Thursday, May 27, 2021

What's the Point of Ability Scores? (Final Random Thoughts)

I make no claims that, in the previous parts of this series, I've covered every aspect of ability scores throughout the history of Dungeons & Dragons. However, I hope I have touched upon enough of them to draw attention to a major flash point, namely the necessity for high ability scores. This necessity is a strange thing, simultaneously the cause and result of a shift in how RPG players imagined the hobby in which they were engaging. 

Starting with Greyhawk, ability scores gained greater mechanical weight. Was this change initiated solely by Gygax, perhaps out of dissatisfaction with the way OD&D handled abilities, or was it a reaction to the desire of players that abilities be more significant? Regardless of its ultimate origin, from that point on, all versions of D&D, most especially AD&D, placed ever greater emphasis on abilities and having high scores in them. Minimum ability scores were established for certain specialized classes or even demihuman races, which fueled the perception that characters needed high ability scores, a perception that Gygax himself encouraged in various ways, as we've seen in previous entries in this series.

This raises many questions, a few of which include:

  1. How mechanically robust should ability scores be? 
  2. If they're not robust, what purpose do they serve?
  3. If they are robust, is it reasonable (or "fair") to generate these scores randomly?
  4. Mechanically robust or not, should they be used to limit entry into classes or races?
I don't think there's a "right" answer to any of these questions, but I do think some answers make more sense, depending on one's conception of RPGs. Prior to looking into these matters, I had an uncritical view of abilities and ability scores. Nearly every roleplaying game I'd ever played included them and, in many, if not most, cases did so simply because all previous RPGs had included them. This is a common pattern in the history of game design: because Dungeons & Dragons included X, then all subsequent games must also include X. Of course, the history of RPGs offers many examples where some feature of D&D was modified or even rejected. I can, for example, cite many RPGs – many quite early – that don't include hit points or armor class or alignment, but I have a much harder time thinking of RPGs that don't include ability scores. For the most part, ability scores are almost universal in roleplaying game rules.

I'm not necessarily advocating the abandonment of abilities. Even divorced of mechanics, abilities serve the potentially useful purpose of providing a player with insight into who his character is. Does he have a low Intelligence and a high Wisdom? Perhaps he lacks formal education but possesses considerable common sense or insight. High Intelligence and low Wisdom? He's a book learner with little experience of the wider world. As I said, this is genuinely useful – but only for a certain approach to roleplaying games. If, as Gary Gygax said, it's important that a player be able to "identify with" his character, does that run counter to, for example, random generation? What if I can only identify with a high Charisma character? Or elves or paladins? 

Random generation of ability scores are not universal nowadays, even within the sphere of avowedly "old school" games. AD&D, as we've seen, doesn't outright abandon random generation, but most of its approved methods of score generation stack the deck very much in favor of player preference. These methods assume the player already knows the kind of character he wishes to play rather than letting the dice decide. Couple this with multiple minimum ability scores for races and classes and we see a much narrowed scope for using abilities as a means of differentiating between characters. Look at any two AD&D rangers, a class with above average minimums in four of the six ability scores, and you won't see a great deal of mechanical difference. That's not necessarily a bad thing, but it does suggest that perhaps ability scores aren't the best – or at least only – way to differentiate between characters.

The tying of ability scores to "survival" is another development that contributes to my questioning their role. If, as Gygax suggests, character survival depends, even in part, on multiple scores well above average (perhaps even beyond those simply needed to meet class/race minimums), then the scope for difference between characters narrows even further. This line of thought takes me to the brink of wondering: do we even need ability scores at all? Would it not be simpler – and more honest – to get rid of ability scores completely, folding their benefits into class abilities, so that, for instance, all fighters gain a damage bonus as they increase in level? 

Alternately, one might prefer to eliminate ability score minimums entirely. Under this scheme, if one wishes to play a paladin or range, one simply does, regardless of the character's randomly generated ability scores. This approach doesn't eliminate the powerful pull toward desiring high scores, given the genuine benefits they have in most versions of D&D, but it does lessen it somewhat. Replacing random generation with point-buy or the so-called "standard array" is another approach, since it ensures that, if a player wishes his character to have high scores in certain abilities, he will have to compensate by having lower ones in other abilities. This certainly addresses some of the questions arising out of this series, but at the cost of some of the variability random rolls provide. It's a trade-off and each referee will need to consider the matter carefully.

