From time to time, I like to look behind the curtain – to see what makes the mechanics of the games that I play tick, and what the implications are. Sometimes this leads down unexpected byways, and at other times it yields a nugget or two of insight. And sometimes, it just goes nowhere. So: In a d20 system (whether it be D&D, Pathfinder, d20 Modern, or whatever – what are Feats?

What Are Feats?

The 3.5 PHB doesn’t define them in its glossary. Chapter 5 of the PHB describes them as “a special feature that either gives your character a new capability or improves one that he or she already has.” The PHB then goes on to define Feats in terms of the differences between Feats and Skills, and then confuses the issue by dividing Feats up into several different types, with different rules applying to each type. Bonus Feats, Class-restricted Feats, Racial Bonus Feats and Feat Slots, Special Feat Lists… It’s a very flexible game mechanic, and it’s been used in all sorts of different ways as a result.

The Pathfinder SRD is even more vague: “A Feat is an ability a creature has mastered. Feats often allow creatures to circumvent rules or restrictions. Creatures receive a number of feats based off their Hit Dice, but some classes and other abilities grant bonus feats.”

One website defined them in terms of character options, used to customize a character. Another speaks of them as a metagame mechanic used to alter the way a character interacts with the rules. A third suggests that they are a way to change the rules of the game as characters become more powerful. A fourth describes them as a way to differentiate the capabilities of representatives of the same race/class combination at a metagame level. Still another talks of evolving characters from a generic common standard to a customized state that is more tightly integrated with the campaign world. And a sixth describes them as a way to give characters a bonus beyond what the normal character gets.

A fellow GM I was chatting to about this a few months back described them as “a way to customize classes or races as a means of adjusting game balance between these conceptual entities”.

And one of my players talks about them in terms of restoring the balance between humans and non-humans with racial abilities, and between Fighters and Mages.

These definitions run the gamut from the hypothetical to the min-maxing character crunch, from the simulationist to the pure roleplaying, from the campaign perspective to the metagame. And in any given campaign, any or all of them may be true – and there are some serious implications and repercussions buried beneath the surfaces of some of them.

Innate vs. Learned

One of the more interesting ideas that I came across in researching this article was the suggestion that all feats represented an innate natural skill or talent while class abilities were all things that the character learned, or learned to do, in the course of their professional development. I’m not entirely sure that I buy the notion, but it certainly raised an interesting question for consideration: is a Feat something that you learn or something that you can do? Or do Feats encompass both? And is that simply because later writers didn’t understand what a Feat was supposed to be any more than I do? In other words, has the original concept been contaminated – and if so, should feats that violate the definition that we arrive at be banned from the game?

It didn’t take very long to get into radical, even controversial, territory, did it?!

Because we have not defined exactly what a Feat is, no answer to these questions is possible. They are just something for us to keep in mind as we examine possible definitions.

If this were the purpose of Feats, progression would be as shown on the left - not as on the right

Balancing Acts

The notion that Feats are a means of tweaking the game balance between races and character classes doesn’t hold water, in my book – though this can be a secondary usage of merit. If this was the concept, only Humans and Fighters would receive feats, and they are not so limited. Of course, it’s possible that this was the original intention, and that the designers decided to raise the bottom line from zero to the default HD-based allocation mechanism.

I’m afraid that this theory doesn’t match up to reality. If it were correct in terms of race, then we disregard the standard Feat every X levels and look only at what the two factors contribute. Can anyone seriously argue that one Bonus Feat is enough to counterbalance all the racial advantages and abilities that Elves receive, or Dwarves? And, if it was correct in terms of class balance, then the receipt of Feats would be more in line with the geometric increases in power that Mages experience instead of a bonus feat every X levels. Either the workmanship is slipshod, and no-one’s ever noticed – yeah, right – or the actual distribution of Feats doesn’t match either pattern.

Nevertheless, the fact remains that there IS a bonus Feat for humans, and there ARE regular bonus Feats for Fighters. So these are either manipulations of the original intent, or the actual definition of what a Feat is must encompass this usage.

