Sunday, September 9, 2012

When Is Hairsplitting Worth It?

True Grit:

Mattie Ross: Some bully shot! That was four hundred yards, at least!
LeBoeuf: Well, the Sharp's carbine is a...
[just then a rock is brought down on his head]

...and this happens:

"If I ever meet one of you Texas waddies that says he never drank from a horse track I think I will shake his hand and give him a Daniel Webster cigar."
-Rooster Cogburn

...and both of these moments of dialogue are brought to you by the same phenomenon--worldbuilding of the hairsplitting variety.

They revolve around the characters making fine distinctions that matter in their world--in the first one LaBoeuf is talking about the difference between his weapon and every other weapon (which has been a comic subplot all along) and in the second Cogburn is making fun of LaBoeuf's character class (ditto a subplot all along). Because Cogburn's a bounty hunter (fighter) and LaBoeuf is a Texas Ranger (paladin).

These are both things that--in a Western tabletop RPG--could totally happen, and for the same reasons. Players are discussing events on the ground that need to to get done and that in turn has them talking about the fine distinctions that define the world and that in turn makes the story seem more real and fun.

So fine details and distinctions are fun right? Especially if they have mechanical weight which makes them come up in the game automatically? ("You're a burglar, burgle something" "What the fuck do we have a cleric of Loviatar for? We need fucking healing dammit."


But then so ok at the top there there's the famous polearm picture which is like an example of the opposite. Like jesus Gary a polearm is a polearm can we just say it does d10 and has extra reach and move on with our lives? Do we need to define the Sharp's Guisarme vs the Colt Semiautomatic Ear Spoon?

So that got me thinking about when something is playable depth and when it is just like kill me now.

The first thing I thought of was Noisms 20 scorpions which I love as background (what world should not have at least 20 kinds of scorpions?) and I hate as monsters. Or I would if I wrote them all up--because I am not going to get to use all 20. If I write them up, they won't all come up, and way more importantly, even if they do, that 4th 5th 6th 7th encounter is, in the end, in danger of being mundane anyway--we're fighting more scorpions? Really? Again?

Monster speciation often sucks. Because monsters are supposed to be eerie and unique and the image in the players' minds should be memorable and new and, for me anyway, going "Well now this is a red hydra" is just kinda...meh. Wasted opportunity.

Same problem with character generation options. With too many options, it gets annoying and intimidating to give new players a gazillion options and, as an experienced player, in variety-packed systems I often feel like I am just making a character concept up out of whole cloth and then doing a lot of homework afterward to make the system fit it. I get the point of the whole customization thing, but too often it seems like rather than giving me ideas, it just sends me checking to see if my own ideas are doable in there. The crunch behind each distinction isn't helping, just slowing shit down.

And so the bill-guisarme--at least during character generation--I want weapons to be simple enough that by just picking weapon A (which fits my idea and gets character gen over with) I am not screwing myself because weapon B does what I want better.

I like that there is a Sharp's carbine but I don't like having to look for the Sharp's carbine.

So what's a kind of speciation that works? I think that Noisms idea about different breeds of dog was neat and the dog table I made after I read it works pretty well at the table.

Why? So I'm thinking:

-It isn't something players have to sift through. It's random. So you just roll and that's your dog, write the numbers down. Nobody has to read that whole table--the richness built into the system is not a burden on the player.

The Warhammer random career system works the same way--richness without burden. And this is why the polearms and their speed factors keep not getting used by generations of players--the players don't get to find out about the difference between a glaive and a spetum, they have to. Or at least get the feeling they might die if they don't.

-You're not fighting all these dogs on this table. This isn't a table that suggests that in the campaign you'll be fighting a Demon Dog one week and then a Hell Hound and then a Devil Dog and then a Moon Dog and then a Shadow Mastiff and you'll start to wonder why you have the most boring GM ever. The dog is just there, it's a characteristic of the setting, but it's not foregrounded. It's not a monster. The variation on last week is not the whole idea for this week's fun. But if one does become a monster--like it goes rabid and you go "Well that's an Azumchefe climbing dog, so hiding in the tree won't help!" then that's a neat moment in the game--the background distinctions have weight and generate plot.

