Are lack of skill points really the issue?

[tedcahill2]

A very popular house rule is to give classes a minimum of 4 skill points, but I'm not sure that even touches on what makes D&D 3.5E skills so flawed.

I think managing skill points is one of the most difficult things to manage as a player. here are the issues I see:

DC Knowledge (knowing how many ranks you really need in a skill)
Hard to audit multiclass character (obvious to anyone trying to back track their skill selection through 10+ levels of multi-classing)
Cross-class skills (tracking different max ranks and dealing with skill points from different classes being valued differently)


Here's my thought, and this is not thought out per se, but I wanted to see if anyone had opinions to share on the topic.

DC Knowledge is one that affects players the most I think; because it really sets the foundation of what makes skills so annoying. Not all skills are created equal; there is simply no good reason to have 23 ranks in tumble when 10+bonuses will be more than enough to hit the hardest DCs. Same is true for balance, climb, jump, etc. The only skills worth maxing are the ones opposed by other skills, spot/listen, hide/move silently, etc.

Most players, and definite my players, don't understand the "needs" of a given skill. So they always err on the side of caution and get max ranks in whatever skills they can and that's it. The fighter is a great example here because most, if not all, their skill really don't need to be maxed out. Giving fighters more skill points doesn't fix that issue, it just let's them max more skills.

I think something along the lines of skill proficiencies should be used instead. No more managing skill points, instead classes will all have a skill proficient bonus built into their chassis that determines their bonuses with the skills of that class. In that way, some classes will be notably more skills without having to manage dozens of extra skill points.

For example (and these are beta thoughts):
A rogue will have 8 skill proficiencies at level 1, and will start the game with a proficiency bonus of +4. They choose 8 skills which you can add your proficiency bonus to. All other skills on the rogue's list of class skills do not gain the proficiency bonus, but gain a bonus of 1/2 the rogue's level.

A fighter will have 4 skill proficiencies at level 1, and will start the game with a proficiency bonus of +2. Again all other skills on the fighter's list of class skills do not gain the proficiency bonus, but gain a bonus of 1/2 the fighter's level.

A character can also dabble in a number of additional skills equal to their INT modifier. They can add 1/2 of their proficiency bonus to skills they dabble in.

To me, this structure solves the issues I outlined above. The fighter gets more skills, but isn't as skill at them as a rogue is. To me that makes tons of sense, a rogue simply spends more time perfecting their skills and should not only have more of them but also be better at them than a fighter type.

I still need to work out a way to handle multi-classing though.

[EldritchWeaver]

I would suggest to import PF rules instead. The very least they make calculating the actual skill ranks easier. Maybe even consolidate the skills, too. If there is still something missing, you can change things from there.

[Telonius]

If you're really doing a full overhaul of skill points, you would also need to give a bit of attention to Prestige Class prerequisites (and, to a lesser extent, feat prerequisites). Some PrCs set high skill point prereqs to give a hard limit to how early a character can enter them. You might just want to scrub the skills and put a level requirement, or actually require the character to dabble or be proficient in them; but either way it's something that should be considered.

[VisitingDaGulag]

Math is hard. Use the UA max ranks variant. Easy

[tedcahill2]

I would suggest to import PF rules instead. The very least they make calculating the actual skill ranks easier. Maybe even consolidate the skills, too. If there is still something missing, you can change things from there.

I liked a lot of what pathfinder did with skills, but I don't think it did anything to solve the issue with DC knowledge, which I think is one of the greatest issue with the skills system.

Without a solid grasp of the DCs that you're likely to face for a given task, not to mention decent statistical modeling, it's impossible to tell if you are putting enough points in a given skill.

Even beyond that, if someone chooses to put only 2 points in climb instead of 4, then they miss a DC by 1, they are then going to regret not maxing the skill, which only further encourages players to get max ranks in every skill.

[tedcahill2]

Math is hard. Use the UA max ranks variant. Easy

True, but overly rigid. I want to strike a balance between full customization (the core rules), and total rigidity (max ranks variant).

[Pex]

For 3E the problem is cross-class costs. It kills any incentive to invest in skills not on your class list. There's no room to expand your character.

