## Pages

### On Skill Confusion: The Accuracy of the Roll

Smarter people than I have written on the failure of growth in conflict resolution ability in games such as Dungeons and Dragons. Here is another article, talking about the linear nature of the d20 check. Here is an article with lots of interesting statistics!

This post covers two topics:
There are a large number of issues with the d20 as a resolution mechanic for conflict resolution.

Topic one, the d20 as a resolution mechanic. Pathfinder example:
"I'm going to arm wrestle the ogre!"
"I have a strength of 8, which is a -1 modifier!"
"Ok, roll a d20, while I roll for the ogre who has a strength of 21 for a modifier of +5."
"I rolled an eight!"
The DM looks down at his 1 and wonders how the world went wrong.

Ok, so here's the simple math - the noodle armed wizard will beat the ogre in a strength contest of arm wrestling 22.8% of the time and tie 3.5% of the time.
[Edit: Someone pointed out these numbers are wrong - I apparently forgot to calculate the 0 value for the difference between the skills. The actual percentages are 19.5% the person seven ranks lower then his contested opponent will still lose. Apparently ties (even though they are mathematical ties) always go towards the person with the higher total in 3.5/pathfinder? Either way, in any contested roll, If I have 7 ranks more than you, I still lose a full 1/5th of the time. Pick whatever contested roll you wish. Notice the underlined point below.]

One quarter of the time the ogre will not win in an arm-wrestling contest. [Edit: one-fifth of the time]

The d20 has no mode value. Given enough rolls the average result (mean) will be 10.5, but on any given roll the chance of receiving any result will be equal to the chance of receiving any other result.

This is such a large and known problem, that the game explicitly acknowledges it and includes a rule for bypassing the roll completely (take 10).

This can be an advantage for something like combat, where highly variable results can be extremely exciting and entertaining. It is chaotic, and huge swings in luck are expected. The high degree of randomness in combat is a large benefit for creating a very real threat and fear of engaging an enemy, leading to interesting role-play to avoid combat and exciting comebacks against overwhelming odds. But for resolution of activities where increased accuracy is expected with increased ability it fails totally.

There are a large number of issues with characters having the ability to increase their skills used for conflict resolution.

Topic two, the scaling mechanic.
Here's another example with level 15 players.
"The bridge is liquid, shifting before you. Heavy winds buffet the path. It's a dc35 skill check to cross!"
"Sweet! I've got a (+9 stat, +15 ranks, +3 class, +10 magic item) 37 in my athletics!"
"Damn! I've got a (+3 stat, +0 ranks) 3 ranks - I can't make it even with a 20!"

Again, simple math. Escalating DC's and increased specialization create situations where characters can be whole statistical probabilities ranges away from the next tier of a successful roll, while other characters can have auto successes. The degree of specialization afforded high level characters easily breaks the game. (and the above 37 rank PC is not one that is even particularly specialized in athletics, just what you would expect a reasonably designed 15th level rogue to have)

This is an artifact of the range of results on the d20 being low compared with the range of possible skill modifiers. This issue intersects with an issue with 'skill tax' skills such as concentration (eliminated in pathfinder) and perception. Someone always keeps these skills maxed, questioning the advantage of 'increases at level'.

These are known problems, and there are a large number of suggestions and house rules created to address them.

So, what advantage is gained by having skills that increase with level increase, given the problems they cause above?

and what advantage is gained by having a mechanic for something as chaotic and random as combat be used for something that expects a more consistent result?
[edit: removed the word 'universal' in front of mechanic, since if the mechanic were universal it would be used for skills, natch.]

1. The only real advantage is that the drastic scaling of high level combat would feel strange if it didn't accompany scaling of out-of-combat effects. It would be weird to look back and realize that, while you have an easy time fighting dragons these days, you're not significantly better at climbing a wall. If combat and magic grows well beyond the human norm, why would other potentials remain limited to it?

I'd argue that this follows from *combat and magic* scaling too far, too fast out of human bounds in recent D&D editions, but that's a much harder problem to solve.

But you've basically hit on the same problem I have with skills in D&D: they're grafted on and don't really use anything similar to the core mechanics other than rolling a d20 with bonuses. D&D's combat roll can be as swingy as it is because:
* You usually get a lot of rolls in a given combat to normalize the result
* It's really hard for attack bonus between two PCs to diverge as drastically at higher level as the bonus in a given skill (and, if it does, it's probably because the character with the lower bonus uses magic that doesn't need an attack roll)
* There's no real margin of success on attack rolls (other than critical hits), so it becomes a flat percentage pass/fail rather than the fail by 5, just hit, succeed by 5, etc. the skill system wants to use

Using the d20 for skills but then removing the effects that make that work for combat introduces a lot of the problems you're discussing.

2. The d20 has no mean result value. Given enough rolls the average result will be 10.5, but on any given roll the chance of receiving any result will be equal to the chance of receiving any other result.

