Monday, May 27, 2013

Level Progressions: History and Experiments

Alright, I am on a history kick for some reason. This time I am looking at level progressions for the four "core" classes in the various "official" editions of D&D. I am mostly trying to get a feel for what the "traditional" progression should be for the first 10-or-so levels, but I'll make some comments on higher levels as well. We start, of course, with OD&D from 1974 (including the thief from Supplement I: Greyhawk; I'll use "wizard" instead of "magic-user"):

LevelFighterWizardClericThief
10000
22,0002,5001,5001,200
34,0005,0003,0002,400
48,00010,0006,0004,800
516,00020,00012,0009,600
632,00035,00025,00020,000
764,00050,00050,00040,000
8120,00075,000100,00060,000
9240,000100,000200,00090,000
10360,000200,000300,000125,000

There are a few things open to interpretation here, for example it is unclear that the progression for clerics is really supposed to be by 100,000 at high levels. The most obvious "problem" with OD&D is that wizards, while starting out slower, eventually progress faster than fighters and clerics, something that doesn't seem in line with their supposed power curve. There's also the weirdness that thieves are described as requiring 125,000/level after 10 which is higher than the implied (sadly not documented) requirements for any of the other classes? Let's look at AD&D from 1978 next (ignoring the useless +1 point issue):

LevelFighterWizardClericThief
10000
22,0002,5001,5001,250
34,0005,0003,0002,500
48,00010,0006,0005,000
518,00022,50013,00010,000
635,00040,00027,50020,000
770,00060,00055,00042,500
8125,00090,000110,00070,000
9250,000135,000225,000110,000
10500,000250,000450,000160,000

Although all classes get slowed down, there's still the noticeable "reversal" where things get easier for wizards at level 8. But we're also told that the "high-level advancement" is 250,000/level for fighters, 375,000/level for wizards, 225,000/level for clerics, and 220,000/level for thieves, so wizards slow down again quickly after that. Overall this doesn't seem less "broken" than OD&D to me. Let's look at B/X D&D from 1981 next:

LevelFighterWizardClericThief
10000
22,0002,5001,5001,200
34,0005,0003,0002,400
48,00010,0006,0004,800
516,00020,00012,0009,600
632,00040,00025,00020,000
764,00080,00050,00040,000
8120,000150,000100,00080,000
9240,000300,000200,000160,000
10360,000450,000300,000280,000

Finally we get a progression that makes some sense. First there's no "reversal" for wizards anymore, they now consistently require more experience points (XP) than other classes. Second we're told that "high-level advancement" is 120,000/level for fighters and thieves, 100,000/level for clerics, and 150,000/level for wizards, so there's consistency here as well. (The fact that thieves advance like fighters at high level is a small wrinkle at first sight, however notice that in B/X fighters and thieves both receive 2 hit points per level after the hit dice stop, clerics and wizards get 1 hit point per level instead.) Later developments (BECMI, RC) keep the same progressions. Let's look at 2nd edition AD&D from 1989 next:

LevelFighterWizardClericThief
10000
22,0002,5001,5001,250
34,0005,0003,0002,500
48,00010,0006,0005,000
516,00020,00013,00010,000
632,00040,00027,50020,000
764,00060,00055,00040,000
8125,00090,000110,00070,000
9250,000135,000225,000110,000
10500,000250,000450,000160,000

What I find amazing here is that despite knowing of a "better" (in my book anyway) progression from B/X, one that could even be made more like the original AD&D progression, the people designing the 2nd edition chose to stick almost completely with the original.

You might say at this point that comparing level progressions from widely different versions of D&D is not something that should be done, after all the abilities of the various classes differ a lot as well. But if you look carefully, those differences seem to have very little impact on what the designers did! The AD&D thief, for example, has almost the exact same abilities that a B/X thief has, but at a much lower XP cost; the AD&D thief even has a larger hit die! But it's even weirder than that: The person designing the "sane" B/X progression was the same person keeping the "insane" AD&D progression for 2nd edition AD&D: Dave Cook!

In any case, just to be "complete" let me mention the unified progression that came with 3rd edition "D&D" as well:

LevelXP
10
21,000
33,000
46,000
510,000
615,000
721,000
828,000
936,000
1045,000

This is obviously extremely different from all the previous progressions. For one thing the XP requirements are a lot smaller starting at about level 6. But also this progression makes it a lot harder for new characters to catch up to characters that avoid death and keep collecting XP. Interestingly this is the same exact progression that a German roleplaying game called "Das Schwarze Auge" used in 1984 (except that they divided by 10).

I am sure you can tell that if I had to pick one of these progressions as the basis for a new D&D variant, my choice would be B/X. However, in the end, I would probably do away with the "different tables for different classes" approach entirely: If you average across the B/X table above, you'll see that the required XP/level across all classes is always slightly below that for the fighter, and the fighter works out by roughly doubling the required XP starting from 2,000. That's probably close enough for all of them if you tweak the abilities of each class appropriately.

As a final note, check out what happens if we start with the AD&D requirements and then strictly double them each level:

LevelFighterWizardClericThief
10000
22,0002,5001,5001,250
34,0005,0003,0002,500
48,00010,0006,0005,000
516,00020,00012,00010,000
632,00040,00024,00020,000
764,00080,00048,00040,000
8128,000160,00096,00080,000
9256,000320,000192,000160,000
10512,000640,000384,000320,000

In my book, that's a mighty fine progression. If we care to extend it to higher levels in the old-fashioned way (switching to a linear progression), we can simply go with 256,000/level for fighters, 320,000/level for wizards, 192,000/level for clerics, and 160,000/level for thieves. That's not too different from the original AD&D progression either, but it follows out of a very simple formula with no arbitrary patching. A winner if you really care to maintain the feel of different XP progressions for different classes.

I simply use the fighter column for all classes but keep doubling forever, which means that for all intents and purposes, the game is capped at level 12. Individuals with higher levels will be extremely rare, and anyone who really wants to be an arch-wizard (able to cast a level 9 spell) essentially must consider the path of the lich (and keep adventuring or build a huge empire) or be an elf: level 17 requires a breath-taking 65,536,000 XP!

Works just fine for me. :-)

1 comment:

  1. The original 1974 version of Empire of the Petal throne gives all characters the same XP chart which matches the BX Fighter chart up until level 10 (where it is 340,000 instead of 360,000).

    An interesting note that in the published version of EPT, the XP charts are modified to give the Magic User a slightly lower required XP for 8th and 9th levels while keeping the Priest/Cleric and Fighter the same.

    Somebody thought high level MUs where just too darn weak.

    Also, Dave gives a formula for earliest XP/level in his First Fantasy Campaign. It was 1,000 XP to get to 2nd level and then each level was 50% more than the previous: 1,000 -> 2,000 -> 3,000 -> 4,500 -> 6,750. That's all he gave.

    Dave's formula matches the XP/level values in TSR's Warriors of Mars listed there as 1,000 -> 2,000 -> 3,000 -> 5,000 -> 8,000 -> 12,000 -> 17,500 -> 24,000 -> 32,000 -> 45,000 -> 60,000.

    So, it looks like there are plenty of choices.

    ReplyDelete