activity. It's readability.

A spam poster is hard to read because they're talking about a bunch of meaningless shit.

A wall poster writes a novel before giving the most important info.

But wait, there's more!

There's a particularly slimy kind of poster I'll introduce you to. You've probably met them out in the wild. They're called NAI posters. A NAI poster is so bad at playing scum (imagine this in the current year ) that they will cleanse any of their town posts of anything interesting or readable in order to further their scumgame.

Lastly there are lurkers, that either shouldn't have been in the game, aren't being allowed to get into the game, or don't know how to play and certainly haven't bothered to ask for help.

In the end it's about getting what you want. People don't necessarily want 100 posts of everyone each day. They want information. It would be a better idea to call for a meta shift where readability becomes king. Instead of respecting every special ray of sunshine people should accuse others that make it hard to form reads on. Townies will adapt and give people want they want. Scums lose a tool and have to perfect a new one.

[popsofctown]

Well the problem is that it's not optimal play to lynch a NAI player if there is a scummy player in the thread.

Policy lynching NAI players is playing to your wincon if no one in the thread has scumtelled, then it becomes arguably optimal, other players will become more readable on future days and the NAI player will not. But if someone has scumtelled you can't policy lynch without violating a common rule most mods post, "play to win the game".

For some players policy lynching won't cause an adaptation because the reason they are adopting the NAI playstyle is because they don't care that much about winning. Causing them to not win is not the right stick. You need a carrot. Or a blacklist.

[Lycanfire]

I'm not suggesting a policy lynch. I'm suggesting a meta shift. Identify a set of behaviors as scummy. Accuse player of being scum. They either stop roleplaying as scum or make friends with rope.

[Psyche]

meta shift requires changing people's mind about what's scummy which itself requires the presence of evidence that said behavior is scummy

[Lycanfire]

I'm sure people have scumread obfuscation. I certainly have before.

The only new part to it is that I'm suggesting to categorically group all kinds of obfuscation together as scummy.

Give people posts to care about. See if more people want to play.

[Psyche]

still missing the part where you evince as much

[Eddie Cane]

ignoeing outside factors (ie a cop guilty), if theres a player thats statistically proven to be unreadable, its actually optimal to never lynch them

but on the topic of your thread, the only real way to fix this is to change site culture. theres better mafia sites where this happens rarely if ever. a second lynch is dumb, and anybody whos willing to sign up for a game where low activity people get lynched twice as often is probably not a low activity person.

[Lycanfire]

There's a difference between "has tells" and "says anything that you could possibly read into".

Compare:
"I think there's scum in A, B, C, and D."

and:
"A is obvious scum. Don't believe me? Post 666 is never made by town A. This is today's lynch, and if it isn't, it's tomorrows lynch."

What's more helpful if A flips scum?

[Lycanfire]

psyche i agree everyone is running boring games so please don't take your frustration out on me. not only am i blameless but i was born innocent and free of sin.

[RadiantCowbells]

if theres a player thats statistically proven to be unreadable, its actually optimal to never lynch them

: t h i n k I n g :

[Psyche]

i really just think setup design is the problem and that telling people to read stuff as scummy bc thinking thst stuff is scummy will help site meta is telling people to play against their win condition

[popsofctown]

if theres a player thats statistically proven to be unreadable, its actually optimal to never lynch them

: t h i n k I n g :

If you townread everyone else in the thread, don't you PoE lynch them?

[Ramcius]

if theres a player thats statistically proven to be unreadable, its actually optimal to never lynch them

: t h i n k I n g :

He's right, you vig them instead

[callforjudgement]

Callforjudgement's Micro Deadlines were invented to solve the problem of low activity in Micro games. There's an extensive discussion of all the reasoning behind them on the wiki page, but a summary is "prod quickly, be reluctant to replace, deadlines are initially short but players can extend them".

They've been through a few iterations now and seem to work pretty well. I'm not sure whether they'd apply well or not to larger games; it might be worth experimenting. (My guess is that they'd solve the problem that they were designed to solve, but it may not be a relevant problem for Larges, and thus different solutions may be needed due to the different problems.)

[TDC]

For some players policy lynching won't cause an adaptation because the reason they are adopting the NAI playstyle is because they don't care that much about winning. Causing them to not win is not the right stick. You need a carrot. Or a blacklist.

Pretty sure the site rules say you need to play to win. Obfuscating one's town play to help future scum play is the definition of not playing to win.

The rule is just hard to enforce without ruining the game for everyone else (same goes for policy lynch in non-larges).

I'm in team blacklist.

[TDC]

One could also generally have one more signup slot than actual number of roles, and have the cast vote out (non-public obv.) one player before the game is rolled.

I realize this may have undesired side effects, if the choice is not based on "do I think this player is playing to win", but "do I like this person".