Sunday, January 11, 2009

On Rules

I still want to do some kind of 2008 in Review article at some point, but one thing that struck me was that it was bookended by two very promising wargames with quite possibly the worst sets of rules I have ever seen - and believe me, as one who has ranted about game rules before, that's saying something. So I thought I'd take an opportunity to rant some more and maybe to offer some solutions.

The games in question are The Devil's Cauldron (MMP) and Pursuit of Glory (GMT). I had every reason to like both.

In fact, maybe I do. I almost can't tell. Because the #1 reason I threw up my hands in despair at The Devil's Cauldron was not gameplay, but utter, blinding frustration with the rule book. I could never remember the Assault Sequence, for example, and every time we needed to confirm some small detail we needed to wade through the rules' endless nattering before finding (hopefully – there is no index) what we were looking for. The insanely verbose and conversational style makes actually using it during a game to look stuff up an infuriating exercise.

Conversely, the Pursuit of Glory rules read more like a rough draft than actual rules. Spread over almost 50 sprawling pages, the rules are constantly re-stating themselves, presenting things out-of-order, clarifying the blindingly obvious, and getting bogged down in minor details. This game is almost certainly not be as complex as it looks, but with 50 pages of meandering and incomprehensible rules, I can virtually guarantee it will never be played around here. Nobody I game with regularly will look at that kind of page count and even bother to try, no matter how much one insists it's very like the modest-complexity Paths of Glory and the rules volume is due mostly to clarifications and keeping the most obtuse players on ConsimWorld happy. Page count may be a crude metric, but 50 pages = no go unless you're OCS or ASL. I take that back, even OCS has "only" 38.

Although these were by far the worst, there were plenty of bad rulebooks this year, particularly from repeat offenders GMT and Fantasy Flight. The Unhappy King Charles! rule book makes a moderately complicated game look daunting, Warriors of God uses opaque and non-standard terminology to make a simple game needlessly confusing, and Tide of Iron's rules turn a light wargame into a major undertaking, with the Desert Fox expansion rules being even worse.

Enough ranting. On this particular occasion I'm here not just to complain, but to offer some suggestions. I'll admit I've never written a rulebook. But I have spent a great deal of time explaining rules to people, and certainly have read more than my share. Some of this stuff seems very basic to me, but apparently it needs to be said.

First and foremost, I think it's important to keep in mind what we're trying to accomplish here. What we are trying to do is to build a model of the game in the player's mind. The player has to have a working model of the game in his or her head in order to weigh the options and make the decisions required to play it. So the goal is to build up these mental systems in a way not unlike you would assemble anything else.

The 100% Rule: When writing rules, one must bear in mind that there is actually a big difference between explaining rules and creating a working rule book. If I explain the game rules to you, I only need to get far enough for you to have a solid enough mental model to begin playing. Things that are initially either not relevant or negligibly relevant can be explained later. As an explainer, I can also rely on the players to ask clarifying questions when their mental models seem to have gaps. But to formally describe a game in a set of rules, 90% is not good enough, you have to have 100%. The same conversational techniques you would use to teach rules in person can fail to fully and concisely convey the complete details of a complex system when read. Sometimes catastrophically, as The Devil's Cauldron demonstrates. It has an acceptable 90% rulebook if you have access to someone who knows the game 100%. Unfortunately, such a person does not appear to be included in the box.

Short Rules Are Better: It's a fact of life that our brains' short-term storage buffers are small, and a rule has to pass through that buffer before it has a chance of being retained long term. Simple, straightforward rules can frequently be made vastly more difficult to retain through over-explanation. Rule 12.5 in Pursuit of Glory is a good example. This is a simple rule: All full strength regular units have to roll a die when activated for attack in certain situations, and if they roll >= the round number, they are reduced. But then when you spend 4 (short) paragraphs clarifying that means that reduced units don't roll, irregulars don't roll, that yes, “when activated” really is before combat, so you have to use your reduced combat strength, that rolling > 2 is more likely than rolling > 5, and that there might possibly be cards out there that alter all this, all of a sudden you've actually made the transition from page to memory far more difficult than if you had just bolded the word regular and been done with it. Plus you've completely broken the reader's rhythm. The simple version is perfectly clear and concise. If you feel you absolutely must preempt possible misunderstanding of an otherwise perfectly clear rule, put it in a footnote, side-note, or appendix. As a corollary, write your rules for the average reader, not some nut-job on ConsimWorld who is out to willfully misinterpret your rules or question your design decisions.

