On the Ancient Feat

Ok, so at some point we have to admit that picking a new ability or power can be pretty cool for a player.

Yes, it carries with it a ton of baggage and problems.

  • Creation requires picking from huge lists instead of rolling and going
  • It effectively is design by denial; it takes things everyone should be able to do and puts them behind walls. "No Bob, I can't let you do that, because that will invalidate spending a feat on that."
  • They show up in splatbooks in great numbers. The quantity causes a lot of them to be stupid.
  • The fact that feats are chained (you need x to get y) mean that you have to plan and build your characters out.
I will seriously Paypal you 1$ if you can find anyone who ever bothered to take Multitalented Mastery.

Multitalented Mastery (Half-Elf)
You are adept at numerous disciplines.
Prerequisites: Character level 5th, half-elf, multi-talented racial trait.

Benefit: All of your classes are considered favored classes. You gain either +1 hit point or +1 skill point whenever you take a level in any class. Apply these bonuses retroactively for all class levels that have not yet gained one of these bonuses.

Normal: Half-elves with the multi-talented racial trait have two favored classes.

Imagine the circumstance this is useful, you have two classes and take a prestige class, right? Then you're getting a few extra skill points (at least 3) and it totally seems useful.

 Except prestige classes can never be a favored class.

What a nightmare.

Of course, there are traps by design also. Monte Cook wrote an essay called Ivory Tower Game Design, now erased. But nothing is gone forever, here's the link from the way back machine Justin Alexander notes in his thoughts on the article that it is impossible to have many choices in a game and have those choices all be balanced from one another. Something is going to be better than another thing. But if you don't realize how useless whirlwind attack is, well, you're going to have a bad time.

So, what is it that makes a good feat? What guidelines can we use to create abilities like this in rule light games that avoid all the problems inherent in the current modern system?

Here are the guidelines for old school feats.

  • No chains. Possibly prerequisites, such as "Must be level 4" might be allowed, but only in ways that segment whole groups of feats, not little dinky separate and unique prerequisites for each and every feat. "Anything off this page at the start. When you reach level 4, you can pick from both pages".
  • The feats must not be something anyone could attempt. This is actually setting design. Is there a roll based chance to discover what a potion is? Or is it something that can only be done with training? It can provide a large bonus on something everyone can do.
  • If feats are available at character creation, there should be very few (7) options, or there should be the option to select the feat later (after the first adventure).
  • The total number of feats, along with their mechanical descriptions should be very few. Think about narrowing your design space. Will this feat fold into another one? Can you just re-fluff another feat to serve this purpose? Does this feat work better as something a class can just do? Is this a feat that everyone will take? Is there another feat that is superior in all or most ways to this one? If so, eliminate it.
  • No feat should be designed to provide something background or character oriented. You can say "I'm a sailor" and you don't need some special "sailor" feat to make it true. That customization should come about from skills (if you use them), ad hoc bonuses, and re-skinning the flavor of feats.
  • The benefit must be substantial, which is likely a side effect of flattening the list. +3 hit points and +1 to AC vs. one opponent are right out. You will only get 3-10 of the things anyway.
  • Flavor is superior to power
  • Feats are useful for providing customization to fighting and thief classes that wizards normally gain through spell selection, so their feat selection options should be greater and more frequent, while wizards should be less.


  1. This appears to be the template WotC used to design the feats in D&D Next, which is interesting. I would also add that if you are going to have class abilities that you can select, they should follow a similar format in terms of concrete advantage, relatively limited number of options, lack of chaining/pre-requisites and specific skills.

  2. Feats as design by denial is something I never bothered with when I dm'd 3e I'd let players try whatever they wanted, they just wouldn't do it well without a feat if there was one designed to do it.

    1. I have the same feeling. Like if my player wanted to sacrifice some precision for damage on a foe, I'd let them get a +1 damage bonus in exchange for -1 attack roll penalty, instead of the 1 for 2 in Power Attack.

