Specialist Rogue

[Celebrim]

This is why every adventure I've ever written or run involves at least one evil undead spike trap that really just wants to be loved.

His name is Jimmy, and once my players get past it, I can run whatever game I want.

Ha!

More seriously, I would say there's some real gains to be made in not having an ability that isn't used at least 1/session.

I feel pretty much exactly the opposite. Much of the value I find in class based systems is it forces people to take abilities that aren't used at least 1/session. There is a lot to be said in the favor of having abilities you don't use all the time, and the reverse is less than pleasant in actual play.

Though I'd just be tempted to redesign some of those trap-based abilities to apply to other kinds of "traps" that are not strictly mechanical (natural hazards...

I never realized that someone didn't consider natural hazards to be traps. How do you key naturally arising hazards (say a floor that has decayed to the point it will collapse if walked on) if not as traps?

, trap-like ambushing monsters like piercers, cursed items, enemy bluffs, hidden creatures, invisible critters, etc.) and fluff it like a general ability to recognize danger when they maybe see it.

To a certain extent I think trope is captured by Uncanny Dodge, perception skills as a class skills, generally high dexterity scores, and good reflex saves already, but it would be interesting to write into the description of things like Mimics, Piercers, etc. that their surprise attack counted as a trap for the purpose of trap sense.

 

[Ahnehnois]

With regard to your rogue sense variant, my problem is first that it doesn't address the real problem with Trap Sense (narrow, as it applies to only a subset of traps) and secondly that its likely to lead to a bunch of finicky fiddly variation for its own sake. Tons of situational +1 modifiers is the bane of play. I'd far rather very slightly expand the range of things your rogue sense applied to.

The bookkeeping concern is legitimate. However, I think the rogue class is one of the least bookkeep-y to begin with, and I find that situational modifiers encourage creative, rogue-ish play wherein players try to contrive the situation that activates the bonus. It's amazing what people will do for a little +1.

Moreover, one of the main goals of mechanics in general and my goal for this ability in particular is differentiation. What is it that makes a character special? Different? A broader bonus would defeat that point.

I think it is a 'no brainer' to choose between Dauntless, Guarded Mind and Sixth Sense, as they are clearly far more generally useful than the rest.

Well, like all house rules it is a draft and they may not all be perfect, I don't know that I would have chosen those three as the best. In any case, that's why they're written as examples. If some player has some thematic justification for a set of bonuses, I'd say go for it. That supersedes the list.

With regard to your rogue talent variations, I think that the reverse of the problem we've been discussing is generally true. Trap Finding is so hugely useful when it is useful, that all the variant talents pale in comparison. I'd pretty much always choose trap finding at 1st level, and then pick up the bonus feats at higher level to taste (though probably Stand Up would be a no brainer 2nd choice).

Very campaign dependent, I'm sure. I think this approach makes it fairly unlikely that any individual rogue will have trap sense, which is really the goal the OP was getting at. If you want to call yourself a trap specialist, you choose it. Otherwise, it's not worth much of anything.

In other words, while I approve of your goals as well, most of your variation doesn't address the actual problems IMO, there is poor balance between the choices, and most of the choices amount to 'NPC only flavorings'. There is nothing wrong with NPC only flavorings - character building is for DMs too - but I think you end up with too many of them relative to the number of choices.

You're welcome to your opinion; I make no claims as to perfection. However, I think the basic approaches (making trap sense the generic rogue sense and making trapfinding just another special ability) are simple ways to accomplish the OP's goal without having to call out a special archetype for each of the fairly basic rogue subtypes.

 

[I'm A Banana]

I feel pretty much exactly the opposite. Much of the value I find in class based systems is it forces people to take abilities that aren't used at least 1/session. There is a lot to be said in the favor of having abilities you don't use all the time, and the reverse is less than pleasant in actual play.

I'm coming at this from a more practical, in-play standpoint. An ability you don't use at least 1/session is mental baggage (gotta keep it in mind for when it comes up on that rare occasion!), inefficient use of space (this line on my character sheet that I never really read), and lacks the capacity to be character-defining (since we don't usually see your character doing that). It could wind up being an ability you NEVER use. It's not worth considering as on-par with more frequently used abilities. It's kind of empty and pointless, since it just sits there doing nothing much of the time, and it's a hassle when it comes up because it's not used often enough for people to get familiar with how it works.