The other problem is opposed rolls. It's probably more player perception than actual play. They're thinking a fighter's dismal Spot compared to a rogue's phenomenal Hide the fighter can't do anything. A fighter trying to Spot something shouldn't always be having to see someone with a phenomenal Hide. The one doing the Hiding should also sometimes be not such an expert on it. The party's fighter won't Spot the enemy rogue Hiding for an ambush, but maybe he'll Spot the enemy fighter and not be surprised after all. He could still be flat footed with respect to the enemy rogue after he acted against the enemy fighter even in the next round, but at least he got to act that round against the enemy fighter.

In Pathfinder the issue isn't as harsh. No cross-class costs means a fighter can invest in Perception and still have a decent chance to see the Stealthy rogue. The rogue is better at Stealth than the fighter is as Perception as a feature of being the rogue, but the fighter is not pathetic about it. Traits help non-class skills become class skills. A feat can help if it's really important to the character. Skill Focus still gives +3, but it increases to +6 when you have 10 ranks in the skill. It's not difficult for the fighter to be equal or even better at Perception than the rogue is at Stealth.

If skills are important for you, a human fighter with 12 Intelligence can get 5 skill points per level including favored class bonus. That is not shabby.

[Zombimode]

A very popular house rule is to give classes a minimum of 4 skill points, but I'm not sure that even touches on what makes D&D 3.5E skills so flawed.

Skills are a complex matter in D&D 3.5 and you do raise a good point that DC knowledge is quite important. I disargee that skills are flawed in 3.5. I actually like that they are not equal and their granularity.

You are quite correct in that placement of skillpoints is not fire-and-forget. Its not easy. You actually have to put some thought into it if you. It is in this sense not accessible.

Now you could say "Well, it should be! So thats is a flaw." But let's reconsider. The 3.5 skill system rewards you for system mastery: you can actually get more out of if by puting some effort into. And this is in line with the 3.5 rule set in general. This is why you (should) play 3.5 in the first place.

I've not completely understood your alternative system (skills with "proficiency" never get better?), but it seems it will reduce the impact of system mastery in managing skills.



The other problem is opposed rolls. It's probably more player perception than actual play. They're thinking a fighter's dismal Spot compared to a rogue's phenomenal Hide the fighter can't do anything. A fighter trying to Spot something shouldn't always be having to see someone with a phenomenal Hide. The one doing the Hiding should also sometimes be not such an expert on it. The party's fighter won't Spot the enemy rogue Hiding for an ambush, but maybe he'll Spot the enemy fighter and not be surprised after all. He could still be flat footed with respect to the enemy rogue after he acted against the enemy fighter even in the next round, but at least he got to act that round against the enemy fighter.

Like you said, this is an imagined problem, not an actual one. Looking over the monster manual there are lots of creatures that would try ambushes without having sky-high hide modifiers.

That specialists can not easily (or at all) be detected by run-of-the mill dudes is a feature, not a bug.

[Grod_The_Giant]

I seem to recall 4e doing something similar, and I know 5e works along such lines (although without dabbling rules, essentially). In that one, multiclassing into a high-skill class after character creation gives an extra proficiency or two--not as much as if you started, generally, but more than you had before.

In any case, removing the granularity and switching to maxed skill/half-maxed skill/no skill simplifies the skill system somewhat, but doesn't necessarily solve the other two major problems: the high cost of buy-in (you need to invest in multiple skills to cover a single role) and the general inefficiency of skills compared to spells. ("Base 4+Int" is generally seen as a way to help with the former, though if you construct your proficiency rules correctly, you can do something similar.)

[BoutsofInsanity]

There are a whole host of issues. Lack of skills in the current system are one of them. If you maintain the current system.

This gets complicated, so hang on.

You have to manage both the intent, and the RAW interpretation of the current skill system. It seems like the intention was to make it so each class had a unique feel to bring to the table. That if you play a rogue, you didn't have to worry another party member (Say a fighter) who could already outfight and out tank you, also out stealth and out rogue you.

Hence class skills and limited skill points. Sticking with an archetype allows for diverse roles to be developed, and classes feel unique.

This worked for about the first long term game run by new people. Everyone is excited to stick to the pre-conceived archetypes in traditional fantasy. The Rogue was rogue-like, the fighter could fight and jump, the Wizard was knowledgeable, and the Cleric wise.

