What's in a feat?

[Dsurion]

You only get so many feats over the course of 20 levels. In the PHB alone, we find a very large gap in the level of usefulness of the feats presented, from Stealthy (and other skill feats) and the Weapon Focus line, to things like Quicken Spell and Natural Spell.

By now, 3.5 players across the internet have collectively picked apart which of these feats are good and which are utter rubbish and not worth taking except in certain circumstances, leaving the best among them most often chosen, and the others consistently left in the dust. So what do we do with them?

I've only really seen two particular solutions:

Give Out More Feats

This usually leads to either just a flat increase in the amount of feats one gets, like in Pathfinder, or a measure of designating certain feats "lesser" and making up a separate progression for them for everyone to take and not waste feat slots on "greater" feats.

Redesign Feats Altogether

FrankTrollman and K have the distinguished trait of being either hailed or completely reviled in most places on the internet (with little in between), but most people tend to agree that their work on feats was at least innovative, making feats scale with BAB or Skill Ranks.

My questions are:

What is a feat really worth? (I realize this one will be subjective, based on particulars, but I mean in the more general sense.)

Which feats would work better as Traits in a Trait system?

Are feats that give only numerical bonuses ever worth it (aside of those that are prerequisites)?

Are feat trees good for the game?

And ultimately, what is it you feel a feat should do?

[SowZ]

I wish feats didn't feat trees but turned into better feats over time. (Feats that are really good or generally improve spellcasting or naturally get more powerful as your character improves can be stand alones.) So Cleave eventually becomes Greater Cleave eventually becomes Whirlwind. Weapon Focus eventually becomes Greater Weapon Focus.

So if you get a feat, the next step in the tree is given automatically. I would have no problem with options. As in, feat A. can turn into feat B. OR feat C. This could increase the playability of the martial classes quite a bit and make it so things like fighter could be played without uber levels of optimization. (Isn't it strange that the classes with the least complexity and have mechanics that require little to no understanding of deep gameplay issues are supposed to be given to the most experienced players to build right?)

There would have to be more feats total to do this.

[DeAnno]

Are feats that give only numerical bonuses ever worth it (aside of those that are prerequisites)?

Are feat trees good for the game?

And ultimately, what is it you feel a feat should do?

I think on the point of numerical bonuses, they can easily be worth it, they're just usually underperformers. For example, Power Attack, Leap Attack, and Shock Trooper all provide numerical bonuses (to damage and hitting respectively) and are quite strong.

If Stealthy for example gave +10 to Hide and Move Silently I think it might attract some more takers (perhaps not in *highly* optimized games, but still).

As for feat trees, I think they have value only when the second feat enables you to do something new or interesting with the first feat. For example I dont think Combat Expertise -> Improved Trip is good mechanics, but I think Power Attack -> Leap Attack is good. "Scaling bonus" feat trees are particularly weak, such as the Weapon Focus/Specialization line, and I think the scaling effects should probably be written into one feat.

[Eurus]

A lot of people dislike feats that simply provide numerical bonuses, myself included. I think that ideally, feats should give options. Power Attack, Empower Spell, Flyby Attack; those are feats that are actually interesting, because they allow choices and strategy during combat. Heck, Dodge would be another good example if it wasn't so weak.

[Jude_H]

I really don't understand why every feat should be useful, but I'll throw in my two cents on the other stuff.

As feats are, they are game-defining - they provide characters with explicit choices and options in a game where the default assumption is that a character can't act without explicit rules.

Feats that give numerical bonuses can be worth taking. Law Devotion has a huge impact on play, Improved Initiative means extra actions (and that's an extremely significant thing), Craven adds tons of damage and Improved Toughness provides a healthy buffer between survival and death. I also hate playing an archer without Precise Shot, but I know opinions vary on that one.

Feat trees exaggerate one of the worst aspects of the game: entangled prerequisite chains and metagamey character directions. These bribe players to spend time outside the game planning characters' developmental courses, which can be kind of fun for its own sake, but is just obnoxious to actually impose. If a feat chain is thematically similar, it makes just as much sense to combine the abilities into a scaling feat as a series, IMO.

Ultimately, I think feats do too much. I think they should be boring numerical bonuses, but I also think 3e's base "No, unless..." assumption is backward. I think the options to "Try pushing someone," to "Try to recklessly whack somebody" or to "Invest extra power into a spell for added effect" should be freely available, and feat abilities should only increase those options that should already be there.