Technically, this means it has no mode value. The mean value is the average of all possible results weighted by probability, so 10.5. Mode is the most frequent result. Pedant ho!

Although I'm against skill-heavy systems myself, in their defence I'm sure no sensible DM would call for a roll in your arm-wrestling example. Conflict resolution mechanics are for when there's an actual conflict, not for when the outcome is a certainty with little to no leeway for chance.

For topic two, that issue is with the range of the scale, not with the skill system itself. Are you arguing against the skill-heavy position in its entirety, or against the specific implementation seen in Pathfinder and similar systems? If it's the former, then the problem's inconsequential since it's specific to the latter.

3. @John:
Fixed

Also, the implementation regarding the automatic increase at level. I do not have a problem with skills increasing, just the endless scaling up to silly numbers.

Obviously the 4e hack of +5 to 'trained' skills and having everything just auto-scale (at 1/2 level) is a bit of a fix, but the whole thing seems like a result of the premise of 'have the same mechanic' and not objectively valuable.

4. @ Samhaine
"The only real advantage is that the drastic scaling of high level combat would feel strange if it didn't accompany scaling of out-of-combat effects. It would be weird to look back and realize that, while you have an easy time fighting dragons these days, you're not significantly better at climbing a wall."

I agree but would go one step further: It seems weird to me that characters start out within such a close percentage of each other's abilities in many 3E skills even without training. Only NPCs really practiced enough to get good at many skills before they went out to murder-hobo about?

And this whole business of small incremental advances in skill bonus at each level and trying to decide which skill to stick them in is just a tedious, undramatic pain.

But I do like the idea if skills as a means to customize a character, as shorthand for common abilities which characters are likely to develop in a setting. I've read the arguments about just discussing with the DM what sort of background you want the character to have. But a blank canvas can be tough to fill. IMHO sometimes lists (and random tables) can be a nice source of inspiration.

This whole discussion has got me musing on some skill based house rules to both have and eat the skill cake:

At each level characters get a "skill advancement". This either gains them a new skill or a greater bonus in the old skill as follows:

Apprentice/Dabbler +5
Journeyman/Trained +10
Master +15
Legendary +20
Superhuman +25

You might get three or four advancements at first level with no limitation on how they're spent and maybe one advancement every couple other levels. (I see the skill categories as being more like "secondary skills" of previous editions, rather than narrower 3E skills).

These are explicitly intended to be used in one of two ways:

1) As a general descriptive gauge of ability. Are you just a total neophyte in all your skills or actually pretty impressive to begin with? Are you up to forging The One Ring yet, or still getting your fingers dirty inking scrolls? This also might be more of an indicator to the DM what sort of setting and descriptive elements to throw your way without specific prompting.

2) If you really must: As a quantitative bonus to add to dice rolls. But specifically only for situations when you really think rolling the dice might really add interesting suspense, drama, etc. to the situation.

5. I don't think that either of your complaints really do a good job of attacking the idea of having skill systems in general. Let's hit them:

1. First off, I'd give the Ogre a +4 for having size Large, which makes most of the problem go away. But still, the problem is that you're using a random game mechanic (rolling dice) to model something that isn't random (arm wrestling). This sort of thing applies to anything in which the result is generated randomly, not just skill rolls, so the problem here isn't really specific to skill rolls and is solved by "don't roll dice for shit that isn't random" no matter what kind of skill system you are or aren't using.

2. For #2 the problem isn't with a skill system as such, so much as the 3.5ed D&D implementation of the skill system. As you say, "this is an artifact of the range of results on the d20 being low compared with the range of possible skill modifiers." This is exactly correct. This problem goes away if you restrict the range of possible skill modifiers. It's far far far too easy to get your modifiers to go up into the stratosphere in 3.5ed since there's vastly too many stackable modifiers, which applies right across the board to things like spell DCs, AC and to-hit right along with the skills. This is a problem (and a very serious problem), but it comes from 3.*ed modifier bloat, not having a skill system as such, since it infects pretty much all of the system, not just the skill system.

6. @peter, I'm a huge fan of "Skills: The Middle Road" which is a simple and elegant way of doing something similar to your suggestion.

@David,
1)The example is RAW, unless you want to define arm wrestling as a combat maneuver. Also, it's an in game conflict - isn't that specifically what we use dice for? Conflict resolution? I'm confused about how arm wrestling is less random than, say, wall climbing.

2) my point exactly?

However, those are sort of already a given - which leads to my two actual questions of

Why do we want skills to increase at each level
and
Why do we want our skill system to be part of a universal mechanic?

The thing is, I know I don't and I know why. I'm curious why someone who likes the pathfinder skill system does want things to be this way.

7. -C:
"Also, it's an in game conflict - isn't that specifically what we use dice for? Conflict resolution? I'm confused about how arm wrestling is less random than, say, wall climbing."

