Tuesday, 17 December 2013

What do I look for in a rulebook?

So Shannon asked for opinions on rulebook contents, mostly on setting content, but it was way too late by the time she asked that. Okay, what do I look for in a rulebook? Some of this stuff I've been mulling over for a while and will probably expand on, but since you asked...

Disclaimer: this is entirely personal opinion, and what's useful is going to vary wildly between genres, game systems, groups with different playstyles, GMs of varying experience, cultures and whatnot.

Second, later disclaimer: this is (as so often) very long, and I basically wrote it stream-of-consciousness with minimal editing or indeed reviewing for making-any-damn-sense. You get what you pay for.

The basics

I like to see a quick and accessible overview, so I can tell what to expect in terms of content, genre, tone, setting and mechanics. About a page (two for small books) feels right. As little as possible of this should consist of self-congratulation, assertions of the game's Special-Snowflakehood or digs at other games. It could also be nice to have an even shorter overview that sums up the game: "D&D 3.5 is a fantasy adventure game, where a party of heroes or mercenaries seek adventure, riches, honour or just to do a bit of good, in a land of magic and monsters. It's designed for 2-6 players and a Game Master who manages the setting, antagonists and non-player characters. It has detailed mechanics and uses a d20 resolution system."

The "what is roleplaying" section can seem cheesy, especially given that many games really aren't likely to be played by anyone new to the hobby. But it can help to highlight which features of RPGs the designer wants to focus on.

Informationwise, I’d say the main issues are:

  • Things that you really need to know to play in the game
  • Enough general information to get a sense of how things work and the broad sweep of the game
  • Things that are unexpected or counterintuitive, particularly if they go against the grain of “common knowledge”. Obviously how important this is will depend on the game. In a historical or semi-realistic setting, if you want to stick to the facts, this is quite important.
  • Any bits of fine detail that will enrich the game, make it more interesting and offer fun play opportunities.
  • Enough pointers, hints and tantalising glimpses that the GM has some starting points for doing their own research or innovation.


Ideally, I want an intuitive character generation process, located at a point in the rulebook where you can expect to know basically what the choices mean, without having to constantly look things up in order to make sensible choices. I don't need every detail, but I do want a general understanding that (say) making a charming, graceful and very dim wizard in D&D is a completely terrible idea. When making a character, steps should be in an order that feels right in terms of creating a concept as well as making mechanical sense. Cross-references to more detailed parts of the rulebook.

Before I move on: cross-references wherever they might be useful. If you decide to spread the rules for character attributes arbitrarily across several books (I'm looking at you, AD&D) then you should damn well include something explaining that your caffeine resistance percentage can be found under the "weapon maintenance" chapter while your blood type is under "health care".

Also: put the chapter titles and numbers prominently on every page so I can tell where I am, rather than referring to chapter numbers but never telling you what chapter you're currently in. If possible, use the trick with a mark on the side of the page that moves down (like a set of tabs) for easy flipping.

You know what, just go and read this post and don't do basically any of those things.

Have a quick reference section for easy access to tables and stuff if they're needed, even if it means duplicating information from elsewhere.

Back to the point...

Rules need to be presented in some compromise between order of importance and dependence, so that the stuff you really need comes early in an order that makes sense, and things that are optional, occasional or can be improvved if the GM forgets the rule are shuffled off a bit into later pages (see also cross-references). Sidebars are fine, but easy to overdo. Make sure sidebars are obviously sidebars. Make sure "helpful voice" sidebars are obviously distinct from rule sidebars. It's basically impossible to create a perfect arrangement of rules, but there are many bad options.

An equipment section mostly covering things I am likely to use at some point, though sample objects for guidance in the long tail of things players will come up with are also good. Some suggestions for guessing costs of other things can be nice. In Shannon's case, there's the advantage of historical exchange rates to use, though these need to be used with caution as the relative costs of different things can change; at one time a calculator cost as much as a holiday, but this no longer the case. Some hints on this could be useful.

Oh, another thing is potentially types of NPC you might run into, and what they do. Just a two-line entry is enough to give a hint, but this can offer ideas you'd never have thought of, both in terms of interesting careers ("travelling sheet-music salesman" is my new favourite from YSDC) and plot-relevant activities. They can also be borrowed for PC ideas.


Creatures, now... Let's start with a few basics just for pointers.

