Increasing Player Activity / Density
Posted: Sat Jul 25, 2015 6:04 pm
Much of me rebels at the notion of writing this post. Windy public posts are generally a bad way to make things change. I will try to structure this one differently, and hope that I get different results.
But we get these monthly-to-quarterly windy posts asking ALFA to restructure itself, and my response is always the same: read the post; parse the activity logs; look for similar things we've done in the past; decide whether it's worth the invested time. The answer is usually no, and I thus really don't want to have people trying to make me do it. Or the answer is that I don't care enough to comment more complexly than to say how possible it is.
Now I admit that there's some assumptions in there, but I hope that people can at least understand my angle. Yes, I assume that people who play more are happier with the game, and that's not necessarily the case. Plenty of emotions can prompt people to play-- but I'm hopefully not too far out of line to say that greater activity, in aggregate, is happier people, in aggregate, even though individual circumstances may be significantly different. I also hope that I'm not too far out of line to say that happier people is a more-successful game.
And so I'm left with some broad statements that I can make-- if you don't care about policy, this section is skippable. I finish this with things that normal folk can just do.
Things We've Tried That Don't Work
I can be more confident in these statements than the others; it's true that correlation doesn't imply causation, but the absence of correlation does imply the absence of causation.
Making the Game More "Pure" -- yeah, either nobody cares enough to change their behavior or they do and the change is a ragequit. Our repository is full of examples of us fixing things that are just not in NWN2 what PnP does, and went in service of the Charter's mission statement (a persistent world o' D&D in Forgotten Realms). Activity patterns seem to not change. Banning stuff tends to be unpopular. Lethality tends to be popular in design and unpopular in practice. And yes, you can find many examples of me doing this a few years ago. It didn't work; that's why I stopped. Sorry about that.
Relaxing Travel Rules / Trying to Give People More Options to Gather -- this includes our multiple PCs rule. It doesn't seem to make people play any more (or, if it does, it's trivial compared to the other forces that make our population fluctuate). I can't find any evidence that it hurts any thing, either, to be fair; I'd guess (of course, put a little [citation needed] here, this part is conjecture) that it's because making travel easier and giving people more PCs also makes it easier to scatter.
Giving People More Rewards -- I have to distinguish this from giving people more content: this is more about taking existing content and/or events and just giving more loot and XP for them. We've had DMs get allowances to fart truly sickening quantities of XP on players before, so we have a pretty-wide range of trends and players to look at, which we can't really parse numbers from. Folk playing with Wynna or HEEGZ as DMs tend to play a lot. Folk playing with me or Xanthea as DMs don't. We wouldn't expect these patterns if XP/loot was a noticeable motivator.
Cutting Space -- disclaimer, of course, is that only one of our servers has undergone significant pruning during a time when we have worthwhile logs to measure outcomes-- that'd be TSM. That incident didn't do anything. Maybe we could do it again and better to get better results.
Things That Seem To Have Worked Before
That "seem" is important. I can guess at mechanisms of causation, but we don't have the resources or population necessary to do some proper double-blind evaluation that could actually demonstrate those mechanisms. I'm mostly interested, of course, in things we've tried more than once. Once might be a fluke; several times is probably (but not necessarily) something more.
Advertisement -- seems obvious, right? The largest and sharpest incline in our activity came from us posting about our world on other forums. We saw clear and pretty-immediate results from just emailing old members. These aren't very longlived boosts, of course (to be fair, nothing is; managing a game is a constant fight against attrition), and I'm sure there's some point at which promotion will result in exhaustion in our potential audience. I'd let people who are more familiar with advertisement talk about that. I'm much more familiar with watching game metrics to study how game changes make people use the game differently.
Adding Content -- our most-sustained increase in activity, and also our highest level of activity, was associated with BG acquiring its more-diverse herb spawns, its hide drops, its mail quests, goblin ear bounties, and its billion isolated fetch quests. There's plenty to criticize in the design of these quests, of course, but they did a few things really well (namely, they required that people be at least minimally aware of their surroundings and kept pointing them in opposite directions down the same road). This does have to contrast with some other instances of content-- like TSM has a lot of content. More than BG did during its peak, I'd estimate, but most of the additions of that content don't seem to be associated with differences in activity (asterisk here, see two points down). My guess is because it has a lot of nodes and doesn't seem to be very good at putting random players in the same areas.