There's a big gulf between "use all the time" (like an attack roll, forex) and "use at least 1/session," and I don't count "making you have largely pointless abilities" as a value-added element of the class system. It's the old tiger repellant scheme: if you don't ever actually have to use it, there's no real value in it.

Which isn't to say that rogues shouldn't be good trapfiders, it's just that this element of rogues might be better as a specific application of a more general skill.

How do you key naturally arising hazards (say a floor that has decayed to the point it will collapse if walked on) if not as traps?

I specifically had in mind effects like sudden changes in weather or venomous patches of unknown plants or hillsides in rain that would give way with a bit of pressure. The general danger sense of a rogue would mean they get chance to detect and a bonus to avoid being affected by those hidden dangers -- a rogue just has this sense of what might be a risky gambit.

 

[Celebrim]

The bookkeeping concern is legitimate. However, I think the rogue class is one of the least bookkeep-y to begin with, and I find that situational modifiers encourage creative, rogue-ish play wherein players try to contrive the situation that activates the bonus. It's amazing what people will do for a little +1.

Since most or all of your senses are reactive, I don't think this justification applies.

Moreover, one of the main goals of mechanics in general and my goal for this ability in particular is differentiation. What is it that makes a character special? Different? A broader bonus would defeat that point.

Differentiation is great. I approve. But a deeper bonus makes the differentiation far more salient than a +1 in six different abilities would anyway. And a bonus can be broad while still allowing differentiation. For example, my suggestion of 'Pick two, but you have to stick with them' provides more than 100 possible choices, while still being simple to track. It also avoids the problem where, if you must choose one, you are pretty much certain to pick primarily on the basis of mechanical utility.

Well, like all house rules it is a draft and they may not all be perfect, I don't know that I would have chosen those three as the best.

Compare Guarded Mind with Astute and Daring. All fear effects are mind effecting, so all you gain with Daring is a minor bonus to a minor skill and what you lose is a bonus to all of those other mind effecting powers. Most illusions that provide for a saving throw are also mind effecting, but guarded mind also protects you from those fear effects and most enchantment spell spells (crucially compulsions). Guarded mind is less strictly better compared to Astute than it was versus Daring, but its still the easy winner.

Sixth Sense is amazing. Bonus to initiative of up to +6 and also a +6 insight bonus to AC when surprised. Both are going to be frequently game changers. If you win surprise, you'll likely go twice in a row (hello sneak attack!!) and if you lose surprise then between this and Uncanny Dodge your AC jumps up to the point you are a bad target. Go pick on someone else.

And Dauntless protects against a enormous number of save or die effects that show up routinely in high level play.

Compared to say Resourceful. Not only is resourceful not even a 1 per session type power, but it seldom is going to be a life and death situation when you use it, and the advantage it gives you is easily replaced by a minor spell effects like 'create food and water' or 'endure elements'. It's nice and flavorful, but I'm not going to be wasting my +1 bonuses on it.

You're welcome to your opinion; I make no claims as to perfection. However, I think the basic approaches (making trap sense the generic rogue sense and making trapfinding just another special ability) are simple ways to accomplish the OP's goal without having to call out a special archetype for each of the fairly basic rogue subtypes.

In general, I agree with the idea of having customizable class abilities. I disagree with the notion of creating specialists. In fact, while I didn't pick on it at the time, I think the idea of reducing the breadth of the rogue's class skills while making them super good at 'that one thing' is a terrible idea. It's all the worst features of point buy and all the worst features of classed based at the same time.

 

[Ahnehnois]

Since most or all of your senses are reactive, I don't think this justification applies.

? That I don't see at all.

Differentiation is great. I approve. But a deeper bonus makes the differentiation far more salient than a +1 in six different abilities would anyway.

It's true that +1 isn't much. But one other thing I'm fighting that's inherently a challenge with class/level systems is determination. When do you decide who you are? My belief is that each level should bring new developments and choices. If you start out as the trap rogue and you decide you want to be suave and charming later, you change course not just in your skill selection, but in relevant class abilities. I don't like the idea of giving out large bonuses all at once at lower levels; I'd rather spread them out.

Compare Guarded Mind with Astute and Daring. All fear effects are mind effecting, so all you gain with Daring is a minor bonus to a minor skill and what you lose is a bonus to all of those other mind effecting powers.

