Stacking extra attack

[JellyPooga]

How would you rank the classes, in terms of a general "how impactful is this class, how often do you get to meaningfully contribute, how useful is that class' tools," etc.

I mean, what's the metric here? In terms of general impact, frequency of meaningful contribution and utility of the Classes tools, I rank the Rogue as one of the highest, with only the Bard being definitively higher under most circumstances, because they get to contribute in a similar (but different) manner to the Rogue with spellcasting slapped on top. In specific scenarios, where other Classes features come into play, yes, Wizards, Druids etc are frequently more powerful in that scenario, but for overall utility and contribution across all encounters of all types? Yeah, Rogue has it in the bag. Number one reason? They have more Actions. They literally have more facility to do stuff at any given point in time. Is it going to do as much damage as a Fireball or PAM/GWM Fighter or have the utility of giving the entire party +10 Stealth or entirely removing a creature from combat for a minute? No. But that doesn't mean that their contribution is actually less meaningful.

As a Rogue player, I'm participating in every scenario in a way that very few other Classes can or do, whether it be because of build limitations or resource management. To pull one of your examples, yes, in theory a Ranger can use PWT to enhance the Stealth of the entire party, but he's spending a significant build resource (Spell Known) as well as a resource (Spell Slot) each time to activate it, that could have been used on a different spell, for a bonus to the specific scenario of needing the entire party to be good at stealth...which might not ever come up in an adventure.

To pull it back toward the OP a little, do Rogues need Extra Attack? I'd argue not only do they not need it, they don't even really want it. Of course I'd slap it on any character if given the option, the same way I might accept a free Cantrip, but Extra Attack isn't required for the Rogue to function not only well, but in many regards better than other characters, because attacks aren't what the Rogue does best. Rogues aren't the best at dealing damage, they're not the best for imposing control effects, they're not the best at buffing...they're not the best at doing much of anything, to be perfectly fair. But that doesn't make them underpowered, because if they were the best at anything they'd be overpowered if they were to maintain their resource-lite nature.

Of course, all of my arguments fall down in games where resources like spell slots are never really depleted; the infamous 5-minute adventuring day. In campaigns that allow resource dependent classes to always have use of their resources, the resourceless class is left in the dust. So it depends on what kind of game you're used to. Me, I play and run tight games with high stakes and short time limits, so the Rogue really shines as a vector to get the big guns to where they're needed. Milage varies.

[GeneralVryth]

I feel like you're making an argument that rogues aren't unplayable (they aren't!).

I'm saying they're underpowered. Not unplayable, or useless. Underpowered.

Edit - to expand on this point

How would you rank the classes, in terms of a general "how impactful is this class, how often do you get to meaningfully contribute, how useful is that class' tools," etc.

Is rogue better than wizard? Absolutely not
Is it better than sorcerer, cleric, druid, or bard? Again, no
Is it better than paladin or artificer? Nope!
Is it better than fighter? Ehhh fair amount of divergence of what each class brings, but I think the fighter is definitely better. Generally, what the fighter brings is more useful than what the rogue brings - especially rune knight
Ranger? A lot of overlap with the rogue. Except the ranger natively gets half casting, has better subclasses, and simply by virtue of having pass without trace, can make the whole team a "rogue."
Warlock? Warlock has some bad weaknesses, but the best they bring is a whole lot better than the rogue
Monk? Barbarian? Oh there's no classes left?

Ok so this is my point. Rogue is arguing with the bottom quarter of classes which one is better. That's.... Not really much of prize, is it?

Ranking classes is extremely wish-washy because it matters a great deal how they are played, and how the DM plays. Just the simple question of whether casters can do 5MWDs has a massive impact on rankings. Expected level range can also play a part. How often do games go from 1 to 20?

Could Rogues benefit from light buffs? Probably. In fact just looking at their powers again, a solid buff could simply take the form of moving a few of their powers earlier (Slippery Mind and Blindsense), and maybe creating weaker versions of some powers to be gained earlier (a once per short rest reliable talent, a once per long rest Stroke of Luck). Then maybe figure out something to fill the empty slots (could easily add another ASI there for one, and the once long rest Stroke of Luck the other).

Personally, as a player I love lots of options and room to be creative, so I would likely play a Rogue over just about any Barbarian, and most Fighters. Rangers and and Warlocks become more interesting, and frankly I have never particularly liked Clerics. So rogues are probably in the lower mid quarter for me, but that has a lot more to with taste than power.