Character Options

The notion that Feats are a way of customizing characters by presenting them with a palette of choices and options is one that has played no small part in my approach to the subject, and one that several of the tentative definitions offered by different websites and GMs also touch on. But in order for this definition to work properly, it has to be assumed that all, or almost all, feats are of roughly equal value, and also that there are an arbitrarily large number of Feats for any given character to draw on.

The first condition is problematic. There is a standard employed in setting the effectiveness of Feats in the PHB, though it is one that I have inferred from the details instead of something that has been explicitly stated. That standard is:

  • +4 to one skill or roll
  • +2 to two related skills
  • +1 to four related skills
  • +2 to a type of saving throw
  • +1 to a combat-related numeric value eg Critical threat range
  • An ability that is normally useful once per round in combat
  • A more powerful combat ability that is only useful under specific conditions or is otherwise constrained

Metamagics, of course, fit into the last category, and introduce a sub-mechanic to the game – level adjustment – that is designed to contain the relative effectiveness of the Feat to something close to the appropriate standard. Unfortunately, not everyone has followed this standard – possibly due to poor analysis of the standard, or error – and there are some outright violators out there, some of them in “official” WOTC publications. By and large, though, the worst offenders tend to be home-brewed feats.

Side-Note: Class Abilities
It’s worth noting that most class abilities also meet this standard, but there are some even more outrageous exceptions. Possibly the worst offender is “evasion” and “improved evasion” which can make a character all-but-immune to damage-dealing magic regardless of the circumstances.

Where such magic has to target the character, evasion seems fair enough, but when it comes to Fireballs and other area-effect spells the logic starts to get shaky. Add to that the fact that the check involved is one in which those characters with access to these abilities are naturally good, and that the ability is absolute and not relative to the class levels of the characters involved, and it becomes a right mess.

Evasion and it’s improved counterpart virtually force the GM into the use of “Save-Or-Die” spells, which I loathe.

Fixing “Evasion”

“Evasion” and “Improved Evasion” are not that difficult to fix.

  • First, the DC for an evasion check should be increased by double the Spell Level, to make it a little bit harder to save against, especially with higher-level spells;
  • “Evasion” should be amended to read “Half the damage otherwise indicated on a successful Evasion save, Full damage otherwise”; and
  • “Improved Evasion” should be amended to read “One Quarter the damage otherwise indicated on a successful Evasion save, Half damage otherwise”.

These changes have the effect of permitting these abilities to continue making a major difference to the character’s capacity for surviving such spells while not making them a complete “get out of jail free” card. It also establishes a differential in which high-level spells are harder to Evade than low-level spells.

Finally, I would rule that using evasion leaves the character prone, requiring them to spend a move action to get back to their feet – unless they have an appropriate feat to let them do so more quickly.


Side-Note: Class Abilities and Feats
Another pet peeve that is also relevant to this discussion is the example of character customization offered by the DMG (Sidebar p175). Many players have interpreted this passage to mean that if the character has a class ability that is identical in description to a Feat, that class ability can be swapped out for a different feat with automatic approval by both the rules and the DM. Those who customarily wear medium or heavy armor are, according to these players, better off swapping out all armor-use feats bar the category of the armor they are actually using.

In theory, this is fine, but in practice, it plays hob with low-level game balance by giving characters as many as 4 additional feats (3 armor types and shield use). The result, when applied to archery-oriented characters or dual-weapon characters, is a hugely disproportionate capacity for inflicting meyhem.

Fixing Armor Proficiencies

This peeve is also susceptible to an easy fix. Simply designate Light Armor Proficiency as a prerequisite for Medium Armor Proficiency and Medium Armor Proficiency as a prerequisite for Heavy Armor Proficiency.

Further refinements are possible, such as:

  • Ruling that characters cannot fail to take an armor Proficiency when it is offered and then take it again at a later point. In the long run, this would make no difference; but at low levels in a campaign, the effects can be considerable. And/or,
  • Ruling that Magical Armor counts as one Proficiency Type less unless used with a shield, i.e. no armor proficiency is needed to use Magical Light Armor, Light Armor Proficiency is sufficient to permit use of Magical Medium Armor, and Medium Armor Proficiency is sufficient to permit use of Magical Heavy Armor – unless the character wants to use a shield with this armor, in which case they not only need the Shield Proficiency, they need the correct Armor Proficiency.