Building Player Groups -- ALFA's also seen really successful pockets of activity coming from players who are just good at making player groups. The most successful iterations have a builder and a recruiter in their membership (sometimes they're the same person; doesn't have to be). I can draw some parallels to the previous point as to some probably-effective mechanisms here, but I haven't actually been in a successful player group here, so this is an indirect observation: but player groups usually settle on a place as "home" and tend to return there after adventure, leaving them with an obvious place to find other people.
Plots With Out-Of-Session Elements -- while not as strong an effect as advertisement, we have seen abrupt and measurable increases in people just being on and doing stuff when DMs run plots and then leave things that have to do with their plots (and impact their plots) just... out there... for anyone to deal with. Some examples would be Wynna's Dean's Quest plots (which actually gives us two discernible data points, because she did it, went inactive, and then came back and did it again) or HEEGZ's demon gnoll plot and its reprise.
Stuff You Can Do
Recommend ALFA -- ... presumably you would? Folk are here playing. There must be a reason. Just telling people about the things you like potentially introduces them to an enjoyable hobby, and probably reminds you of the things you like about this place. It's easy to get too deep in the problems, because they demand more of our attention.
Foster Player Groups -- making a new character? Ask to attach to an established character, or to an established group, or both. Got an existing character? Welcome those new characters. Builder for a player group? Make sure you've got enough space for new players to show up.
Include Non-Session Elements in Your Plots and Invite Everyone -- is your plot about some brutal invasion? Maybe a lieutenant is camped somewhere on the server, and he's going to make a mess of things next session if nobody does anything about that. Or maybe the backdrop of the session just happens to include an infestation, and its server coverage determines how difficult you make the fights during the session. Is your plot a mystery? Maybe there's consistently a clue to find between sessions (though you might need to cheese a hint to the session-attending group if someone they don't normally talk to finds it -- that's probably good for the health of the world, though; it's players interacting with players while you're off doing other stuff). You probably also want these things to have a noticeable impact on your plot; we're a bit short on datapoints for this one (reference back to BG's static content: zero narrative value, so it seems to be weird that the narrative value of this would matter while it doesn't with herbs and mailbags)-- we did have a brief time when BG just had a new infestation every time one was defeated because why not. They seemed to be losing credibility toward the end.
Avoid Plots that Seal or Isolate Player Groups -- it's good writing, I know, to have a dangerous and unmovable army on a major road down the middle of a server, but you make a dramatic cut to the number of potential interactions on a server if players can't reach each other (which would count "certain death on the way" as "can't" -- and noting with the point above, that the dropoff seems to be between "can get there" and "can't get there." "Quick to get there" and "tedious to get there" doesn't seem to get us any results)-- and those interactions seem to be good for play.
Make Content That Pushes People Toward Few Nodes and Common Paths -- This is probably best to reference with example. The Trade Way on BG (especially between Baldur's Gate and Beregost) and the choke point on the road in front of the Silver Spires. Pick a spot and make people cross it often. If the narrative of your quest requires people to disperse, make them return to a hub at points along the way. I've had mixed results with using the content to also get metaphorical "common paths" (e.g. encourage people to make compatible characters); it doesn't seem to do any harm, at least, so I guess whatevs? Faction or alignment restrict if it makes sense?
Make Content That Includes Social Challenges -- Hard for me to pick the right header for this. I don't mean to just sprinkle diplomacy checks on your static content. That can cause there to be a social challenge, if built well (how does the 6 charisma orc get through a thing that requires talking?)-- so the idea is that there are mechanical challenges ("This enemy has 18 AC, 90 HP, +10 to hit, and does 2d8+5 damage. It has the same speed as me and only has a melee attack. I'd like to kill it, but will settle for surviving.") which have mechanical solution ("If I'm at least level 5, with average wealth for my level, I can get my armor class up to 30. Though this will reduce my ability to hit, this enemy is relatively easy to hit and will just take a long time to kill. If I'm not, it has average speed and I have a potion of Expeditious Retreat."). There are also social challenges ("I need to lift this very-heavy rock to reveal a target that must be struck with magical frost. I can't just move it out of the way because the walls serve like a kind of track. Even if I'm a powerful spellcaster, I'll need to concentrate on levitating that rock, and so a second person has to shoot the target.") with social solutions ("I'm a pretty big guy. I think I can pick up that rock. I need a bard or rogue with a wand or a wizard with a cantrip.") -- D&D makes this something of a pain ("I'm a high-level wizard. I disintegrate the rock and then ray of frost the target. I then grab my crotch and flip off the builder.") but you can still build for the common solutions and then let the corner cases be less than ideal.
