[D&D '74] Experience Benefit Analysis

[Marshall Burns]

This is the first part of a project that is an exploration of the currency of older forms of D&D, and some new ones derived directly from those older forms. Specifically, it's an analysis of the experience point, and of the amount of Effectiveness and Resources an experience point is worth. The issue of domains, with all the Resources and Positioning contingent on them, has been excluded, because its impact on the game cannot be examined in abstract terms -- the benefits of a domain can only be analyzed in terms of a specific setting in a specific campaign at a specific table. I concede that I may very well be remiss in excluding domains from my analysis, but I don't see a way around it.

My goal in doing this analysis was more to arrive at a personally satisfactory currency based on the experience point for my own OSR-type purposes, rather than to calculate the precise, canonical worth of the experience point in old-school D&D. I figured that this analysis would make clear to me the bits that I liked and didn't like about how the textual D&D currency already works, thus giving me information on which to base my own construction. I intend to continue this analysis with AD&D1 (or at least OSRIC), LotFP, and ACKS. I'd like to do B/X and/or BECMI as well, but I don't have access to them, so that'll have to wait indefinitely.

As for why I'm bothering to work within the OD&D-based OSR-type framework rather than designing something new from the ground up: please put the question aside for now. While I am not particularly averse to discussing the why of it, I don't want this thread to be bogged down with it. So, even if you have misgivings on that front, please assume that I have thought about it, that I have my reasons, and that, generally speaking, I know what I'm doing. We're here to talk about the experience point.

D&D '74
I began with D&D '74 because it's the oldest, and it's the only official D&D game that I actually have access to. I have summarized the levels and the benefits they provide in this pdf:
https://docs.google.com/file/d/0B9k6jY73glEKZmRUYW9Gd19tUHc/

So, what can we take from this?

Let's get the obvious out of the way: D&D '74 is a hot mess. I already knew that (from playing it as by-the-book as you can), and you probably knew it too. And now it's here in black and white, where anyone can see it at a glance rather than having to decipher its rather arcane text. Now that's settled, let's move on to specific details.

A Note on Variable XP Requirements
The conventional wisdom is that Magic Users require more XP to level up because high level magic users are more powerful than anyone else, and that Clerics require less XP in order to encourage players to step into a thankless support role. Well, for this version at least, the numbers do not bear this out: MUs are harder to level at first, but by level 8 become the easiest by a rather striking margin (so much so that I suspect typographical error, or something more insidious, about which more later); as for Clerics, up through level 8, they are hands-down the best damn class in the game, even without the lower XP requirements. The Fighter doesn't significantly outstrip them in combat until level 10, and their only real disadvantage before that is no arrows (magic arrows are the only weapons that deal more than 1d6 damage -- well, besides the +3 warhammer, but that's only when a dwarf uses it). But things like the best saving throw profile (until high levels) and the frankly stunning potential of Turn Undead make up for that handily. The Cleric here is a straight-up metal-as-hell holy warrior, not a namby-pamby Band-Aid box.

Still, in theory, I can see value in the concept of variable XP requirements, if you wanted to reflect _actual_ disparity in power level, or if you just wanted to create differences in difficulty level such that, if somebody wants to play on hard mode, they pick a class with higher XP requirements.

The Devaluation of XP
This is something that nobody complains about: as you gain levels, individual experience points decrease in value, because the amount you need to level up increases each time. We accept this because it makes sense: you leveled up and got stronger, which means you're now capable of handling bigger challenges, so you should have to go handle bigger challenges to get enough points to level up again.

But here's a thing: in this version, that's already taken care of. If a level 1 character loots 1,000 gp worth of treasure from the first floor of a dungeon, it's worth 1,000 XP. If a level 2 character does the same thing, it's only worth 500. To a level 3 character, 333; 250 to a level 4, and so on. To keep the ratio of gp to XP at 1:1, you have to venture to floors that are equal to or greater than your level -- floors that are expected to be more dangerous (and to some extent are guaranteed to be by the wandering monster charts).

So if you're going to decrease XP gain relative to challenge by level, why also devalue the XP themselves? It would seem more elegant to do just one (especially to do only the latter, since the former requires division, and nobody wants that).

On the Correlation of XP and Levels to Benefits
...There isn't one, really. It's a mess. Spell gains are all over the place. The way that to-hit bonuses and saves increase every so many levels (in groups of 3, 4, and 5 for Fighters, Clerics, and MUs, respectively) creates an appreciable pattern, but also creates dead spots in the progression, particularly for Fighters, who have to spend two out of every three levels gaining nothing but basically a hit die (which boils down to just one more blow they can survive, since HD and damage dice are both 1d6).