[Witty Username]

They still don't have much to do offensively, have very little presence on the field, have no resources or emergency buttons, and another class can do everything they do and more.

Resources is a disadvantage,

Take uncanny dodge vs rage, both offer damage resistance, but a barbarian is limited to a few times per day while the rogue can always have uncanny dodge available.

[Theodoxus]

OTOH, Rage is against every attack the barbarian faces while raging. The Rogue either has to know which attack to UD or be very lucky (I don't dodge the big axe swing because the Oathbreaker is about to hit me with a big old smite!)

So, it's not exactly resourceless... you're burning your sole Reaction a round on the hopes that you chose wisely.

[Slipjig]

Not sure what the point of Mountain Dwarf armor proficiency would be then. Basically, rogues are the only ones who would take advantage (and apparently not ATs). And how would you rule the AC provided by tortles and others?

Feats seem likewise problematic. You're basically stating that casters should never take an armor boosting feat; which again basically nixes the need for them. /shrug.

I'm responding to Skrum's prompt to put together language cutting down on armored casting. We can debate whether any given provision is good or not.

But in my personal opinion, racial armor and weapon proficiencies are almost always bad deisgn choices.

I *do* think a feat is a sufficient investment that a caster ought to be able to up-armor when using it. But I was struggling with how to phrase it for weird cases, like a Wizard-Cleric. How about this, rolling all the armored feats into one:

"Feat: Armor Training. Every class or subclass you have (or gain in the future) gains proficiency in the next heavier weight of armor. (e.g. Wizards and Sorcerors gain proficiency with Light Armor, etc). This includes the ability to cast spells from the appropriate spell list while wearing that heavier armor.

This applies to each class individually. So a Wizard-Cleric who has taken this feat once could cast Cleric spells in Heavy Armor, but would need to be wearing Light Armor to cast Wizard spells.

This feat may be taken multiple times, gaining proficiency in the next heavier weight of armor each time. e.g. A Wizard who has taken this feat twice may cast Wizard spells in Medium Armor."

[Theodoxus]

I don't disagree with your premise that racial proficiencies are bad, or at least wasting design points/space. I was just looking at it in situ; modifying races to cut down on proficiencies is fine.

I think your first crack at a feat is quite wordy. Not surprising, first cracks tend to be, since you're trying to cover all the bases - easier to pare down anyway.

I do think, since the idea would be to make this wholistic by reworking the races, that most of the wordiness in the feat can thus be moved to the classes. If the primary idea of casting in armor is moved to each class, then the feat itself doesn't need to be self-referential, and can simply state something along the lines of "Armor Training: Taking this feat increases the type of armor you can wear by one step (none->light; light->medium; medium->heavy), and allows the casting of spells while wearing that type of armor. This feat may be taken multiple times."

So classes talk about the specific type of armor proficiency they get. The feat elevates that by one step. Then you just need a blurb in the multiclassing rules about armor type interactions and casting sans proficiency.

[Dr.Samurai]

I feel like you're making an argument that rogues aren't unplayable (they aren't!).

I'm saying they're underpowered. Not unplayable, or useless. Underpowered.

Edit - to expand on this point

How would you rank the classes, in terms of a general "how impactful is this class, how often do you get to meaningfully contribute, how useful is that class' tools," etc.

Is rogue better than wizard? Absolutely not
Is it better than sorcerer, cleric, druid, or bard? Again, no
Is it better than paladin or artificer? Nope!
Is it better than fighter? Ehhh fair amount of divergence of what each class brings, but I think the fighter is definitely better. Generally, what the fighter brings is more useful than what the rogue brings - especially rune knight
Ranger? A lot of overlap with the rogue. Except the ranger natively gets half casting, has better subclasses, and simply by virtue of having pass without trace, can make the whole team a "rogue."
Warlock? Warlock has some bad weaknesses, but the best they bring is a whole lot better than the rogue
Monk? Barbarian? Oh there's no classes left?

Ok so this is my point. Rogue is arguing with the bottom quarter of classes which one is better. That's.... Not really much of prize, is it?

The rogue, for me, doesn't really have a lot of combat oomph. Sneak Attack is nice, but it's not great damage, and it's one attack. And we get by without rogues in our games just fine when it comes to out of combat stuff. I don't think that's necessarily a class issue, more a table/style issue. But skills should really be an "everyone" thing anyways.