The Test of Context: I've talked about this a little on a recent thread about explaining Race for the Galaxy on BoardGame Geek. This gets back to the whole mental model thing. When you're trying to help someone build a mental model of the game systems, you want to build the systems in a logical order, such that a player doesn't have to do a lot of work to hook them up once the explanation is done. In many games, the sequence-of-play order is the way to go. But there are some dramatic examples where explaining things in that order actually makes it significantly more difficult, like Race for the Galaxy or Through the Ages, because understanding things that happen in the first phases requires understanding what's going on later - but the opposite may not be true.

For wargames, what this translates into is that you have to start with the victory conditions, because that is the overall context. Paths and Pursuit of Glory get this right; the victory conditions are up front. Unhappy King Charles! gets it wrong and puts the victory conditions right at the end, so you struggle through the rulebook with little understanding of what you're trying to accomplish and then have to retrofit your mental model when you find out the answer. From there, you might argue for explaining rules that are critical to victory before breaking into the sequence of play; for Unhappy King Charles! and Hannibal: Rome vs Carthage, for example, one might make an argument to explain political and provincial control second rather than in sequence of play order.

The other thing this argues strongly against is up-front terminology explanations, which have become distressingly common in GMT rulebooks. Pursuit of Glory has two dense pages of terminology up front which are almost totally incomprehensible since you have no context for understanding what they're going on about. Pursuit of Glory is actually a multiple, severe offender here. We get terrain effects on combat on page 4 before we've even gotten to the sequence of play. Detailed unit descriptions are on page 6 and 7 before we have any way of understanding what these unit distinctions actually mean in game terms, so the rules of course have to repeat everything again later, which itself becomes incredibly problematic. There is absolutely no reason to introduce a rule like this before the reader can possibly understand it.

You get or lose players in the first five pages or so, and almost definitely by page 10. If players can get get a running start on what the game is all about – if you get them a solid context to work with - they'll have hope, even if your game is complicated. If they're on page 8 and still haven't got past the component overview, as is the case in Pursuit of Glory, you're screwed. Put the glossary at the end.

Tell 'em Once: There is an old rule of business presentations: "Tell 'em what you're going to tell 'em, tell 'em, and tell 'em what you've told 'em". Unless you intend your rules to be read by disinterested and bored people who don't really want to be there, this is terrible advice for rules-writers. If you're repeating rules (like the Blockade rules, repeated several times in Pursuit of Glory), you're re-building a part of the mental model you've already built. Unless it's absolutely unavoidable, do it once and cross-reference.

The Once per Game Test: Rules need momentum and continuity. You need to build up an understanding of the core game systems before you deal with flavor rules. Any rules that applies only once per game should not be in the main body. Unhappy King Charles! is an offender here, with a couple pages of one-off rules right smack in the middle of the rulebook, breaking up the coherency the game system explanation. One-offs should be at the end, in their own categorized sections, unless there is a compelling reason otherwise.

In the same vein, one of the absolutely critical strengths of these card-driven wargames is that they can put a lot of these sorts of one-off rules - which are great for flavor but hugely problematic in terms of increasing real complexity - in the cards without burdening the player with learning them up front or having to remember them (in fact, for many folks not knowing exactly what is in the decks is a desirable feature of the first few games). One should leverage this. If a card explains its effect(s) perfectly clearly, it doesn't need a rule. Pursuit of Glory is again a repeat offender here, including (for example) rules 7.4.1, 7.4.2, 17.2.2 & 18.2.2 which, while admittedly short, are still unnecessary.

The Less than Once per Game Test: Any rules that take effect less than once per game (on average), whether they are rules that cover oddball situations that rarely come up or are chrome, should also be removed from the main rules. If they are patching up the rules to cover rare but awkward situations they should be in footnotes or something similar. If it's a real rule that has an application of less than once per game, you should first consider if you really need it, then put it somewhere where it isn't going to bother anyone.

Bad Game Systems Make Bad Rules: If you're having a hard time explaining something, it may be the fault of the game system, not the rules. If, as in Pursuit of Glory's section 11.2.2, you find yourself apologizing that seriously, this rule is actually really simple and just hard to explain clearly, you have officially entered the swamp. Which Turkish and Bulgarian LCUs can't do, apparently.

Designer's and Historical Notes: I love designer's and historical notes. I don't love them when they break up the flow of the rules. Too often they just serve to provide historical rationalization for bad rules, and they rarely, if ever, serve to help learn, clarify, or remember things. Put them at the end as a serious piece, like Avalon Hill used to do. Alternatively, do what Columbia does and have a three-column format, two with rules, and one with historical notes, designer's notes, optional rules, and other interesting tidbits where you can both see them (if you're interested) without having to delve into the rules, and also delve into the rules without being distracted by them.

