I've always used the original method of calculating XP by counting up gold pieces and kills. Over the last few years I've also started awarding XP for magic items using the charts in the AD&D DM's Guide. A lot has been written about the pros and cons of this traditional approach, so I won't go into it here.
I have to admit, though, that I am certainly not alone in finding it frustrating when mid-level characters have to acquire swimming pools of gold pieces, impossibly huge gemstones, or Satan's Pitchfork itself to advance at a reasonable rate (at least given my gaming frequency, which is only two or three sessions per month). I don't have a specific objection to massive wealth accumulation - it worked for Conan, right? - but sometimes it doesn't jibe well with the flavor I want to go for in my games. I mean, goblins wearing jewelry is pretty weird and cool and I'm definitely down for that, but it gets progressively more difficult to keep monster stashes fresh and in line with a gygaxian naturalist universe. Also, the arnesonian approach of squandering gold for experience doesn't work in my games since characters are almost always traveling through some manner of proverbial Night Land, not hanging out in taverns.
There are a huge number of XP tweaks and alternate systems out there. My head starts spinning when I think of trying out new house rulings. There's a tricky trade off spanning multiple parameters including complexity and ease-of use, transparency, rigor and consistency, and meta-gaming effects (i.e. players change their character's actions in order to game the XP system). I see three common approaches to houseruling advancement:
XP-free leveling systems:
This probably works fine, but I have never tried it. This approach would completely nullify traditional class-specific advancement. Advancement tables are such a fundamental aspect of Dungeons & Dragons, I find it difficult to walk away from them. I don't give a crap about game balance, but class-specific advancement rate is an important stylistic component of D&D that underlies much of the flavor of the game.
Ad hoc XP systems:
ad hoc or abstract XP systems assign experience on a purely subjective basis at the end of each session. This approach is nicely described thus at the Paper & Pencils blog:
"Almost every game I’ve run as a GM has used a kind of ad hoc experience distribution system. I look up how many experience points are needed for the characters to reach the next level, and I give them whatever percentage of that number which I feel like they’ve earned. Most of the time I base that percentage on what speed of progression is optimal to keep the players in-step with events in my game world, rather than basing it off of challenges they have overcome."
In my opinion this opaque system is not very fair to the players. If a player chooses to be driven by character advancement, they should have some objective handle on what will work for them. I think effective refereeing requires some degree of objectivity, even if it's partially an illusion.
Going back to the very beginning of the hobby individual referees have tweaked the specific categories of experiences that can result in XP awards. Gygax's personal choice was to keep it simple - treasure and kills - but David Hargarve's Arduin, for instance, had a different, expanded set of categories that included interestingly sensible things like XP for being cursed, resurrected, or serving rear guard. Tweaking XP categories to fit one's campaign is time honored, old school tradition. It forces some important considerations, though:
Individual vs. Party XP: Moldvay was very specific that XP should be given to the party, and then divided equally among characters. This is what I have always done. It's so fast and simple. I also think it discourages tedious treasure grubbing conversations among players during or after a session. Individual XP is attractive because you can reward specific characters for noteworthy experiences or achievements, however routine implementation can result in serious bookkeeping overhead and arguments about referee fairness.
Slippery slope of categories: The obvious danger that Gygax worked to avoid by sticking to the original XP system is the increased complexity associated with adding categories. It's easy to write down dozens of things characters should get XP for, but what's the tradeoff in terms of bookkeeping, "realism", and enjoyment? I don't know...
Motivation: Your selection of XP categories could drastically influence your player's decisions. XP for gold encourages tomb raiding. XP for completed quests encourages railroading. XP for being cursed encourages self destruction (kind of cool, actually). None of this is necessarily good or bad, but it's real.
So... IF I decide to start houseruling XP to some degree, here's what I would probably do:
- Keep treasure and kill XP categories in place
- Add a few more categories to reward exploration, NPC interaction, and discovery
- Stick with party XP
- Possibly have a very small list of individual hargravian rewards, like for resurrection and curse
- Make a simple checklist to track party XP for each session
Hopefully I'll get a chance to think and post more on this topic. It's something I want to resolve in my mind. I'd be interested to hear about other ideas or approaches, especially if they've been playtested.