I've been thinking about posting this thread for a while. I've been prompted to do so by @The-Magic-Sword's interesting thread about neo-trad RPGing: Thinking About the Purpose of Mechanics from a Neo-Trad Perspective
Back in 1981, Lewis Pulsipher gave the following advice about designing character classes for D&D (it's from White Dwarf 25, though I know it from the Best of White Dwarf Articles v2):
When I first read it, I took it as gospel - not that I designed very many classes, but I think it probably helped me build my sense of what counts as broken in AD&D build options.
But in more recent years (probably the past decade or two) I've started to change my thinking: particularly about the "19s and 20s" bit. It seems to me now that I want the ordinary play of an ordinarily-built PC to emulate (more or less) the ordinary feats of the literary or mythological inspiration. This doesn't mean auto-success - after all, those inspirational figures don't always succeed. (Even Gandalf is thwarted by cruel Caradhras.) But it means not requiring extraordinary luck to achieve, in play, feats that emulate the source material.
4e D&D was the first RPG that I grasped as revealing this possibility, though now I know of many more, some earlier than 4e (like Prince Valiant, Over the Edge and HeroWars), some more recent (like Agon, Marvel Heroic RP and even Torchbearer after a fashion). That's not to say that Pulsipher was wrong - his model in his article for RPG play is classic dungeon crawling, and this sets tight parameters around permissible player-side moves (eg Shadow Cat-style intangibility, or a Conan or Spider Man-level of trap avoidance, become broken in that context). But I haven't played that sort of RPG as my main thing for nearly 40 years, and so it doesn't need to frame my thinking today. In a RPG with more "open" fiction, and with non-dungeon techniques for establishing adversity and consequences, there's no reason why the play of a character can't and shouldn't reliably emulate the source material that inspires it.
Back in 1981, Lewis Pulsipher gave the following advice about designing character classes for D&D (it's from White Dwarf 25, though I know it from the Best of White Dwarf Articles v2):
Create a character class you could believe if you read about it in a good fantasy novel. . . .
Begin by giving the class powers at the high or "name" levels, say tenth or eleventh, equal to those you see in the tradition or story on which you base the class. Find some evidence of how the character fared against creatures or dangers already defined in AD&D. Say the character fought a bear - did he have much trouble? Even if the eleventh level Eldar or whatever killed the bear in two rounds in the story, a first level wont necessarily do as well! . . .
When you model a class after a group or character from a particular story, there are several things to keep in mind. . . .
[R]emember that protagonists of epic fantasy are "born lucky". They roll 19s and 20s for saving throws, and stumble into good positions. The character class should be able to reproduce the greatest feats of the model only when the character gets lucky, not as a standard action.
Begin by giving the class powers at the high or "name" levels, say tenth or eleventh, equal to those you see in the tradition or story on which you base the class. Find some evidence of how the character fared against creatures or dangers already defined in AD&D. Say the character fought a bear - did he have much trouble? Even if the eleventh level Eldar or whatever killed the bear in two rounds in the story, a first level wont necessarily do as well! . . .
When you model a class after a group or character from a particular story, there are several things to keep in mind. . . .
[R]emember that protagonists of epic fantasy are "born lucky". They roll 19s and 20s for saving throws, and stumble into good positions. The character class should be able to reproduce the greatest feats of the model only when the character gets lucky, not as a standard action.
When I first read it, I took it as gospel - not that I designed very many classes, but I think it probably helped me build my sense of what counts as broken in AD&D build options.
But in more recent years (probably the past decade or two) I've started to change my thinking: particularly about the "19s and 20s" bit. It seems to me now that I want the ordinary play of an ordinarily-built PC to emulate (more or less) the ordinary feats of the literary or mythological inspiration. This doesn't mean auto-success - after all, those inspirational figures don't always succeed. (Even Gandalf is thwarted by cruel Caradhras.) But it means not requiring extraordinary luck to achieve, in play, feats that emulate the source material.
4e D&D was the first RPG that I grasped as revealing this possibility, though now I know of many more, some earlier than 4e (like Prince Valiant, Over the Edge and HeroWars), some more recent (like Agon, Marvel Heroic RP and even Torchbearer after a fashion). That's not to say that Pulsipher was wrong - his model in his article for RPG play is classic dungeon crawling, and this sets tight parameters around permissible player-side moves (eg Shadow Cat-style intangibility, or a Conan or Spider Man-level of trap avoidance, become broken in that context). But I haven't played that sort of RPG as my main thing for nearly 40 years, and so it doesn't need to frame my thinking today. In a RPG with more "open" fiction, and with non-dungeon techniques for establishing adversity and consequences, there's no reason why the play of a character can't and shouldn't reliably emulate the source material that inspires it.