      In a way, I kind of like feats that do this because it modularizes such mechanics in feats to allow a designer to streamline the combat system. Otherwise, there'd be a ridiculous number of extraneous options littering up the combat rules.

      However, I do see -C's position and agree.

  3. Interestingly, I'm working on a system of old school feats, and I got to the same conclusions. One thing I also decided is that a good way of balancing more powerful or useful feats is to charge different costs, ranging from 1 to 3 points from a pool of points that the character receives as it advances in level.

  4. This doesn't step back far enough from the "problem".

    It starts with an artificial mechanic (the feat slot) and tries to fill out the resulting design space in order to solve the original "problem," plus other "problems" that never really existed in the first place. It's mechanic-first design, instead of naturalism-first design. Gygax-tenure flavors of D&D tend more toward naturalism-first design than later versions of D&D, though admittedly there's not really a hard dividing line.

    What is clear is that this kind of design pattern undermines the "seek things during play" properties that emerge from Gygax-tenure flavors of D&D.

    Why not start with a naturalism-first approach? A player says he wants his character to be able to do something that transcends the bounds of commonality? Great! Think of how someone in the game world might actually acquire that ability, and then put something in the game world that can potentially produce that effect. Maybe it's as simple as tracking down a barbarian trainer and working out a deal, but maybe it's more complex, requiring a trip to the lower planes to make a bargain with a demigod. Turn the player's desire into adventuring motivation! Why throw that free adventure motivation away by granting feats every X levels?!

    The "feat every X levels" paradigm brings along concern about balance, slowness, wasted expenditures, and distraction from play. By contrast, ad-hoc naturalistic solutions get to ignore all that mechanical baggage.

    1. You might be very interested in the post tommrrow.

  5. http://paizo.com/threads/rzs2q4se?Secret-follower-of-Norgorber Someone wants to use it for a funky build - complete with a debate over exactly how to interpret the rules. It's reminds me of Magic the Gathering.


    What if different feats have different level requirements for different classes? Like fighters can take stunning blow at 1st level, clerics at 4th, no one else can take it? Or a feat that helps all saves vs. magic that wizards can take at 1st, clerics at 4th, everyone else at 7th? This lets you achieve the "no more than 7 choices at first level" while still having a decent number of feats.

  6. As lists of things people can pick, it might work, though I still worry about the tendency towards complexity leading to "builds".

  7. Ugh, is this "trap" nonsense still going on? I'm sick of it! There were no traps, just feats that didn't have the same general level of utility. That's commonly used in good game design. Some weapons might get used more often while others are great for specific purposes. Trying to "balance" everything to be equal for everyone under every circumstance leads to bland, boring design, which I have absolutely no interest in

    Feats were great before all the errata and splatbooks (and especially that bastard of a revision!). They represented specialized training. You weren't denied something you'd otherwise be able to do, but you could be better at it. For instance, you might not provoke an AoO. And chains represented a link in their training. It made sense that, for you to learn one technique, you had to have mastered another first. And benefits were kept small to make choices tough, and therefore more interesting, while avoiding too much inflation

    My ideal for feats would look alot like the original (pre-errata) 3E system, minus a few like Weapon Specialization, and no feats at CharGen. Choosing them during level advancement's fine. Better yet, I'd like get rid of the generic "slot" idea and simply have players ask to swap class abilities for those of another class or one found in a magazine or whatever. You want your ranger to only have one favored enemy, plus the paladin's warhorse? Ask the ref and they'll decide if it's a fair trade. Don't want to fiddle around with character customization? No need. Effective choices have already been made for you

    1. No. It's common game design, but not good game design. If you design something with scarcity as a core feature, then providing options that are substantially inferior to other options is exactly a trap. It wastes limited resources. It's still going on because it is a valid criticism of how 3e (and 4e, and Pathfinder) are designed and expanded upon.

    2. It's inferior for some characters, but a great option for others. Not every feat was designed for every character to take, nor should they have been. And honestly, a poor feat choice shouldn't be much of a hindrance unless you're playing in such a way that you have to optimize just to get by, which isn't a D&D I'm interested in. Your choices as to what your character actually does in a session matter far more than what your character is

    3. This sounds like a good idea, but some feats are so terrible that in actual use, it's not functional. The above named "Whirlwind Attack" as a good example.