As I said earlier, there's no right answer to the questions raised by looking at the development of abilities and ability scores in Dungeons & Dragons. That said, what seems clear to me is that no version of the game has ever properly balanced variability with utility. I'm not even certain that that's possible. Nevertheless, I think these are questions worth pondering and we should be willing to consider unorthodox solutions, such as the complete abolition of ability scores. This all assumes one actually cares about these matters. Until recently, I certainly didn't and I can't blame anyone who is content with the way the matter is handled in their preferred edition of the game. However, I am no longer in that frame of mind. Exactly where my thinking will lead me, I don't yet know, but I will certainly share the fruits of my thoughts in future posts.

Part I | Interlude | Part II | Part III | Part IV | Part V


  1. An interesting series, though as you mention, I have a way I like to do things no matter the edition (standard array) and don't give it much thought otherwise as to the method of generation.

    What I *would* like to see is a different way of handling scores...


    A) the removal of the score itself- and only the modifier remain. The business end of the gun, if you will


    B) Just use the score more prominently and remove modifiers.

    As Ken St. Andre has mentioned in his initial reaction to D&D- why have scores and not use them in some way? Why have HP, and not use the Constitution score itself? Saving Rolls were based off scores. Spells and weapons had minimum scores required to use.

    More recently....Very early on in the 5E playtest, WOTC were considering the use of Ability scores primarily. IOW Rating tasks as having to have X score to overcome the situation e.g. this heavy wooden door will yield to STR15 or better. You will need a WIS of 14 to find this trap, etc. No roll needed. I'd like to see that, instead of "roll these dice to get this score- now write the score down, and check this chart/table to see what modifier the score gives you to tasks, and write that down too- BTW that's all you will pretty much use in games, the scores themselves are pointless after this". Added complication/steps with no benefit.

    Some modern OSR games, like The Black Hack, and Relics & Ruins have done away with one or the other- TBH of course uses the score itself for everything. I don't care for roll under systems, but I love the idea of using scores and not modifiers. R&R lets you roll the dice, but it (like many Dungeon World hacks), just gives you a modifier as your score, what you rolled is never written down. So your STR is "+2" not "17".

    1. "the removal of the score itself- and only the modifier remain."

      Don't get this. If the score is removed to leave just the mod, doesn't the mod then just become the score? It's just a variant of your idea B.

      "why have scores and not use them in some way?" Except that they are being used: as a way to imagine and describe a character. Agreed they're not being used as a direct part of the mechanic, but that's not everything . . . .

      "No roll needed." That seems to be pretty much the same thing as in Amber Diceless Role-Playing: contrast the stats; whoever has the higher, wins. Unless the _player_ gets really imaginative (and RPGs really shouldn't be about _the player's_ abilities: not every player is as imaginative as the next), there's no chance whatsoever of success (in a scenario you mention for even fairly mundane things like forcing open doors). I find that incredibly frustrating. That sort of automatic, no-win situation should be reserved as a "guard rail", to keep PCs from going where they shouldn't at the power levels they currently have, not as a feature of the system as a whole.

    2. Exactly, it removes the score which is essentially unused in the game, and the modifier becomes your "score".

      I'm not familiar with Amber, though it seems highly regarded. I'll have to check it out.

  2. I'm maybe in another room with this but I really like ability scores. I think it's great to have a meassurement of the 'natural traits' of the character, separated from his 'professional knowledge' (a.k.a the character class). And I really like when this abilities are fixed. Warhammer RPG is one of my favourites rpg's and the only thing I don't like it's the fact that the abilities of the PC's grow, and a lot. I don't really know to say why I love this, I'll need to think about it.

    1. I kinda hate WHFRP, (1e, which is all I have played) cause your starting character sucks.

      "I swing"
      "I fire my bow"
      "I try to talk him into joining us"
      "Failed fellowship roll, he hates you"

    2. It was my first RPG and I like it just for that. The starting characters suck, yes. There is no more to it, they are scum. In the following editions this was mitigated in some form, but the 1st edition the characters grew very fast.

      It's a matter of tastes I suppose.

  3. In the D&D OSRsphere I think OSE is the game that has the "best" relationship with how ability scores are handled.
    3d6, BX modifiers, very low requirements.
    But as I said in a previous comment to your series on ability scores, I'm increasingly of the idea that rpgs in general can do without ability scores, and especially without the additional distinction between score and modifier.

  4. I wonder how often early players used “roll against ability” to do stuff without being told it was possible. It always seemed natural to me, and every group I was in seemed to do it without discussion. We had scores ranging from 3 to 18, we had d20, why not roll less than or equal to ability score to do something related to that ability? But I started with Holmes, not OD&D, and quickly “graduated” to AD&D.