At least for me, I'd be leery about rolling dice for non-random stuff and save the dice for when things become sufficiently random (like combat). Not breaking out the dice in all cases doesn't mean you don't want to break out the dice in no cases. There's a middle ground here.

I'm not familiar with how Pathfinder has changed 3.5ed in the details, but as a 3.5ed DM I would've slapped a size modifier on any situation (like this) in which size is directly helpful (leverage helps a lot with arm wrestling). I'm not sure if It's RAW, I don't feel like going down the rat warren of 3.5ed RAW especially when you're looking at Pathfinder.

"Why do we want skills to increase at each level"
For this one, it feels weird if a character gets better at hitting stuff with swords but gets not even the slightest bit better at a huge swath of other things related to adventuring (note: this is a problem with d20 as well since fighters don't get any better at knowing which monster is which at level 20 than at level 1, in most cases). At least for me, you need at least some kind of skill advancement unless you're going to reduce the advancement in combat, otherwise things start feeling out of wack as you hit higher levels.

"Why do we want our skill system to be part of a universal mechanic?"
I think this would be better filed under "we we want a universal mechanic to begin with." If you want a universal mechanic is should be universal, if you don't want a universal mechanic then the question becomes moot.

8. @David,

Well, that's the thing about RAW, is that it's RAW. If we're talking about how things are broke, pointing out how I would avoid the problem doesn't make them any less broken.

Re: Skill increase.
My question was every level. I am not (and have not) made the statement anywhere that skills should increase as the character gains experience - I'm talking about the problems inherent with adding points every level.

So I'm down with skill enhancement and improvement - is there any reason it has to occur every time you gain a level?

Also, I feel compelled to point out things already feel out of whack at higher levels.

Re: universal mechanic

ah, ipso facto of course. Yes. I think a universal mechanic is a bad idea.

9. Wall-climbing is more random than arm-wrestling because it involves more outside variables. A given handhold might be slippery, or jagged, or seem secure but crumble when the character puts their weight on it, or there might be a sudden gust of wind, etc. Arm-wrestling doesn't involve that kind of chance; it's a simple game with straightforward rules intended to ensure no one has an advantage over another. Unless one player is cheating or has "competitive arm-wrestler" written on their sheet, I see no reason anybody would play it as anything but a direct comparison of strength scores.

A more reasonable example of a contest of strength would be if the wizard and the ogre are fighting over control of a weapon. The ogre is much stronger so is likely to easily tear the weapon out of the wizard's grasp, but maybe the ogre doesn't get a good grip on it, or the wizard twists it just right, or some other piece of luck. There's room for random chance in that scenario, so the wizard winning doesn't seem quite so absurd.

10. Simple D20 resolution does choke when one does roll vs roll contests. Unless of course one wants the underdog to succeed in unlikely situations now and again.

The example arm wrestling contest can have the unpalatable result of -1 STR mod Wiz beating +4 STR mod Ogre by increasing the number of relevant modifiers. A +4 size bonus for the Ogre is reasonable and in scale with the grappling rules used in the game, a non proficiency penalty would be fitting as the WIZ class isn't noted for having STR as a prime requisite or being overly skilled applying physical strength. The WIZ isn't going to win unless the Ogres rolls a 1 (lucky thing that fly landed on the ogres nose and distracted him).
The PC winning a contest doesn't have to represent the PC prevailing in a directly linear manner.

Of course why bother with the Ogre rolling at all? Have the Ogre's STR score be the target number the Wiz must beat and the limp armed WIZ is never going to prevail.

11. As other people have noted, if you treat the arm wrestling as a grapple instead of a strength check, the wizard has little chance of winning, but that doesn't get to the real nub of the issue.

What you think is that there should be different gradations of randomness:

A. Things that aren't random at all.

B. Things that are somewhat random, but not really.

C. Things that are highly random.

d20 D&D can cover A. (due to the Take 10 rule) and can handle C. at low levels but then C. stops working after a while due to how bloated the modifiers get.

Basically you value this kind of gradation of randomness since you don't want arm wrestling to have a small random component but not be anywhere near as random as combat. However, I don't think that breaking things down by combat vs. not-combat is necessarily the best way of doing that. Is combat inherently more random than anything else?

12. Have you ever arm wrestled? This is a good example of where the oft misunderstood skill challenge would be useful. It would take so many successes to win, and for every 5 you beat the opponent by you would earn a success.

So in this case if you set it at 3 successes necessary and the ogre rolled a 1 with the wizard rolling a 20, the wizard would get 2 successes and still not succeed. That’s not to say the wizard couldn’t succeed, but obviously some random fluke occurred (the ogre suddenly got a cramp, was distracted by someone yelling his name, etc) which is why we roll the die in such situations.

