Sunday, March 28, 2010

d6 Based Thief.. er, Specialist Skills

(I don't see the "thief" class as a "thief," and in the case of the game I'm putting together, I very much think of the Specialist as the only normal guy of seven classes. Yet, obviously, 35 years of the "thief" kind of makes it roll off the tongue... but I've always thought of him as an Indiana Jones type. Professional explorer rather than criminal.)

Right now, in the LotFP rules, I'm using percentile skills for the thief, as has been done for a long time.

I've renamed a couple skills, combined two of them, and am using the 2nd Edition "allocate points" inspiration because otherwise, the "thief" means "guy who can climb walls and do shit else for a bunch of levels."

But some of this stuff should be able to be done by everyone. That's one of the complaints against the thief class, right? Makes it sound like nobody can sneak, nobody can climb, because it's this guy who has the skills and the rules for it.

So how to address that?

Well, one of the things I'm already doing is giving the Specialist is a 2 in 6 chance of finding secret doors (like an elf), instead of the usual 1 in 6.

Why can't all the skills be like that?

How's this sound?

All Specialist skills (Climb, Find Traps, Read Languages, Sleight of Hand, Stealth, and Tinkering) start at 1 in 6. In fact, all characters can do these 1 in 6. The Specialist will then get however many points to allocate at 1st level, and more every level they gain, just like thief skill percentile points now. You know, "I put one point in Read Languages so I can do that 2 in 6 now."

The secret door thing can then be one of the skills. Everyone starts at 1 in 6, but the Specialist can choose to be better at it. Eventually, he can be a 6 in 6 secret door finder if he so pleases, but of course at the expense of other things.

This would solve a few things: The idea that the very existence of "thief skills" takes away things that all characters should be able to do, and that thief skills are the only rule that uses percentile dice.

Fighters would be the only characters that get better at fighting. Clerics get their spells. Magic-Users get theirs. And so the Specialist will be the guy that improves at having skills.

(my only initial hesitation is Read Languages, maybe that should start at 0 in 6 because any old guy rolling to see if they can read the ancient text written in a dead language seems wrong).

I don't think this would hurt compatibility, because in oldschool play and modules, how often are modifiers really given to skill use? Not much as I recall, and I know I never really call for modifiers.

I know this is not an original idea. Others have done d6 thief skills. But am I on to something here?


  1. I developed a similar system.

    One issue: The skills quickly ramp up to 5 in 6 & 6 in 6 without some other rule to temper them.

    My solution: All skills top off at 5 in 6, but once a certain level is reached, a skill with 5 in 6 chance of success gets a side skill of something more advanced (like Stealth allowing a 1 in 6 chance of Vanishing in plain sight).

    Of course, all this can be avoided if you develop some skill modifiers with bonuses and penalties (like a -1 to Read Languages if the language is ancient, etc.)

  2. I like the solution.
    It is elegant, straightforward.
    It does minimize value of the thief somewhat with Expected value statistics. Suppose, for instance, to open a lock everyone has a 1 in 6 change. Across a party of five, none of which are thieves, someone is bound to roll a one... sort of like buying the infantry as the Soviets in Axis & Allies... Granted, their could be consequences, like a trap or the like.

    I think one of the key "problems" many have always had with the old school game is a lack of fungiability across the classes - e.g., what if a fighter had to climb a wall, or, what if a wizard wanted to follow footprints?
    Moreovr, the lack of understanding of how to resolve things in a manner not seen as capricious when you "do things" besides attacks with weapons and casting of spells.

    This is handled in both the 1st ed. Dungeoneers & Wilderness survival guides - but also fairly effectively with Gurps-esque skill systems of 3.x and beyond.

    So, I like the 1 in 6 inrementalism solution.
    I just become concerned that with many incremental moves, in aggregate it results in hybrization of the game... which then forces question of comparison to 3.X system, and then begs question as to why not just go all-in selling out to 3.x, and focus upon setting and house rule tweaks that are appropriate.

    I guess if you preserve the very essence of an OS system, some incremntalism ain't that bad. Is there a line that is ever crossed?

  3. This would solve a few things: The idea that the very existence of "thief skills" takes away things that all characters should be able to do, and that thief skills are the only rule that uses percentile dice.

    Based on me doing a similar thing in the Majestic Wilderlands.

    1) You will still get criticism that you have skills in D&D.

    2) You will get way less criticism and more people looking at using it.

    As for the d6 mechanic what you are doing is having a skill system like this percentage wise.

