Welcome to part four of our series of posts on our Roll to Advance alternative experience and advancement system for our fantasy RPG. In short: At the end of each playing session, the player rolls a d20. If the modified roll exceeds a specified number based on the character’s class, race, and current level, the PC advances to the next level. Accumulated XP, awarded at a rate of 1 (one) per gaming session, provide a positive modifier to this roll. Traditional experience point awards and tracking are eliminated. We’re using it in our modified Labyrinth Lord game, but it should work similarly in any old-school version of the game.
Part 1 introduced the system and outlined its basic operation. Part 2 looked at the specific numbers for each of the standard character classes. Part 3 discussed demi-human races and the penalty paid by these characters, plus the elimination of the racial level limits. Today we’re looking at multi-classed characters.
To say that I’m not really a big fan of multi-classed characters would be a bit of an understatement, at least as they’ve been run before. And the dual-class option available to human characters in several editions of the game is even worse. So I’m taking this as an opportunity to made widespread changes to the multi-class system. This new approach looks like it will work quite well in our game, but others may not be fans. So I’m offering two options. No doubt there are others. My recommendation would be to decide upon one of these (or another) and stick with it rather than allowing multiple methods of multi-classing. But that, of course, is up the to players in each campaign.
The first alternative would be for the player simply to choose which class he or she wants to attempt to roll to advance at the end of each session. This is quick, easy, and will give results similar to the traditional method of multi-classing. Accumulated XP could be used toward either class, and a limit could be placed on how far apart the classes could be. Say, no more than three levels may separate them. Alternatively there would be no limit. Nothing would stop a player from making a 10th/1st level fighter/magic-user.
Example: A magic-user/thief may elect to roll to advance in level in either magic-user or thief, but not both. If the magic-user/thief reaches level 5/2, she may not attempt to advance further in magic-user until she gains at least one more level in thief unless there is no limit on the gap between classes. Attempting to advance to 6th-level magic-user would require a 28, while advancing to 3rd-level thief only requires an 18. XP used to gain levels in one class are used and may not be applied toward the other class after a later session.
When creating the character, roll both hit dice and divide by two. When rolling hit dice upon advancing, roll the die appropriate to the class being advanced and divide by two. Fractions could be retained to be used later, dropped, or rounded up. Another option would be to round the larger die down and the smaller die up.
Something that I think is cool and fun about this approach is that the player has control over exactly what sort of character they work for. I can see the 10th-level fighter/1st-level magic-user being interesting to play, if there is no limit to the gap between classes, and it could be used to re-create characters from literature. In fact, this would not be all that different from the somewhat odd approach taken with elves in the original (pre-supplement) D&D where the player had to choose each day which class to play and earn experience for, except that all earned abilities from both classes would always be available. A problem with this, though, would be a 1st/10th fighter/magic-user who would be able to use the weapons and armor of a fighter while only bothering with the rolls to advance of a magic-user.
The second approach, the one we’re trying out now in our own game, does away with tracking separate levels for multi-classed characters. A fighter/cleric begins at 1st level, then advances to 2nd level, then to 3rd, and so on. The abilities in each class equal those of a single-classed character, and abilities in both classes go up a level upon a successful roll to advance.
Hit dice for this single-level multi-class option alternate, with the larger hit die being rolled on odd-numbered levels and the smaller hit die being rolled on even-numbered levels.
Example: A starting fighter/cleric of 1st level rolls a fighter hit die (d8 in Labyrinth Lord) during creation and has all the abilities of a both a 1st-level fighter and a 1st-level cleric. After a few sessions, the player successfully rolls to advance to 2nd level. At that point the player rolls a cleric hit die (d6 in LL) to increase hit points and gains all the abilities of a 2nd-level fighter and a 2nd-level cleric. Upon gaining 3rd-level, the player will roll another fighter hit die, and so on.
The required roll to advance in this single-level scheme is calculated by taking the two target numbers for the two separate classes and deducting 12. Triple-class characters deduct 24. Several of the more common multi-class combinations are listed below:
As the numbers increase at higher levels, the deduction of 12 per additional class becomes an increasingly smaller proportion of the total. This represents the growing difficulty in mastering more than one discipline when compared to a single-classed character who does not have to split his or her attention.
Remember that demi-human characters will have to add their racial modifier to this number at each level if using the system which eliminates level limits.
Speaking of racial modifiers, I’m thinking that in our game we will open up at least some of the multi-class options to human characters. The final decision on this is still to be made, though, and it is the sort of thing that will probably vary from campaign to campaign in any event.
A final word about a special multi-class combination: the Bard. In our game, bards are going to be multi-classed fighter/thief/illusionists with some additional musical abilities. Exactly how this is going to be worked out, I don’t quite know yet. But the potential is there with this system.
NEXT: A final summary and wrap-up, including discussion of some of the feedback I’ve received so far.