  • Include at least one creature of every major game-mechanical type. If you have Minions, Hordes, Monsters and Overlords, we need at least one of each. The more variable a creature type is in how it plays, the more examples are useful. If Hordes have basically the same sets of abilities, whereas Overlords can vary wildly from small children with immense psychic powers to sentient teleporting cars, we need more Overlord examples.
  • The more common a creature type is, the more examples are useful as long as they are meaningfully different. There's no point offering five full examples if Goblins, Kobolds, Gnomes, Pixies and Imps are 99% identical, you can just include a line saying "use these stats for..." and maybe a note on the one difference. But if crazed elementals are supposed to be the most common creatures you'll deal with, include lots of crazed elemental examples.
  • Include some examples of vanilla creatures, without weird abilities beyond the usual. This allows GMs a starting point in understanding balance and creature construction, without having to understand and strip away an unusual ability that completely changes the power and playstyle of a creature. It's really hard to work out how useful some odd abilities are, and so to decide what to replace them with.
  • The more esoteric creatures' abilities are supposed to be, the more examples you're likely to want, because coming up with ideas that are interesting, actually work out in play, and aren't massively unbalanced is hard work. In contrast, if creatures have exactly one stat and no special abilities, that ain't hard.
  • Include at least one creature of each distinct power level, for reference and sanity-checking (in the management consultant sense, not the Call of Cthulhu sense).

For what it’s worth, I prefer a creature compendium to offer enough examples that I feel like I’m seeing slices of a coherent world, with obvious links or similarities between some creatures, predator-prey relationships between others, and patterns emerging. One of the issues I had with Numenera was that on my quick glance through, the creatures all just seemed to be isolated entities completely distinct from one another. In contrast, D&D is a bit more careful to establish (for example) relationships between goblins, orcs, wargs and ogres, or different types of dragons. Of course, Numenera is a more unique setting; but that’s precisely the problem faced by less generic settings, of firmly establishing their ‘ecology’ in the minds of the players. They don’t necessarily have to be an ecology as such, but some kind of consistency and coherence is nice for getting a sense of the world. That being said, some systems (superheroes, for example) are far more suited to unique creatures with completely different abilities.

The amount of advice needed will also depend on the system in use. I think it’s a reasonable plan to present a sample creature or two and explain what its abilities imply, both mechanically and in terms of presentation.

Another point maybe worth highlighting is if the mechanical consequences of an ability aren’t immediately obvious. Something that buffs its allies can exponentially raise the danger of an encounter; a fear aura that sends players fleeing can be unexpectedly dangerous if they’re trapped or trying to stay hidden; a creature that can become invisible can be a real pain even if it’s not that threatening because it’s hard to escape or destroy; abilities that reduce key player stats can make the rest of a scenario unexpectedly difficult; immunity to certain effects can be a real problem if that’s all the players have; and so on.

Expanding on the last point: D&D is prone to this problem, especially in its CRPG incarnations. Creatures immune to non-magical weapons, and trolls needing fire or acid to finish off, crop up earlier than in P&P and in numbers that are frankly ludicrous. Right now I’m replaying Icewind Dale, in which my 5/6th-level party have just hacked their way through somewhere around fifty wights and a couple of dozen trolls. There are a mere handful of fire vials around, otherwise you’re stuck to what the wizard can do, which is usually kill three trolls per night’s sleep, nowhere near useful when you’re fighting ten at a time. Fire arrows technically solve the problem, but actually don’t, because even unconscious trolls get a saving throw against the fire damage and usually pass! That’s poor implementation, but you get the point.

Similarly, the wights are immune to mundane weapons, which immediately makes the ranged half of the party useless as you don’t have more than a handful of magical ammunition at this point. If you picked the right weapon specialisations (mace, hammer and large sword), you’ll have a few characters able to use what you’ve looted. If you happened to pick spear, flail and greatsword for your fighters, you’re dead.

The point here isn't to moan - although I appreciate the soapbox - but that mechanics traps aren't always obvious. Try to make sure that the GM will be aware if something can make it physically impossible for the players to complete a scenario.

If the system is simple, then you can probably get away with simple advice on how many ‘points’ to spend on a creature. If it involves complex special abilities, you need some discussion of how they interact with the basic stats to determine a creature’s power level. If there are distinct classes of creature in the game, you want some advice on how to make new creations feel like they fit in where they should – as well as on how to avoid blurring the boundaries and compromising those distinctions. It’s also worth considering who the book is aimed at, because new GMs will want more support in creating new creatures than experienced ones, while experienced GMs may need more warning if their expectations from other games are likely to mislead them. On the whole, I’d suggest offering an example of building a creature from scratch to fill a particular niche; possibly one simple creature and one with more complex abilities. Hints on reskinning creatures and powers are also useful.