      It requires four feats to reach, one that is terrible (Dodge) one that is of questionable utility - especially if you are specced for whirlwind attack (Combat Expertise). Of course once you have it, you can - oh probably make fewer attacks then you can make normally.

      It's not that it's useless, it's that so many other options are so superior that anyone who takes it is going to be way behind other paths. There's no situation where it's a 'great option'. It's kind of shit in pretty much every situation. That's just off the top of my head. There's other winners like "Improved swimming" (either useless or a feat tax for a water based campaign) or "Brachiation" (You can move through undergrowth at your normal speed. . . if you're 20' off the ground.)

      You're argument appears to be A) That every feat is useful for someone which is pretty clearly not true or B) Even if they are bad it shouldn't matter because if you're playing in a game where you have to be good, you wouldn't like that game.

      Look, it's not about being a mega-optimizer. It's that it is very likely that you will take a feat that will only come up once or twice in the entire campaign, and someone else will take one that is useful 30 times a session. That's not good design.

    4. No, I'm not arguing that it's OK for a feat to be crappy. I'm saying that, if a feat isn't right for you (though it may've been a good choice for someone else) it's not that big a deal because feats aren't a huge part of the game in the first place (or at least D&D as I prefer it. There are certainly lots of people who play it differently). They're just extra flavor

      I'm not going to defend Improved Swimming or Brachiation, as they aren't in my PHB and I hated most changes made to the game since the first printing (including some of the errata, like changing Skill Focus from a +2 to a +3). Whirlwind Attack sounds pretty good to me. Being totally surrounded, while not something that'll occur every encounter, is a common enough situation in Swords & Sorcery (and 3E works a lot better if you're using swarms rather than one or two big baddies), and this lets you take out every adjacent foe with your full BAB! You don't even have to knock 'em out first like you would with Cleave. And when you're taking on a couple dozen (or hundred) orcs, you can bet your ass you'll want this ability

      As far as the prerequisites go, you need to think of those less as a tax and more as an indicator of who this feat's for. Assuming those other feats are useful for some characters, it seems like a pretty sweet deal for them to also get this kind of power for only one extra slot. And I don't see what's so bad about Expertise. Not only does it give you a better (double!) bonus:penalty ratio for fighting defensively than normal, but it also gives you fine control over how big you want those penalties to be. That's a seriously huge advantage. Oh, and it stacks with the bonus from Dodge. Yeah, the +1 from the latter isn't huge, but you can switch foes each action at no cost. Honestly, these are starting to sound almost overpowered to me

    5. Sorry, that should say "modifiers" in the sentence about fine control, not "penalties"

    6. The fact that whirlwind attack sounds good to you is the problem.

      The way the game is designed and actually plays, is that by the time you get whirlwind attack, creatures of appropriate CR never will surround you.

      Yes, once or twice in the entire campaign, you might find yourself in a situation where a)all the spellcasters are out of their AOE attacks, B) you're fighting a rare enemy at a high CR where more then three of them is a likely encounter and C) the DM is a bad DM and creates a situation where all three are in your reach.

      Now you can address the problem by building a specific character to take advantage of whirlwind attack (natural race with reach that has a low ecl, monkey grip or choke up to gain reach with a weapon) but that is the problem.

      The feat is a trap. It's utility is so rare that unless you've played a lot of 3.X and realized that, no, past sixth level it will not regularly occur that you'll fight more than three to five monsters and the situation where they are all without your reach doesn't exist and you have three attacks by that time anyway Meaning the bonus is about equal to a plus three weapon for one round oh, and the spellcasters basically have to be out of spells - that someone will look at it and go "that sounds great!"

      It's not. In play, you dot get to use it. It's a trap. Meanwhile "craft arms and armor" doubles the gold spent on magical weapons! Power attack turns into an extra 100 points of damage easy a battle. How can the trap feat even compare?