[Big Fau]

Feats that lack feat prerequisites really ought to do 2 things:


Provide a scaling bonus to an ability you can use in a wide number of situations. (Frank&K's works have a ton of these that are good inspiration)
Provide a unique effect that you could not do before. (The Devotion feats are a good example)


For the second one, these abilities should not be like a feat tax (Adaptive Style for Swordsages, Natural Spell for Druids), as I really consider that bad feat design.

Feats that have other feats as requirements should:


Make using the prerequisite feat easier or make the prerequisite feat more powerful. (Shock Trooper)
Provide a useful effect that remains relevant in the high levels. (Stormguard Warrior)

[Eldariel]

Give Out More Feats

This usually leads to either just a flat increase in the amount of feats one gets, like in Pathfinder, or a measure of designating certain feats "lesser" and making up a separate progression for them for everyone to take and not waste feat slots on "greater" feats.

I think regardless of what you do, you should give out more feats. If feats are as character defining as they are (and should be), having to wait multiple levels for one is simply stupid; this practically means level 1 characters only do one thing, level 2 characters two things, level 6s 3 things and so on (outside class kit, anyways); which really doesn't flow well with the world building and requires way more high ECL characters than the world should contain.

[Coidzor]

Are feat trees good for the game?

They're theoretically neutral, but I find too often "feat trees" become "feat taxes" to "balance" getting an actual benefit from the end of the "feat tree" by having the earlier feats in it be absolute crap for a mediocre final benefit if that.

So I would have to say that too often they're a trap for the inexperienced and complacent to be good for the game or have done good for the game in the past.

[Divide by Zero]

A lot of people dislike feats that simply provide numerical bonuses, myself included. I think that ideally, feats should give options. Power Attack, Empower Spell, Flyby Attack; those are feats that are actually interesting, because they allow choices and strategy during combat. Heck, Dodge would be another good example if it wasn't so weak.

This is my opinion as well. No matter how good a +X to Y feat may be, it's boring, so I never take such feats except as prerequisites.

In fact, making feats give large numerical bonuses is worse, if anything. It means you pretty much have to take it to be relevant, and it still doesn't actually add anything new to your character.

[Groverfield]

This reminds me of something I'd been working on for an epic game system, where you ceased advancement after level 30 (or maybe that's where the game just starts.) In addition to the weath advancement, players are rewarded with Feats(Abbreviation for Feature) for Feats (Abbreviation for Feat of Strength).

For example, an archer hits a threatening target in the 10th range increment, he's rewarded with an epic feat that must be related to archery. My only fear is that the high rollers will advance a lot quicker than those less fortunate with dice.

[Hazzardevil]

With two-weapon fighting, (A very good example of a feat tree) you basically pay a lot to gain little overall benefit.

Because if theres 2 charecters with 18 str, one with a greatsword, 1 with twin shortswords, theoretically they will deal the same damage. But the problem is the twin shortswords guy is less likely to pull off both of those attacks.
Now folding the standard tree into 1 feat and then making anything else in the tree that scales, (IE: Two-weapon defense) then suddenly two-weapon fighting becomes a much more viable option.

[ffone]

This reminds me of something I'd been working on for an epic game system, where you ceased advancement after level 30 (or maybe that's where the game just starts.) In addition to the weath advancement, players are rewarded with Feats(Abbreviation for Feature) for Feats (Abbreviation for Feat of Strength).

For example, an archer hits a threatening target in the 10th range increment, he's rewarded with an epic feat that must be related to archery. My only fear is that the high rollers will advance a lot quicker than those less fortunate with dice.

You could call it "E30".

[Tenno Seremel]

The problem with flat bonuses is also fluff. Weapon focus – I swing my sword a bit more precise than you vs Up The Walls – I'm Spider-Man now.

[DeltaEmil]

The problem is that there are feats that give a small bonus or slightly change an effect to have a small added or different effect, like toughness, the saving throw-modifiers, skill bonus, weapon focus and specialization, dodge, mobility, perhaps improved summoning and such stuff.

And then there are feats that are practically class features, as they do actually define the abilities of the character. Those are metapsionic/magic feats, power attack, improved trip, martial study, the devotion feats from Complete Champion, and other far more radical changes.

The second category is practically almost better, because it drastically improves your abilities or gives you new options no matter if its 3.0, 3.5 or Pathfinder and any other d20-system that uses feats. In fact, without those feats, your character is useless.

