Time to talk for a bit about one of my favorite RPG subjects — skills and skill systems.
A few weeks ago, game designer Rob Donoghue posted an article on his blog about skills and skill failure, and it’s gotten me to pondering a bit. Rob’s a talented game designer and analyst, so he’s good at making you think about such matters. ;) So I thought I’d riff a bit on Rob’s ideas, offering my own spin and opinions and hopefully more food for everyone’s brains.
How Long To Succeed?
I’m really intrigued by Rob’s idea that maybe the foundation of a skill system could be not the success/failure binary, but the idea of how long it takes you to succeed. As in the Gumshoe system, success is assumed; the issue is how much time it takes a character to get there.
Obviously this would work very well for any sort of skill that involves the application of a character’s abilities over time. This includes research, many acts of deduction, various engineering and gadgeteering projects, and so forth. That gives you a great dramatic tension: will the hero succeed before something negative happens? Of course, the GM has to set up the “something negative” and give it proper play during the game, but that usually isn’t too hard.
On the other hand, this wouldn’t work at all for situations where you need to use a character’s skill to determine success or failure at this very moment. To take Rob’s example of a Riding skill, what a character usually needs Riding for is to determine “do I fall off the horse or stay in the saddle?”. That’s not an issue where “he’ll succeed eventually” matters — it’s one where determining what happens to the character immediately as a result of his use of his skill (or lack of skill) is crucial. If he doesn’t succeed now, he falls, and that’s going to have serious repercussions right then and there (e.g., he gets trampled by an enemy’s horse).
So maybe the first step is to divide all skills into two categories — for discussion purposes let’s call them “Action” skills and “Work” skills. Action skills are ones that are most often used in crisis situations (typically, but not always, combat), where success or failure matters right then and there. They’d be resolved in more or less the same success/failure binary way as attacks in combat are (though possibly allowing for a “degree of success or failure” system such as many games have). Work skills are those where the game assumes characters will succeed and the real issue is how long that takes.
One question that occurs to me is, are there considerations a game would want to take into account for Work skills other than “how long it takes” (or as Rob puts it, “before”)? Offhand I can’t think of any, but I have no doubt the collective brain of the Internet will have some ideas. Maybe using up some limited resource would count, but on the other hand that could just be an expression of “before we run out of so-and-so.”
As Rob notes, there are plenty of other potential issues involved (like skill-versus-skill resolution), but since I’m writing a blog entry and not designing an RPG I think I’ll move on to a few related points. ;)
The Value Of Failure
Rob calls out the general fail/succeed binary of most games’ skills as an “implementation failure” — as an adaptation of the general combat mechanic to skill resolution, which isn’t a good idea per se. I don’t necessarily think Rob’s entirely wrong here, but I think perhaps he goes too far.
As I’ve written elsewhere in this blog, I think it can be argued that the chance for a character to fail at what he does in an RPG is one of the most fundamental, and fundamentally good, things about RPGs. It creates a true dramatic tension that rarely exists in movies or books, and it gives us a stake in our characters that we can never have in a fictional character.
If one accepts that argument, then at least some of Rob’s ideas become unnecessary — they’re attempts to fix a “problem” that doesn’t really exist. Or, to put it another way, if you think this is actually a problem, perhaps that should dictate the choice of game you play. You probably want a “storytelling game” (which isn’t really a game, or is barely one; it’s more about forming a narrative, I suppose you’d say) instead of a “roleplaying game” (which, IMO, has to mix equally elements of “roleplaying” and “game,” or it’s something else entirely).
When To Roll
Of course, a lot of this gets back to the age-old discussion of “when should you make a skill roll?” Early RPGs often led people to believe, directly or indirectly, that you had to succeed with a skill roll to do practically anything. “We’re going to mount up and ride down the road.” “OK, everyone make a Riding check.”
Later games have gotten away from that, and have often taken some pains to explain that there are plenty of times when succeeding with a roll should be unnecessary. All this essentially boils down to a simple dictum, expressed by many people before me: don’t bother with a skill roll unless the outcome of the roll changes the game (or the flow of the story) in some way. If characters are just riding their horses down the road, there’s no need for a roll, because even if they fail it won’t affect the game (the character who fails will just remount and keep riding).
Naturally, there are a couple caveats here. The first is that sometimes an unexpected failure on a roll can create humor for the group, or can give the GM ideas for ways to direct the narrative that otherwise would never have occurred to him. Some of my gaming group’s best-loved stories involve failed rolls, and I wouldn’t be surprised if the same’s true of your group.
Second, particularly in grittier/harsher settings, it may be that virtually any roll can affect the outcome of the story. Character fails a routine Riding roll? OK, roll to see how badly he hurts himself in the fall. Oooh, broken arm — too bad, pal, no combat for you for awhile. That unquestionably affects the outcome of the game/story. It’s just an issue of how far the GM wants to push things and how much he wants to make himself and his players slaves to the random outcomes dictated by the dice.
But those exceptions aside, I think the wisdom of the dictum holds — and if properly applied, the dictum can eliminate a lot of the problems Rob discusses in his blog.
...what do you think?