[Skrum]

Resources is a disadvantage,

Take uncanny dodge vs rage, both offer damage resistance, but a barbarian is limited to a few times per day while the rogue can always have uncanny dodge available.

I really think both the designers and many players have this backwards - being able to use an ability all day every day puts some necessary restrictions on how good it can be. Uncanny Dodge is a good example - hope you guess right!

And yes, in the hypothetical 100 encounter day, the character with Uncanny Dodge instead of Rage looks to be in a great position. But there is no 100 encounter day. There's almost never a 10 encounter day. Hit points get depleted. The story moves on.

I agree that "resourceless" rogue is great for certain types of games. But if it almost never comes up because *the entire rest of the party* needs to rest, ergo that's not how the game is played, well what good is it?

Being able to use an ability over and over and over is much less useful, in practice, than having a powerful, easily applied win condition. Fireball the goblins. Hit the ogres with slow. Smite-crit for 120 damage. Slow and steady in fact does not win the race, not when combat rarely exceeds 4 rounds.

[GeneralVryth]

You guys are still dancing around shields, which in practice are likely to be the bigger problem. Also, class specific proficiencies are just destined to get messy.

Besides adding/increasing strength requirements, if don't want to change classes much I like the idea suggested earlier in the thread of just breaking Medium and Heavy Prof into a Simple and Complex (or Martial) option. All of the racial traits that give armor profs become simple martial profs. All class features that give them also become simple. And then when you get to level X (5?) in a class that gave a simple Medium/heavy Prof, you get the complex version. Still need to deal with Shields though.

[Skrum]

I mean, what's the metric here? In terms of general impact, frequency of meaningful contribution and utility of the Classes tools, I rank the Rogue as one of the highest, with only the Bard being definitively higher under most circumstances, because they get to contribute in a similar (but different) manner to the Rogue with spellcasting slapped on top. In specific scenarios, where other Classes features come into play, yes, Wizards, Druids etc are frequently more powerful in that scenario, but for overall utility and contribution across all encounters of all types? Yeah, Rogue has it in the bag. Number one reason? They have more Actions. They literally have more facility to do stuff at any given point in time. Is it going to do as much damage as a Fireball or PAM/GWM Fighter or have the utility of giving the entire party +10 Stealth or entirely removing a creature from combat for a minute? No. But that doesn't mean that their contribution is actually less meaningful.

As a Rogue player, I'm participating in every scenario in a way that very few other Classes can or do, whether it be because of build limitations or resource management. To pull one of your examples, yes, in theory a Ranger can use PWT to enhance the Stealth of the entire party, but he's spending a significant build resource (Spell Known) as well as a resource (Spell Slot) each time to activate it, that could have been used on a different spell, for a bonus to the specific scenario of needing the entire party to be good at stealth...which might not ever come up in an adventure.

To pull it back toward the OP a little, do Rogues need Extra Attack? I'd argue not only do they not need it, they don't even really want it. Of course I'd slap it on any character if given the option, the same way I might accept a free Cantrip, but Extra Attack isn't required for the Rogue to function not only well, but in many regards better than other characters, because attacks aren't what the Rogue does best. Rogues aren't the best at dealing damage, they're not the best for imposing control effects, they're not the best at buffing...they're not the best at doing much of anything, to be perfectly fair. But that doesn't make them underpowered, because if they were the best at anything they'd be overpowered if they were to maintain their resource-lite nature.

Of course, all of my arguments fall down in games where resources like spell slots are never really depleted; the infamous 5-minute adventuring day. In campaigns that allow resource dependent classes to always have use of their resources, the resourceless class is left in the dust. So it depends on what kind of game you're used to. Me, I play and run tight games with high stakes and short time limits, so the Rogue really shines as a vector to get the big guns to where they're needed. Milage varies.

Truly and unsarcastically, I'm happy the class you have the most fun playing is also the class you think is the best. But I'm simply not going to agree with you on that point.

[Skrum]

You guys are still dancing around shields, which in practice are likely to be the bigger problem. Also, class specific proficiencies are just destined to get messy.