After that game though a player said "Hey, I want to play a Knight. A well educated fighter!". Great, you will need knowledge Nobility, History, Local, Animal Handling, Ride, Diplomacy, Perform Dance, etc... And the skill system fell apart. Because people realized perception was important, stealth was universally useful, and people didn't really figure out what DC's were for because we were all a bunch of ***** in highschool.

In this instance, having more skill points solves the "problem".

But then, what if you want to play something out of the stereotypical class archetype? Then class skills ruined you.

Game design has advanced past that. The thing that is actually being dealt with is that 3.5 and Pathfinder operate under the idea that Classes are in-game constructs. IE: Your profession and being is "Fighter" or "Wizard". Which implies a specific training and history behind becoming what the in-game meta-narrative believes you should be.

Which we as players who have advanced 10 years of game design realize how limiting that can be. Which is why in later editions such as 5e or even 4e they got rid of class skills and skill rank advancement, and added proficiency which gives a controlled automatic advancement of Skill Proficiency. They realized that a "Fighter" could be a knight, hulking brute, gladiator, hunter, or a bandit. That a Wizard was a thief, explorer, war-mage, intellect, or diplomat. And that skills added a way to customize your character into your own unique construct that fulfilled the player's desires of seeing their concept in-game.

This all leads into the issues that you are struggling with. Your playing in a retro edition with a mindset of a modern gamer with all the design breakthroughs we have had since then.

What you want to do, reducing the DC's is actually on the right track. Paying attention to the number bloat is important because those high numbers trivialize not the DC's which can be set higher. But other players. If you have an enemy that can't be beat except by maxed out ranks, then the players are forced to put their ranks maxed out.

But if you set DC's in the 10's other people can make those, but are trivialized by the player who maxes out his ranks.

The best would be to control the max ranks skills have. ("BOUNDED ACCURACY") Which would allow you to determine what is the best of the best and what is reachable with a benchmark. Then develop your skill system around that. Like 4e or 5e. To Fix what your issues are I would do the following...

Remove ranks entirely. You have a proficiency Bonus for being proficient. Say 4. This never increases. You then have a class bonus of 2. For being in class and a stereotypical archetype skill. Items, races, feats will only give a +2. Magic (Spells) are the only exception. Lastly, proficiency per class should be increased to a minimum of 4, I prefer 6.

This says the maximum a character with an 18 in a relevant stat will achieve is 4+2+4+2+2+2 = 16. Which means the upper echelon, performed by the best of the best with magic DC is 20+16 or 36. You can set your DC's accordingly. Your opposed rolls will look much more reasonable, because the chief diplomat of a nation, all though not high level will look like the following: 4 (Prof.) + 2 (Class) + 2(Stat)+2 (Statesmen Lapel :Non Magical Uniform) + 2 (Skill Focus) of +12. Which means engaging the diplomat in a battle of wits might not be a good idea.

Of course the Swagalicous Bard might be able to do it, naturally being more charismatic, and boasting some magic items. Also hit the Diplomat with an intimidate (not Physical Violence, but with your reputation) to throw him off his game, say a -2 and the odds have swung in your favor. Of course the Diplomat might still put up a good fight, but you just stacked it for your ends.

Just my thoughts.

[ExLibrisMortis]

Maybe you can make max skill ranks increase along a pattern similar to saves: good skills give you +2 at level 1, and then 1/2 at each level; bad skills give you +1/3 at each level; skills you don't have get no bonus.

Each class comes with a list of primary class skills that are always good skills. For example, rogues would always get Hide, Move Silently, Open Lock, Disable Device, Spot, and Listen.
Each class comes with a list of secondary class skills that are always bad skills. You pick a subset of these as additional good skills. For example, a rogue might pick Tumble, Bluff, Use Magic Device, Diplomacy, Disguise, and Sleight of Hand.
You also pick a number of skills equal to your Intelligence modifier to be moved up one notch; a skill you don't have becomes a bad skill, or a bad skill becomes a good skill (or you spend two picks to turn a skill you don't have into a good skill). For example, a rogue with 14 Intelligence might improve Search and Gather Information to good skills, or Spellcraft and Knowledge (arcana) to bad skills, or Autohypnosis to a good skill.