4th edition went the way that most feats are rather in the first category (which do give quite strong bonus) and a few of the second category that aren't must-haves.

You must decide if you want a game where feats should be in the first or second category.

Seeing as classes in 3rd edition should have class features, feats are better if they're in the first category too. The fighter of course will still remain a suck-tasticly designed class.

[ericgrau]

The thing is you could give away 17 feats like stealthy and it wouldn't make a difference, it'd still be useless. At the same time I heard of a DM where every last one of his players too skill focus(swim) because he liked water hazards so much.

I believe the solution is simply to categorize feats as generally useful or situationally useful. There may be some confusion where to draw the line, but all you have to do is decide which feats you don't want people to even look at unless they're going for something specific. Ok, bunch those together and bam, you've saved headaches for a million new players.

Here I'll do it: All the skill feats, all proficiencies, endurance, eschew materials, great cleave, run, spell mastery, toughness.

Yes, I left out tons of borderline feats, heck all the core feats won't hold up to a lot of splatbooks, and even the feats I listed can be useful in very specific situations like I said. I think keeping these feats but placing them in their own category will be enough to save many new players a bit of trouble. Beyond that I wouldn't give these out in bunches or even for free. Like I said they're either utterly useless and 50 of them won't make a bit of difference or in special circumstances they are worth every penny.

[SowZ]

They're theoretically neutral, but I find too often "feat trees" become "feat taxes" to "balance" getting an actual benefit from the end of the "feat tree" by having the earlier feats in it be absolute crap for a mediocre final benefit if that.

So I would have to say that too often they're a trap for the inexperienced and complacent to be good for the game or have done good for the game in the past.

Of course they are neutral! Very few feats have alignment restrictions. (;

[NecroRick]

From a game design point of view there are numerous problems.

Here are the most egregious:

Feat trees are good in concept, but poor in execution. In general a feat tree with two levels should give you a net combination of powers that is as powerful as any two other feats, but not more powerful than that. In other words, if you specialise you should be good at your area of specialty, but not so good that everyone has to specialise.

Feats should be something special, but the designers give them out like candy. Characters who can have 4 or even 5 or 6 feats at first level are essentially playing a different game from those who only get one (yeah, but you got the ability to see in the dark for 15' as compensation. Whoopy frickin' dooh).

To fix the above point, kill flaws, and replace the human (and strong heart halflings) racial bonus feat, and replace it with an extra skill point per level (net of 2 for humans) or really hairy feat - er I mean feet for the halflings.

As various people have noted, fixed numerical bonuses are a trap. They look good early on when your Artificer really needs those extra points of UMD simply to activate a wand (more on this in a second) but later on they are a waste of space. The classic example of this one (Weapon Focus) is actually not as bad as people think (on paper) because while your pimped out Barbarian might have +25 to hit - and the +1 from your sucky feat is hence only 4% of his total, if the number you need to hit at +24 is a 20 then having Weapon Focus actually doubles your damage output. On the other hand, anything that is that hard to hit or with a sufficiently high DR you are better off defeating through your plan B (for Batman). So it does suck after all, because the only time it is worth it is when you are so bad that you need umpty-tillion goes at it.

The various DCs are inconsistent (and retarded). Take using a wand (for instance) it has a DC of 20. The same difficulty as making a masterwork item. Does that even make sense? So we see that at low levels some skills don't 'perform as advertised', whereas at higher levels they never fail. Or people take only exactly what they need in order to pass it when they take 10. But in those cases they have automatic success, whereas the person who falls 1 point short of the take 10 DC, and needs to roll an 11 or better fails 50% of the time. There is a massive difference, and no gradiation between the two. One point either way is the difference between sucktitude and genius.

Or, to put it another way, the skill system is borked. Hence feats based on skills are borked.

Not enough attention to detail when assigning pre-requisites to feats. Most of the fighter feats are actually pretty well specced out for pre-reqs, but the stuff that makes casters rock is laughably easy to get. This is an example of the game designers being scared of fighters but rolling out the buffet for the casters.

And the above also explains why most caster feats are more powerful than most fighter feats.

However, it is also too easy to game the system by combining things to get too much power. If someone can put together a melee build that will kill the dragon in one hit (on any to hit roll except a 1) then there is a problem with the system. But that leads to varieties of power level creep where winning or losing initiative equates to a boolean I win/I lose scenario.