An Index: If you have more than 12 pages of rules, have an index. Seriously. It's not that hard these days, and it has the added bonus that if your index is a mess, your rules are probably a mess too. It's insane how many complicated games don't have indices. Like Pursuit of Glory and The Devil's Cauldron.

Write Rules: While a game may have a goal - to teach some history, to espouse a theory of mobile warfare, to explain why things happened the way they did - once you cross the threshold from light to medium-weight, the game’s rules’ only goal must be to build the player’s mental model. That’s it. The rules are not the place to defend your design decisions, put across your point of view, or explain the history. The rules must be designed to cleanly and clearly explain the game system(s), nothing more. The systems themselves are, after all, supposed to be the vehicle through which you do all that other stuff and should stand on their own. Anything else belongs in footnotes, Designer’s Notes, Developer’s Notes, Historical Notes, More Notes, Appendices, Further Reading, Historical Booklet, Further Notes, or whatever.

As I finish writing this piece, I realize my goal - of setting down some hard and fast rules for writing rules to more complicated games - is obviously bigger than I could hope to tackle. So I ask you to help me out here. What are the worst mistakes you’ve seen rules-writers make, and what would you do to correct them? What are some of the best rules styles you’ve seen? For my part, I think Ted Raicer does good job - his original rules for Paths of Glory and WWII: Barbarossa to Berlin stand as good models for card-driven wargames of this sort. Even though they’ve become a bit needlessly bloated in recent updates of the living rules, they’re still pretty good. Mark Simonitch and Vance Von Borries also do a good job. While I might sometimes quibble with the follow-through, I think Richard Berg knows how to do this stuff properly. I like Columbia’s format a lot for their higher-end games like EastFront and Rommel in the Desert. On the other hand, GMT’s line of card-driven wargames has a lot of entries with painfully bad rules.

Saturday, January 10, 2009


Reiner Knizia's and Valley Games' Municipium is one of my favorite releases of 2008. It's a classic recent Knizia game, reminiscent of Blue Moon, Blue Moon City, Tower of Babel, and Beowulf in its ability to cram so many interesting decisions and so much flavor into such a relatively small set of rules and a relatively short but high-intensity playing time.

When I think about the overwhelming majority of games, I think about them having a couple or a few distinct game systems that interact in interesting ways. Take Agricola: you've got game systems for growing crops and maintaining herds of animals and playing occupations, but those game systems interact only lightly, in the sense that you have limited actions to spend on one or the other, but your farm and your herds are managed separately.

In Municipium, there is a lot of stuff going on – competition for citizens, the Praefect, building special powers, and turn order – but everything interacts heavily with everything else, and it's hard to pick out individual game systems. Even thought it might look like a worker placement or area control game, it's not; it seems to me really just a single large system which has some elements of both.

Which leads me to what is my biggest problem with Municipium, and that is how to pitch it when people ask you what you want to play. Games are easier to pitch when they are like something. The classic example for me is Agricola, which to some people can be sold as "a lot like Caylus, but actually fun". When a game fits into nice categories, like tile-laying or auction or negotiation or area-control, or more recently worker-placement or role-selection, it's easy to sell. You can get 80% of the way there using a few words to describe the basic idea, and then talk about what makes the game unique or unusual (like Agricola's asymmetric player positions and diversity of cards). This doesn't work here.

Interestingly, I've decided that the best way to sell recent Knizia games like Blue Moon City, Beowulf, and Municipium is to go directly to the theme and not try to pitch mechanisms at all. After all, the large majority of gamers buy and play games for their themes, however expressed, not their mechanical workings. Even though these recent Knizias are fairly simple rules-wise, the mechanisms are too involved or ambiguous to explain in a brief pitch. Trying to sell Beowulf as an auction game is not the way to go, even though the central driver sort of looks like it might be an auction, and the same goes for selling Municipium as an area-control game. But if you describe it as influence-gathering in Imperial Rome, talk about influencing the citizens or the Praefect or going to the Baths to hobnob with the rich and powerful or the Tavern to get your opponents drunk, that's something you can get traction with. And, helpfully, it's what the game basically delivers.

Sunday, January 4, 2009

Battlestar Galactica

I'm a little late to this particular party, but I finally had a chance to play the new Battlestar Galactica a few days ago. I was conflicted going into the game: the reviews had been good (they usually are), but Fantasy Flight Games' track record isn't always, and I love Lord of the Rings but hate Shadows over Camelot. I also wanted to like the television show, but couldn't.