Besides adding/increasing strength requirements, if don't want to change classes much I like the idea suggested earlier in the thread of just breaking Medium and Heavy Prof into a Simple and Complex (or Martial) option. All of the racial traits that give armor profs become simple martial profs. All class features that give them also become simple. And then when you get to level X (5?) in a class that gave a simple Medium/heavy Prof, you get the complex version. Still need to deal with Shields though.

Could bring back arcane spell failure. Lol.

Clerics can cast in armor because they cast divine spells. Ditto for paladins and druids.

But wizards, sorcerers, etc., cast arcane spells and that means they have spell failure for using a shield or wearing armor. Bards would have to suck it up until they get to their subclass or something.

Even though (I think) I was the person to bring this up, it occurs to me that I might be looking at the wrong thing. I think maybe my problem isn't with armor - it's with the shield spell. A wizard with 20 AC? Ok, that's cool. Hit points are still low, still gonna avoid melee. Doesn't really change much.

A wizard with 20 AC that can selectively raise it to 25? That's getting ridiculous.

[Rukelnikov]

I really think both the designers and many players have this backwards - being able to use an ability all day every day puts some necessary restrictions on how good it can be. Uncanny Dodge is a good example - hope you guess right!

And yes, in the hypothetical 100 encounter day, the character with Uncanny Dodge instead of Rage looks to be in a great position. But there is no 100 encounter day. There's almost never a 10 encounter day. Hit points get depleted. The story moves on.

I agree that "resourceless" rogue is great for certain types of games. But if it almost never comes up because *the entire rest of the party* needs to rest, ergo that's not how the game is played, well what good is it?

Being able to use an ability over and over and over is much less useful, in practice, than having a powerful, easily applied win condition. Fireball the goblins. Hit the ogres with slow. Smite-crit for 120 damage. Slow and steady in fact does not win the race, not when combat rarely exceeds 4 rounds.

This, if the typical adventuring day has about 20 combat rounds, you need to fill 20 turns worth of actions, the closer you are to having buttons for those 20 actions the less relevant resourceless damage gets.

[GeneralVryth]

Could bring back arcane spell failure. Lol.

Clerics can cast in armor because they cast divine spells. Ditto for paladins and druids.

But wizards, sorcerers, etc., cast arcane spells and that means they have spell failure for using a shield or wearing armor. Bards would have to suck it up until they get to their subclass or something.

Even though (I think) I was the person to bring this up, it occurs to me that I might be looking at the wrong thing. I think maybe my problem isn't with armor - it's with the shield spell. A wizard with 20 AC? Ok, that's cool. Hit points are still low, still gonna avoid melee. Doesn't really change much.

A wizard with 20 AC that can selectively raise it to 25? That's getting ridiculous.

Spell failure just adds more complexity and helps kill off character archetypes. The problem isn't that Wizards (or what have you) can use heavier armors, or shields. It's that it's too easy. If you have to dip 3 or 4 levels in another class to do it, that's a significant cost. Same if you have to spend a feat (and ideally some stat points). It's Mountain Dwarf, and the various 1 level dips that are the issue. 1st level is too loaded for most classes (and in some ways not loaded enough). Ideally the 1st level of most classes would be cleanly broken in to 2 categories. Stuff you get if its truly your first level, and stuff you get if it's just your first level in that class. And the system does this halfheartedly, but it doesn't really work because there isn't enough in the first category for most full casting classes. Which makes multi-classing in to them too cheap (if you were always going to want 1 level from that class and another).

As for the Shield spell. It's really not that powerful. It only seems that way when combined with other stuff that is powerful.

[Slipjig]

I think your first crack at a feat is quite wordy. Not surprising, first cracks tend to be, since you're trying to cover all the bases - easier to pare down anyway.

So classes talk about the specific type of armor proficiency they get. The feat elevates that by one step. Then you just need a blurb in the multiclassing rules about armor type interactions and casting sans proficiency.

I agree it's wordy, that's why I just gave up and said "No Feats" in my first draft. I think one unified feat probably occupies fewer column inches than having a separate version for each weight, but I might be wrong.

The last paragraph is trying to preclude an interpretation where a Wizard/[Any Med Prof Class] could take the feat once and jump straight to unlimited Heavy Armor casting.

[Kane0]

Even though (I think) I was the person to bring this up, it occurs to me that I might be looking at the wrong thing. I think maybe my problem isn't with armor - it's with the shield spell. A wizard with 20 AC? Ok, that's cool. Hit points are still low, still gonna avoid melee. Doesn't really change much.