The skill challenge is sometimes useful and sometimes needed. It can create tension in dramatic situations (such as a fierce ogre nearly losing an arm wrestling match to a wimpy wizard or climbing a challenging 50 ft. cliff) but just like anything else, if the players aren’t enjoying the tension or don’t agree, change the way you handle things. That doesn’t make the d20 system wrong as a method of adjudicating. Also, it is a matter of the DM. Yes, if the DM plays it out as such:

“Alright the ogre swaggers up to the table and glares at you with bloody eyes. Roll a die.” :roll: Ogre 1 Wizard 19 “Okay roll again” :roll: Ogre 4 Wizard 12 “No successes, Roll again...” :roll: Ogre 2 Wizard 14 “Wow you beat the ogre!”

Uh, yeah that’s gonna be boring and not particularly dramatic.

“The ogre swaggers up to the table and glares at you with his bloody eyes. ‘I’m gonna wrench your puny arm off fool!' Roll”
Ogre 1 Wizard 19
“As your hands touch, you immediately apply pressure. The ogre’s hand nearly touches the table as you catch him off balance! Roll again.” Wizard goes “If your brain were any smaller, only a zombie flea would find it to be an appetizing meal.”
:roll: Ogre 4 Wizard 12
“The ogre isn’t able to fully recover from your sudden success, but you seem to have lost your momentum. He glowers at you and mutters epithets under his breath. Roll again.” The wizard quips “What was that you were saying, I couldn’t quite hear you.”
Ogre 2 Wizard 14
“And the wizard manages to touch the ogre’s hand to the table to the surprise of all the onlookers as the ogre jumps up and lets out a fierce bellow. ‘Trickery! This puny magic man could never best me without magic!’ His companion steps forward and puts a hand on his shoulder. ‘The rules were set. You accepted the duel. He cast no magic while here.’”

Just like all of D&D, it’s in presentation and DM ability on whether or not you find skills to be fun. Personally, I find nothing fun about “Okay, the ogre has a 21 strength and you have an 8 (or say an 18 just because that’s a bit more realistic when looking at modern day arm wrestling competitions.) His strength is higher so he wins… Uh… what?

13. One thing that helps with combat is that it rarely comes down to one single roll. So even if the d20 has no mode, if combat takes 5-10 rounds of rolls on each side, you're looking at 10-20 rolls at least. That will help with averaging out results.

Skill tasks, on the other hand, used to be settled with single rolls. Highly random, not good. One improvement 4e offered was skill challenges that required multiple skills, but the idea remains that sometimes you use a single roll to test the use of a skill.

What if, instead of one roll, you treated skill use like combat? To climb a wall, you roll against your skill, and if successful, you advance d6 feet upwards. If you get a 20 it's 2d6; if it's a 1 you fall (and this adds drama, since falls from greater heights become more dangerous). Pick locks? There are four springs on this lock and you have to dismantle all of them to open it. Alchemy? You have to add each of four ingredients in just the perfect way or it won't work. There are very few skills (mostly knowledge-based ones, I think) that couldn't be simulated this way.

14. @ -C:
I'm a huge fan of "Skills: The Middle Road" which is a simple and elegant way of doing something similar to your suggestion."

Huh, the link to Skills: The Middle Road appears to already be purple in my browser, suggesting that I'd read it when slogging through the recommended OSR reading a few weeks back. So I might have gotten the idea, or part of it, there.

My adherence to the +5, +10, +15, etc. bonuses in particular though was for 3E backwards compatibility purposes. There being a few things I like about 3.x related games, I never know when it will be handy to cannibalize some d20 related mechanic.

15. I'm not really a fan of padding out something by adding a bunch of rolls where you're doing the same thing over and over, I'd rather deal with the conflict by breaking it down into the specific task the PCs are doing rather than something more abstract.

For example:

The PCs see a bunch of skeletons in a hallway and on one of the walls there are some round holes, an obvious arrow trap. If the PCs still blunder in anyway they get hit by arrows. If they notice the arrow trap there's a few things they can do:

A. Try to sprint past it. Result: they get hit by arrows.
B. Jam a ten foot pole in front of the arrow trap, trigger the arrows and then sprint past before it can reload. Result: automatic success.
C. Stick their finger into the arrow holes and try to wiggle their fingers around and disarm it that way. Result: they get hit by arrows.
D. Try to worm a wire under the pressure plate in the floor and disarm it. Result: PC makes disable device check, failure may result in getting hit by arrows.

Note that only in D. does a skill check get made.

16. I agree with your points, but if the example is from D&D3.5 it's a terrible one, and doesn't match what a fair DM might put PCs up against in that game. I'll elaborate more, unless the example is from 4E, in which case I don't know the rules well enough to bother.

I can't resist this though - your DC35 check in D&D3.5 would be the equivalent of a 35 foot jump or a 30' charge along a greased 2" wide pole (that's actually only DC30, but a 60' charge would require two checks). A challenging roll for the specialist, but an absolute dick move to throw at a whole party.