Monster speciation can be cool if the variety of monsters are all part of the some overall idea. Like: "You fought pink ooze, but this week it's orange ooze!" is just a lack of imagination, but "When you set pink ooze on fire it turns into orange ooze, which can talk!" is fun. Devil Dog and then Doom Dog and then Danger Dog is boring but Northern Goblins at war with Lava Goblins is fun. In other words, if you're gonna do variations on the same thing, make the fact that they are variations part of the setting.

-They come up a lot, so you're going to use them. Kimberly's character, for instance, has a dog whose name is "Sueno #7". This table has already generated more time worth of in-game dialogue than the time it took to write. It isn't "99 Magic Hats". Dogs are a normal part of the setting and a good idea, tactically. They're gonna come up.

-You can tell Dog A is different from Dog B because of mechanical differences you have to actually talk about during a game to bring into effect. Every time you want your dog to climb, you have to point out he's an Azumchefee Climbing Dog. It's not like a fauchard where you just write down some numbers and never have to bring it up again.

So when thinking about places to add variety to a setting, I'm wondering what other things fit this profile...

Horses, for sure.


Religions maybe--giving someone a random horse or dog is fine, you'd need few enough religions at least in the beginning that choosing wouldn't be a pain in the ass.

Fighting styles? Same problem as religions, really. You can see where splatbooks come from doing this...

Monsters in hierarchies, definitely. Like if you can put in the effort to make Right-Handed Hobgoblins do one whole thing and Left-Handed Hobgoblins do another thing and they interact in an interesting way, that'll be worth your time when you bring it into play.

What else?


  1. In defense of monster differentiation: it is true that my players won't fight 20 kinds of scorpions. That said, as the DM, I want a range of options to pick from. "Oh, a grabby scorpion with focus on claws? Oh a scorpion with poison? No no NO wait, this scorpion has hallucinogenic poison, that is DEFINITELY the one I want to use."

    1. That is true--but I was thinking less in terms of designing products for other people and more in terms of stuff you mae for yourself

    2. These kinds of differences seem both trivial in fluff and crunch to create and use mordicai.

  2. Not presenting a burden is a good criterion, but not all gamers will experience extensive differentiation as a burden. There's a particular subtype of nerd who loves to work out every detail of a system --- they want a Klingon dictionary and the blueprints of the Enterprise, and they feel it enhances their experience of watching Star Trek.

    Since I'm some other subtype of nerd, I haven't mapped out the whole system subtypes, hee hee. Anyway, if I were playing with a group of such players I'd probably be more of a cataloger, just to give them more to chew on, but with a less-OCD group I'd gladly avoid it.

  3. I grappled with this issue in Two Fisted Tales (a pulp RPG by Precis Intermedia Games I've considered sending you in PDF form; I'd love to see what you and the ladies think of it). The pulps are filled with gun details such as a "a Webley Fosbery .45 Automatic," but they never seem to make much difference during combat. Ultimately I just used "little handgun," "big handgun" and so forth.

  4. This comment has been removed by the author.

  5. There should be just enough options so that both you and your players can be surprised; and options both common and rare so that you have expectations to set the surprises against. Critical hits are an example of something that works well with this approach.

  6. I think it REALLY depends on context. Critical hits, sure, but if your campaign has 40 different kinds of horses that are all a little different every single time that are all equally common? Not a problem. Interesting in itself.

    It depends on the narrative context in which different ideas come up.

  7. I once did a post about oozes and slimes -- not going to look up the link, since I ought to write an updated version -- where, instead of the "this color slime is immune to this damage and takes half damage from that", I just said "each ooze variety has these common characteristics and 1 invulnerability per X HD, plus 1 vulnerability to a common substance". And that random roll was *deferred*, so the players had to experiment with a newly-discovered variety to figure out what would work. You throw salt at the mottled ooze? Roll a d6, and on 5+, the ooze is vulnerable to salt.