Another thing to bear in mind is to tell the GM what the designer was thinking when they set about creating creatures. Is the most important thing to find a novel set of abilities, and then flesh it out with fluff (arguably, D&D 3.5? Is the mechanical role and abilities of the creature (4E)? Are you trying to model existing entities from fiction or folklore, or more broadly, to evoke particular impressions rather than worrying about mechanics? Do you devise creatures to fulfil a narrative end more than a tone or mechanical one? Do you start out from "in this genre, X and Y happens, so we want a creature that will do X and let the PCs do Y"? While GMs may not want to take exactly the same tack, understanding what the designer thought was important can be a big help in creating satisfying creatures.


Advice for GMs

Tell the GM what their job is. How do they fit into a satisfying session of Game? How does their role contrast with and complement the players? What (broadly speaking) are they trying to achieve - what does it mean to be a "good" GM in this game? What should they watch out for in general? Are there any specific snares or misunderstandings they should beware of?

If they actively work against the PCs, how much should they push, and when should they stop? Are they supposed to work against the players as well? Are players supposed to be always unified, in direct competition, antagonists, or what? How can, should and shouldn't the GM support this playstyle?

What are the most crucial rules, both mechanically and in terms of play structure? For example, there may be strict rules governing the way players take turns, when suggestions can be made, or when you can use Astral Knowledge to influence your Flashbacks (only before rolling) and to change the terms of a Contest (after both players commit, but before consequences are revealed). Structural rules may seem like they can be handwaved, but it could be that this sabotages the play experience for reasons that aren't immediately apparent. Some rules are easily overlooked by people working things out as they go along, but can come back to bite you at the end when you realise you missed them - make sure to highlight these ("You weren't tallying the 1s you rolled? Well, pants.").

Include examples; preferably get someone else who's never played it to indicate where examples are needed, and someone who has played it and misunderstood it to do the same. Examples should be no more complicated than they need to be; it's unhelpful if they introduce several new ideas at once, or when it's not clear which of the six situations involved is the deciding factor. Examples are good for showing what rules counter, stack or overrule each other, but only if each case is clearly explained. Examples like "because the centaur is Frozen, Incompetent, has the high ground, charged, doesn't have a melee weapon and is coming to terms with the loss of his hedgehog, he can't Manoeuvre and rolls three dice for Grappling" are rarely helpful unless you already clearly explained each of those points.

Actual Play is great. Really great. New games can't usually offer game recordings, but authentic game text in a rulebook is very helpful. It helps to illustrate not only how rules work, but also how the game is expected to flow. Perhaps most usefully, it helps to show (in no particular order):

  • when, how and why the GM makes particular rulings
  • when, how and why the players take particular actions
  • how the GM responds to those actions
  • how often dice are rolled, abilities are called in, and in general the balance of mechanical versus freeform play
  • how players interact with each other, both socially and mechanically
  • how the players and GM invoke the rules to achieve particular things, and which way round that goes
  • which rules and mechanics are common and important, which isn't always what you expect

If the GM is supposed to use characters against themselves, by calling in Phobias, Weaknesses and so on, give them a hint as to how much and how often they should do this to achieve the result you want. Groups will have their own preferences, but it's nice to have a rough idea. It can also be important mechanically, or a major tool for sharing control of the plot direction, and in that case misunderstanding the designer's intention may seriously affect how the game plays, and indeed how good it seems.

If you have an important mechanic that is unusual or counter-intuitive, spend as much time as you can drilling it into the GM's head. Give them rule examples. Give them multiple in-play examples. Highlight the rule and how it can be used in the introductory scenarios. This will help avoid them completely flubbing the rule and then thinking your game is a bit pants.

Make sure your introductory scenario is a good introductory scenario. This is emphatically not the same as being a good scenario. In some cases they might even be incompatible goals. More on this later.

Specific advice for introductory play is helpful. Point them towards ways to help players learn the game. Encourage them to give leeway where it's needed, while highlighting which rules they should stick to firmly to avoid sabotaging their game. Suggest some general options for recovering a scenario if things go badly wrong, so that you end up in at least the right genre.

Advice for Players

Some of this can fall under general introductory stuff, but make it clear what the game expects players are trying to achieve, both in-character (serve the Queen? make a living? make a name for yourselves? show them all, the fools? endure eternity as painlessly as possible? shape the cosmos to your whim?) and OOC (gain XP and levels? see cool new stuff? fight for narrative control? get the most Angst points? create slapstick hilarity? mimic a particular genre?) and how these goals are related. In some genres, IC and OOC goals tend to be opposed: horror is often one of these, but action-adventure can also inclined that way. Consider whether it's necessary to suggest suitable and unsuitable PC goals that work well with your IC/OOC balance. Not all games have strong expectations about what the PCs' in-character goals are, but they tend to have assumed limits on what they aren't.