These changes not only expand the available choices of armor for a character (a little honey to make the restrictions more palatable), they restrict the benefit that can be achieved by gainsaying proficiencies. The character is forced to choose between being restricted to a lighter armor permanently and having more feats, or having access to the heavier armor types and the full range of protections, with only the standard number of feats. There’s even a middle ground, for those who like to compromise!

They also make the nature of an encounter less prone to telegraphing by means of the armor being worn. That guy in scale mail – is he a poorly-equipped fighter, a fighter wearing some fantastical enchanted armor, or a mage or rogue in enchanted armor? The beefy guy next to him in full plate – is he a Fighter, a Paladin, or a Cleric? If it’s not quite so obvious what character class an NPC is, it is also not quite so obvious what his vulnerabilities are, or what threat he poses!


Side-Note: Feats from multiple sources

A third (relevant) pet peeve that I might as well get off my chest while I’m talking about them is the assumption that players can draw feats from any published, compatible, sourcebook without GM approval, and that two different feats that do the same thing but have different names can automatically stack unless the bonus is of a Named Type.

This opens up all sorts of game-unbalancing possibilities. Two feats that are perfectly satisfactory in isolation can combine to create and exploit a rules loophole through which all sorts of game-unbalancing effects can crawl.

In 99% of cases, there is no problem, but that last percent – which min-maxers always seem to locate – annoys the heck out of me.

Fixing Multisource Feat Problems

Thankfully, yet again, this is easy to fix.

  • Any Feat that affects a given subsection of the rules, e.g. Flanking or Charge Maneuvers, is deemed to be mutually exclusive to feats from other sources that affect the same subsection of rules except with GM permission. Such permission is given on a case-by-case basis and never as a blanket ruling.
  • Any feat that confers a bonus to a given ability or score is deemed to be the same as any other feat that confers the same bonus, and therefore the benefits do not stack with that feat.
  • The “Flavor Text” that describe feats, including any personality traits, are considered rules just as much as game mechanics are; the referee is entitled to force them to be applied in roleplay, to take them into account in relations with NPCs etc, and/or to require the feat to be replaced with another if the character acquires a feat or other ability that inhibits or controls the non-game-mechanic consequences of a feat.
  • No feat or character class is permitted unless the referee also gains access to a copy of same for use with NPCs.

These four simple house rules permit the characters to utilize any game supplement that the players might have, from any source – within appropriate and reasonable limits.

Getting Back To Our Knitting

We’re still trying to figure out exactly what a feat is. I’ve been looking at the various definitions I could find, in reverse order, so far without finding a complete one. The best we have so far is “a way to customize characters”, which is at roughly a 2nd-grade level so far as definitions go!

The next one is “A way to give characters a bonus beyond what the normal character gets”.

This immediately raises the question, what is meant by the term, ‘the normal character’?

Two possible interpretations come to mind:

  • An NPC without character levels
  • A character without the feat
NPCs without character levels

This assumes that characters don’t automatically get class levels, that there is something extraordinary about those who do. The majority of NPCs encountered will be 1HD peasants, in other words, with no extraordinary capabilities whatsoever.

This interpretation actually stems from older versions of the D&D game system, especially AD&D, which stated outright that most NPCs never gain class levels. The problems with this notion are that it’s hard to set up social infrastructures for the advancement of PCs when they are so rare, and that it becomes difficult to explain where the villains come from.

If the campaign world is set up along these lines, it becomes something very different from the majority of modern campaigns. When creating the world, it becomes necessary for the GM to spell out exactly who the high-level characters are because they would be famous figures throughout the civilized world – and the same goes for the bad guys. PCs become tethered to the base of operations because that is where their character class has its resources – they can go out and adventure but must periodically return to home base to utilize those resources. If characters need to train in order to go up levels – another element that has fallen by the wayside in modern campaigns – that can only occur in locations where there is the infrastructure for such training. This is an elitist model in which the PCs are the elite.