For multiclass characters, skills that are uniformly good/bad/not are easy to keep track of, but varying statuses are a bit of a nuisance. Then again, it's often hard to reverse-engineer skill points from current rank totals after a dozen levels with varying skill point amounts and rank caps, so that's not entirely a new problem.

[Zaq]

I have a severe love-hate relationship with the way 3.5 does skills. There are some aspects of it that I love more than the way they're handled by any other D&D or spinoff that I've personally touched, and there are some aspects of it that call out for cleansing fire (and yet simply removing them with a hatchet won't solve the problem entirely).

Like, I love the fact that taking cross-class skills is an option. I love that you can just brute-force your way into being decent at something (never great at it, and likely not even especially good at it, but decent enough that you can roll without wasting everyone's time) even if your class structure doesn't support it. You don't have to spend separate build resources like feats or your race or anything to be allowed to take cross-class skills; you're already losing out on efficiency by doing so, so that's a sufficient cost to just let you do it without tying up something else. It's good for niche builds and good for just taking weird oddball flavor options. I don't like that it's so ragingly inefficient to take cross-class skills that you'll basically never do it, but I love having the option as a backup. (I also hate the fact that if you take the wrong number of levels in a class that has your chosen skill as a cross-class skill, you can end up with a dangling half-rank that you can't fix even once the skill becomes a class skill, though that's mostly a concern in, like, Iron Chef.)

I also like the fact that you get skill points every level. Most other post-3.5 D&D-based systems I've looked at (4e, Legend, 5e, though not PF) treat skill proficiency as being one-and-done, with certain exceptions. You make a character, you pick which skills you're proficient with, and you call it good. I like the fact that you can shift gears midway through a build, I like the fact that you've basically got "currency" to spend, and I like the fact that you can, if you choose, just keep getting better at your current skills (or at some of them but not all of them, spending the remainder elsewhere). And I really love the fact that it makes INT relevant even if you aren't a Wizard-style caster or otherwise interested in INT-based skills; someone who is quicker to learn things gets more skill points than someone who's slower to learn things. That makes perfect bloody sense. But the flip side of the fact that skill points are a thing is that if you take advantage of 3.5's super flexible multiclassing rules, it's hella finicky to track how many points you get each level and how many of those points you can spend on what. I like leveling up and getting a batch of points to spend, but I don't like making an existing build with complex multiclassing and spending all the points in delicate layers. These two things are basically inextricably entwined, though. You can't elegantly have one without the other. Like I said, there's some love-hate going on.

I love that there are a million sources of little nickel-and-dime bonuses to skills. +2 from your race, +2 from synergy, +2 from a masterwork tool, and so on. The fact that there are so many ways to collect little bonuses means that many of those bonuses aren't very expensive in terms of build resources, so if you choose to focus on skills, you can cobble together the ability to become pretty darn nice at a few specific skills even as sort of a side thing. You can make it the centerpiece of the character, of course, but you don't have to, and I love that flexibility. (In contrast, just try becoming good at skills in 5e. Even being a Rogue or a Bard only gets you so far, and heaven help you if you aren't one of the Approved Skillful Classes.) You can use these bonuses to turn someone from good at a skill to great at a skill, or you can become good at something you otherwise wouldn't be very good at. This is similar to, but distinct from, my first point. This can be frustrating when magic gets involved if the magic overshadows (Glibness gives plus what?!) or obviates the skills, but I like that it's a backup. I won't even call out the finickiness of gathering the bonuses as a downside, because it's pretty much entirely optional—you can just have a few skills that you max out and then you don't go searching for bonuses to them, and that's a valid character choice much of the time.

I agree that there are way the hell too many skills, and I'm firmly on the side of the fence that says that a good pair of first steps is to just bump up skill points (which is kind of the topic under consideration) and to condense things as much as possible. There's a large number of skills that I've literally never rolled or made a character capable of usefully rolling, even ignoring the Craft/Profession/Perform/Knowledge fracturing. Having two separate rolls for sneaking around is unnecessarily cumbersome. We don't need both Open Lock and Disable Device. Use Rope is kind of a punchline most of the time. I've never seen a real justification for why K: Arcana and Spellcraft are two separate skills. The difference between Search and Spot is too granular by half most of the time. Heal doesn't do nearly enough mechanically. I'm not saying anything new here. But insufficient skill points are a big part of it.