It's possible that Daring needs another bonus. Perhaps to Tumble. That would make a difference.

Most illusions that provide for a saving throw are also mind effecting, but guarded mind also protects you from those fear effects and most enchantment spell spells (crucially compulsions). Guarded mind is less strictly better compared to Astute than it was versus Daring, but its still the easy winner.

I don't know about that one. SM is a pretty useful skill.

Sixth Sense is amazing. Bonus to initiative of up to +6 and also a +6 insight bonus to AC when surprised. Both are going to be frequently game changers. If you win surprise, you'll likely go twice in a row (hello sneak attack!!) and if you lose surprise then between this and Uncanny Dodge your AC jumps up to the point you are a bad target. Go pick on someone else.

This one I don't see as being that powerful at all. I mean, yes, it is really good at getting you the upper hand, but in any situation other than the beginning of combat, it does nothing. +4 to initiative is one feat. Initiative is worth less than a skill.

And Dauntless protects against a enormous number of save or die effects that show up routinely in high level play.

Compared to say Resourceful. Not only is resourceful not even a 1 per session type power, but it seldom is going to be a life and death situation when you use it, and the advantage it gives you is easily replaced by a minor spell effects like 'create food and water' or 'endure elements'. It's nice and flavorful, but I'm not going to be wasting my +1 bonuses on it.

You may be right. This is cross-listed with the barbarian replacement for trap sense and I wrote it to give the barbarian a standard option. That said, I don't know what else I'd add to the survivalist option to make it better. Dauntless is pretty situational as well.

 

[Celebrim]

I'm coming at this from a more practical, in-play standpoint.

No, no, I'm coming at this from a more practical, in-play standpoint.

You are clearly coming to this from a theoretical 'this is how you should design things' tag line that someone wrote in the subheading of an essay with a title like 'Six elements of good design'. I can hear your formalism rolling around in the background quite clearly, but my preference comes from practical, play-tested, organic, evolutionary play.

One of the reasons I have a hard time getting excited any more about point buy systems is that they fail hard on this test. Novice players create interesting well-rounded characters. But players eventually evolve to create characters that efficiently use the space on the character sheet and have only character defining powers. This is not a good thing. You end up with wacky balance. Everyone has a hammer, and they have to use the hammer for solving every problem that they have. Either the hammer just smashes the problem because well, with every single point invested in being good at that one thing, they are really really good at it, or else the hammer doesn't work and they are completely helpless. Every situation becomes basically binary and the GM is forced to metagame as a consequence. Having nothing on your character sheet that isn't immediately useful is a disaster in play for everyone at the table.

There's a big gulf between "use all the time" (like an attack roll, forex) and "use at least 1/session," and I don't count "making you have largely pointless abilities" as a value-added element of the class system. It's the old tiger repellant scheme: if you don't ever actually have to use it, there's no real value in it.

No no no. It's hugely valuable. It means that I as the GM never have to think, "Can I use tigers? Because I'm not sure anyone is going to have a relevant ability and if I use tigers, it could mean a TPK" or conversely, "What's the point of using tigers? Tigers aren't even a challenge since player X has fully invested in Super Tiger Repellant". It also means that players occasionally have to go, "Crap, the hammer doesn't work. I wonder if this screwdriver might be effective." Or sometimes, "Hmmm.. I have tiger repellant. Maybe that will be useful."

I specifically had in mind effects like sudden changes in weather or venomous patches of unknown plants or hillsides in rain that would give way with a bit of pressure.

In the case of say an avalanche or a flash flood or similar hazards, I would tend to have the Trap Sense apply to any relevant AC or Reflex saves to avoid it (if any), because well, it's functionally equivalent to a planned trap that sends some hazard sliding and bouncing the way of the rogue. Gravity is gravity. I might sometimes use a different skill to recognize the hazard than 'search', but I'm probably literally recording the physical hazard as a trap when I make my notes. Just because someone didn't do it on purpose, doesn't mean it's not a trap.

I'm undecided/unconvinced on the value of replacing 'rogue sense' with a generic 'danger sense'. I toyed with embracing the option that I gave Ahnehnois of 'pick two' for a while just because I like the diversity and flavor of his list, but I think at this point I've tentatively abandoned that due to the complexity of balance and the fact that most do something redundant with some other character building resource.