More modern campaigns are more generalist. Removing the need to return for training before a character can go up a level permits more flowing narratives (the need to train being a constant handicap and interruption to an ongoing storyline). The result is a more “novelized” approach. It also means that many more characters have class levels, which in turn makes the integration of their class infrastructure more ubiquitous. It doesn’t completely solve the problem – for that you need something like the Shadow Levels approach that I offered in Shadow Levels: A way to roleplay the acquisition of Prestige Classes in D&D 3.x, one of the more popular articles here at Campaign Mastery.

Removing the restriction on the acquisition of class levels makes adventure easier to come by, but it makes a profound difference to the game world. It also necessitates the creation of character “classes” such as “Noble”, for those characters who don’t go out and adventure to maintain some level of authority over those who do. Without it, the NPCs can quickly be forced to dance to whatever tune the PCs call. (In my original AD&D campaign, I ensured that every nobleman had at least 10 class levels purely to justify their positions of authority – and higher rank had higher-level requirements. That campaign was the tale of the son of the King going out into the world to ‘prove himself worthy’ of his right of succession).

So, there are advantages and disadvantages to both, and a very different campaign flavor. This one assumption takes a modern game and gives it a very ‘old-school’ flavor. But the very fact that it is necessary to distinguish between the two indicates that this is not a correct base interpretation, though it is one that can viably be made in a campaign’s house rules.

That leaves us with:

NPCs without the feat

Restating the proffered definition of a feat as “A way to give characters a bonus beyond what a character without the feat gets” doesn’t seem to get us very far. It seems to be a tautology, adding up to “a bonus is an advantage over anyone who doesn’t have it”.

But there is a hidden implication there, one that removes the tautological overtone. In order to make this definition sensible, the assumption has to be made that not everyone receives everything that’s on offer.

If everyone has access to class levels, but NOT everyone has access to feats, we have a blended compromise between the ‘old school’ approach suggested in the previous section and a fully-democratized model in which everyone has at least theoretical access to everything – class levels and feats.

To be honest, I’ve never seen anyone write up such a campaign structure, in which Feats are the difference between elite (PCs & their Arch-nemeses) and ‘mundane’. But it makes a certain amount of sense, in terms of preserving the infrastructure benefits of the world and still making the PCs elite. It’s not much of an edge, especially at low levels, but it would become massive as the characters progressed.

The only problem with this approach is that I haven’t seen it done anywhere before. In fact, the opposite is true – the trend has been to make Feats more ubiquitous, not less.

Monsters gain feats just by increasing their hit dice, for example – preserving some semblance of balance between PC capabilities and the difficulty of encounters. Can it be seriously suggested as logically-consistent that Monsters can get feats and a 14th level NPC cleric can’t? I’m not saying it can’t be done, but the campaign setting would need to justify this discrepancy.

Examination of this possible definition of a Feat has led us down some interesting byways, but the very fact that standard usage – even within official publications like adventure modules – doesn’t fit the resulting models demonstrates that this is NOT the correct definition.

From Generic To Unique

The definition that I actually use is the next one to be considered: “A means of evolving characters from a generic state to a uniqueness that is more tightly integrated with the campaign world.”

Again, there are some implications and hidden nuances to this definition that are worth taking the time to explore, contained within a couple of loaded phrases in the definition.

A generic state

The suggestion here is that all characters start out being carbon-copies of every other character (other than differences in statistics). That being the case, character stats become the primary differential, in game mechanics terms, between suitability for this career vs that, without actually blocking characters from undertaking a career for which they are unsuited in ability.

The implications for roleplay are enormous. A character who is desirous by personality of a career in a particular character class – any character class – can adopt that class regardless of ability. There will be especially pious clerics with a Wisdom of six who think they have heard “the call”. There will be fighters with glass jaws. There will be rangers who couldn’t follow a ploughed furrow in the ground, and Wizards who can barely light a candle, and Thieves who can trip over their own shadows.

Naturally, few of these will progress beyond 1st level, and many may die trying – but from time to time there will be some who survive by sheer luck or by virtue of more capable companions.

