Are there any balance issues in allowing two half-feats to be taken without the Ability Score Increase instead of a feat?

Question

Some feats such as Actor and Keen Mind provide a single Ability Score increase along with some other benefits. These feats are sometimes called half-feats (and I am using this definition for this question).

Half-feats are sometimes difficult to incorporate into a build unless you plan ahead and even then you probably need to spend at least part of your adventuring career with an odd Ability Score.

In order to somewhat ease taking these feats, I was considering including this house rule in a campaign:

When you would gain a single feat, you can instead gain two half-feats but do not receive their Ability Score Increases. This counts as taking both feats for the purpose of being able to only take each feat once.

At first glance, this seems naturally balanced because the non ability score benefits of a half-feat seem to be equivalent to a single Ability Score Increase, meaning you are still obtaining the "value" of two Ability Score Increases.

One difference I notice is that this is a bit better for half-feats where the Ability Score improved is fixed (such as CHA for Actor). If you spend 1 ASI on 2 half-feats and then spend the next ASI to increase two ability scores, you've effectively taken two half-feats but gained the ability to reassign their associated ability score increases however you like¹. I am okay with this benefit.

Am I overlooking any other benefits or balance issues that this house rule might introduce?

---

^{1. As written by Ryan Thompson who helped clarify this in comments.}

Answer 1

Generally, this should be fine, but is strong.

You're speeding up the rate at which characters can get certain features and allow them to really spread out their utility.

A Warlock could get up to heavy armor at level 4 (level 1 with a variant human) by taking the half feats for moderately armored and heavily armored (which could be countered by not allowing prerequisite half feats to be paired).

A Dragonborn could get both Dragon Fear and Dragon Hide benefits, should they desire.

It's hard to quantitatively compare the utility effects of two of these "half feats" vs just a +1.

Watch out for Fighters and Rogues

Fighters get enough ASIs that they could max out their primary stat and then get 10 half feats; Rogues could do 8 half feats. Again, it's hard to say what mechanical effect that would have and I'm not sure why someone would go for that, but there it is.

Fighter with 8 half feats:

• Actor
  • pass as another person
  • mimic voices and sounds
• Observant
  • read lips
  • increased passive perception and investigation
• Tavern Brawler
  • proficiency with improvised weapons
  • increased unarmed strike damage
  • more grapple opportunities
• Athlete
  • easier standing
  • easier climbing
  • easier jumping
• Durable
  • more HP from hit dice
• Keen Mind
  • You always know which way is north
  • You always know the number of hours left before the next sunrise or sunset
  • You can accurately recall anything you have seen or heard within the past month
• Linguist
  • learn additional languages
  • create ciphers
• Heavy Armor Master
  • take less damage while wearing armor

---

Addendum: Racial Feats

• Dragon Fear
  • frighten with breath weapon instead of damage
• Dragon Hide
  • natural armor
  • natural weapons
• Dwarven Fortitude
  • spend hit die when dodging
• Squat Nimbleness
  • walk faster
  • skill proficiency
  • advantage to escape grapple
• Fey Teleportation
  • learn Sylvan
  • learn misty step
• Elven Accuracy
  • reroll advantage die
• Fade Away
  • turn invisible
• Orcish Fury
  • extra critical damage
  • reaction attack
• Second Chance
  • force a creature to reroll

Answer 2

This would lead to extra min-maxing, and can be slightly overpowered.

The balance of "half" feats assumes that if one feat grants a special ability and +1 to an ability score, and that this costs the same as +1 to two ability scores, that we can mathematically reduce the value of special ability to +1 to an ability score, or half a feat.

However, this comparison is flawed because feats give a fixed ability score increment, whereas the alternative is two freely-chosen ability score increments that can be assigned optimally. For example:

• Take a cleric with Str 14, Dex 10, Con 13, Int 11, Wis 15, Cha 13.

  • If they take Heavily Armored, the +1 Strength does nothing for them at this point because their ability score modifier does not increase.
  • If they take Keen Mind, the +1 Int will do relatively little because a Cleric does not rely on Int.
  • If they take half-feats of Heavily Armored and Keen Mind, they gain two significant abilities in one level, and sacrifice only ability score increments that are useless to them. This is clearly a superior choice to either feat.

• Take a high-level fighter with Str 20, Dex 14, Con 15, Int 10, Wis 11, Cha 10.

  • If they take Heavy Armor Master, the strength increase does nothing because they've hit the cap.
  • If they take Keen Mind, the +1 Int is of little use.
  • If they take half-feats of both, they gain both abilities. This is a clearly superior choice.
  • Had they spent freely on ability scores, they could have had Con+1 Wis +1. Therefore, two freely-chosen ability scores make this character more powerful than the Str+1 or Int+1 Int mandated by the feats.

Another issue is that you significantly reduce the cost of armor proficiency feats. It may even be possible for a character to take two ranks of armor proficiency at once, which would let e.g. a variant human wizard start with medium proficiency right away, which is not how it's intended.

You also radically decrease the price of Observant, which is already an exceptionally good feat. If I didn't benefit from another point in either Int or Wis, I can now take Observant and another feat. Or, I can take an Armor Proficiency feat and throw in Observant for free.

Allowing half-feats may also allow the character to take a larger number of unique abilities than expected. Increased versatility is valuable to a character.

Yet another issue is that it impedes the thematic meaning of feats. You can now have a character able to wear heavy armor despite being a complete weakling who has not physically gained any strength on his adventures. Historically, there are real-world records of knights would would train for strength and physical fitness while wearing plate armor in order to move around in it well. Letting you get good at armor without improving your fitness suggests it is just a matter of technique, which doesn't make sense thematically.

And an important but subtle balancing factor in, say, the armor proficiency feats is that since a wizard doesn't rely on Strength, it's significantly sub-optimal for him to buy his way up the armor tree because the Strength bonuses won't benefit him very much. This reinforces the important connection between heavy armor users and melee combat.

In short, it's not entirely game-breaking, since feats that grant +1 to an ability score are generally secondary abilities that don't raise your character class efficacy and merely add some useful additional talents. However, allowing half-feats is slightly more powerful than the default, since players will only take combinations that are more optimal than normal, and this will offer an additional avenue for min/maxing.

Answer 3

It depends on the campaign and the feats

Many campaigns are very much combat focused, so combining half-feats like this for roleplay focused feats seems like a reasonable house rule. Merging the non-ASI bits of Actor and Keen Mind seem like a very cool and fun pairing to increase the non-combat utility of a character.

However, if your campaign is heavy on RP, then this may be an unfair advantage that normally wouldn't be available.

Combining more combat/general focus feats like resilient, athlete, durable, many of the racial feats, etc can vastly increase your capability in combat to an unfair advantage. It's especially meaningful if your stats are already good and you can gain more 'feat' features by doing this without necessarily needing (or wanting...thanks bounded accuracy!) the ASI.