17. Ok, before my reply I'd like to state two things as a given.

A) Running the arm wrestling contest as a grapple in pathfinder or as a skill challenge in 4e would mitigate the issue of die randomness. Running it as a single strength check is also a valid method of resolving the conflict in game. This is beside the point - the point is the lack of a mode of the d20 is a known issue with contested results.

B) Related to my earlier post, the issue is not with the skill of the DM. Clearly a good DM can make bad things not bad, and good things even better. The fact that they can is tangential to the the disscussion of skill mechanics in this post.

Games (and the rules to those games) can be interesting, fun, consistent, and help maintain suspension of disbelief on their own merits without the aid of a player. My evidence for this is that there are many fun games where a narrative by a participant is not necessary for enjoyment to be had.

Basically we want to avoid the Rule 0 Fallacy. (i.e. if you can fix a broken or in this case not fun rule, then it's ok that the rule is broken)

18. @David:

Excellent summation. Yes, you have outlined my thoughts more clearly then I could. I desire mechanics to address different levels of randomness.

Also Combat might not be the best way of splitting it up, but I do think that it is more random then other activities. I go in deeper depth on this point tomorrow.

19. @DomDem

I'm positive you are aware that skill challenges have their own mathematical issues.

4e treats these as known problems and invents several 'kludgey' solutions to address them, such as all skills increasing at 1/2 level, and limiting the bonus to trained skills only.

This is just a patch that doesn't address the core issue involved with the skill system and the desire to have the skills increase every level and sharing a mechanic across the skill system and combat

My core point is that skills increasing with level is ('ahem') dumb. They should increase with training and it should come with a cost - but associating it with level is a bad idea.

Also, the mechanics should be representative of the subsystem and they should be fun on their own merits. They should not just be the same as combat because they are representing different things.

20. @Menace

yes but will it be fun for your players to "fight" the wall every time they want to climb? As a special occasion I could see it being enjoyable, but many of the skills will come up frequently.

@David

That seems like an excellent way to resolve the situation.

@Danger-dino

Yes, but at high level this comes up nearly every time a skill needs to be used. This is the root of the problem with the esclating DC's. Either you spend a lot of time trying to engineer situations where skill challenges don't apply to the whole party, or you have to address the core issue.

21. My core point is that skills increasing with level is ('ahem') dumb. They should increase with training and it should come with a cost - but associating it with level is a bad idea.

Levels in D&D are already an abstraction, and nobody uses the training rules as it is. If I had to use a skill system, I wouldn't want a separate method to keep track of improvements. If inflation is a problem, reduce the amount by which skills improve. That seems simpler than creating a whole separate method of advancement. Maybe someone who actually likes skill checks disagrees with me.

22. @ John:
"Levels in D&D are already an abstraction, and nobody uses the training rules as it is. If I had to use a skill system, I wouldn't want a separate method to keep track of improvements. If inflation is a problem, reduce the amount by which skills improve. That seems simpler than creating a whole separate method of advancement. Maybe someone who actually likes skill checks disagrees with me."

I think the problem is that "level" is shorthand for "all character experience". But in D&D "level" is strongly tied to combat ability (through HD, To Hit, magical ability, etc.). This makes sense since improving at combat is a significant component of the D&D experience. But to a large extent non-combat skills are just along for the ride (especially with 3E & 4E advancement).

This does support a certain style of play: the idea that a character starts out with humble origins and gradually becomes epic in every regard. But it does negate the style whereby someone can be an awesome thief with no real combat ability.

Hence the "middle road" style of advancement: skill is handed out in chunks, thereby avoiding the tiny, incremental increase in every ability with level.

The original middle-road suggests skill being obtained via pricey training, but I see no reason that there couldn't be a more-middle middle-road, where the chunks were handed out with level. To some degree skill increase would still be associated with level, but could progress at a different rate from combat ability.

23. @Peter K:

My thoughts exactly.

24. @ -c: I personally only had problems with it the first time I ran higher level D&D3, mainly because I didn't have the hang of the numbers yet. The examples of skill DCs don't have anything to do with DM skill either, they're right there on the DM screen and in the book.

Here's where I pointlessly go on about the d20 example. For anyone that doesn't care about D&D3 minutiae, skip the bit between the lines.

_____

+9 ability bonuses won't usually come in until 16-20th level, and then only for a character with a maxed out Dex belt that started with an 18 or 20 at 1st level. That's a pretty big assumption. Remember that D&D3 is balanced against either the elite array (15/14/13/12/10/8) or PCs with 28pt buy. If you're using 32pt buy or 4d6-drop-lowest, you're basically giving PCs a free level.