I was hoping for some rhyme or reason here. I was hoping I could look and see things like, "ok, level is worth X fraction of a to-hit bonus, Y amount of spell power, etc." Or maybe that, if the benefits per level were going to be uneven, then the XP requirements could be uneven to match -- that would be cool, too. And there's one instance of that: the MU has a short slump where the Cleric surpasses him in spellpower (at least by the SxC reckoning I cooked up), and it starts right when the MU's XP requirements start dropping. But when the MU pulls out of the slump (with aplomb), the reqs don't swing back up to match.

Clearly they just winged it all as they played, and that all just went almost directly into the books. I can't help but have less-than-charitable suspicions of the MU -- I can imagine that the MU requirements started out as 125% of the Fighter's right down the line, but since the scary spells don't kick in until later levels, some asshole was like "waaaah I'm never gonna get any sixth-level spells because magic users are so squishy" (nevermind that they really aren't particularly squishy in this version; their only real disadvantage at low levels is the lack of armor) and so they lowered the requirements to shut that asshole up and maybe also because they were eager to see those spells get into play, then just threw that into the book.

And, y'know, none of that is necessarily "wrong." But it is all rather wonky, and unfortunately doesn't help my mission of developing Unified Experience Point Theory except by showing me lots of things to avoid doing.

Next up, I'ma do OSRIC. I hope it will be more comprehensible.

[Eero Tuovinen]

For what it's worth, I am largely of the same mind about the '74 text and how it came about - it's pretty clearly just a snapshot of the current practice of the campaign, dressed up as "rules". I find that I write something about as coherent by merely vomiting up random notes about whichever D&D campaign we're currently playing. Nothing wrong in descriptive writing, of course, but as I've said before, I wouldn't mind it if they had paid attention to the processes of play and written more about that instead.

To play the devil's advocate for a moment, a couple of suggestions:
- Maybe the benefit of different xp profiles for different classes is not about difficulty at all, maybe it's all in there merely to be different and thus cause fruitful asymmetry between a group of PCs that at least initially might be gaining XP at the same rate with each other. Asymmetry of how much you still need to level up, specifically, as asymmetry of actual power is already well taken care of by the different and differently scaling class features. For this purpose anything at all goes, as long as the numbers particularly at 1st level are different enough so not everybody levels up all at once.
- Maybe the point of devaluing xp by several different means at once is to create an impenetrable fortress of regulations that can be tweaked by the GM flexibly, without excessive player oversight; if the mechanism of devaluation were clear and singular, it would be much easier for the players to get a cry on about any fiddling with it. A multi-layered solution allows you to choose your point of attack in such a manner as to leave whatever the players obsess over untouched.

I could also put down my own notes on Unified XP Theory (I obviously have some pretty developed ideas on it), but I'll desist for now, as I'm sure you've got your own scheme well in hand here. I'll read the next sequence with interest, to be sure.

[Marshall Burns]

It has occurred to me that, if each class is ensured to have one unique thing about them that is crucial (even the Fighter here is the only one who makes suitable dragon bait), even the most random and asymmetrical of progression patterns is can contribute meaningfully to play -- if you really need character X to gain level Y so that you can do Z, it becomes an element of the challenge to figure out how you're going to get him the gajillion XP he needs to make that happen.

While that does contribute to the game board, however, I can't help but think that a principled progression could do so just as well and probably better.

[Callan S.]

I think you've got it with 'character X to gain level Y so that you can do Z', but just take it to a larger picture - D&D is disassociated with it's overall ending. That being, presumably, level 20. No one gets there (Though a side note: I know of someone who's GM'ed one group of 4th ed to 30th level and is on the way with a second). It's worth mulling over an idea of '20 in an evening' - go up to level 20 in just one session/evening (yep, work out much multiplied XP awards), or at most, two sessions. Just do it. Get it out of the system. Witness the end.

I guess a broader question of 'isn't it about more than that?' comes up. "That'll just be gobs and gobs of XP - where's the meaning, man?". But I think a game like The Riddle of Steel with it's spiritual attributes is a game that does indeed go on to being about more than that. D&D, sans such mechanics, has level 20 as the end. But it doesn't seem like the end, because no one gets there - so one is left wondering as to the value of every XP point, because each XP point has become dissasociated with any sort of goal/ending in particular. Since XP is disassociated with anything in particular, it can seem significant in and of itself somehow. A sort of remnant feeling.