But we get these monthly-to-quarterly windy posts asking ALFA to restructure itself, and my response is always the same: read the post; parse the activity logs; look for similar things we've done in the past; decide whether it's worth the invested time. The answer is usually no, and I thus really don't want to have people trying to make me do it. Or the answer is that I don't care enough to comment more complexly than to say how possible it is.
Now I admit that there's some assumptions in there, but I hope that people can at least understand my angle. Yes, I assume that people who play more are happier with the game, and that's not necessarily the case. Plenty of emotions can prompt people to play-- but I'm hopefully not too far out of line to say that greater activity, in aggregate, is happier people, in aggregate, even though individual circumstances may be significantly different. I also hope that I'm not too far out of line to say that happier people is a more-successful game.
And so I'm left with some broad statements that I can make-- if you don't care about policy, this section is skippable. I finish this with things that normal folk can just do.
Things We've Tried That Don't Work
I can be more confident in these statements than the others; it's true that correlation doesn't imply causation, but the absence of correlation does imply the absence of causation.
Making the Game More "Pure" -- yeah, either nobody cares enough to change their behavior or they do and the change is a ragequit. Our repository is full of examples of us fixing things that are just not in NWN2 what PnP does, and went in service of the Charter's mission statement (a persistent world o' D&D in Forgotten Realms). Activity patterns seem to not change. Banning stuff tends to be unpopular. Lethality tends to be popular in design and unpopular in practice. And yes, you can find many examples of me doing this a few years ago. It didn't work; that's why I stopped. Sorry about that.
Relaxing Travel Rules / Trying to Give People More Options to Gather -- this includes our multiple PCs rule. It doesn't seem to make people play any more (or, if it does, it's trivial compared to the other forces that make our population fluctuate). I can't find any evidence that it hurts any thing, either, to be fair; I'd guess (of course, put a little [citation needed] here, this part is conjecture) that it's because making travel easier and giving people more PCs also makes it easier to scatter.
Giving People More Rewards -- I have to distinguish this from giving people more content: this is more about taking existing content and/or events and just giving more loot and XP for them. We've had DMs get allowances to fart truly sickening quantities of XP on players before, so we have a pretty-wide range of trends and players to look at, which we can't really parse numbers from. Folk playing with Wynna or HEEGZ as DMs tend to play a lot. Folk playing with me or Xanthea as DMs don't. We wouldn't expect these patterns if XP/loot was a noticeable motivator.
Cutting Space -- disclaimer, of course, is that only one of our servers has undergone significant pruning during a time when we have worthwhile logs to measure outcomes-- that'd be TSM. That incident didn't do anything. Maybe we could do it again and better to get better results.
Things That Seem To Have Worked Before
That "seem" is important. I can guess at mechanisms of causation, but we don't have the resources or population necessary to do some proper double-blind evaluation that could actually demonstrate those mechanisms. I'm mostly interested, of course, in things we've tried more than once. Once might be a fluke; several times is probably (but not necessarily) something more.
Advertisement -- seems obvious, right? The largest and sharpest incline in our activity came from us posting about our world on other forums. We saw clear and pretty-immediate results from just emailing old members. These aren't very longlived boosts, of course (to be fair, nothing is; managing a game is a constant fight against attrition), and I'm sure there's some point at which promotion will result in exhaustion in our potential audience. I'd let people who are more familiar with advertisement talk about that. I'm much more familiar with watching game metrics to study how game changes make people use the game differently.
Adding Content -- our most-sustained increase in activity, and also our highest level of activity, was associated with BG acquiring its more-diverse herb spawns, its hide drops, its mail quests, goblin ear bounties, and its billion isolated fetch quests. There's plenty to criticize in the design of these quests, of course, but they did a few things really well (namely, they required that people be at least minimally aware of their surroundings and kept pointing them in opposite directions down the same road). This does have to contrast with some other instances of content-- like TSM has a lot of content. More than BG did during its peak, I'd estimate, but most of the additions of that content don't seem to be associated with differences in activity (asterisk here, see two points down). My guess is because it has a lot of nodes and doesn't seem to be very good at putting random players in the same areas.