In short, the problems with feats are the problems of 3.5 - it is hard to fix because it is like the loose thread that unravels the whole jumper.

[Coidzor]

Of course they are neutral! Very few feats have alignment restrictions. (;

*froths at the mouth, eyes roll back up in head, and something mostly unintelligible except for exalted splutters out*

[SowZ]

*froths at the mouth, eyes roll back up in head, and something mostly unintelligible except for exalted splutters out*

...

there...

is...

no...

exalted!!!

[NNescio]

*froths at the mouth, eyes roll back up in head, and something mostly unintelligible except for vile splutters out*

[NecroRick]

Also, another thing wrong with feats is that many of them are purchasable as magic items.

This means that anyone who spends their limited feat selection on those ones is taking a suckers bet. However, it is worse than that, because having the feat now might give them an advantage (e.g. overcoming the penalty for firing into melee) in the short run, but then when everyone else throws a couple of grand of gold at the problem and makes it go away their character is now bad by virtue of 'wasted' and 'useless' feats. Note - this is the exact reverse of the Frank and K solution, wherein feats get better as you go up levels. In 'normal' D&D there are some feats that don't just provide a static benefit (which diminishes proportionately in usefulness as you accumulate other modifiers), but that actually get strictly worse as you go up levels.

This puts the player into a "suck now or suck later" scenario. And these are *always* a result of bad or flawed game design.

[ericgrau]

One thing that should be pointed out is that in the d20 system hitting DC 11 with a +1 modifier is absolutely identical to hitting DC 1011 with a +1001 modifier. And a +1 untyped bonus in either case has an identical impact. And in core it's rather astounding how precisely so many things scale at 0.75-1.25 per level, with DCs that progress at a matching rate, but only if you use all the options available and not outside options. If I have a +1 to my +4 to hit the AC 13 orc it's identical to a +1 to my +31 to hit the AC 40 pit fiend.

At 16,000 gp it's a long time before I take a scabbard of keen edges over a feat. That a huge cost for a minor benefit. Maybe by level 17 I'd do it. The keen enchantment on a weapon OTOH tends to give less damage than other options. It seems more like in core they were over-conservative with feats in items so that you never grab them except at the very end. Or you can get alertness for 10,000 gp. The item becomes almost pointless. Certain feats might be available cheaper outside of core, which ya really is a slap across the face. I mean what if I made a magic item that granted permanent mind blank for 1,000 gp? Suddenly one of the better spells would become a joke; no caster would ever prepare it and everyone would laugh at how pointless the spell was.

[NecroRick]

One thing that should be pointed out is that in the d20 system hitting DC 11 with a +1 modifier is absolutely identical to hitting DC 1011 with a +1001 modifier. And a +1 untyped bonus in either case has an identical impact. And in core it's rather astounding how precisely so many things scale at 0.75-1.25 per level, with DCs that progress at a matching rate, but only if you use all the options available and not outside options. If I have a +1 to my +4 to hit the AC 13 orc it's identical to a +1 to my +31 to hit the AC 40 pit fiend.


+1 Excellent synopsis.
The problem is when you fall off the +0.75-1.25 range.

Which players will always attempt to do. How do you plan for this? You get whacky things like the trap in a level 2 module with a DC of 30. For the maxxed out character this is no problem. For the 'normal' level 2 rogue this is utterly impossible.

I think it is inherent in a system that has a flat 'bell curve' for success that it is too easy to get these odd results. Of course even with a probability curve that is more 'clumpy' you can still get oddities - like Shadowrun's target number 6 is mechanically exactly the same as a target number of 7.



At 16,000 gp it's a long time before I take a scabbard of keen edges over a feat. That a huge cost for a minor benefit. Maybe by level 17 I'd do it. The keen enchantment on a weapon OTOH tends to give less damage than other options. It seems more like in core they were over-conservative with feats in items so that you never grab them except at the very end. Or you can get alertness for 10,000 gp. The item becomes almost pointless. Certain feats might be available cheaper outside of core, which ya really is a slap across the face. I mean what if I made a magic item that granted permanent mind blank for 1,000 gp? Suddenly one of the better spells would become a joke; no caster would ever prepare it and everyone would laugh at how pointless the spell was.

With the criticals example, I could imagine that what you say is absolutely true in the general case. But then along comes some intrepid adventurer determined to break the system by overloading the critting. Now his crit range is something like 12-20 and he gets an extra attack every time he gets a crit threat... and he adds the various burst enchantments.