[Eero Tuovinen]

How is 20 the end? I mean, I can understand how it is the end in 3rd edition of D&D, because that's where a class description ends and epic levels (an arguably alternative rule) begin, but almost any other version of the game has different logic:

Original '74 version arguably ends at around level 10, at name level, although the game has an opening for the occasional post-name level adventuring, and of course the resolution to end around there apparently didn't last long (as can be seen in e.g. the introduction of 7-9th level spells, which almost single-handedly I think are responsible for the abominable level inflation). Still, I would say that it's the technically correct way to play the game: you play to name level, set up your keep, start up a new cast of characters maybe, have your name level superheroes patronize younger adventurers, play some dollhouse with them, and for the most part don't adventure with them anymore because they've already "arrived". The extra levels aren't there to be obsessively reached, but rather merely to have some rules for when characters happen to settle down at somewhat different pace; some take until 12th to retire, some until 14th, but sooner or later you've accomplished what you set out to accomplish in the campaign world.

Basic D&D has more of an "there's never any ending" ethos to it, if you ask Mentzer (I mean, that's literally what he says - that he doesn't feel like the game should have an ending), and the game does have 36 entirely ordinary levels in his treatment (vague as it is, it does suppose that you'll continue adventuring long after getting your castle and keep, although the nature of the adventures should change vastly) before getting to Immortal play. Earlier editions of Basic were at the other end, they ended after 3 levels and urged you to switch to AD&D.

AD&D I suppose is technically closest to the idea of 20 levels, although from what I've heard of older AD&D play (1st edition, basically), many people seem to have played it roughly like original: play to around name level, retire characters, start new ones. And that's the functional groups that manage to maintain interest in legitimate play to level up characters over a time-span of years, of course for the rest the game features mere short arcs at low and medium levels (or artificially inflated high-level characters), with the high and heady 10th level forever out of reach.

Considering the above simple overview, if I had to name a natural ending level for D&D, it'd be the 10th, not the 20th. Goes to show how optional the very idea is, that we don't even really agree about where the game ends :D

[Marshall Burns]

It's been my experience in playing D&D and D&Dalikes that the end condition has nothing to do with levels or experience. Perhaps I'm just weird, but we've never sat down at our table without some sort of in-game end goal in mind, whether it was a short-term scenario goal or a long term PC ambition.

Which is perhaps why I don't think that XP are the big-R Reward of D&D, but more of a means to an end. They are one of the cogs that contributes to the turning of the big gear that is the Reward Cycle, but aren't the RC themselves. They are good as proof of work -- evidence that someone is capable of playing smart enough to get as many XP as they have -- but ultimately they aren't an accurate way of keeping score. Yes, when somebody levels up, everyone congratulates him (or he boasts about it and rubs it in everyone's places, depending on how frenemy-friendly the Social Contract is), but the guy with the most XP is not necessarily the same as the guy who is currently winning. I know it'll make Callan twitch, but everyone "just knows" who is winning D&D, with or without XP.

All of which is to say, in the process of Unified XP Theory, I don't think issues like winning or end conditions belong among the considerations.

[Marshall Burns]

Actually, having just said that, something occurs to me. If XP come with a fixed endpoint in the manner of, say, Transcendence in the Solar System, that could conceivably be of use to the "game board" of D&D -- it creates a constraint and with it a challenge. Can you accomplish what you want your character to accomplish within foo-thousand XP? The more XP you get, the easier it becomes to accomplish things, but it brings you closer to the point that you're required to remove your character from play. That's a potentially fruitful tension.

This is, as far as my D&Dalike experience goes, a completely untested concept, but it's a very interesting one.

[Ron Edwards]

This is great stuff. I really, really wish you'd do it for the 1977 Holmes D&D, a game which many people in my age group learned "D&D" from (in combination with auxiliary materials), and which has apparently almost completely disappeared from the myth-tale of the origins of the game and unless I'm missing something, is diminished in discussions of the "OD&D" construct.

(People who are into the recent discussions - am I right in thinking that Holmes is sort of the precursor to Moldvay, and that Mentzer drew heavily on current AD&D developments, like skills, in modifying Moldvay? In which case Holmes and Moldvay are very much the odd man out as a unit, and tragically so, as they, again as a unit, are probably the best functional update of the original three booklets.)

