I recently made a
thread that covered a number of pain points I've experienced when playing the full release of BG3, many of which included suggestions for spells and abilities. For visibility, I'm also posting the ones specifically related to spells here since it's stickied.
Minor Illusion:
Bit of a shame that it's only useful as a distraction, but the real issue for me is that there's no way for me to end the effect early. I've had a number of occasions where I use it to benefit from it for a round or two, but then need it gone. As it stands, I have to just wait until it vanishes, which sometimes causes issues with whatever I'm trying to do.
Mage Hand:
Both the EA and full release versions of this spell bug me, for different reasons. In EA the fact that it could be cast as many times as I wanted as long as I could afford to concentrate on it meant I was able to use it in early levels since I didn't need concentration for most of my spells, and it helped in combat a bit. Problem was that once I started really needing concentration for other spells, I stopped using it almost altogether. In full release, it's a different but similar problem. Now it doesn't require concentration, but you can only cast it once every short rest. This caused me to never use the cantrip, even at low levels, because now I feel the need to keep it on hand (hah) in the event I need its utility to grab a trinket or something. Outside of those niche instances, it's basically entirely forgotten about. IMO, there's no need to balance how frequently a player can use it, as the power given to it by its implementation in BG3 is roughly equivalent to that of 5e. Essentially, it traded out-of-combat utility for in-combat utility. Even if I could cast it endlessly, I don't think I'd always want to use my action that way.
Arcane Trickster Rogue's Mage Hand Ledgerdemain:
This trait completely lies to the player, and does not reflect how it's supposed to feel in the TTRPG. For Arcane Tricksters, Mage Hand Ledgerdemain is a
core aspect of the class. Not only does it really need to be able to be cast as much as the rogue wants, it also needs to perform as advertised and honestly stick to the source material better. Its invisibility type should be Greater Invisibility, so just as in the TTRPG it always stays invisible if the rogue chooses to make it so. It also doesn't currently do much of anything other than what the normal Mage Hand does, despite what the traits claims. The only extra utility I've seen from it is being able to use it to lockpick or disarm traps at a distance, but unlike in the TTRPG doing so uses the Mage Hand's (terrible) stats rather than the rogue's Dex and Sleight of Hand proficiency/expertise, making this utility next to worthless.
Arcane Lock:
This is already a niche spell in the TTRPG. The way it's implemented in BG3 makes it even more so, for not really any good reason. It needs to last much longer (e.g. 100 rounds), and also when initially cast prompt the user for what kind of allow list they want. At the very least the player should be able to choose between "nobody passes through" and "only party members can pass through".
I feel it's pretty telling that this is a terrible spell as-is when absolutely no NPCs ever make use of it, and the scroll is hard to come by (as is the same with Knock).
Knock:
This is another terribly implemented spell. In the TTRPG this is meant to be a counter to Arcane Lock, following the general trend that spells with persistent effects have an opposing spell that can directly counter them (e.g. any cloud spells vs wind spells). This needs to be able to counter Arcane Lock, and it should also be bad for stealth if it isn't already, as it's supposed to be very loud. As-is, it's never really worth a 2nd level slot to make use of this since individual locked containers don't commonly have loot that's worth the spell slot cost. The value of this spell also hits 0 if you have a Rogue or similar in the party, because they can basically do this for free all the time.
Find Familiar (all variants):
Something I feel is missing from the Find Familiar spells is the ability to have your familiars deliver items and throw lightweight stuff. Having a tiny inventory and very low carrying capacity but the ability to toss water bottles and healing potions short distances for certain forms would be a welcome addition, as would being able to fit through tiny holes (currently restricted to Druids) to grab items or whatever. These would also fit the flavor that they have in the TTRPG. I realize this would be a bit odd to balance, but it would go a long way to making familiars feel more involved. As they're currently implemented, familiars feels pretty useless as anything but a small meat shield in combat, and they have limited utility outside of combat as well.
Starving for Rituals:
There appears to be very few rituals in the game, which causes it to easily be an overlooked mechanic, and also turns Ritual Caster into a feat with questionable value. Would be nice to see more rituals implemented, or more existing utility spells converted into rituals.
Polymorph Effects Shouldn't Prevent Ability Activation Circle of the Moon Druids are the biggest offenders here. In the TTRPG, the way you're generally meant to play a Moon Druid is to use a mixture of their Wild Shape and their magic simultaneously. A first turn for a Moon Druid generally looks like using their Action to cast a concentration spell (e.g. Flaming Sphere) and then use their Bonus Action to Wild Shape, allowing them to continue concentrating on that spell. BG3 sort-of allows this, but with a catch: if the concentration spell can be re-activated using an Action or Bonus Action, you cannot do so while in Wild Shape because for some reason that counts as casting a spell and in the beast form you are silenced. This prevents comboing Wild Shape with classic druid spells like Moonbeam, Heat Metal, and Call Lightning. It feels more like a developer oversight than an intentional design decision, honestly.
Transfuse Health + Warding Bond Interaction is Broken Warding Bond is supposed to halve all of the player's incoming damage so it gets shared with the person who made the ward. Transfuse Health damages the player by half of their current health and then heals their target for the same amount. The problem here is that Warding Bond does not halve this incoming damage, but still responds to the damage in turn. So if I'm warded with 20 HP and use Transfer Health on another character, I would expect that both my warded character and the character that's warding them to take 5 damage (10 total, half of the HP) and then the target of Transfer Health to get healed for 10. Either that, or for Transfer Health's self-damage to not be eligible for Warding Bond at all, and act as though the effect isn't there. What happens instead is that in this same scenario, the warded character using Transfer Health takes 10 damage, the character who used the ward take 10 damage, and the target of Transfer Health heals 10 HP, effectively doubling the downside of Transfer Health and making it never worth using, which is rough because it's already pretty niche. As with my previous gripe concerning concentration spells and polymorph effects, this feels more like an oversight than intentional design.
Goodberry Needs to Remember Who Cast It This mostly matters for gear interactions, such as with Wapira's Crown and The Whispering Promise. If my druid has these sort of items on, I'd really expect any character who consumes the Goodberry to trigger the effects of these gear items, because the druid cast a spell that healed an ally.
Wood Woad's Regeneration to Work as Advertised Recently got ahold of Conjure Woodland Beings, and while experimenting with it I noticed that despite the Wood Woad supposedly healing 10 HP per turn that it's in a vine surface (provided it doesn't take fire damage), this doesn't actually seem to work at all. I hadn't tested it in combat, though, so perhaps it's combat-only. But, if it is combat-only, this needs to be messaged to the player.
Polymorph is Extremely Disappointing With all of the other spells that offer options, why was Polymorph restricted to turning an enemy into a sheep? Why not just treat it like the Druid's Wild Shape, and give creature options that scale with the target's level? The main draw of Polymorph in the TTRPG is how extremely versatile it is, and how it can be used by arcane spellcasters as a "pseudo-heal" for allies in danger. The implementation in BG3 doesn't offer any of this, instead opting for easily the least-used application of Polymorph, which is uncommon to see use because in the TTRPG other spells do the same thing but more effectively and with a better stat target (e.g. Banishment targeting Charisma). Polymorph
can be used to do this, but it's worse than a spell that specializes in it. Polymorph is meant to be a spell that's a multitool, and that's its main draw. By taking out the soul of the spell, it's been completely neutered and is basically never worth using.
