Domains and Discontent

Brutorz Bill of Green Skeleton Gaming Guild asked for the blogosphere's thoughts on ACKS recently, and I commented to the effect that most of ACKS is great, but the domain system has left me disappointed.  I was asked to elaborate, and decided that it warranted a post.  I may catch some flak for this, but keep in mind that I'm aiming for healthy, constructive criticism which may make ACKS a better game.

Which begs the question: "What is best in gaming, Conan?"

Conan's answer, I suspect, is "something other than spreadsheets."

The trouble is that running a domain in ACKS is not fun.  It is paperwork.  ACKS' domain rules have the following property: an extrinsic incentive exists for PCs to want domains (primarily XP from running a domain, but also some cash from taxes and ready availability of a market), but there is nothing intrinsically rewarding about the actual mechanics of operating a domain.  First campaign with domains, we had a character with a domain who was milking it for XP, and other players decided to establish their own to keep up, and were disappointed with the experience.  Second campaign with domains, we decided to avoid the loss-of-cohesion trap of individual domains and have a jointly-held party domain.  This meant someone had to be the Spreadsheet Guy.  Nobody wanted it.  Sure, I could run the domain, but I'm already swamped running all the other domains and building adventures and trying to remember to eat and sleep.

(Incidentally, the following analysis owes much to Alexis/Tao's book, How to Run.  I have not yet finished it, but it articulates many things which agree with my experience.  I also owe a little to Collins' Interaction Ritual Chains, which is really good)

In the dungeoneering phase of the game, there are a number of events that are emotionally charged.  Fighting is exciting because you almost die, and then there's the thrill of victory.  Falling below 0 HP is ACKS is exciting, because it puts everyone in a state of tension for the mortal wounds roll, which is delayed (this is one of ACKS' finest contributions).  Getting treasure is exciting, because maybe you'll find some sweet magic shit or maybe your friend will pick up a cursed sword and you get the satisfaction of having dodged a bullet.  Exploring is tense and anticipatory, because you might find combat or treasure or trapped death at any time.  Even tracking torches can contribute to excitement under the right circumstances - it builds tension towards that moment when you run out and are trapped in the dark.  Basically, any of: gaining power, finding yourself in deep trouble, or things likely to lead to either of those outcomes has emotional merit (combat which seems unlikely to lead to either gain or loss is usually deeply boring).  Dungeoneering, of all of the phases of the game, does the best job of concentrating these emotional charges in the hands of an unskilled DM.  The rules make it practically unavoidable.

The wilderness phase of the game is weaker, but similar.  Getting chased by worgs is high-tension in an immediate sense.  Getting lost in the woods, or becalmed at sea, or running low on drinkable water in the desert, all build tension.  Coming over the ridge and seeing home in the valley, knowing that you are safe and your travels are over is a release of tension, a relief.  The wilderness exploration rules support these sources of rising and falling tension.  However, while the dungeoneering rules suffer more abstractions and nonsense for this purpose ("We can move how many 10' squares per 10-minute turn of exploration movement?"), wilderness play tends to get a little more bogged down in realistic overland speed and encumbrance, and the tension-release cycles feel much longer than in dungeoneering.  I'm not entirely sure why this is - might have to do with ability to restore resources by resting in the wilderness, so combat loses center stage to logistics attrition, which has a longer cycle?  Anyway, that's another post.

Domain play in ACKS is even worse at supporting emotionally-charged play.  The peaks of domain play (those moments when your players go "haha yeah, suck it!") might include: crushing an army in the field, storming a castle, assassinating a rival, being crowned emperor, sacking Rome, and winning the tournament to marry the princess.  The deep lows of domain play (those moments when your players say "oooh shit, we're boned", followed shortly by swearing revenge if they survive) might include: having their army crushed in the field, having their castle stormed and their town sacked, waking up with assassins in their bedchamber, being called to court to answer for their crimes, Mongol horde or great dragon politely requests submit or die, being excommunicated/fatwa'd by the head of a major religion, plague, natural disasters.  ACKS' domain rules...  don't do any of these things particularly well.  Assassination is supported, but it's one roll.  Being summoned (though not for crimes) and marrying the princess are on a random events table that you're supposed to roll occasionally.  Domains at War adds the crushing/being crushed, storming, and sacking, but Battles is a relatively high-detail system which can make pacing it difficult (and is in general error-prone for the inexperienced).  Moreover, the rules don't really provide ready paths to such situations.  There is no clear state-machine and path of play like in dungeoneering or the wilderness ("Go to the place, kill the thing, take the stuff, deal with complications along the way, and don't die").  The domain XP threshold rules strongly encourage expansion, but it's less clear how to go about it in practice.

Instead of focusing on these emotionally-charged events, the domain rules focus on taxation and population.  It is true that war and skullduggery cost money and manpower.  It is true that gathering an army and preparing an invasion build tension and anticipation of that action. But the balance between means and ends is tilted too far towards the means in ACKS; the tension-release cycle is all buildup (part of this may be a failure of our group's collective patience, but when a campaign lasts in the 3-6 month range on average, it is worth considering).  There has been a failure in choosing what to abstract away - taxation, population, and recruiting are high detail, while many of the interesting parts of domain play are either a single roll, random chance, or just left to DM improvisation.  Unfortunately, so far our luck with DM improvisation in the domain space suggests that it is hard to get right, and terrible unintended consequences often ensue from changes that seemed initially reasonable.

In dungeoneering, the lazy, exhausted, or incompetent DM (yours truly) has a host of systems for generating dungeons full of monsters and treasure from random tables.  Do these dungeon ecologies make sense?  Not particularly.  Do your players care?  Not as a rule.  They tend to be perfectly happy with plausibility and consistency, which is achievable with a little massaging of table results, rather than deep, realistic simulation.  You can paint a dungeon by numbers, stuff it with random monsters and treasure, and your players will probably have fun (the emotional content of the dungeoneering ritual is only loosely related to the literal content).  We also have such systems in wilderness play, though they are not as well-developed.  But when you get to the domain level, ACKS throws the lazy DM, who has relied heavily on tables for the preceding part of the campaign, to the wolves.  No random tables to be found, just demographics and statistics.  And it is possible, from those demographics, to derive reasonable size and population and troop strength estimates for NPC domains.  It's doable, but substantially more labor than rolling some dice and going "yeah that'll work," and demands a much deeper understanding of what those numbers actually mean.  If one is a compulsive automator, it requires a very different approach than tabling there as well.  And unlike with dynamic lairs in the wilderness, very few abstractions and affordances are made to DM ease of use in this area.

If ACKS' support for dungeoneering play is at GUI-levels of usability, where you can kind of bumble your way through it ("push button, receive dungeon, insert players, enjoy"), domain play is closer to the command line; you have to be really willing to read the manual, consider many options, and glue lots of obscure pieces together.  And god help you if you err.

There are also some issues with thief and wizard domain play.  Thieves are overpowered gold-fountains that obliterate suspension of disbelief in ACKS' economies.  Wizard domain play is a comedy of errors at the mid-levels where we've played it, because you can try to research the function of a magic item for two solid months, easily fail all four of your tries, and be left with nothing to show for all that time that you could've been adventuring.  It's very frustrating.  I don't know much about the operation of cleric-specific domain abilities (blood sacrifice, divine power) because nobody plays clerics, but their main domain game is functionally the same as a fighter's.

So there you have it: domains and discontent.  At some point maybe I should write some houserules / abstractions aimed at making the interesting parts of the domain game actually happen.