It may also be worth mentioning the default assumptions about group dynamics, and about how the players and GM interact. Are the players fending off trouble created by the GM, actively seeking out interesting challenges (or things to do that will involve challenges), or do they devise their own troubles for the group? If there are rules for things like causing yourself trouble in exchange for later benefit, mention briefly how much and how often you expect players to do this. If there are abilities that players "call in" (such as Traits in Mouse Guard), drop a hint as to whether you expect them to do this as much as humanly possible, whenever it's obviously reasonable, or on rare and special occasions.

Reassure them. If they don't need to sweat the small stuff, tell them so. If their choice of abilities is significant but they can't screw themselves by Doing Chargen Wrong, point this out. If there are simple guidelines they can follow in most situations that should make things work out satisfactorily, point that out. If a lot of the rules don't see use in low-level play, say so. These things won't always be true! For example, some games rely on the players understanding and using a list of abilities, trying to argue the case for various Traits to be relevant, and so on. The GM should help new players with such things, and allow them leeway, but the GM may not know them either; in particular, they don't usually have a copy of everyone's abilities to look at.

Setting information

Oh hey, it's the bit Shannon actually asked about! Finally! Sorry Shannon...

Broadly speaking, I think the most important considerations for deciding how much setting to give are:

  • Whether the setting is familiar to players
  • Whether the setting is generic or specific
  • Whether the game aims for realism


A familiar setting generally calls for less general background. You don't need to explain how people obtain objects or support themselves in a modern-day developed country, even if it's one your players haven't been to. Familiar systems of government, law enforcement and education can often be skipped entirely. Even fantastical settings that are generic can be skimmed over: a fantasy game can drop in a few sentences about the King, the College of Wizards, the Knights of the Silver Spear and the court of scheming nobles who form the government, and most players can extrapolate comfortably.

Whether they're real or not, settings that are substantially different from what players know require explanation. This may be a wholesale issue: life under a system of tribal elders, ruling wizards or benevolent alien domination is hard to imagine in detail, while everyday life for vikings was rather different from the pillaging stuff everyone knows about. It can also be specific details: mediaeval fantasy is one thing, but very few people have a solid idea of what mediaeval life was actually like day-to-day, or even what "mediaeval" actually entails in terms of culture, technology and society. Coming back to the modern day, while some things can be assumed, the audience for an English-language RPG probably aren't that familiar with life in Turkey, Ghana, Nunavut, a croft in the Orkneys or various remote and far-flung parts of their own country. A game may also be set in social circles that they don't normally frequent, be that the military, a Cree reservation, Buckingham Palace, a boarding school, a crime-ridden estate or a football fan club. Obviously there are gamers from... I'm going to say, most of those backgrounds (I can't see the Windsors gaming, honestly) but they're only a small proportion of the audience, and a game can't assume you have intimate knowledge of those environments.

If your setting is completely made up from whole cloth, you have a ton of work on your hands. I'm still waiting to read the Numenera book I took off Arthur's hands, hoping to make sense of the setting, because I couldn't during the game. If you can't rely on existing expectations and assumptions, everything needs describing. What is the technology level? Is there magic, and what does "magic" mean? What types of creatures exist? What is society like? What's the PCs' place in society? What's the legal system like? What are the cultures like? How much do people travel, and how? How equal is society, and how does inequality manifest itself? How dangerous is life, and how much does anyone care? What are the most significant influences on society: arts, technology, religion, magic, war, money? Where do the lines fall in the setting that are important to establishing genre - for example, the lines between "vicious thug", "heroic warrior" and "physical humour", or between "completely normal behaviour", "distasteful attitudes" and "criminal offensiveness"? Is tampering with the Spellwall a prank that'll get you a night in the cells, an act of undescribable lunacy, or a capital offence endangering everyone? What do distances mean in practice? How much freedom, chaos and unpredictability is there? How much wilderness (or equivalent) versus urban sprawl versus tightly-organised city-state is there, and which are the expected play environments?


A generic setting (that is, a setting designed to be generic) tends to call for a lot less detail. One reason is that it benefits from familiarity - a generic setting is pretty much by definition one that players should be familiar with, or at least most of the group probably are, otherwise why did you decide to play it? As such it offers the chance to incorporate whatever elements are wanted from that genre, and ignore others. The other reason is that too much detail can constrain GMs as well as inspiring them, depending where you put it. It's a good idea to give some detail of at least one major location of each main type, a few sample NPCs to help illustrate the kind of characters you expect to fill the world, and some possible issues that PCs might deal with (perhaps tied into those locations or NPCs). If guilds, nations or companies are important to the setting, examples of those are also sensible.