Also, I'm trying to remember anything about the level limits. Marshall, what is the precise text concerning those, in the 1974 books? Damned if I can remember whether any such thing was ever dictated for play prior to recent times (e.g. 30th is over for good, in 4E) - and worse, I can't remember whether I forgot about it, if that makes any sense.

[Marshall Burns]

I'd love to do it for the Holmes D&D (that's one of the Basic editions, right?), I just don't have a copy. If anyone does and wouldn't mind sending me all the information, I'd tackle it as well.

The '74 text claims that there are no level limits for humans: "There is no theoretical limit to how high a character may progress, i.e. 20th level Lord, 20th level Wizard, etc. Distinct names have only been included for the base levels, but this does not influence progression." But it doesn't give any guidance as to the XP requirements; as you can see from the pdf, it's anyone's guess what those would be, so the furthest you can go without doing your own design is F9, C8, and MU11. What's odd is that attack rolls and saves are listed well beyond that point; at least, to me it seems a very strange oversight. Dwarves max out at F6, elves at F4 and MU8, and halflings at F4 (only humans are allowed to be clerics).

Another interesting thing: there is no name level for Magic Users in this version. The closest thing they get is the ability to make magic items at lvl 11. They never get their own wizard tower thang going on.

[Ron Edwards]

Argh! See what you made me do?! DnD publications and dates through 1989 (AD&D2). Various other TSR titles and other companies' games are included for reference and perspective.

The colors are an artifact of doing it hurriedly, which I'll clean up a bit when I have time. (just did it - RE)

(Small rant not directed at Marshall) If tomorrow is the day when I live on without ever again hearing the words "red box" as a D&D reference again, it'll be too soon. For one thing, there were three fucking red boxes. For another, ignorant buttholes are forever bragging, "I've been playing D&D since Basic -" [pause] "Red Box." Implying "since the beginning," the pustulent fools.

[Rafu]

Argh! See what you made me do?! DnD publications and dates through 1989 (AD&D2). Various other TSR titles and other companies' games are included for reference and perspective.

Thanks for putting this together! It's a much handier, cleaner reference than whatever I already have available to work with.

For another, ignorant buttholes are forever bragging, "I've been playing D&D since Basic -" [pause] "Red Box." Implying "since the beginning," the pustulent fools.

Please cut some slack to, of all things, non-native-English-speakers bragging like that. For example, one (and only one) "red box" was the first D&D product to be translated to Italian. I've always assumed it was the Mentzer one, as the whole -ECMI line of variously colored boxes followed it. Anybody already playing D&D in Italy before that would be an adult hobbyist using English-language materials from parallel import, while if you started as a kid then that particular "red box" is the earliest you could have started, thus deserving some relative bragging rights I suppose. It was already a hard-to-find collector item by the time I was in high school, IIRC.

[Moreno R.]

@Rafu: yes, the "basic" italian edition is the Mentzer one.  But even at the time the English hardbacks were easily available in game shops in bigger cities

@Ron: why did you write "level 4-14" on the 2th edition?

An interesting turning-point in AD&D is not marked by a new edition, but by a simple reprint: when the very crude cover images of the three 1st edition hardbacks were substituted by the much more elaborate and refined cover images better known today.  It was a total change of image for the game, to a much more "serious" tone.

[Ron Edwards]

Italians are forgiven for past bragging in that manner. No more!!

Moreno, you're right about the new covers, which I think of as the "orange stripe" books. I'll double-check about the levels 4-14; my understanding is that the Cook sequel to the Moldvay 2nd edition was limited to those levels.

I can't get everything into one diagram. If I could, I'd include crucial adventure modules like the Giants series, B1: In Search of the Unknown (which is in Holmes), B2: The Keep on the Borderlands (which is in Moldvay), The Village of Hommlet, the A series, and the N series.

In the new revision of the file to be uploaded soon, I'm adding more horizontal lines and making them heavier, to indicate publications which were both economically discontinued and rendered "unknown" in the hobby subculture.

[Moreno R.]

I'll double-check about the levels 4-14; my understanding is that the Cook sequel to the Moldvay 2nd edition was limited to those levels.

Ah, I think you are conflating the Cook "Expert" D&D (1981) to Moldway's Basic with the 2nd edition ADvanced D&D (1989) (both by David "Zeb" Cook)

[Moreno R.]

Reading again the diagram, probably is simply a copy-and-paste error from the 1981 book box to the 1989 one that brought that part. You can cancel these last posts of mine after the correction, they are off-topic to the discussion...