As I said, my current version of the rogue has a Rogue Sense which is equivalent to having Ahnehnois's 'Trap Sense' and 'Spell Sense' simultaneously. In play, I find this plenty broad enough to be useful while avoiding trampling on anyone else's space and also avoiding problems I see in wording a generic 'danger sense' well and simply so as to avoid arguments as to when it applied. The rogue has used 'rogue sense' in both manners multiple times at this point, and the prior rogue using my current version of the rules also used both on occasion. Play testing is convincing.

I also largely think that making the rogue's 'trap sense' as generic as a 'danger sense' risks a double mechanical dip of trope establishment (doing two slightly different things to establish the same skill), given that the rogue already has advantages like Uncanny Dodge, good Reflex saves, and so forth that apply in most or all cases that 'danger sense' would apply. More broadly, Ahnehnois's selectable +1 bonuses do very little that couldn't be accomplished by replacing the whole 'sense' concept with 2-3 bonus feats from a generic list of +X feats (Iron Will, Courage, Endurance, Skill Focus, etc.). So I'd need both a compelling reason to give the rogue the equivalent of a few feats, and a compelling reason to do this instead of something like 'trap sense'.

The Rogue Talents I like better, but have much the same issue. Basically you are just adding 2 rogue bonus feats to the progression, and unlike Ahnehnois I think one is going to amount to a feat tax in any campaign where you aren't certain ahead of time that traps won't come up because someone in the party needs to be able to deal with traps. Unlike the other things on the list, Trap Finding is an absolute. If you don't have it, every trap with a DC above 20 is undetectable. To be able to deal with a whole class of hazards for the investment of 1 feat is a no brainer. Even if you don't plan on disarming mechanical traps, just finding them and locating them has huge value.

Again, unless the rogue is poorly balanced and not a baseline, giving the rogue a few feats seems like 'Power creep is the solution to everything' thinking. If you want to be Resourceful and Daring, why can't you take Endurance and Courage as feats? Why must the solution always be, "I never want to forgo anything. I just need more stuff!" Thinking about my own Rogue, I think that I have in fact added the equivalent of one bonus feat to the progression, but backloading it rather than frontloading it because a) at low levels rogues provably rock already and b) the main help martial classes need is above level 10 and c) rogues are already arguably tier 3 or so (and certainly are in my game with increased skill focus).

 

[Celebrim]

But one other thing I'm fighting that's inherently a challenge with class/level systems is determination. When do you decide who you are? My belief is that each level should bring new developments and choices. If you start out as the trap rogue and you decide you want to be suave and charming later, you change course not just in your skill selection, but in relevant class abilities. I don't like the idea of giving out large bonuses all at once at lower levels; I'd rather spread them out.

I can sympathize with all of that, particularly if you are coming from this from Pathfinder which tends to lack this flexibility. For example, I very much dislike the Pathfinder approach to fighters where you choose what sort of fighter you are up front and that defines you for all 20 levels.

But for the rogue, if you want to be suave and charming later, it's just a matter of allocating the skills points. With the most skill points of any class in the game (actually my rogue has a base 11 per level to reflect that there are slightly more skills in my game than in core), it's not really an either/or problem. You can be both a good trap finder and suave and charming. Heck, if you play an 18 int human rogue in my game you can maximize 16 skills without making sacrifices, so you are fast, charming, perceptive, etc. all at the same time.

I don't accept that the only suave and charming characters have Skill Focus (Diplomacy) or Deceitful and that's how suave and charming are defined. You don't need a special marker for everything you can do, and thinking that you do leads to 'Power Creep is the solution for everything'. And to the extent that you want to be extra suave and charming, you always could take 'Deceitful' as a feat.

It's possible that Daring needs another bonus. Perhaps to Tumble. That would make a difference.

I don't know about that one. SM is a pretty useful skill.

It's not bad, but I'd rather take a bonus to saves versus compulsions than a bonus to sense motive. For better or worse, sense motive is often a character skill that can be replaced by player skill. I find my players mostly use it for confirming things that they already know.

This one I don't see as being that powerful at all. I mean, yes, it is really good at getting you the upper hand, but in any situation other than the beginning of combat, it does nothing. +4 to initiative is one feat. Initiative is worth less than a skill.