A specific setting has a lot in common with the Whole Cloth point above, but can be real as well as fictional. A game set in Genericville in the Modern Day is very different from one set in Ashby-de-la-Zouche in 1994. The more specific a game is, the more detail you tend to need to illustrate and use. The GM is more likely to care - and to be asked - about specific points, especially if the game is relatively static and so consistency is important. While the details themselves don't necessarily matter, you don't want to paint yourself into a corner by offering the PCs elevation to the nobility when the nobles' descent from a powerful sorcerer is both a diamond-hard law of the setting (they maintain the Spellwall as the price of their titles) and a plot point vital to the campaign that doesn't allow you to secretly give the PCs sorcerous descent (his evil ghost starts taking control of them in the climax of the metaplot).

Moving back into fantastical settings, if Our Elves Are Different then you need to explain how, and possibly why. If there are astrological rules governing when magic can be cast, or common mythological tropes don't hold true, those are important. In a sci-fi setting, there's a big difference between settings with and without warp and time travel, and even between incarnations of those technologies. It matters whether your robots are basically humanoid machines, Asimovian Three Law models, or sentient artificial beings with independent moral codes, because those will affect how they behave and how players interact with them.


Finally, realism. This isn't just restricted to real settings, though those can encourage a thirst for details. Is there a bank in the south part of town that opens after 5pm? Can you carry a shotgun in this state, and what licence do you need; can foreigners get one; how long does it take? If a major fire erupts in the town hall, is it near anything that'll cause huge problems, like a fuel dump or hospital? How likely is it that speaking Welsh is a safe way to have a discreet conversation in a pub, and also won't get you punched in the face? Did the HMS Murderiser even have coal-fired engines whose ashes you can use to melt ice on the deck, or were that class all oil-powered by 1946? Which side does the driver sit? You don't necessarily actually need to know that stuff to play, but people often care. Even in fictional towns, there are questions about the plausibility of various political and social systems, how the economy works, and so on.

Sci-fi realism is a common problem that just does call for handwaving, so here the question is what gets handwaved, especially if the rest of the game is relatively hard science. Are there space bandits, even though it makes no sense? Does time-travel move you in space and adjust your velocity so you don't end up suffocating in a vacuum or flying through a wall?

A game that tries to evoke a setting without insisting on realism doesn't necessarily have an easier time, but does call for less detail.

Okay, taking a step back... in general I would say offer at least one example of (in Shannon's case) paranormal society, relic and strange place, so that GMs have something to draw on to see how it's actually done. As with creatures, it helps to include some information on the choices made in designing them, and how those achieve the goal of the design. The more important each element is in your setting, the more examples are likely to be useful. If interacting with a whole set of paranormal societies is crucial to the game, we need a lot.

Tool kits can mean all kinds of things, and here I'm not entirely sure what Shannon's getting at. I certainly think brain-prompting tools are useful, things that prompt you to make the right kinds of decisions as part of creating your new thing. Templates can be useful, especially for very common or significant elements, and where precise adherence to mechanics is important. In general, though, I feel like prompts and rule-of-thumb guides are the most immediately useful once you have an example or two to look at. For example: "Consider WHY the antagonist is here, their MAJOR GOAL, their MORAL OUTLOOK, and their likely ATTITUDE to your PCs in particular, before determining abilities" and "as a rule of thumb, one Fate will make for a decent encounter, three Fates is suited to a major conflict and the end of an arc, and five Fates offers a drawn-out conclusion to a campaign, perhaps split over several locations or meetings". Or "combining abilities from more than two Talent pools is not recommended, as the NPC may be unexpectedly weak or strong due to Talent synergies".

As I think I mentioned earlier, a quick (and dirty, if it's simpler) guide to reskinning non-useful content is also good. Can't use the Brotherhood of the Rat, who specialise in summoning ghosts to assassinate political leaders? Well, the Onyx Guild summon elementals to rob banks.

Okay, that's a lot of writing and I'm tired now. Hope it's at least vaguely helpful!


  1. I don't suppose that once I have it all edited and largely laid out, you'd be willing to take a look at it to give me advice?

    1. Sure, absolutely, but I don't claim any special expertise :)

    2. You have the language and analytical skills to describe what works and doesn't work for you. That's all that's needed to do a good review.