Saturday, June 6, 2009

Ability Checks

Lots of people have asked me how I handle ability checks in my Dwimmermount campaign, so I figured I'd make a short post about it. In brief, I don't use ability checks in my game. That is, I have no formal system whereby I allow a player to roll dice based on his ability scores to determine if he succeeds at some action not covered explicitly by the rules.

The reasons for this are twofold. Firstly, from a historical perspective, I never used ability checks back in the day. I don't recall AD&D, Holmes, or Moldvay Basic -- the three rules sets that had the most influence over me back in the day -- allowing for such a thing, though it's possible there's some reference to them I somehow overlooked all these years. Regardless, I simply never made use of the concept. Secondly, my attitude nowadays is that any action not covered by the rules is one that falls within the bounds of referee adjudication. If one of my players wants his character to try something for which there are no rules, I make a decision, often in consultation with the player, about its likelihood for success. If there's a significant random factor in whether said action succeeds or fails, I'll roll some dice to see what happens. In my Dwimmermount game, I've sometimes used the saving throw mechanic as a way to gauge an individual character's "luck," but most of my rolls are purely ad hoc.

So far, this "system" has worked very well for me and I've seen no need to introduce a formal ability check or universal mechanic to the game. Before I began, I thought I might need one and contemplated several different ways I might introduce one, but, when push came to shove in play, I ignored them all and just used common sense and (maybe) a dice roll. Nothing more has really been necessary. As I noted in my earlier post of today, having a high degree of trust with my players is a huge asset. If my players didn't trust me, I might try to "objectify" the process a bit more in order to give the illusion that such matters were out of my hands as the referee, but that's not my preferred approach and it's certainly not the approach that I feel is most consonant with the philosophy behind OD&D.

I'm curious: does anyone know when ability checks first get introduced into any version of Dungeons & Dragons? My guess is that it happens in Mentzer, but I may well be mistaken.


  1. 'There's always a chance" on page B60 of Moldvay.

  2. Thanks! I don't think I'd ever noticed that before you pointed it out to me.

  3. I know I recently saw ability checks in an old 1e module I was reading, but I don't remember which one. I will go back and check, but the wording of it made it sound like this was not something that was presented in the rulebooks (it didn't just say "have the player do an ability check" as if referring to something in the player's handbook or DM's guide, it described how to roll against the required ability score, which I think was Dexterity in this case, to see if the character succeeded at the task). I don't know if this would predate the Moldvay reference mentioned above, I will try to find the text that I am remembering and repost.

  4. Yep, B60 of Moldvay, circa 1980. I know there are NO ability checks in White Plume Mountain (first printed in 1979), even though I've always house-ruled some for jumping across the suspended disks.

  5. I also believe that they appear in some old Judges Guild adventures that would likely predate Moldvay.

  6. But even if you're not using a universal mechanic or coherent set of rules, you must have some set of loose guidelines that you use to help you in making your decision.

    When you call for a die roll, for instance, are you more likely to set it as a percentile roll or as d6 roll? If a character's ability score is a factor in the particular chance, how do you tend to take it into account?

    It seems to me that even if you are making ad hoc rulings (which I completely agree with as the best method by the way) you still have to have some basic assumptions or principles in place to guide your decision making process.

  7. Ironbeard,

    Most of my rolls, when I use them at all, are 1D20, with high being a good result and low being a bad one. I sometimes I also use 1D6, with the reverse being the case. Mostly the dice roll is there just to give my imagination a jolt rather than as a final determinant of what happens.

  8. I like the use of ability checks (or in my case skills) becase I like to do more than just swinging a sword at monsters or casting a spell at them. If my character can do a lot of things that are not combat related then I feel more involved with the dungeon. I like to have the ability to climb, hike, open locks, detect for traps, create oil missiles, stealth, hide, analyse magic and potions etc. Of course this is why I play GURPS but I still love the old school dungeons. I just play them with skill based rules.

  9. First I saw the mechanic was in I3 “Pharaoh”, but obviously there were earlier examples.

  10. I like the use of ability checks (or in my case skills) becase I like to do more than just swinging a sword at monsters or casting a spell at them.

    The lack of an ability check mechanic (or skill system) doesn't mean characters can't do things besides fight or cast spells. It simply means that, when they do those things in my games, I don't determine their success or failure at them by means of a formal mechanical system.

  11. like the use of ability checks (or in my case skills) becase I like to do more than just swinging a sword at monsters or casting a spell at them. If my character can do a lot of things that are not combat related then I feel more involved with the dungeon. I like to have the ability to climb, hike, open locks, detect for traps, create oil missiles, stealth, hide, analyse magic and potions etc. Of course this is why I play GURPS but I still love the old school dungeons. I just play them with skill based rules.