He might only hit on an 11+ - so the vast majority of his hits are crit threats.

[Larpus]

As others have said, the problem with numeric feats is that they're usually too weak or get weak quickly; there's nothing wrong in they existing per say, but the initial bonus needs a nice bump and it must scale with HD/BAB/Skill.

What is worth a feat? Well, basically something that changes significantly your character.

Take the metamagics for example, a single metamagic feat is enough to change considerably how your character plans his spells and their effects. Can the same be said for combat and general feats? More often than not, no, it can't.

Similarly, I dislike feat trees, not due to grouping the feats or anything, but ever so often you have to pass over some stinker feats in order to get a nice one, sometimes to the point that you truly ask yourself "is this worth 3 feat slots?", so not only I'd couple together some feats (Point Blank Shot + Precision Shot, for example) and tweak the prereqs, I don't think that any feat should be any further than 3rd in a tree.

The combat maneuver feats could also get some extra love, either by giving their bonuses to more than one maneuver from scratch or have more maneuvers added to them every X levels (so, for example, you start with Improved Trip, whenever you have 5HD, you automatically gain Improved Grapple [or some other better suitable maneuver] as well).

I've thought about changing the trees to pools, but I don't think that would make much difference and could lead to more trouble than solution, so I've abandoned that line of thought.

[NecroRick]

What is worth a feat? Well, basically something that changes significantly your character.


This is true. This is also why there should be less feats handed out (especially at first level), not more.

Feats make you special. If everyone has super-powers then no one is special anymore.



Take the metamagics for example, a single metamagic feat is enough to change considerably how your character plans his spells and their effects. Can the same be said for combat and general feats? More often than not, no, it can't.


I think most of the "stupid combat tricks" should be feats.

Grapple, Bull rush, Charge, Trip, Tumble etc. You should have to take a feat even to be able to do any of that stuff.

That will make your fighter-type different and unique. Also, people might actually use sword and board if they have to spend feats to do all the stuff they're doing instead of shield-bashing.

You could trial these two melee powers:

Empower: use as part of a full attack. One of your iterative attacks does 50% more damage, but you do one less attack.

Maximise: use as part of a full attack. Your iterative attacks for this round do maximum damage, but you do two fewer attacks.

----

In any case, the obvious melee feats are about doing 'moar damage'. But that isn't what makes the tier system tick. What you'd need to do is to introduce battlefield control. One of the ways of doing this would be to simply not let monsters move through the squares adjacent to you. That way you could block them from getting to "mr squishy and his invisible space hamster" in the back.

Introduce feats that encourage the battlefield to form into an actual proper battle line. That way you can talk about front row and back row combatants, whereas right now it is such a big pot of mess that those distinctions are meaningless.

[Tenno Seremel]

This is also why there should be less feats handed out (especially at first level), not more.
That makes you more generic (bland), however.

[SowZ]

This is true. This is also why there should be less feats handed out (especially at first level), not more.

Feats make you special. If everyone has super-powers then no one is special anymore.



I think most of the "stupid combat tricks" should be feats.

Grapple, Bull rush, Charge, Trip, Tumble etc. You should have to take a feat even to be able to do any of that stuff.

That will make your fighter-type different and unique. Also, people might actually use sword and board if they have to spend feats to do all the stuff they're doing instead of shield-bashing.

You could trial these two melee powers:

Empower: use as part of a full attack. One of your iterative attacks does 50% more damage, but you do one less attack.

Maximise: use as part of a full attack. Your iterative attacks for this round do maximum damage, but you do two fewer attacks.

----

In any case, the obvious melee feats are about doing 'moar damage'. But that isn't what makes the tier system tick. What you'd need to do is to introduce battlefield control. One of the ways of doing this would be to simply not let monsters move through the squares adjacent to you. That way you could block them from getting to "mr squishy and his invisible space hamster" in the back.

Introduce feats that encourage the battlefield to form into an actual proper battle line. That way you can talk about front row and back row combatants, whereas right now it is such a big pot of mess that those distinctions are meaningless.

But then if you haven't taken the feat, you can never trip anyone in combat for your entire life. It is something anyone could reasonably attempt. Also, this takes us to the scenario where a Minotaur wearing full plate with 26 Strength cannot ever knock down a 6 strength halfling wizard.