I don't think that giving out more points is necessarily a cure-all (I'd call it necessary but not sufficient), and yet I think it does more than any one other single fix towards making the skill system flow more smoothly. You get flexibility on characters even after the "tax skills" are paid (seriously, Rogues are "skillful," but if you're expected to be sneaky, nimble, and "the trap dude," you'll lose 1/3 to 1/2 your points by the time you've paid your dues in Hide/MS/Search/Disable/Tumble. You can choose not to do some of those, but there's still kind of a baseline expectation, you know? And we haven't even talked about face skills, Spot/Listen, or mobility skills other than Tumble), you make cross-classing slightly less painful, and you get fewer characters who don't feel engaged with the skill system because there's nothing fun they can do with it.

[ExLibrisMortis]

There's a large number of skills that I've literally never rolled or made a character capable of usefully rolling, even ignoring the Craft/Profession/Perform/Knowledge fracturing. Having two separate rolls for sneaking around is unnecessarily cumbersome. We don't need both Open Lock and Disable Device. Use Rope is kind of a punchline most of the time. I've never seen a real justification for why K: Arcana and Spellcraft are two separate skills. The difference between Search and Spot is too granular by half most of the time. Heal doesn't do nearly enough mechanically. I'm not saying anything new here. But insufficient skill points are a big part of it.
I agree with you; the finnicky nature of skill points is, in some ways, a pleasure. (My suggestion two posts up is not something I would prefer myself, it's only a suggestion.)

As a sort of mid-point between 3.5 and PF, there's the option of, as you gain ranks in a skill, awarding half-ranks (or even full ranks) in "associated skills". For example, Search, Spot, and Listen form a group of associated skills; buying ranks in one gives you half a rank in each of the others.


Hide, Move Silently, Sleight of Hand
Balance, Escape Artist, Tumble
Spot, Listen, Search
Disable Device, Open Lock, Use Rope
Craft, Profession, Appraise, Forgery, Knowledge (architecture and engineering) - mainly the engineering/craft connection. Forgery requires careful Appraisal and precise Craft, right?
Concentration, Autohypnosis
Knowledge (arcana), Spellcraft
Knowledge (psionics), Psicraft
Bluff, Disguise, Intimidate
Use Magic Device, Use Psionic Device - assuming transparency
Climb, Jump, Swim
Handle Animal, Ride
Gather Information, Knowledge (local)
Knowledge (history), Knowledge (nobility and royalty), Martial Lore - not sure on ML, has to go somewhere though
Knowledge (religion), Knowledge (the planes)
Knowledge (dungeoneering), Knowledge (nature), Heal
Survival, Knowledge (geography), Speak Language
Perform, Diplomacy


Of course, if you're going to do this, you should probably review synergy bonuses. For example, you might want to award synergy bonuses simply for having one, two, or three skills in a particular group at 5 ranks--bonuses of +2, +3, and +4 respectively.

[Darth Ultron]

DC Knowledge is one that affects players the most I think; because it really sets the foundation of what makes skills so annoying. Not all skills are created equal; there is simply no good reason to have 23 ranks in tumble when 10+bonuses will be more than enough to hit the hardest DCs. Same is true for balance, climb, jump, etc. The only skills worth maxing are the ones opposed by other skills, spot/listen, hide/move silently, etc.


Well, DC is the game do always keep going up. Assuming your not playing a one shot short adventure, and do intend to play the game for a year or more, the DCs can climb high.

Skill are much more of a Roll Playing problem for Roll Players; the people that think that their character in the game can't do anything unless they have to Roll for it. The type of player that has a character that encounters a locked, looks down at the character sheet and sees they don't have the skill open locks, then they just shut down and do nothing.

[Fizban]

've never seen a real justification for why K: Arcana and Spellcraft are two separate skills. The difference between Search and Spot is too granular by half most of the time. Heal doesn't do nearly enough mechanically. I'm not saying anything new here. But insufficient skill points are a big part of it.
Know: arcana and spellcraft are separate because having one skill for *everything magic* would make it even more ridiculously good then the two skills already are. More skill fracturing is actually needed in some places: arcana gives you both monster and general arcane knowledge (which is more valuable than most other types of knowledge), while spellcraft lets you read and analyze and invent spells of every description. Search and spot are different because looking for a trap trigger and seeing someone sneaking up on you are completely different things. Heal, uh, actually does do a ton mechanically when people actually make a point of it- it turns delay disease and delay poison into as good as removed.