    Why do you need rules to do these thing? Can my fighter climb, sneak, hide, find traps, etc? Yes. You don't need ability checks or skills for that. Just assign a probability or find away for the other game mechanics to handle the result.

    For example, the fighter is trying to sneak up on the Orcs. He has average dexterity and has taken his platemail off and covered himself in soot. The orcs are drinking and being rowdy around a bonfire. To the DM it sounds like he has a pretty good chance to sneak up on them. The DM decides that the Orcs will be surprised on a roll of 1 to 5 on a d6.

    Or the fighter wants to climb a cliff to get to a ledge above the path so that he may push some boulders on some bandits that are pursuing him. He is wearing his platemail but does take off his helmet and gloves. The DM thinks that the climb will still be pretty tough and that the fighter has a 10% chance of success. The player reminds the DM that his character grew up in the Western Hills and spent a lot of time climbing as a kid. The DM then decides that the fighter has a 15% chance of success.

  12. The way I see it is that if a character has trained in a certain skill then he will have a higher percentage of accomplishing it. Say a PC has lerned to analyse a potions at 15, the if he rolls 15 or under he can detect what the potion is. Another character who has not learned as much about ananlysing potions will have say a skill of say 10 so in order for him the analyse the potion he will have to roll a 10 or under on 3d6 to be able to determine what it is. Skills are better, at least for me, to be able to assign a probability to accomplish a task.

  13. "Say a PC has lerned to analyse a potions at 15, the if he rolls 15 or under he can detect what the potion is. Another character who has not learned as much about ananlysing potions will have say a skill of say 10 so..."

    The prospect of recording all that stuff for every possible task makes me puke a little bit in my mouth nowadays.

  14. When I was playing in junior high school lunch breaks we had some pretty acrimonious (tiresome, win at all costs) players and I found that using rolls on attribute scores was a good way to adjudicate the success of actions whose success was uncertain. No one could argue with attribute checks because they are an external, non-GM defined, (ie. less open to accusations of bias, everyone saw you roll 13 for STR, etc) criterion for success once any modifiers are applied if required: eg. for slipperiness of an object, say, or any other factor that will obviously affect the outcome.

    Also you have to keep in mind that the writers of the rule-books at TSR were definitely not lawyers or particularly great editors and some things had to be inferred. For example, in my 2e Gamma World rule-book which is heavily based on D&D it suggests a Strength check where feats of strength are required (pp 6 and 63). I just dug out my "What is Dungeons & Dragons" Penguin paperback (my introduction to the game) and it suggests something similar on p 89.

    Mutatis mutandis we applied attribute checks for other kinds of feats and it worked with a minimum of book-keeping.

    We just went for a (possibly) modified d20 roll instead of multiplying by 5 (why do more maths?) but looking at the Penguin "What Is" book I note on p115 they also recommend combining two relevant attributes and doubling the result for a score to beat on a d100 roll which sounds like a good way too.

  15. While it was not an official rule, of course, Wesley D. Ives published a system for resolving ability checks in the very first issue of Dragon Magazine way back in June of 1976.

  16. Hi James,
    I've been wondering how to handle ability checks lately myself. I've been reading the OLD SCHOOL PRIMER by Matthew Finch that you put links to lately, and I like the examples he gives as to reasons way not to go over board with ability checks; a real focus on common sense reasoning and planning when it comes to traps. still there are times when I think some sort of check make sense. I guess the ad hoc approach works if your with a group you've been with for awhile, but IF I'm able to get a group together it will probably be with people I'm just meeting, so some sort of standardize system would help. I like alot of the ideas that have been coming up in these comments. One of the ideas I was throwing around was to set my own difficulty on a d20, and allow the plus 1-3 bonus depending on the ability score (maybe somehow incorporating a skil idea, not sure yet). Does this sound too much like D&D 3e d20 system, because I never played 3ed, or read their books, but i did play 4e once and they had something like this in their system.

  17. The first place I saw something like this, which is not to say that's where it first appeared, was the skill system introduced in GAZ 1 (Karameikos).

  18. I don't see why skills are too difficult to run. Even swinging a sword or casting a spell is a skill. All you have to do is have the players keep records of their skills on their player sheet and then have them look it up when needed or you can keep a copy of their player sheet. In some ways D&D is more complex because you have to roll against a chart based on level that is somewhere in a book. The same is true of saving throws. In GURPS you just roll against health score if physical or roll against intelligence if mental.

  19. We always used the d20 under attribute = success. In a few games I've run I let players come up with a background career or skill and allow them to reroll checks that are related to the background once a session.

    In my last Carcosa game I used a stealth rule proposed by Snorri on the OD&D boards; 2d6 vs AC + Level + Dex Mod. It worked well for low levels, but had the game continued I imagine that it would have become problematic.

