In a recent Rule-of-Three article (via Erik Tenkar via Keith Davies), Rodney Thompson wrote about varying character generation complexity by starting above first level. As the mechanical complexity of building a 3E or 4E character is one of my biggest complaints about post-TSR D&D, this is of great interest to me. To quote:
Additionally, we’re looking at having the classes gradually layer in more capabilities over the first two or three levels, rather than providing a large number of class features at level 1, so that players new to the class have a short period of time to learn the basics of their class through play. Experienced players could simply start at 3rd level if they want to leap right into a more advanced starting experience.
I think this is an excellent approach, and one that I have been advocating for some time. I have observed before that in terms of both power and complexity a first level Fourth Edition character is approximately equivalent to a 4th – 6th level traditional D&D character. So why not let every group start at the place that they want to?
The important principle here is to expose complexity slowly through play rather than all at once at the beginning. That also makes the game more social, as other players will experience your character’s progression. The magic-user, for example, is a very complicated class, even if you restrict available spells to the core rules of whatever edition you are playing. The only reason that this complexity is tractable (and why the class remains playable by people not interested in system mastery) is because the spells are introduced gradually during level progression (see also this post by Jeffro for the apotheosis of this insight).
I would also like to see more advancement options tied to events within the game world. Thus, a fighter who wants to learn a special whirlwind attack ability might need to seek out a master to learn from, or discover an ancient manual of techniques. Then, they could use their next advancement point (or whatever) to learn that ability (maybe in place of a base attack bonus increase). This increases the uniqueness of individual characters and combats bonus inflation while not adding any complexity. Positive effects all around.
The framing here is critical though. It’s important that players not see the first three levels as training wheels that experienced player should want to bypass. Rather than capabilities layered in over levels 1 through 3, I think capabilities should be layered in slowly over the entire level range.
I also think there is an inverse relationship between potential setting deadliness and work required to build a new character. If you want save or die with teeth you can’t have character generation as complex as 3E or 4E. Players simply won’t stand for their special snowflake PC (which took an hour or more to build) being killed in the first ten minutes of play because they did something stupid. To be honest, I really can’t see any other way that you can get both players who are interested in quick character creation and players who are interested in complex character builds to play the same game.