Building Player Groups -- ALFA's also seen really successful pockets of activity coming from players who are just good at making player groups. The most successful iterations have a builder and a recruiter in their membership (sometimes they're the same person; doesn't have to be). I can draw some parallels to the previous point as to some probably-effective mechanisms here, but I haven't actually been in a successful player group here, so this is an indirect observation: but player groups usually settle on a place as "home" and tend to return there after adventure, leaving them with an obvious place to find other people.
Plots With Out-Of-Session Elements -- while not as strong an effect as advertisement, we have seen abrupt and measurable increases in people just being on and doing stuff when DMs run plots and then leave things that have to do with their plots (and impact their plots) just... out there... for anyone to deal with. Some examples would be Wynna's Dean's Quest plots (which actually gives us two discernible data points, because she did it, went inactive, and then came back and did it again) or HEEGZ's demon gnoll plot and its reprise.
Stuff You Can Do
Recommend ALFA -- ... presumably you would? Folk are here playing. There must be a reason. Just telling people about the things you like potentially introduces them to an enjoyable hobby, and probably reminds you of the things you like about this place. It's easy to get too deep in the problems, because they demand more of our attention.
Foster Player Groups -- making a new character? Ask to attach to an established character, or to an established group, or both. Got an existing character? Welcome those new characters. Builder for a player group? Make sure you've got enough space for new players to show up.
Include Non-Session Elements in Your Plots and Invite Everyone -- is your plot about some brutal invasion? Maybe a lieutenant is camped somewhere on the server, and he's going to make a mess of things next session if nobody does anything about that. Or maybe the backdrop of the session just happens to include an infestation, and its server coverage determines how difficult you make the fights during the session. Is your plot a mystery? Maybe there's consistently a clue to find between sessions (though you might need to cheese a hint to the session-attending group if someone they don't normally talk to finds it -- that's probably good for the health of the world, though; it's players interacting with players while you're off doing other stuff). You probably also want these things to have a noticeable impact on your plot; we're a bit short on datapoints for this one (reference back to BG's static content: zero narrative value, so it seems to be weird that the narrative value of this would matter while it doesn't with herbs and mailbags)-- we did have a brief time when BG just had a new infestation every time one was defeated because why not. They seemed to be losing credibility toward the end.
Avoid Plots that Seal or Isolate Player Groups -- it's good writing, I know, to have a dangerous and unmovable army on a major road down the middle of a server, but you make a dramatic cut to the number of potential interactions on a server if players can't reach each other (which would count "certain death on the way" as "can't" -- and noting with the point above, that the dropoff seems to be between "can get there" and "can't get there." "Quick to get there" and "tedious to get there" doesn't seem to get us any results)-- and those interactions seem to be good for play.
Make Content That Pushes People Toward Few Nodes and Common Paths -- This is probably best to reference with example. The Trade Way on BG (especially between Baldur's Gate and Beregost) and the choke point on the road in front of the Silver Spires. Pick a spot and make people cross it often. If the narrative of your quest requires people to disperse, make them return to a hub at points along the way. I've had mixed results with using the content to also get metaphorical "common paths" (e.g. encourage people to make compatible characters); it doesn't seem to do any harm, at least, so I guess whatevs? Faction or alignment restrict if it makes sense?
Make Content That Includes Social Challenges -- Hard for me to pick the right header for this. I don't mean to just sprinkle diplomacy checks on your static content. That can cause there to be a social challenge, if built well (how does the 6 charisma orc get through a thing that requires talking?)-- so the idea is that there are mechanical challenges ("This enemy has 18 AC, 90 HP, +10 to hit, and does 2d8+5 damage. It has the same speed as me and only has a melee attack. I'd like to kill it, but will settle for surviving.") which have mechanical solution ("If I'm at least level 5, with average wealth for my level, I can get my armor class up to 30. Though this will reduce my ability to hit, this enemy is relatively easy to hit and will just take a long time to kill. If I'm not, it has average speed and I have a potion of Expeditious Retreat."). There are also social challenges ("I need to lift this very-heavy rock to reveal a target that must be struck with magical frost. I can't just move it out of the way because the walls serve like a kind of track. Even if I'm a powerful spellcaster, I'll need to concentrate on levitating that rock, and so a second person has to shoot the target.") with social solutions ("I'm a pretty big guy. I think I can pick up that rock. I need a bard or rogue with a wand or a wizard with a cantrip.") -- D&D makes this something of a pain ("I'm a high-level wizard. I disintegrate the rock and then ray of frost the target. I then grab my crotch and flip off the builder.") but you can still build for the common solutions and then let the corner cases be less than ideal.