  20. For my own part, I favor ability checks, because I'd like those six numbers that are the central focus of the character sheet to have some mechanical consequence. And I'd like that reference to the ability to consider the original number, not some derived modifer, because then we might as well just use -3 to 3 (or similar range) as the ability. Which is a perfectly fine game, but somehow doesn't feel like D&D to me. (FUDGE maybe ;)

    It's cool to hear people mention some of the early ability check mechanics that showed up in various places. Gives us an insight into how groups back then were approaching the issue.

    Certainly the generalized Saving Roll mechanic in T&T was always something that drew me to that system, and they are ability checks in all but name. Interestingly, the 5th edition ruled were almost as coy about the idea as the Moldvay example cited above, but their use was more generally accepted in the community of players.

    I'll also admit that ability checks in D&D was not a feature of our early play (Holmes leading into AD&D). But then we were the autodidacts James mentioned in the last post (myself being patient zero) and such things did not occur to us. The appearance of the concepts in other rulesets led me to embrace those rulesets, rather than grafting the idea back to D&D.

    Having recently rengaged with the mechanical simplicity of early D&D via S&W, but with the experience of other ways of doing things, I find that the introduction of a simple ability check provide the type of play I am seeking while still allowing me to appreciate the old rules for what they are.

  21. If you're thinking about changing stat bonuses because the stat score levels don't feel right, I suggest that actually you HAVE been missing something by not having ability checks.

    And they don't have to be a spelled-out mechanism or part of a core mechanic. I do them by asking a player, e.g., "Make a Dex check and tell me how much you make or miss it by," as I'm judging the situation. How the number is used is still up to me and can still get fed into the black box. The number means whatever I want it to mean.

    If you're rolling "D20 and wing it" a fair bit, a stat check does not have to be any more than "(STAT-D20) and wing it."

    Finally, for a lot of players (especially newbies), stats are quite important. Stat scores are usually the first thing players know about their characters, and is often the whole key to imagining them. Calling for stat rolls can make them feel like their ability scores are doing something.

    All just M.O. of course. Good gaming...

  22. For a middle ground between skills and 'everything is a d6' I like Castle's and Crusades SIEGE engine.

    All skills/saves/and various actions are rolled against one of the 6 stats. Each character has a few prime stats that provide a bonus to these rolls. [there's slightly more to it than this see quick start]

    It provides for a nice variety of characters while still having archetypes. A fighter who DX as prime vs one who picks CN as prime.

  23. The earliest place I know of where an attribute check is mentioned is under the dig spell description in the first edition AD&D PHB (1977, p. 76):

    Any creature at the edge (1’) of such a pit uses its dexterity score as a saving throw to avoid falling into the hole, with a score equal to or less than the dexterity meaning that a fall was avoided.

    Whoever wrote that spell description was certainly using them, in whatever form [e.g. 1d20, 2d6, 3d6, 4d6, etcetera...]

    The AD&D DMG (1979, p. 110) has this to say on the subject of task resolution in general:

    There will be times in which the rules do not cover a specific action that a player will attempt. In such situations, instead of being forced to make a decision, take the option to allow the dice to control the situation. This can be done by assigning reasonable probability to an event and then letting the player dice to see if he or he can make that percentage. You can weigh the dice in any way so as to give the advantage to either the player or the non-player character, whichever seems more correct and logical to you while being fair to both sides.

    ...which obviously does not preclude an attribute check of some sort, but is more reassuringly generalised and restates what Holmes says (1979, p. 41):

    You are sure to encounter situations not covered by these rules. Improvise. Agree on a probability that an event will occur and convert it into a die roll - roll the number and see what happens!

    This also appears in the introductory boxed set to AD&D 2e, First Quest:

    * DMs decide what happens. If necessary they just make it up.
    * DMs decide what percent chance an action not covered in the rules has of working. If a Dungeon Master's d100 roll is less than or equal to that number, the action worked.
    * DMs are always allowed to decide that a character's action automatically does or does not work. DMs are always allowed to change a die roll if they think there is a good reason.
    * Dungeon Masters should always be fair

  24. Interesting discussion. I'm in the camp that likes ability checks; they feel more "organic" to me, more connected to the character, than a percentage made up on the spot. It was their explicit use in Chaosium's BRP that drew me to that system and away from xD&D (among other reasons). Simply multiply the relevant stat by a modifier that represents difficulty (5= very easy, 1= very tough) is my quick and easy resolution system. It has the benefit of consistency, too.

    In the end, it's really just a "what you prefer" moment, no right or wrong. (Though I have to admit I'm surprised I missed the references to ability checks in early D&D products back then.)

  25. For Labyrinth Lord, I use D6+/-ability modifier, with a high roll being a success. It seems to work quite well without being too intrusive a subsystem.