    1 - 16.7%
    2 - 33.3%
    3 - 50.0%
    4 - 66.7%
    5 - 83.3%
    6 -100.0%

    That means you going to have to space out advancement a lot. There will be a lot of dead levels where little is gained. This is why I went with a d20 roll high system.

  4. It's one idea, but if you go this route may I make a suggestion:

    Change the die type to a d10 or larger. Yes, that includes on secret doors and surprise and all that. As has been pointed out d6 ramps quickly. At even just one point per level a thief can max out four skills by level 20.

    I'd recommend using a d12. It's eleven levels to max out (or 10 if you rule a 12 always fails). It is ridiculously easy to convert existing items (like finding secret doors) by merely doubling them. It would take 60 points to master everything so you can be fairly generous. For example, giving level number up to name level (9th) and then point there after gives 45 points at name level and 56 at level 20. Have the pre-name progression (level/2 rounded up) gives 25 at name level...enough to specialize two to mastery or be about 50% in everything.

    You could also look at converting everything to stat rolls, which would probably be my option.

    Stuff you're bringing over can be translated to your new die type

  5. >>As has been pointed out d6 ramps quickly. At even just one point per level a thief can max out four skills by level 20.

    Skills get maxed out across the board far earlier than that in traditional systems.

    I think a 6 in 6 skill chance shouldn't be an automatic success though. Maybe you roll 2d6 and snake eyes still fails.

  6. re: skills for everybody: sounds nice and is a good answer to the question "why my elf can't climb a tree?". my answer to that is "everyone can climb easy walls, but if you want to do it fast or climb hard walls, you need to be someone that practices it - a rogue".

    re: skillscap: granularity and cap are a problem with a 1d6 system. I'd either raise the dice type or (as we discussed, more complicated) go for rerolls/exploding dice.

    re: read languages: I make it start at 0 too. And use magic items (for scrolls) at -4.

  7. I use similar concepts, but different mechanics.

    However in this case I see a problem with the scaling as the specialist levels, that has been brought up before.

    My suggestion would be that the thief rolls more and more d6 and chooses the best result. it never gets to 6/6 but rolling 7 or 8 d6 is pretty close.

  8. Rolling more dice would give a better success curve, but still not perfect. The problem with that system is, that its a new dice mechanic previously unknown to the game, and I get the impression Jim is going for streamlining mechanics, not adding.

    The question is how fast do you want the specialist to max out his specialist skills. If there are eight skills, which all start at 1/6, thats a total of 40 points. Assuming a level cap of 20, that's two "skill points"/level.

    My idea on how to handle skills of 6/6 and not autosuccessing all the time: Whenever a 6 is rolled, it fails, except when the specialist in question has a skill of 6/6, in which case an additional d6 is rolled. If that one also comes up as a six, its a failure. This is basically the same as Jim's snake-eyes -idea, except boxcars is a failure, not snake-eyes. Making snake-eyes the failure is inconsistent with usually having to roll low to succeed.

    Another solution would be to max out the skill at five, meaning no one could have more than 5/6 in a skill.

  9. I've come to see thief skills as of two types: things that everybody ought to be able to do but thieves can do better, and things that only professionals can do.

    Anybody can probe the floor with a pole/examine the chest without touching it to search for traps. Anybody can climb a wall.

    OTOH, disarming a trap without getting killed takes training and expertise. Ditto picking pockets.

    So there are some things thieves can do somewhat better, and then there are some things that you've got to be crazy for anyone other than a thief to try it.

    So maybe the key is to use two different base values - 1 in 6 for commonplace stuff that anybody can try to accomplish (find traps), and 1 in 12 for stuff that really should only be attempted by highly trained experts (picking pockets, disarming booby traps). Anyone can still try to do anything, but some things should really only be tried by a thief because nobody else is going to stand much of a chance.

  10. If you really want to slow it, why not use a d12, but default to 2 in 12.

  11. That's exactly how I've done it for a while now, except that I've switched everything over to a d12 (so Elves have a 4 in 12 chance to notice secret doors). That way I can hand out +1s a little more freely, such as for high stats and Thief levels (which I refer to as the Delver class, for the reasons you describe so well in this post).

  12. I don't think changing the die type would be a very good solution, due to the reason I already mentioned: It would introduce a new mechanic to a game that does just fine with less mechanics. There's nothing wrong with a d6-resolution as such. All you need to do is decide how fast you want the specialist to max out on one of his skills and how fast you want him to max out all of his skills. If you give a specialist a base chance of 1/6 on all of his skills at level one, and two ranks bonus at every level including first, he will have a skill at 6/6 at lvl 3 at earliest and he will have 6/6 in all of his skills when he hits lvl 20.