What’s more, this raises the prospect of differentials between temporal authority and levels of expertise – knowing the right people, or having the right relatives, can get people promoted to positions of authority their abilities do not warrant. And of members who go ‘bad’ (which means different things when you’re talking about Rogues and Paladins, of course). The result is more akin to the real world with which we are all familiar, where nepotism and low-levels of corruption are routinely expected by the populace (whether they occur or not).

If the GM recognizes these implications, they can build a consistent world around them. Eventually, that will lead to the players becoming aware of the implications (assuming the GM doesn’t tell them outright as part of the campaign briefing) – and once they know, they can begin employing them as tools in their interactions with the society around them. For example, identifying a discrepancy between demonstrable capability and level of temporal authority implies that the position was achieved by virtue of something more than competence – something that’s useful to know when characters are seeking an avenue to political influence.

A uniqueness

To a min-maxer, there is an ‘ideal solution’ that maximizes the power of any given character class by stacking slight differences in relative effectiveness of every nuance open to the character in their favor. If every character class and feat and class ability is equal in effectiveness to every other, their constructions are no better than anyone else’s. The more players subscribe to this philosophy, the more the characters at higher levels become carbon-copies of all others within their class and character level.

Equality in diversity is naturally the enemy of min-maxers and the ally of the GM who wants the PCs and NPCs in his campaign to be more interesting than this carbon-copy approach. The more viable options, and combinations of options, that are available, the more diverse and distinctive characters become even with exactly the same race, stats, character class, and class levels.

One of the many definitions of “game balance” – or, more properly, game imbalance – can be “the capacity for successful min-maxing within the system”. A less negative definition of game balance might be “the capacity for reflecting character persona in ability options without detrimental effects to the character’s abilities relative to characters who have made different choices.”

As a general statement, the more “equality in diversity” there is, the more character construction becomes an adjunct to roleplay, as opposed to “rollplay”. Clearly, the two types of activity in synergy produce a greater effect than if they are in opposition. Ideally, you would want players to be able to identify what a character can (reportedly) do and be able to extrapolate to a personality.

In practice, that might be an unachievable ideal; but the more closely it can be achieved, the healthier a campaign will be in many respects.

So the sine-qua-non of this definition, in practical terms, and with respect to feats specifically, is the type of parity standards that were offered in “Character Options” above. Without such a standard, the capacity for inequalities exist – which undermines the potential for uniqueness by defining “must have” feats for any given character class.

A plurality of equality

The implications go further. Another is that there be many feats available for characters to choose from – at least five times as many as a single character has the capacity to receive, and the more, the better.

To some extent, this actually undermines the arguements and justifications I employed in my “pet peeves” boxed sections above. Every ability that is traded out makes a character more different from the standard, by definition. A character who has expended a feat slot doubling up on an advantage – an initiative modifier, for example – has not used that slot to gain a different advantage or ability, by definition.

So, which “Pet Peeve” solutions do I actually use?

  • Evasion/Improved Evasion
    I’d love to implement this solution but my players won’t hear of it. They argue that it is one of the few mechanisms counterbalancing the excessive power of high-level Wizards – which is true. So, until I develop some other counterbalancing influence over the spell-slingers, this is a non-starter. It’s ironic that one game-unbalancing element’s removal should be countermanded by another game-unbalancing element.
  • Exchanging Class Abilities for feats
    The only reason that this fix is not in place in my campaigns is that I hadn’t thought of it at the time! Right now, there’s a blanket ban on the practice in my campaigns – but that’s subject to change without notice if my players approve (it’s their campaign, too).
  • Open-sourcing of feats
    The actual restriction in place in my campaigns is more strict than this proposal. While I couldn’t always put my finger on the source of the problem, I felt that certain feats caused game balance issues, and so set up an approvals process† that let me approve, reject, or modify feats – and prestige clases, and spells, and so on. I have given ground (reluctantly) in the latter case – the Spell Compendium is just too convenient a resource – but in other ares, the process remains.
    At the same time, this solution is partially implemented – the “flavor text” part, to be specific. It’s actually taken quite some time to derive a general statement of what I look for in that approvals process. I’d love to introduce this fix in its entirety, but it’s probably too late for the current campaigns.
    Besides, if there is one “pet peeve” that is undermined by the arguement given in “A plurality of equality” above more than any other, it’s this one.

