This is part of my RPG series of entries here at SOB. See the inaugural entry in the series for more details.
In running high level ROLEplaying games, I said:
The biggest problem of bookkeeping in a roleplaying oriented campaign isn’t on the player’s part, by any stretch of the imagination. The GM has a tremendous burden, dealing with high level NPCs and trying to manage the game in light of what the entire group of PCs can do at high levels, and this is only made worse in cases where the campaign is focused on roleplaying because of the greater numbers of NPCs required as well as the greater depth of NPC development that is needed. The only fix for this sort of thing seems to be inspiration plus elbow grease.
Bookkeeping isn’t really as big a problem for PCs as one might think, at least in games where characters aren’t already too complex at starting power levels, and at least in cases where the PCs developed naturally over the course of many sessions of play. When playing a character for a long time, starting out at simple, low-power status and developing from there, a player develops a relationship of sorts with his or her character, and the player’s understanding of the character and its abilities grows with the character. It is typically only when starting characters at high levels that players really have a huge problem, in a roleplaying sense at least, with bookkeeping — because the player isn’t used to the plethora of capabilities available to the character at those higher power levels, and will often forget about various abilities during play that should come as second nature to the character itself.
This is, in fact, the exact same problem GMs have with high level NPCs all the time, regardless of the style of play. If you, as a GM, intend to keep NPCs true to the rules, you have a tremendous bookkeeping burden. This makes it quite difficult at times to not only play the character appropriately with dozens of feats and class features and spells and so on listed on the NPC’s character sheet, but also to create NPCs quickly enough to keep up with the pace of play. While this problem has always existed in D&D to some extent, it became especially problematic with D&D 3E, and hasn’t really been relieved by 4E.
Bookkeeping may not be much of a problem for players most of the time, but it really is the bane of the GM in high level campaigns because of the complexity of high level NPCs and the sheer number of NPCs that show up in any game that isn’t a pretty strict dungeon crawl campaign oriented toward generic monster enounters. RPG campaigns that focus on playing roles (rather than resolving rolls) are hit especially hard, in part because of the greater depth of characterization needed for NPCs, and in part because of the greater quantity of NPCs required to keep the game moving along. Aside from random NPC generators (and I’m working on writing one myself a little at a time, to suit how I run games), I don’t know of many reasonable fixes for this problem besides hard work and a nimble mind.
There are ways to “cheat”, of course. As Joseph pointed out in a comment responding to running high level ROLEplaying games:
In regards to the amount and depth of NPCs, I find it helpful to keep vital statistics on all past NPCs so that they can be resurrected in a new incarnation. It literally halves the prep time in that regard.
It’s a good technique for lightening the burden, and often with just a few tweaks here and there — tweaks that can be made on the fly — NPC stats can be modified strategically to mask the fact they’re retreads. That, of course, tends to be useful mostly for NPCs that fit within certain common molds and don’t get too tightly integrated with the ongoing game, since it leads to a certain amount of generic feel for the NPCs. For more richly realized NPCs with greater depth, they need to be truly unique, at least within that game world. Changing the names and a couple of minor details (swap out the longsword for a bastard sword, trade numbers on thieving skills with those on social skills, and change colors in the description) only goes so far.
You can also go outside the current campaign for help, though. This works to some extent with borrowing stats from previously used NPCs, like above, of course, but only within certain limits. Sometimes, a set of stats may not translate very well because of setting quirks that may invalidate the basis of an entire character sheet. Using only examples from real homebrew settings in my own life: maybe a set of Wizard stats is inappropriate to a Midian campaign, where a set of variant arcane spellcasters replaces the Wizard; a priest of Oroboros in the Serpent’s Spine may be fundamentally incompatible with the Velesh setting, which doesn’t even have a typical D&D pantheon; an Orcish psionic Wilder in Neiäth would surely be incompatible with almost any other campaign setting. Even when character stats do translate well across settings, though, if you run games in multiple settings, or if you start borrowing from your PCs in other GMs’ campaigns in their own settings, you run the risk of being thought too attached to certain “types” of characters, if you do this more than occasionally.
All of the above suffers the problem that changing some details about character stats, slapping on a new name and basic description, and making up mundane background details as needed, can lead to your NPCs all having a very generic feel, though. Rules help to flesh out and support concepts, after all, and lend a stronger flavor and sense of uniqueness to a character built on a strong conceptualization. As stats get shared around more, they become less uniquely suited to supporting any particular concept, which means that the concepts for the NPCs start looking less unique. With luck, that’ll be the worst of your problems, because there’s always the possibility you’ll try to make the character “unique” with lame, Mary Sue style gimmicks, and superficial alterations. We really don’t need any more red-headed lesbian Elves, who are long-lost heirs to the noble Armand family, with wings and a tragic past in which they were sexually abused. Trust me — it has been done to death, and it wasn’t even good the first time.
There’s an easy way to get roughly instant NPCs with real depth, of course. I’ve been known to actually crib not only NPCs in other games I’m running (particularly when two campaigns are taking place in the same campaign setting), but also my PCs in other GMs’ campaigns, too. This results in NPCs with deep, rich backstories, and it’s trivial to make minor changes — often on the fly — to fit them better into the current campaign if need be. I’ve had both NPCs inspired by PCs of mine and PCs inspired by NPCs of mine. Once they’re introduced to a game world, they begin to diverge from their counterparts in other game worlds, and the players seem to find these among the most interesting NPCs, for obvious reasons. This is definitely something to do only rarely, though.
In cases where the players aren’t already familiar with the character, it can work brilliantly for obvious reasons, and doesn’t run the risk of sapping the concept of all sense of uniqueness when adapted to a new campaign setting. Where the players are familiar with the translated character, you must be more discerning in how you handle things; make sure you don’t have the appearance of creating a “GM’s pet” NPC, you don’t do it so often that the players get tired of never meeting anyone new, and you don’t suck all the fun out of meeting new NPCs by introducing characters the players already know everything about in circumstances where someone mysterious and new would be a better fit. Done well, though, you can definitely enhance the quality of play, not only with an already richly developed character being ready (for the most part) to just drop right into the game, but also with the added excitement amongst your players of being able to interact with a character they’ve grown to know and love under other circumstances.
Then, of course, there’s always the expedient of “borrowing” NPCs from other sources, such as published campaign settings and adventure modules, online accounts of other people’s ongoing games, and so on, with changes made where you feel they’re necessary to add your own flavor to them or eliminate some details you find “dumb” or otherwise unsuitable to your campaign. I think it was Albert Einstein who said that the secret to creativity is knowing how to hide your sources. This can help immensely with the mad scramble to come up with a developed character concept when you need one right now, but it doesn’t really solve the problem of being familiar and comfortable with playing the stats on the sheet. You may end up smacking yourself in the forehead when you realize you’ve been forgetting about the character’s always-on ability to detect psionics.
Ultimately, though, the dedicated GM of a roleplaying oriented campaign is sure to want to actually create NPCs (mostly) from whole cloth a lot of the time. I definitely do. The less your campaigns rely on characters that also appear in other contexts, explicitly or beneath a thin veil of midsirection, the more the NPCs will acquire their own unique flavors. That requires a lot of time and effort for high level campaigns, though. For some GMs that time and effort requirement can be reduced by using some largely random system to come up with the skeleton of a character concept, then flesh out the NPC with off-the-cuff decisions made during play. In the end, though, it comes down to a balance of hard work and a nimble mind.