I don't agree at all. Most D&D combats in either 1e or 3e are over in 2-3 rounds if you play RAW. Winning initiative means that you get in 2 attacks to the foes 1 or 3 attacks to the foes 2 as opposed to the reverse. Add to that that as the rogue, winning the initiative often means getting in a couple extra sneak attacks. The old joke in D&D is that the surprise round is the early game, the initiative roll is the midgame, and the end game is the first round of combat. Improved Initiative is a great feat, equivalent in my opinion to "You have a 20% chance each combat of gaining an extra turn." On a rogue add to that, "On your extra turn, you have a better than 50% chance to do triple normal damage."

 

[Ahnehnois]

But for the rogue, if you want to be suave and charming later, it's just a matter of allocating the skills points. With the most skill points of any class in the game (actually my rogue has a base 11 per level to reflect that there are slightly more skills in my game than in core), it's not really an either/or problem. You can be both a good trap finder and suave and charming. Heck, if you play an 18 int human rogue in my game you can maximize 16 skills without making sacrifices, so you are fast, charming, perceptive, etc. all at the same time.

One thing I don't like from PF is skill consolidation. I'd much rather spread things out than clump them together

I don't accept that the only suave and charming characters have Skill Focus (Diplomacy) or Deceitful and that's how suave and charming are defined. You don't need a special marker for everything you can do, and thinking that you do leads to 'Power Creep is the solution for everything'. And to the extent that you want to be extra suave and charming, you always could take 'Deceitful' as a feat.

There are a number of classes in PF that get skill bonuses above and beyond those types of feats; it's part of how things work. I've expanded the concept so that most classes get them, but I've also slowed them down and made them competence bonuses to lessen stacking issues (and put a dent in the Christmas Tree phenomenon).

It's possible that Daring needs another bonus. Perhaps to Tumble. That would make a difference.

I appreciate the feedback. I went and relooked at some of these and I think that Tumble is quite a useful skill and also something that a daring person would do a lot of.

It's not bad, but I'd rather take a bonus to saves versus compulsions than a bonus to sense motive. For better or worse, sense motive is often a character skill that can be replaced by player skill. I find my players mostly use it for confirming things that they already know.

SM does have some combat applications (in 3.5 at least; can't remember what PF has to say). I find it's something I can often work in very effectively. I did go back to the Astute benefit and add a Search bonus. That should definitely make it powerful enough.

I don't agree at all. Most D&D combats in either 1e or 3e are over in 2-3 rounds if you play RAW. Winning initiative means that you get in 2 attacks to the foes 1 or 3 attacks to the foes 2 as opposed to the reverse. Add to that that as the rogue, winning the initiative often means getting in a couple extra sneak attacks.

Does that mean that Improved Initiative is overpowered, or even obligatory as a feat? It's a decent choice, but I don't think it's that good. +1 to init is 1/4 of a feat. It's not earth shattering. It is most useful for a rogue, but I don't think it overshadows the other ones. Sixth sense is also one of the few benefits that is almost strictly combat-related, and it gives no direct offensive bonuses. I think it's just in the mix.

I appreciate the feedback and it has helped me think about some things, but there's always going to be debates as to which is the best option in any menu of options. I won't be able to get it to the point where every option is exactly equal in my eyes for the next game I run, let alone in your eyes, and that's okay. It doesn't have to be perfect; it just has to present an interesting choice.

 

[I'm A Banana]

You are clearly coming to this from a theoretical 'this is how you should design things' tag line that someone wrote in the subheading of an essay with a title like 'Six elements of good design'. I can hear your formalism rolling around in the background quite clearly, but my preference comes from practical, play-tested, organic, evolutionary play.

Nope, I'm just thinking about what I would want to play with. I don't want to have something on my character sheet I'm not going to use in play, and so if it's not something I'm going to use in each session (regardless of what the DM throws at me), it's not something I can count on to use in play, so it's just taking up space that could be better used. I also don't want to have to remember abilities that only come up once every few real-world months (if that), or bother trying to figure out how to use them.

I want game elements I can play with, not things that just sit there on my sheet.

As DM, I also don't want to give my players these albatross abilities that they just have to hang around their necks. I want a narrow list of effective decision points for my characters to make in actual play, which means irrelevant abilities just gum that process up as they sort through them looking for the thing that applies.