  13. 1. I think d6 skills is definitely the way to go, whether you decide to limit them to thief skills, or broaden them to include some non-thief specific skills (swimming, etc.) is up to you.

    2. I like you're idea that everyone starts with a 1 in 6 chance with maybe some racial modifiers like halflings getting a +1 bump for hiding or dwarves a +1 bump for traps.

    3. I'd keep the dice as a d6, but make the number of skill points per level low, like 2 or maybe 3 if you wind up with a larger list of skills.

    4. No skill no matter what should have a success rate greater than 5 in 6. Nothing should be garaunteed if it requires a skill check! If it's something so simple that DM fiat can declare it accomplished then there's no need to reference a skill.

    5. I think you intend skill improvement to strictly be a thief/specialist function, but if you go with a broad skill list (or others reading thinking of doing so) with skills favored by other classes, if you allow "cross-class" have it to where it takes all of a class' skill points to go up one point in a non-favored skill. If a fighter wants to improve his lockpicking to 2 in 6 then that's all he can do for skill improvement, no improvement to riding or swimming, etc. For this kind of system with a broad list I'd probably give non-specialist classes 2 skill points per level and thieves/specialists 3 skill points per level that they could spend on class favored skills.

    A thief could probably max his skills at 5 in 6 by 10th level if he gets 3 skill points per level, but if the list of skills was broadened to include such things as riding, swimming, tracking and such then there would be other areas that the thief might want to dump his points into.

    What you would also see is probably a lot of thieves becoming expert trap finders/disablers as soon as possible, who slowly build their other skills up to parity. If you want traps to be a challenge for a while keep this in mind, and strongly give thought to maxing out skills at 5 in 6. No point in even having traps if a thief has a 6 in 6 chance of finding them. He will find them, and it becomes boring to go through the motions.

  14. That's how I run the thief as well. It played out pretty well in the Grinding Gear with nor issues popping up.

  15. I think your base outlook/concept on this is right on the money.

    I used both the d6 system and "every class can do most of these" idea in my own White Box Thievery add-on. I added 1 point to two skills every level and they reached mostly 3-in-6 at 10th level, though since my concept was that thief functions were an add-on for a regular class, they started out a bit lower than might be expected.

    For our Labyrinth Lord game, we went up to d12 to allow finer advancement rate while remaining compatible with the standard d6 checks. Which seems to be a fairly common solution to the coarse granularity of the d6 method.

  16. I took a different tack in that I was taught how to make sense of the existing rules rather than overhaul them.

    I’ve come up with various ways to expand n in 6 past 5 in 6. One simply way that actually works OK is switch die and count only the highest roll as a failure. So the progression goes: 5 in 6 → 7 in 8 → 9 in 10 → 11 in 12. Of course, then you hit another limit unless you have a d14, d16, and d18. Lots of places to go from there, but—yeah—you’ve left the simplicity of n in 6 far behind at that point.

    The one thing that keeps me from adopting the 2e point allocation thing is that I don’t think all the thief skills are equal. I’d want to assign different costs to different skills. That something I’d rather not have in my D&D.

  17. I've done a few variants on a d6 ability for "Dungeoneers" and "Scouts," and it has worked well. My method is a bit different from yours, using a flat skill level for all abilities, with slightly different options for whether the character is meant for more of an indoor or outdoor type.

    When I did experiment with a method more like yours - buying different levels in different skills - i ruled that the same skill can't be improved twice in a row, so advancement can't be too quick, but more importantly: a skill of 5 in 6 that is improved becomes 7 in 8; after that, 9 in 10, then 19 in 20.

    The main reason i switched from that method was to keep things quicker and easier: "What level are you now, third or fourth? Third? Okay, you're still 2 in 6 to do this, next level you'll be 3 in 6. Good luck!" It suits me.

  18. Most of a core mechanic's merits and flaws are found in how it interacts with the rest of the game. Seems to me that there's little inherent good or bad in a given die mechanic.

  19. I think Lord Kilgore is right on.
    Use a d12 mechanic for fine tuning the Expert while still being in tune with the original d6 mechanic.
    This also would allow th referee to add/deduct bonusses due to the player being ingenious using the skill or there being negative circumstances (eg. darkness for Lock Picking).