More skill points isn't a bad idea, but the more skill points you have the more you're compelled to take the uber skills (even cross class) because they're so uber and you've got plenty of points. So increased skill points actually needs to come with some better skill divisions for magic, manipulation, and a proper idea of the difference between craft and profession. Whack all the skills down to the same level of granularity (though never all the same *power*, as there should always be non-adventuring skills), and you can adjust skill points without certain classes just getting a bonanza and other barely meeting someone's idea of "par."


seriously, Rogues are "skillful," but if you're expected to be sneaky, nimble, and "the trap dude," you'll lose 1/3 to 1/2 your points by the time you've paid your dues in Hide/MS/Search/Disable/Tumble. You can choose not to do some of those, but there's still kind of a baseline expectation, you know? And we haven't even talked about face skills, Spot/Listen, or mobility skills other than Tumble), you make cross-classing slightly less painful, and you get fewer characters who don't feel engaged with the skill system because there's nothing fun they can do with it.
Yeah, no, this is the problem with CharOp and rogues. Because rogues have the highest skill points, CharOp makes builds that are perfect at everything, and then there's the expectation that every rogue is expected to be perfect at everything, which is bogus. The only expectation the game makes of the rogue is search+disable device for the trapguy role. Sneaking is a choice, you can sneak attack by flanking if you want and scouting is another word for "splitting the party." Social skills are a choice which the DMG doesn't actually give any xp for without DM approval, people just love them for unbounded power. Tumbling, spotting, none of these are required if you have a good grasp on the basics, rely on your magic people to deal with magic, and maybe bring some bat-utility belt.

Then comes in proper knowledge of DCs, where you only need 4-8 ranks or so to do a lot of the things after ability+equipment+etc (DC 20 to climb a good wall, DC 15 to make indifferent people friendly, DC 30 and take 20 to open Good locks, DC 20 to craft masterwork items, DC 21 to sneak past a bad guard without fail at a range of 0', DC21 to conceal your identity from anyone without spot and good luck, DC 15 or 20 for "basic-tough" questions, DC 20 to identify 5th level spells, DC 20 to secure a grappling hook at maximum range, etc), so by 7th level a rogue can have a ton of extra skills in the good enough range.

Its the focus on winning opposed rolls vs other specialists, identifying monsters, and reliably activating UMD (and a measure of laziness) that make people think all skills need to be maxed all the time. Changing the system into a max/zero system means all those numbers in the middle stop mattering and you get the same problem 5e has of "What's the skill DC? Eh, whatever it takes to make your PCs pass/fail the amount you want." 3.5 has a ton of pre-defined skill DCs that allow you to build a character with exactly the amount of skill needed to do a specific thing, and then put the rest of your points elsewhere, and that's good. Keeping those DCs while going to a max/zero system means that instead of varying levels of skill, certain things just have a flat required level, which is boring.

[tedcahill2]

I think we all can pretty easily agree that skills are not even. Spot is not equal in usefulness to disguise is not to forgery, is not to climb etc.

What if skills were categorized into simple and complex skills. Jump, swim, and climb would all be simple skills, and disable device, open lock, diplomacy would all be complex skills. Simple skills can be bought at a rate of 2 ranks per skill point, and complex skills would be one rank per skill point.

In other words it is less of an investment to learn to jump well then it is to pick locks. This sort of makes sense to me, but it might add to, instead of subtract from, the complexity of skills.

[Elkad]

...there is simply no good reason to have 23 ranks in tumble when 10+bonuses will be more than enough to hit the hardest DCs...

While I see the point for some other skills (who needs to swim up a waterfall? By the time you can hit the check, you can fly/teleport anyway), Tumble is important.
L10 Rogue, 26 (22+4) dex, 13 ranks, 2 jump synergy, +5 skillboost springy boots.
That's only +28, so I need to roll a 12 for Free Stand. Not exactly reliable.
Spend 10k on my springy boots (+10 Tumble), and I still need a 7.

Sure, about L15 I can slack off - if I never take dex damage, or don't decide to wear armor with a penalty. At that point, why would I?