Giving several major magic items to one half of the example (+13 of your bonus would be from magic items!), while depriving the other of same doesn't make sense to me. All characters at high levels are going to have access to pretty cool magic items, but it's such a random variable that you can't make assumptions. May as well give one of your examples the wild shape ability, a wind walk spell, or a half dozen potions of climbing, jumping, and flying.

This example fits in better with my feeling for the d20 system:
Level 15 players
DC 25 balance to cross. Failure means they're going over. Quick thinking players that fail but take appropriate actions may be able to take a DC20 Climb, Jump or Tumble check to grab hold of an edge as they go over.

Player A is at +28 (+18 skill bonus, +2 skill synergy bonus from Tumble, +8 ability score bonus)
Players B&C clock in around +6 (say from a combination of ability scores, skills and/or synergy bonuses)
Player D sits at +0 (+0 skill bonus, +0 ability bonus). He's glaring at the DM right now.

(I can sense the eyes glossing over, D&D3 rules are dull, dry things to read!)

In this, one character succeeds without rolling, two need to make tough rolls and the fourth can't succeed (and, role-playing-wise, might be too terrified to move from sheer terror. I would be, considering). He'll need to use something besides his physical skills to get across.
_____

Anyways... I've been enjoying the posts on skill rules and player agency, it's just that after playing high-level D&D3 for years this example sticks out like a sore thumb. You don't create situations where PCs are required to make DCs 20 points above their character level, unless the whole point is to get the PCs thinking creatively how to not even have to roll that check.

25. Blogger ate my big ol' post during its problems earlier this evening.

This is one of the strongest critiques of the D20 system I've ever seen. Strong enough to make me fundamentally question whether or not the D20 system is worth supporting. I'll need to ponder possible solutions for awhile.

I do have two responses I would like to make.

1) With regards to the disparity in skill checks as characters become higher level, I view that as a strength rather than a weakness. In your example, the character who made the auto-success now has to figure out how to help the character with the auto-failure cross the bridge. It's an interesting and organic obstruction. The best kind!

Secondly, unless I misunderstand, you're saying you dislike skills which rise as a character's level rises. I'm not saying there's no reason to feel this way, but the argument you made here seems to work against a system where the dice have a linear progression. If, for example, 3d6 was used for skill checks, I don't see how there would be a problem.

26. @Dangerdino - perhaps my math was wrong.
I assumed a starting value of 20 (+5) and then assumed that the character would be able to boost that to 24 by 15th level (+7 total) and then had an item that provided a +4 bonus to the stat (total +9)

I then assumed a skill boosting item (like boots of striding/springing) which is fairly cheap to create. (+10) (bonus squared X 100 -> fairly cheap at 10,000gp out of 240,000gp at that level).

And then 15 ranks, + the 3 class skill bonus in pathfinder.

Is this out of line for a 15th level character?

" You don't create situations where PCs are required to make DCs 20 points above their character level, unless the whole point is to get the PCs thinking creatively how to not even have to roll that check.
"

I guess in my experience players are building their characters better than this?

When I created my example of the skill - I lowballed it. If I were actually trying to max out that ability, I could get it much much higher in play then +thirtysomething. (mid 40's for a skill you're building - mid 50's if you're really trying to push the limits).

I guess I've been playing at tables where an AC 22 at first level is considered low. The current group I'm part of is not very guilty of this but many other groups I have played with are.

27. @LS

Not sure if serious. . .

I am just sharing the same arguments that convinced me to move back to AD&D and move away from 3e and it's derivatives due to poor design as a DM. (I will play pathfinder all day long because I like the combat game - but I'm not fond of it as "D&D")

I found I enjoy DMing much more after the switch.

1) I have not had my personal play experiences reinforce this. Perhaps it is true, but mostly it seemed to turn into "Let frank take care of this room guies" in my experience.

2)Yes. There are other solutions besides eliminating linear progression, but a bell curve would by definition produce more accuracy as the player became more skilled.

28. I'm going to have to agree with John here, if you're going to have a level-based game tie things to level, if you're not going to tie big swaths of the game to levels, then just throw out having levels entirely. In my opinion, 3.5ed divorces skills and levels too much as it is.

What I'm leaning towards with gaining skills is to have it work like Weapon Proficiencies in AD&D. Basically characters get better at different rates at a very broad class of actions (using weapons) and then get a penalty if they're using one particular sub-set of that broad class of actions in a way that they haven't been trained in (non-proficiency penalty).

So, to apply the same thing to skills, divide skills into a few broad classes for example "sneaky stuff" (which would include traditional thief skills basically) and have different classes get better at it at different rates. Then hit people with penalties if they try to do specific things that they're not trained in.

Not really what you're looking for, but I like it since it's MORE closely tied to level than 3.5ed skills.

By the way, I'm really valuing this discussion of skills, it's making me think about what situations to use skills in. I'm coming to different conclusions than you, but it's being quite helpful.