I’ll write a separate blog post on the approvals process some other time.

Campaign Integration

The final loaded phrase in this prospective definition is “more tightly integrated with the campaign”. What does that mean? In practice, it means that some feats can be designated “only available to race X” or to “Class X” or to “characters of level X or more”, or combinations thereof – in other words, manipulating and extending the requirements list for a feat to suit the particular campaign. It can also permit certain feats to be made available for free to all members of Race X or Class X – sometimes in addition to, and sometimes as a replacement for, racial or class abilities.

You’re really only limited in this area by the amount of time and effort you can put into the campaign ahead of time (it’s generally too late once play starts).


I refer to this practice as “Counter-skinning” because it really is the opposite to the technique of “skinning” one race or monster to create another whose capabilities just happen to match.

It’s when you combine the two that you really develop a powerful tool. “Monster X is exactly the same as monster Y except…

I employed this technique extensively to unique (additional) skills for the different races in the House Rules for my “Shards Of Divinity” campaign – another series of blog posts to be presented in the future – and it worked a treat.


The next potential definition of a feat to be examined – “differentiation between stock examples of a given race/class combination” – has turned out to be just another way of stating the same thing as the previous one, but one with fewer tools for the GM to employ. If the GM employs no manipulations – no Counter-skinning – that affect PC races, the “campaign integration” phrase of the preceding definition goes away and we are left with what is essentially a recapitulation of this definition. So it’s a functional definition, but one that’s less useful than the more verbose one already examined.

Changing Rules with Power Levels

This implies a greater structure to feats than is actually the case, though some feat dependency chains lend it an air of plausibility. Unfortunately, these dependencies are too haphazard for this definition to be correct in a general sense.

Could a more formalized, structured, hierarchy of feats be developed? Of course. Is there any benefit in doing so that outweighs the expense in time and effort of doing so? Are there any lurking downsides?

The potential upside is in fighting min-maxing fire with fire – because that’s what we’re really talking about, here. The downside is the usual one that comes with rampant min-maxing: cookie-cutter assembly-line characters become ubiquitous, the common standard.

This is pandering to the min-maxing crowd – either a GM who has reached his wits’ end and decided “if you can’t fight ’em, join ’em”, or a GM who thinks this is the way the game is supposed to be. I’m sure there are some out there who fall into the latter category.

Unfortunately, simply because the GM is restricted in development time and scatters his efforts over many different characters, he can never compete with the focused (almost obsessive) attention lavished on their characters by the dedicated min-maxer. The GM is on a hiding to nothing, to use an apt Australian expression. (Actually, the phrase comes from the UK but it has fallen into relative disuse there, so far as I can tell, while it remains a common part of the ‘Ocker’ parlance).

The problem is that min-maxing – or “power gaming” to put a more friendly face on the practice – is fairly addictive, something that everyone falls prey to now and then. Weaning players off it can be exceptionally difficult, if not completely impossible. The original premise behind The Knights Of The Dinner Table is that most of the players are incurable power gamers – much to the frustration of the GM and the representative of the genuine roleplayers at their gaming table.

The final definitions

The last couple of definitions to consider don’t actually tell us much more than those already examined – “a metagame mechanic used to alter the interaction between character and rules” (which implies that Feats aren’t part of the rules) and “Character options used to customize a character” – which ignores that Feats are available to more than just characters.

So, where does that leave us?

A feat can be many different things, in many different campaigns – and subtle changes to the definition can have massive effects on the underlying precepts of a campaign. There IS no one “right” answer; instead, we have a tool for manipulating campaign and game elements at an almost primal level, if carried through to their logical conclusions.

Make the choice that’s right for the campaign you and your players want to play, and you strengthen that desired style’s hold on the campaign. Make the choice that’s wrong and you’ll be fighting the game system all the way – and wondering why you can’t get it to work for you.

You can even change from one definition to another to reflect some subtle but fundamental change in the game world, changing the tone and texture without your players being able to put their finger on exactly how you’ve worked the magic.

Have fun…

Related Posts with Thumbnails
Print Friendly