  26. The next step would be eliminating rolls for combat and adjudicating them.

    What's inherently unfun in having ( btw very simple) rules for adjudicating non-combat situation, vs having them for combat situations, or, say, magic? Or save vs. poison?

    I'm not trying to be a troll or anything, James, I'm highly sympathetic with many old school ideas and I enjoyed my AD&D1ed days as much as the next guy (with ability checks!). But this is one of those points where you lose me - not that's important, but anyway...-versatile resolution mechanics (and skills!) do not really restrict play and imagination.

    Maybe it's just that I am primarily a RQ/BRP guy rather than a pure D&D fan. BTW Skills were there in RQ1 in 1978 and ability checks were certainly in Stormbringer in 1981 (I'm not sure about RQ1). I hope there is room for different shades and flavors of old school in the community.

  27. I have nothing against ability checks and I don't begrudge anyone who uses them. I simply don't use them myself and have never seen a pressing need for them. The point of the post was simply to answer a question I get a lot from people, not to suggest there's anything wrong with ability checks.

  28. The reason I do ability checks is because my brain is tired enough by the end of the night from adjudicating things in the game. Plus I think the players enjoy being able to roll, they seem to get rather annoyed when I make rolls for them. With ability checks as an easy, quick method, I try to instead focus more on the game world than on the system.

  29. I think I have to stand with the defenders of SOME kind of unified mechanic. I just rarely trust my players enough to go ad hoc. At what point then does D&D end and Cowboys and Indians begin?

  30. Starting with GAZ1, I found that skills simply served as shortcuts to adjudicate things I would have had to adjudicate myself. Did I need them? Surely not. I started with Mentzer D&D, and I felt the game was complete as it was. When I needed them, I used ability checks as suggested in the Mentzer book itself. When I bought GAZ 1 I did not adopt the General Skills system, which I found redundant, since it did not add rules, but only provided a "layer" to things which seemed to me abstract enough to be handled without rules/rolls at all!
    I have seen with the years of introducing the game to newbies, is that they accept/use Skills only as "parts" of the rules of the game, not as things which are necessary to define what a character is/can do.
    Nowadays, I use a system of Talents which works like the Weapon Mastery rules in the Masters Set, in that it simply modifies rules which already exist in the game, so it has a practical impact. For example, the Alertness talent allows a character to be surprised only on a roll of 1 on 1d6. Or the Horsemanship talent allows a rider to attack at the same time as his war horse and to push the horse harder when running.

  31. I use the S&W saving throw system as a kind of 'general task resolution' system (as I explain in my article in Knockspell #2). I like using it, because it provides a kind of self-imposed consistency on my rulings as a DM.

  32. At what point then does D&D end and Cowboys and Indians begin?

    That's for each referee to decide. For myself, I prefer to keep things as light as I can, since I just don't find much need for rules for many actions. Since my players are comfortable with this approach too, why complicate the game further?

  33. Back in the day, several DMs I played 1E AD&D with used a primitive skill check system which involved rolling a d20 and comparing the roll to a particular player stat. A success was rolling less than or equal to the specific stat. (Critical successes and failures were not used at all). So even with an 18 in a particular player stat, there was still a 10% chance of failure.

    For harder stuff, the DM would adjust the "success" DC to something lower. For example a 1st level magic user sneaking up on a dragon without being noticed, the DM would require a roll of 1. A less silly example would be a fighter in plate sneaking up on a bunch of orcs without being noticed, the DM would require a roll of 7 or less. Basically the DM adjusted the DC to reflect the probability of something being a success. It was largely an ad hoc system based on DM fiat.

    I don't know offhand where these particular DMs learned this primitive skill check system from. Among our groups, we only had the three core 1E AD&D books and none of us had any copies of Dragon magazine. In my own games, I only used the 1E Player's Handbook with the combat tables and monsters from Moldvay.

  34. From the Forgotten Temple of Tharizdun:

    Party members 10' to 30' inside the upper level, and not
    east or west of the center 20' entrance line, might be
    caught in the net trap. Those characters positioned near
    the edge of the area where the net falls must roll a score
    of their Dexterity or less on 4d6. Those in the center area
    are trapped, no saves.

    About as close as it comes to a skill check, I think.

    I used this as a template for an upcoming 1E module I wrote for Goodman Games. It was used sparingly, of course. :)

  35. Mike,

    That's a great find there! I don't remember having seen that before, even though WG4 is one of my favorite modules. Thanks for pointing it out.

  36. Sorry to post in an old old thread, but note that in this interview with Sham, Greg Svenson says that Arneson was using stat checks as well.

    Which I remembered way back when but forgot to post about it until the interview was linked afresh just now on S'sG&B.