In a video game, (or even 4e,) you get special abilities like this as powers but can't attempt them earlier. I don't mind this in video games. You are expected to follow a fairly linear path. But in D&D you can attempt anything. There is no reason reason why anyone couldn't try and trip someone in battle or knock them over.

And you do get an AoO if someone tries to run by you.

[Eldariel]

{{scrubbed}}

[Larpus]

This is true. This is also why there should be less feats handed out (especially at first level), not more.

Feats make you special. If everyone has super-powers then no one is special anymore.
I agree that if feats are re-balanced and actually give a sizeable bonus to your character, then yeah, there'll be less need for each character to have a bazillion feats; for example (a possibly non-balanced one, that is), if most/all feat trees are changed into "get first feat and get the others for free once X condition [not feat-dependent] is met" then hell, a character could actually fare rather well with 3 feats alone or something.

But as the system stands (and seeing how big of a job is to re-balance everything), the easiest "solution" is to give everyone more feats, especially the melee people, but doing so essentially just hurts the Fighter deeper.


I think most of the "stupid combat tricks" should be feats.

Grapple, Bull rush, Charge, Trip, Tumble etc. You should have to take a feat even to be able to do any of that stuff.

That will make your fighter-type different and unique. Also, people might actually use sword and board if they have to spend feats to do all the stuff they're doing instead of shield-bashing.

You could trial these two melee powers:

Empower: use as part of a full attack. One of your iterative attacks does 50% more damage, but you do one less attack.

Maximise: use as part of a full attack. Your iterative attacks for this round do maximum damage, but you do two fewer attacks.
I disagree here, anyone in the correct situation should be able to make a combat maneuver, however those with them feats do it better, so in order to give that flavor, I'd change the maneuver feats to add more options to what you cna do once your maneuver is done, not only "you get an AoO if your maneuver succeeds", but more:

Made the trip? Slap in a free Use Rope check and tie that bastard down!

Success on your grapple? Give the man a Str check to break that wizard's arm, yeah, let's see that pansy dress-user cast with his arms broken!

Bullrushed a foe? Hey, what about instead of pushing him back you don't have the option to toss him prone to the side so you can give your friendos some nice AoOs and even flank the poor evil-doer guy yourself?

The options are limitless! Though yeah, giving all that for the first feat is a bit too much, so I'd only slap a decent (scaling) bonus to the maneuver and have the second feat do the options (and the original thought of having a 2 for 1 deal granting the bonuses to two maneuvers instead of one); possibly left open so it's creative as opposed to restrictive (in the same vein of X Image spells, which can be killer on the right hands).

However, I believe that similarly magic people should be able to attempt metamagic feat effects without the feats themselves, but at a higher cost, such as an extra spell slot (as in, both the enhanced and the original slots) or higher level. At least some of them should, Heighten Spell especially, it has always bugged me and looks kinda out of place.


In any case, the obvious melee feats are about doing 'moar damage'. But that isn't what makes the tier system tick. What you'd need to do is to introduce battlefield control. One of the ways of doing this would be to simply not let monsters move through the squares adjacent to you. That way you could block them from getting to "mr squishy and his invisible space hamster" in the back.

Introduce feats that encourage the battlefield to form into an actual proper battle line. That way you can talk about front row and back row combatants, whereas right now it is such a big pot of mess that those distinctions are meaningless.
I agree, I see nothing wrong with "moar damage" feats, 'cus honestly, they should be there for those who want them, but they should be better than they are and/or less costly.

As for options, they are indeed limitless as I said, the only thing that I believe that people need to stop thinking is that melee = 100% realism, magic = 100% shenanigans.

Not saying that Fighters should now summon a dragon every time they strike someone with their sword, but since magic is allowed, why not some "he's so good at X he can make impossible things" magic?

Think of action movies, the guys there pull impossible maneuvers, hell in LoTR movie there are some sequences that are so absurd they would make Jason Statham and Jet Li scream "impossible sheetz!"; that's what I think when I think of "awesome mundane guy", not a guy who waves his sword around saying to himself "I'm helping" while his magic-loaded friends do all the work, a nice of example of this: the Ice Giants battle in Thor, Thor is some half-caster guy (or melee with the awesomenest equipment ever, that even Elminster would like to have), the rest are just normal melees and Loki is obviously hiding his gold.

Having the melee people have more alternatives over full-attacks, especially alternatives that lead to a full-attack next round (or from team mates), is the way to go with feats, that's what I believe anyway.