29. @David

Thanks for the comments, but I'm curious about what you don't say.

Why is tying skills to level an improvement? I'm not saying it's not, I'm asking about your reasoning. It's pretty clear you're a fan, so why tie this d20 model to level? Or any model really, but we're specifically talking about d20.

I think the most important thing I can here is why you're coming to those different conclusions. That's why I'm posting this stuff - I'm genuinely interested in the other position.

Care to help me out?

30. @david

(Also, some of this has to do with perspective. You mention 'out of whack' increases, but to me most 'on the sheet' character advancement stops at level 8/9 for AD&D. After that point, everyone is essentially the same - they all save on a 2+ and hit nearly as frequently, the differences come in characterization and play.

Who are they as people? and what choices do they make?
Those questions become the interesting thing - not that their climb skill keeps increasing. )

31. @-C I'm absolutely serious. I'm relatively new to the tabletop role playing online community, so I had never seen these issues spelled out before. They are compelling.

Due to your critiques, and those I've discovered myself as I explore the d20 system more deeply, point to some very serious flaws. Some of which I might not be able to house rule away.

Sorry if I made that unclear in my last post. I was very tired, and as I said, blogger had eaten my comment.

32. @LS, there are a hundred solutions, but if you play by RAW, and have somewhat min/max oriented players, expect the game to drastically change around level 7 and then again around level 13.

A character that can jump with some consistency 60+ feet is what I would call a superhero (leaps tall buildings in a single bound)

Again, pathfinder is awesome at being pathfinder - it's super fun to play. It just isn't what I associate with dungeons and dragons. I spend my time running and playing AD&D doing much different things then I do in pathfinder. I enjoy the AD&D things more. The pathfinder things are still fun. I however just cannot handle running a pathfinder game - too much book lookup and numbers prep.

33. I think I see then. Assuming a 20 at 1st level goes against the expectation of the system - it's built with the 28pt buy or 15/14/13/12/10/8 ability array in mind. With that, a human can attain a natural 18 ability score by 12th (+3 inherent), or 20 by 20th.

Now, I know that doesn't ever play out in reality - but if you use 32pt-buy or 4d6/drop, it's much the same as using UA's Method V for everyone in AD&D... you'll have to adjust and juggle for the whole campaign for those higher average scores. Still, high scores happen, novice/foolish players will nerf themselves even in a 28pt-buy to get them, and +6 stat items aren't impossible to have around 15th level, so I allowed for that when I rewrote the example.

About the boots (which actually only provide +5)... I didn't include any competency or circumstance bonuses in my changes to your example because everyone can get those bonuses, making them pointless to add into any generic comparison. I likewise didn't include inherent bonuses because they require manuals or wishes (and with wish you have to cast multiple wishes in quick succession to ever get your inherent bonus past +1).

The cheap availabity of minor items like that is why adding their bonuses muddy the water - any character could have those. Well-prepared players are likely to be as well stocked with one-use items like potions or scrolls as you'll allow as well.

I agree that as you say D&D3 gives a wide spectrum of skill scores at high levels that don't mesh well with linear rolls, just not as wide as you presented. I have been tempted to use the 2d10 example you posted to get rid of the linear range, and in 0D&D/AD&D I use 2-4d6 against ability scores & NWPs depending on difficulty.

"I guess in my experience players are building their characters better than this?

When I created my example of the skill - I lowballed it. If I were actually trying to max out that ability, I could get it much much higher in play then +thirtysomething. (mid 40's for a skill you're building - mid 50's if you're really trying to push the limits)."

On one hand you're giving maxed out scores, skills, and magic items, and on the other you're saying you're lowballing?

I'd like to see those numbers. Are players giving you the bonus, or are those 40's the net result after they roll? Every time someone calls out a number like that across the table at me (the D&D3 campaign I'm running is at 17th/18th level), a quick glance at their character sheet shows either they're stacking similar bonuses or doing the math wrong. Only the druid is getting numbers in that range in our game, and those aren't on skill but grapple rolls (now *that's* a place where your criticism of the unbound and linear nature of D&D3 and other d20-based games is especially applicable).

"I then assumed a skill boosting item (like boots of striding/springing) which is fairly cheap to create. (+10) (bonus squared X 100 -> fairly cheap at 10,000gp out of 240,000gp at that level)."

ACK! You can't do that. That's not even twice the price of normal boots of S&S, and definitely more than twice as useful. Follow the advice in the DMG. You can't just blindly follow the base formula to crank out items, especially when they *double* the performance of standard items. A continual +10 should be 20k minimun. It's not exactly a bow of true strike (price that one out for a grin), but it's getting there. Letting players use DM *guidelines* as some sort of rules crowbar to lever open the game is the way to ruin.

I'd write more, but this is starting to look more like D&D3 evangelism, which is not what I'm going for. D&D3 is a very fun game to DM at the table (though it is a nightmare to prep), but it's definitely not my favorite or most comfortable edition.