So I was some combination of surprised and relieved when BSG was pretty fun.

The game actually makes a bit of a bad first impression, unfortunately. It beats you over the head with a lot of complexity, from the traditional overwrought FFG rulebook to critical references that should be in an easy-to-see place on the board but aren't (it's not like there isn't plenty of dead space) to a lot of critical rules detail that can only be found in tiny type on the board and simply cannot be seen if you lack a high-power spotlight, are viewing at a distance of greater than one foot, or are over the age of 30.

But, if you get past this, there are good ideas here. The foci of the game are the hyperjumps the Galactica has to make as they plot their course to Kobol. These are checkpoints where the board is periodically cleared of threats and the game timer resets, and it is a great way to segment the game and manage tension and ensure a semi-regular restart so the players don't get into a death spiral the way they do in Shadows over Camelot. The hidden loyalties are well-executed for the most part – dealing them out in two batches, at the beginning and mid-game, similarly ratchets up the uncertainty and tension and avoids some problems (and is true to the show).

Where BSG struggles, though, is with pacing. The game is just too long and gets too repetitive, is too much of a kitchen-sink type game and so has too many moving parts, and is too much at the mercy of the draw deck for its tension. Some stages will be terrific as Cylon raiders pile into the Galactica in waves while food shortages develop in the fleet and morale collapses. Some not so much, as you spend half an hour dealing with relatively uninteresting crises that never develop while waiting to jump. The system of crisis cards, where each turn the players draw a crisis from the show which they must resolve using skill cards, is clever but is simply not enough to reliably deliver tough and interesting decisions on its own. Things only really get fun when you have bad guys swarming and interesting crises going on at the same time, and for this to happen, you need things to come out in the right proportions, which they too often don't.

In general, there are just too many moving parts which are not tight enough. To be grossly general, to the extent that we're willing to call games art, they are the art of decisions. Music generates feelings and emotions through sound. Literature is the art of words. Painting is visual art. Games create their impressions, feelings, and emotions through the decisions they ask you to make. Every complaint people make about games ultimately boils down to a problem with the decision-making (i.e., too much luck = my decisions don't make enough of a difference; too much downtime = I make decisions too infrequently; brain-burner = the decisions are too hard; the theme is a paste-up = the decisions I make don't seem authentic; and so on).

In Battlestar Galactica, the players manage many more resources than they do in Lord of the Rings. BSG has food, people, morale, fuel, fighters, transports, dual-use skill/action cards, cylon fighters, cylon mother-ships, cylon boarding ships, cylon boarders, nuclear weapons, and political cards. Plus every player has a once-per-game special power. In Lord of the Rings, on the other hand, the players "only" have skill cards, action cards, ring tokens, life tokens, shields, and corruption. And despite the fact that BSG goes on for 3-4 times as long as Lord of the Rings, and even though it has so many more resource types, it still seems to manage to generate fewer interesting, really tough decisions than Lord of the Rings.

Now, part of this is because the players are secretly split into loyal Human players and Cylons, and much of the tension in the game is figuring out who is who. And that more or less requires a fair number of small decisions for the players to look at, so loyalties can be revealed over time. To the extent that the game succeeds - which is not insignificant - it does so in this aspect of it, as each decision is scrutinized for signs of treachery, and players banter around accusations, threats, and general paranoia. Not unlike in the show. But you simply can't create an interesting game out of a lot of uninteresting decisions, and here the decisions - whether how to resolve crises, or figuring out who the traitor is - are not reliably interesting, compelling, narrative, or evocative. At the end of the day, I can't help but think this game could be much improved if it were the euro that in its heart of hearts it really is, and wasn't trying to be the overwrought super-themey sort of thing FFG specializes in - stuff that always delivers a boatload of rules but doesn't always deliver a plausible theme or a plausible game. Usually less is more, and this applies to theme as much as anything else. Here as much as anywhere, a tighter, tenser game would be thematically far stronger than this kind of kitchen sink game.


As long as we're talking about BSG the game, I can't resist putting in my .02 on the show. One of the reasons that the new Sci-Fi Battlestar Galactica ultimately turned me off was it really only had one tone: grim. It completely lacked any emotional range. Real people are sometimes funny and crack jokes when they're under stress. BSG characters always take themselves so excruciatingly, painfully seriously. This was especially funny in context of playing the boardgame, where each character from the show is brutally, and totally effectively, boiled down to three traits, a characteristic, a special ability, and a flaw. When you put it like that, BSG the boardgame becomes a rich mine of humorous possibilities, and if only the show had been able to capture some of the humor we found in the game, maybe it wouldn't suck.