    Looking at this post and thinking about the trade-offs Brendan mentioned recently with reference to spells gives me vague ideas about doing the same thing in a more general fashion, so that you could have deferred horse breeds, deferred religions, deferred fighting styles. I'll have to think it over.

    1. Follow-up: I took the idea you once posted about keys for doors and applied it to "deferred" dog breeds. Short version: you define a breed as a "Lustful Retriever" or "Aggressive Tracker" or some other combo of general weakness and strength, and roll dice each time the breed encounters a novel situation to determine what it's lustful/aggressive towards or which situations it's good at tracking or retrieving in.

  8. Variant races are high on mt list, but only if the GM goes all-out. Right now I'm working with various groups using the Advanced Race Guide for PF to populate a campaign world with hundreds of races with their own customs. And it has added to the games. When ever they run across a human with light tan skin, they eves drop on him to determine whether he is a smooth-talking Cavasian north islander or a gruff Garian islander, both of which you can identify by language and mechanical traits.

  9. The main criteria I use are:

    Will splitting hairs increase the agency of the PCs? Does it allow them to demonstrate their individual ways of engaging with the world in-game?

    I created more complicated teamwork rules for Openquest that distinguish different ways of helping other PCs because I wanted to encourage PCs to help one another and the current rules only really provided one way of doing that.

    Similarly, to avoid having long lists of weapons, I just created a checklist of qualities PCs could go down to create whatever weapon they wanted:

    My experience has been that if it will actually (and I emphasise "actually" against "theoretically") get PCs to make different decisions and act on those decisions and lacking it will cause them not to do those things, then I think it's meaningful enough to be represented.

    If it's a one-time thing, then I'll just spot rule it, of course. But if it's going to come up a bunch, I figure it's worth investing some time into getting right.

  10. The alternatives LotPF classes you were publishing earlier this summer exemplify wonkiness done right, at least for me. They allow for mechanical differentiation of classes without forcing on the player any burden of choice or desire to optimize based on mechanics.

    You can build similar feature tables for all types of things. Fuck a +2 sword, about about a sword that *clatter* has silver inlaid in the edge and is +2 against all undead and *clatter* has a 25% chance of shattering any metal weapon when parried or used to parry.

    I guess what I'm saying is that when I have something that lets me generate new objects, I think it's awesome. When I have a list of pre-generated objects that I have to choose from, it's more likely that I'll find it wearisome.

    The trick is having tables that generate at least somewhat coherent things. In my games, I want things to be self-consistent at least.

    Applying this to religions, this lets you generate a new one every time you need one, either by rolling or by picking from the list of attributes and coming up with others as the world demands. But, it's not a list of "20 awesome religions that omg I just have to fit into my game". Also, it lets you build alternate sects really easily. The Unitarians are Christians who don't believe in the trinity and are all nice to each other, the Eastern Orthodox are Catholics without a central hierarchy and with less strict food requirements, the early Christians are a messianic Jewish cult who recruit non-jews and preach the overthrow of the Roman government.

    Other categories I think this approach works well for are:

    Shops (both done in Vornheim)
    Magic/special/unique items/weapons
    Social systems (say, religions or guilds or what have you)
    Window dressing on new locales

    Hopefully that wasn't too much preaching to the choir?

  11. Damnit, man. Stop making me rethink everything I know about roleplaying games. Or, you know, keep doing that, because it's awesome. Whichever suits.

    But yeh, players choosing from a good few weapons, while random magic weapons come in thousands of varieties. That works.

    Random spellbooks you choose your spells from. Coo.

    Random weapon training could totally work though, attach a feat chain to each one like 3e Rangers get. Give Fighters a few special weapons and let them change what they can do by switching weapons.
    Rewards in-game choice, infinitely extensible, and can spice up the funky ones.

    Never seen a random roll that wasn't bypassed when the outcome was known ahead of time anyway, so can't really hurt.

    Religion could work too. PC chooses alignment (aka broad church), randoms up a cult that matches, and still gets to define how they use that.