    1. We didn't- I never saw such a thing until Moldvay came out, IIRC. Maybe it was in a module or such before that. We as DMs (all in our group DM'ed our own game and played in each other's games) just made a ruling based on stats and circumstance, and if it really was uncertain asked for a die roll...whatever die was closest usually. Higher better, and we would narrate results based on the roll: failure, success, or like modern games "failing forward" or "success with a cost"- we just didn't codify it in any way or call it that. As players we would say "I have a STR of 18 though- does that help?". Or the DM would say "what's your Intelligence score?" and modify the results based on that. it was pretty loosey goosey, and we really didn't have much issue with it.

      It's only when I started playing with other groups/in afterschool clubs with some rules lawyers did I start to feel a need for codifying things. And Runequest was far better at that, and it became the Fantasy Game of Choice (TM), once I found it late 1981/early 1982-ish.

    2. Agreed. Never heard of it until late TSR/2e.

    3. I started using the 1d20 roll under as soon as I got my hands on a copy of BX around 1982.
      In the previous year or two my DM hadn't used any kind of ability rolls, if something wasn't covered by the rules he allowed or disallowed actions based on what seemed right.
      In the last ten years or so I moved back to what my DM was doing back in the day.

    4. As I commented on one of the other posts, I'm not sure where we got the idea of rolling against ability scores, but we were definitely doing it by 1978 with Holmes and OD&D+Supplements as references.

    5. Rolling vs Ability was there from the beginning:

  5. Umm, we haven't even figured out if there's a problem (it seems there's just opinions on the relationship between balance [whatever you take _that_ to mean], variability, and utility), and you're already talking about the need for solutions?

    Nevertheless, this got me thinking about the point of ability scores in the context of the larger and so-far unmentioned point of RPGs themselves (aside from the obvious stuff about their being commodities): RPGs are basically thrill-generating machines that work through odds-manipulation and imaginative identification.

    That's what attributes are for: one more variable to throw into the calculation of odds for the "success" of the character. And there's nothing more exciting than watching a long-shot ie a supposed "poor-attribute PC" come up on the outside and "win" despite those long odds. That's one part of making a memorable PC and campaign; Gygax himself even alludes to this in his comments on saving throws in the DMG:

    But at least the character had some hope, and he or she
    fought until the very end. Stories will be told of it at the inn, and songs sung of the battle when warriors gather around the campfire. Almost, almost he managed to reach the bend in the passage where the fell breath of the blue dragon Razisiz could not reach, but at the last moment his toe struck a protrusion, and as he stumbled the dragon slew him!

  6. My honest preference is for the 1974 method. Lots of variability in the scores, and little to no mechanical impact. Let them all just be prime requisites for the classes and call it good. Don't use d20 rolls against the ability scores as a miscellaneous task resolution system; just assume that adventurers are ordinary and competent, and judge from there.

  7. As to ability scores or not, modifiers or not. While I can see advantages of not separating these, most of the games I like all have both ability scores and modifiers and I can't see merging them since there are mechanics that depend on the scores themselves.

  8. I realize pathfinder isn't D&D exactly, but I did not play 3e or 4e. but the need to build your talent tree, and the overwhelming importance of your stats was a real turn off, especially if you sucked at it. you could spend entire sessions worth of time (which I did not have) "Building' your pc, so you could be effective.

    which meant that the DM had to up the difficulty, or powergamers ran away with it, while your suck ass was taking "Mountainclimbing" as a feat.

    ask me how i know? ;)

  9. I can see where Ken St. Andre was coming from. It's interesting just how limited in utility the ability scores were in the original game - to the point of wondering why they were necessary at all. I don't think they've ever been fully integrated into the game mechanics of D&D (admittedly my understanding of 5E is incomplete) compared to later iterations of the basic idea like Tunnels and Trolls, Runequest, or Traveller. In the original game it they seem like a good idea in search of a purpose.

  10. I like that 5e allows Characters to improve their ability scores. That feels like something that always should have been part of D&D leveling.

  11. One way to balance ability scores, and so make character classes more diverse, could be to give an in session advantage to higher prime requisites, ie a damage bonus to Fighters with high strength, but provide an in game bonus to characters with lower prime requisites, ie an experience multiplier to Fighters with low strength, on the basis that those with less ability gain more from success than the more able.

  12. Warlock! does away with ability scores and it works brilliantly.

  13. I like the OD&D method the most: 6- is negative, 15+ is positive. Everyone else is fairly average. Lower numbers across the board, including low hp. I don't mind rolling vs attributes for certain checks if one must, but I hate perception/search rolls and most of the time, things are better adjudicated narratively.

  14. I may be wrong, but I don't think you can balance utility and randomization when it comes to stats. Not in a satisfying manner at least. I also never understood the notion that point buy is ''for sissies'' that some OSR groups hold. If the points are just enough to make a character either average, or good at something and bad at eveything else, what is the player's unfair advantage?