34. -C: I think it basically comes down to me not liking skill/level hybrid games. There's a lot of games that are completely skill based that I like a lot and if I want a system in which skills are raised independently, I'll use one of those games. For me at least, hybrid systems are often the worst of both worlds. So to turn the question back on you, why do you want to have levels to begin with? There's plenty of good RPGs that don't have them.

As for why I want skills tied to levels, it's because for me a level is a really broad thing. A level not only means that your fighter is better at hitting stuff, he also has more of whatever hit points represent, is better at dodging fireballs, a whole slew of things. This means that, at least to me, after a character has gained a few levels they start jumping up genres. As the famous dragon magazine article says, most of the Lords of the Rings characters make sense as 1-5th level heroes, that means that with a couple more levels they're starting to inch into legendary fiction territory and with a few more levels, they're starting to look like Demigods. Basically, to me, a 10th level fighter is a lot more like Hercules than Aragorn and I want a Hercules to be better than a random farmer at more than just sword-swinging accuracy, I want Hercules to be better than a random farmer at basically everything.

As for the "out of wack" comment, what I meant is that it feels weird for me for a name level character to be incredibly better than a 0-level farmer at some things (sword-swinging accuracy, not getting killed by shit), but not a whit better at other things. It feels out of wack to me for a character to be the sort of over the top badass who can laugh in a dragon's face, but then not be better at you're average John McClane action hero athletic stunts than a random 0-level farmer. The same sort of thing pops up in 3.5ed in which very often a 20th level fighter is no better at recognizing what a kobold is than a 1st level fighter.

I guess it comes up down to what gaining a level means. Does gaining a level in fighter mean that you're a more skilled fighter and that's that or is gaining a level in fighter pushing you one step further on the gradient from smuck to demigod. For me at least it's the second and it feels weird to me if a high level fighter can do demigod-quality stunts in one area (say, fighting a big dragon in single combat) but be no better than a 0-level farmer in other areas.

tl:dr Is D&D zero to Olympic fencer or is D&D zero to Hercules?

35. Oh and as far as where I'm coming from in general. I played bunch of TSR-D&D as a kid, took a long hiatus, then played a lot of 3.5ed D&D before getting burned out badly on it. Then 4ed took things in a completely weird direction and Pathfinder made most of the things I didn't like about 3.5ed worse, so I was stuck without a go-to version of D&D for a while. Then, fairly recently, I went back to some old versions of D&D and I've found that in a lot of ways they just work better in play than d20 games (basic bread and butter like class balance and speed of combat is a lot better than d20 games), so while I have some philosophical problems with what old games are trying to do, I find that they do what they're trying to do better than d20 games.

36. @Danger-Dino

I think my point is that table expectations are different. What I mean is that the example I provided you would be a baseline for my groups in my previous experience of pathfinder. Yes, I'm assuming a maxed out core stat and those items - but that is a far far cry from maxing out the bonus.

At my table the example I provided is what I'd expect from someone who had no interest in pumping acrobatics. Just the score I'd expect from an average dude. It's low-balled, because for a determined player it's possible to get the bonuses up into the stratosphere.

Some points.

Even with the array you mentioned the starting stat in pathfinder will be 17 because of the +2 stat bonus, leading to an 18 at level 4 and a 20 at level 12.

I agree, A DM who was wary of our wiles would probably require a ring slot to get a +10 bonus - but someone looking to pump his skills wouldn't stop at +10 (ring of jumping etc. is clear example of straight cost equation). Even with the increased cost, that's a sacrifice someone looking to pump a stat is willing to make.

Other bonuses someone looking to add to really boost their stuff might use.

Skill Focus (+3/+6)
Racial Bonus (+2 - +4)
Spells (Heroism, jump, et. al. +2 - +30)
Aid Another
Magic Items X2 (Circumstance, Competence, Insight, Luck, sacred/profane)

Like I said, I didn't include any of those. Adding a skill focus, heroism, a party member helping and one more pricy magic item at +5 (insight) would give another +6, +2, +2, +5 = +15 to my already calculated 37 for a total 52, giving a range of 53-73.

And again - that really isn't maxed. My example was low-balling because I would expect at least some of the above - possibly more.

I am enjoying this conversation - can you imagine now? If this is my experience, can you see what actual skill use might look like at my table? I'm hitting this 'PC's are out of range' problem historically at level 7, not level 15.

37. @David,

I think some of this might have to do with pace of advancement.

I am a big fan of level and class role based games, primary because it insures that at high power levels the players all look different. They are quick to pick up and play, and generally the classes are broad enough to allow anything.

However, when playing AD&D, it can be six months or more of real time for someone to go from level 7 to level 8, shutting down their ability to advance in learned skills during that time can be an issue.

I think that article is from the Alexandrian