  12. Could you make the polearm thing work if you put it in a random table?

    Like, you wander up to yon scraggly merchant and he has d4 weapons for sale. And it's a... grappling pole arm that pulls things, a high-jump pole, a throwing boomerang pole-arm, and a broken stick.

    And you assume that the world is so inherently chaotic and raggedy that you can never just go up to a merchant and take a specific thing off the list.

  13. Possible additional areas where varieties would work:
    Medicinal Herbs and Ailments: Just wrote up something for this. If a minor cold (sneezing) prevents stealth, stinkbulb root might provide one day's relief.

    Favorite Snacks of Monsters: dropping food can distract monsters, but maybe a particular food is more likely to distract displacer beasts. Also could be applied to repellents.

    Cosmetics and Reaction Rolls: maybe cobalt eye shadow is especially alluring to orcs, giving a reaction roll.

    Decorations on Trade Items: Yellow diamonds on an iron helm certainly raise the helm's value, but may make Burgundians specifically more likely to want it.

    I'm determined to tackle the religion problem, but I have a post on horse/animal breeds coming up next.

    1. i don;t know about you but those all seem like things that wouldn't come up very often in my game

    2. I get players asking about the most general form of healing, wanting to cure even 1 hp of damage or looking for a poison antidote. And dropping stuff as distractions is pretty common, so the favorite snacks bit is useful for me to add details. The others are iffy, but since the trick I propose for medicinal herbs is adaptable to the others, this is really just me noting that hey, if I ever get a rare attempt to seduce an orc chieftain or something, I don't need to write up a special table.

  14. You can almost do this with polearms, albeit not enough to justify 20 varieties. If all your polearms have the same damage and speed and so on, that's all the players really need to know. But if you can remember more or less their basic shapes, you can differentiate them in situations where it matters without needing any extra rules at all. Like a pike is really long, a spetum can catch people's weapons, a halberd can chop or smash stuff, a glaive or fauchard could pull someone's legs out from under them, a bill-hook can unhorse someone, etc. It won't come up unless the player with the polearm initiates it, and you don't have to remember anything beyond a few shapes and common sense.

    1. I guess actually that's my preferred approach to most fine details - rather than mechanical benefits, just describe the thing and wait for the description to become relevant in the player-DM agreement process. So, not very helpful for the kind of thing you're talking about, in retrospect.

  15. I decided to think about this for a while, and I came up with a solution that I intend to try in my games. This only applies to weapon hairsplitting.

    First of all, it makes sense that only higher level fighting classes would even NOTICE the distinctions in weapon types. A novice isn't going to catch that this halberd is good for tripping people, or that this sword has better balance. Furthermore, weapon hairsplitting should not be part of the lower levels for the sake of sparing new players the headache of massive choice.

    Solution: every 3 levels, a fighting character can notice a Distinction on a type of weapon. This could be increased balance in a sword, or a grip that allows for a quick hilt-strike in a scimitar, or flexibility on the spear-shaft, or what have you. Then, when purchasing a weapon of the type they can notice Distinctions, they can spend extra money to buy a weapon that has said Distinction (say, x2 price for 1 distinction, x3 for two, x4 for three, and so on).

    These Distinctions could be drawn from a set list for each weapon (a project in and of itself), but allow the character greater control over their combat style, and introduce attachment to weapons (NOOO! Not Jenny, my greatsword-with-the-disarming-hooks-and-weighted-chain-for-tripping! Where will I ever get another sword like that!).

    When finding weapons, you can randomly determine whether it has Distinctions that the character can utilize. If they gain a distinction later, I suppose you could make a retroactive roll.

  16. Fighter? I stat'ed him up as a fallen Ranger: Acrobat.

    Curse this post, now I want my next character to be a dog breeder. What is your ruling on a blink hound from the moon X a celestial hound?

    I'm thinking
    1: looks like one, traits of the other
    2: Looks like both, traits of both
    3: Looks like a dog, traits of none (that's the loveable one that gets lucky combat rolls too, or, possibly horrible rolls)

    man, its everything pokemon was supposed to be but never was.