One of the reasons I have a hard time getting excited any more about point buy systems is that they fail hard on this test. Novice players create interesting well-rounded characters. But players eventually evolve to create characters that efficiently use the space on the character sheet and have only character defining powers. This is not a good thing. You end up with wacky balance. Everyone has a hammer, and they have to use the hammer for solving every problem that they have. Either the hammer just smashes the problem because well, with every single point invested in being good at that one thing, they are really really good at it, or else the hammer doesn't work and they are completely helpless. Every situation becomes basically binary and the GM is forced to metagame as a consequence. Having nothing on your character sheet that isn't immediately useful is a disaster in play for everyone at the table.

Again, big gulf between "not immediately useful" and "use at least once per session."

Also, the myopic specialization thing is kind of a different problem. Making sure PC's have abilities that are broad enough to come into play at least 1/session doesn't necessarily yield hyper-specialization -- it actually fights it, in that PC's won't try to pump up their traps abilities without being aware that traps are a sometimes thing.

No no no. It's hugely valuable. It means that I as the GM never have to think, "Can I use tigers? Because I'm not sure anyone is going to have a relevant ability and if I use tigers, it could mean a TPK" or conversely, "What's the point of using tigers? Tigers aren't even a challenge since player X has fully invested in Super Tiger Repellant".

This is still more about hyper-specialization than about frequency of ability use. You'll get no argument from me that hyper-specialization is generally something you want to avoid, but that's not the same problem as having pointless abilities just hanging out gettin' ignored week after week.

Traps are a sometimes thing, so they shouldn't be something a character is able to define themselves around. Traps fit in the broader category of "stuff that can catch a party unaware," and THAT's something with a lot more frequency, and something that it makes genre sense for a rogue to be good at picking up on before it kills them. That's like giving a character who wants Super Tiger Repellant a suit of metal armor instead. It protects you from tigers, but it doesn't JUST do that. It's a broader ability you'll use more often.

It also means that players occasionally have to go, "Crap, the hammer doesn't work. I wonder if this screwdriver might be effective." Or sometimes, "Hmmm.. I have tiger repellant. Maybe that will be useful."

Again, this is more about the binary quality than about the frequency of use. Yeah, of course, having a narrow pool of things that work is problematic. But that isn't an issue that a bunch of occasional-use abilities solves. Have an ability that is hammers and screwdrivers and tiger repellant, and suddenly you've got something you don't have to wait for the perfect moment to use!

Again, unless the rogue is poorly balanced and not a baseline, giving the rogue a few feats seems like 'Power creep is the solution to everything' thinking.

It'd be power creep, to be sure. And possibly redundant with qualities like Evasion. It's really just an argument for a rogue re-design.

Short of THAT, sure, swapping out one ability you'll rarely use for another ability you'll rarely use is balanced, but the OP seemed like the issue they were trying to SOLVE was the rare use. That swap isn't going to satisfy that goal. If you want something to be used more often than a rogue's trap stuff, you'll need to make it BETTER than the rogue's trap stuff (that is, broader).

 

[Starfox]

So if you play a cleric do you also need on average 1 undead per session?
Do you also need on average 1 evil foe per session if you have paladins?
Do you need to guarantee 1 encounter that can be solved by diplomacy per session if there is a bard in the party?

Its a matter of expectation management, But yes, surely, in any DnD-ish game I've been in in the last decade or two, these were all common. Ubiquitous in the case of evil opponents, merely common in the case of undead and diplomacy. Maybe 1 per 2 sessions. Certainly much more common than traps. If I made a paladin and the GM did not warn me about it, I'd certainly expect a clear majority of all fights to be against evil opponents. If they're not, then my paladin is clearly in the wrong place and should request transfer.

In the case of channel energy, in Pathfinder that is now mainly a healing resource.

In general, I agree with the idea of having customizable class abilities. I disagree with the notion of creating specialists. In fact, while I didn't pick on it at the time, I think the idea of reducing the breadth of the rogue's class skills while making them super good at 'that one thing' is a terrible idea. It's all the worst features of point buy and all the worst features of classed based at the same time.

I agree overspecialization is bad. But rouges ARE already specialists - at traps. The least I feel I want to do for them is give them a choice of several specialties. That actually lessens the degree of specialization in the class as a whole. Thus also goes with how Pathfinder does things, look at bloodlines, revelations, and cavalier orders. Sets of abilities to chose among, sometimes with sub-choices (like revelations). 3E did it too to a lesser degree, with cleric domains.