    7. 3E combat's at its best when you're facing hordes of lower CR monsters which add up to a higher EL (probably still lower than the average party level, but sometimes equal, and even more rarely higher). Not once or twice in a campaign. This is a standard encounter. A single monster with a CR equal to the party's average level should be the exception rather than the rule, and Whirlwind Attack works pretty well in such a game

    8. Again, through play. Let's say you are 10th level and have whirlwind attack.

      512 orcs is not a good encounter.

      Hordes of lower level monsters _in play_ are not most effectively dealt with by hitting three of them once. Assuming you're talking about four to eight creatures of a CR 3 or 4 below the party, AoE SoD spells is the effective response. The smart move for the fighter is to hit each individual until they go down. Not hit each for a little damage. It's the same weakness that fireball has with the HD and Constitution inflated hit points.

    9. Also, not to nitpick, but the 3.x DMG explicitly notes that there are expected to be 4 encounters a day, with a CR of equal to the parties level, consuming about 20% of the parties resources per encounter.

      It not only isn't the exception, it's the stated norm, which is why the numbers are equal.

    10. No, it states that one could take on 4 encounters with a CR equal to the party's level. It later goes on to recommend a wider distribution of challenges in terms of EL, rather than CR. CR is a clear descendent of the old HD+askterisks model and they were simply providing some context for just how challenging it was

      And no, I didn't suggest 512 orcs would be a good encounter (unless these are all lieutenants and captains). How 'bout a colony of ettins (3-5 plus 1-2 brown bears and 7-12 orcs or 9-16 goblins)?

    11. Yes, and in that case, all of the lower level opponents are trivially eliminated with a single cast of (cloudkill, black tentacles, fireball, etc.) whereas the fighter is best served by attacking single opponent to cause severe damage or take it out.

      Whirlwind attack has always been a trap feat. After years of play, it just doesn't see use and has a heavy cost for nearly no benefit, compared to other feats that end up providing hundreds of additional damage a session.

    12. All great spells when you've got quite a bit of distance, but I wouldn't go casting 'em too close to the party. They also come at the cost of the wizard having fewer utility spells at his disposal, which are more valuable in a puzzle-solving/exploration-based game. You're right, though, Whirlwind Attack would suck in PnP WoW

  8. Aren't spells feat-like? They have a progression table based on class. Could a Fighter or Thief have feats, with their own feat progression tables, in the same way Wizards have spells and spell progression?

    The main difference is spell users have tables that tell them how many times their "feats" can be manifested/used by level, and it would take some finagling to explain why a Fighter may only use his Cleave feat once per day, so some other mechanic must be added to make it work.

    1. @J. Alexander

      That is sort of what 4E did with powers. In practice, it feels a bit too homogenized (at least, for me). The schedule of feat/power acquisition could probably be more standardized, but I still prefer martial and sorcerous abilities to use different resolution mechanisms.

    2. I have comment approval turned on for old posts, so it will take me a bit but I'll get them posted.

      I agree. The design for abilities can be similar, but must be fairly unique to each class, otherwise everything becomes a homologous hodgepodge of hogwash.

    3. Mechanics should reflect the game setting. There's obviously a fundamental difference between conjuring an explosion around you and swinging a sword in a circular motion. There should be a difference in how they function and how the character learns to do it. But I can see if combat classes had some kind of pool of talents they could learn from just as mages can draw from spell lists.

  9. I enjoyed reading this article. I found the topic especially relevant to my interests as I'm currently trying to design a classless system that relies on feats for character building. When writing a reply, I realized that I had quickly written a wall of text! So, I posted my own feat design criteria at my blog: http://rhetoricstudios.com/cyrad/feat-design/

    While some feats will always be better than others, most feats in Pathfinder/D&D are simply poorly designed and the game does not benefit from having many crappy/niche feats because it makes character creation a nightmare. Many developers also seem to forget that feats should be fun! If it's not making the character more fun to play, then it doesn't deserve the ink spent to print it.


Related Posts Plugin for WordPress, Blogger...