[Fizban]

Eh, I don't really count free stand for tumble- there's boots that let you swift stand for 500gp and some balance ranks. The main target is 25 to just go through enemy spaces, though if you want to do so at full speed you'll need 35. So tumble is an excellent example of a skill that you don't need maxed: you can take it up to +5 and go 50/50 on avoiding AoOs, or +15 to do that without fail and 50/50 getting through enemy spaces, or +25 to do that without fail and 50/50 doing it at full speed, etc. As long as you're not trying to max every skill, you can still choose tumble as one of the skills you to want maxed, while stopping other skills at lower bars.


I think we all can pretty easily agree that skills are not even. Spot is not equal in usefulness to disguise is not to forgery, is not to climb etc.

What if skills were categorized into simple and complex skills. Jump, swim, and climb would all be simple skills, and disable device, open lock, diplomacy would all be complex skills. Simple skills can be bought at a rate of 2 ranks per skill point, and complex skills would be one rank per skill point.

In other words it is less of an investment to learn to jump well then it is to pick locks. This sort of makes sense to me, but it might add to, instead of subtract from, the complexity of skills.
This is close, but still focused more on expedience. You can't save skill points or go past your maximum, so simple skills end up alternating every other level, unless you also add that you can spend half a point to get one rank. Either way, you still have the problem that some of those complex skills are worth even more than two "simple" skills- but no one will stand for increasing the cost of spellcraft else the poor spellcasters can't afford their skills (and unless I wanted a school based system I'd generally count spellcraft as worth about 3 skills, so it'd have to cost 1.5 ranks under this). And there will always be some skill that someone thinks should be "simple" or rolled together when its not. So I think its better to fix the skill entries than fiddle with the prices. And stay away from any words like "useful,' which have different meanings to different people/characters. There are a bunch of skills because most of them are fairly narrow, except some aren't.

A rating of simple/complex/super complex for goals is useful for thinking about how to divide up the skills though. In the thread about free profession skills, I suggested making profession the planning, and craft the construction. Making something you already know about takes one skill, but planning/designing something you haven't made and don't have an example of takes a second skill. Inventing something completely unheard of could also require a knowledge skill to know the principles required to design it so you can craft it. Three skills to master: knowledge, design, and construction let you make anything make-able of that construction type.

-Manipulating people should also require three+ skills- the skill you use to directly mess with them (bluff, negotiate, threaten), and the 1-2 skills you use to figure out what to say and back up your words so they don't just ignore you (sense motive, gather info, forgery, basically anything outside the box)- the catch is that lots of people are pretty simple so you're usually allowed to guess your way past one of these and lots of DMs refuse to rule certain bluffs or requests of "helpful" friends impossible, so that's another requirement ignored. So instead of doing that, don't, and require an appropriate number of skills for more difficult/complex people.

-Designing a new spell to counter some magical effect? First you need to analyze the effect, then research the spell (or research the effect and then craft the spell, same thing). Identifying spells in combat? Completely different. That's three different forms of spellcraft- and if you want an actual divide between arcane and divine, the research and identify skills should be split into two each, for a total of 5 skills (analyzing effects already in place also applies to non-spell effects and is used for figuring out how to mess with the other guy, its the same skill). That's magical theory, arcane research, divine research, arcane spotting, and divine spotting.

[Florian]

I think the original 3E skill systems gets misunderstood a lot. As someone already stated earlier, the original concept was to make a character a "pro" in what the class is supposed to be good at (class skills), or at least a surprising "semi pro" or "gifted amateur" (cross class skills).

Now while WotC also added "take 10" and "take 20" as an option, they botched that by 1) tying DCs to tasks, 2) overusing skill checks in their modules and 3) making some things like spot and concentrations that should not be skills, but rather be handles like BAB (and don´t get my started about armor check penalty).

So, create four difficulty levels (trivial - standard - hard - heroic, DC 5, 10, 15, 20 (or so, just an example)), then assume automatic success, with class beating heroic, cross class beating hard, no skill still beating standard and no-one ever having to roll on trivial. Try a task that is above your level of competence, you have to roll the dice and beat the DC as usual.

picking up the Fighter and Knight example from earlier shows the difference. A "Profession: Khorne Berserker" or "Profession: Knight of the Inner Circle" should cover basic competency at the chosen profession, while the above skill choice showcases what you're really exceptionally good at.