A wizard with 20 AC that can selectively raise it to 25? That's getting ridiculous.

Have the Shield and Shield of Faith spells not stack with actual shields.

[Skrum]

Have the Shield and Shield of Faith spells not stack with actual shields.

I was thinking something like a cap on temporary AC. Defensive fighting style, magic plate, stack to your heart's content. But the shield spell, shield of faith, haste; those can't raise your AC higher than 21.

[Skrum]

As for the Shield spell. It's really not that powerful. It only seems that way when combined with other stuff that is powerful.

It's absolutely the icing and cherry on the cake. The straw that broke the camel's back.

+5 AC, for an entire turn, as a reaction. Raising AC from 14 to 19, that's totally fine, and I'm sure it was exactly the use-case the writers were imagining. But when the character has a base AC of 22, 23, 24? It's a problem.

[Witty Username]

Spell failure just adds more complexity and helps kill off character archetypes. The problem isn't that Wizards (or what have you) can use heavier armors, or shields. It's that it's too easy. If you have to dip 3 or 4 levels in another class to do it, that's a significant cost.

Maybe tie armor and casting to subclasses.

Take for example, general rule characters cannot cast in armor, but then make it a trait of EK to be able to do so. Then casting in armor becomes a 3rd level dip.

You can also modulate different classes,
Druids have metal restrictions
Clerics don't get to apply theirs ouside there own spell list, same with Paladin and Ranger.
Make bard able to do so generally.
Fighter and Rogue use subclasses like EK and AT.

Now MCs are a balance point of classes to use, Armor and Wizard/Sorcerer is harder, Bards are now a flex pick, and the divine casters are mostly unaffected.

[Kane0]

I was thinking something like a cap on temporary AC. Defensive fighting style, magic plate, stack to your heart's content. But the shield spell, shield of faith, haste; those can't raise your AC higher than 21.

Or just 'one source', like THP and such. Just make it a blanket rule for ease-of-use, no static number on your character sheet can be added to from more than one non-permanent source.
Might need some clipping around the edge cases though

[GeneralVryth]

I was thinking something like a cap on temporary AC. Defensive fighting style, magic plate, stack to your heart's content. But the shield spell, shield of faith, haste; those can't raise your AC higher than 21.

It's absolutely the icing and cherry on the cake. The straw that broke the camel's back.

+5 AC, for an entire turn, as a reaction. Raising AC from 14 to 19, that's totally fine, and I'm sure it was exactly the use-case the writers were imagining. But when the character has a base AC of 22, 23, 24? It's a problem.

How many attacks is Shield actually saving you from though? 1 for sure (depends on GM), and 25% (likely less if you stacked AC) of the remainder before the start of your next turn? What does that actually compute out to? Probably 1 to 2 attacks per use on average. Which really isn't a lot. It's better than many of the very crappy heal spells at low level (and it scales with opponents), but that's not really world breaking either.

Maybe tie armor and casting to subclasses.

Take for example, general rule characters cannot cast in armor, but then make it a trait of EK to be able to do so. Then casting in armor becomes a 3rd level dip.

You can also modulate different classes,
Druids have metal restrictions
Clerics don't get to apply theirs ouside there own spell list, same with Paladin and Ranger.
Make bard able to do so generally.
Fighter and Rogue use subclasses like EK and AT.

Now MCs are a balance point of classes to use, Armor and Wizard/Sorcerer is harder, Bards are now a flex pick, and the divine casters are mostly unaffected.

I had a similar thought up thread about such a thing for shields which I think is more important. The problem in your list is the Clerics/Paladins/Rangers. It just ends up finnicky having a trait that usually is generally applied only apply to certain classes in certain cases.

[Psyren]

Nerf ideas for the Shield spell:

- Reduce to +3 AC (upcastable to +4 / +5).
- Applies to a number of incoming attacks per PB / casting stat, then ends.
- Can't increase AC above {ceiling} (e.g 21 as Skrum mentioned.)
- Doesn't stack with physical shield (I dislike this one as it introduces additional subtraction to the game rather than straightforwardly telling the player what it does each time.)

[Witty Username]

I had a similar thought up thread about such a thing for shields which I think is more important. The problem in your list is the Clerics/Paladins/Rangers. It just ends up finnicky having a trait that usually is generally applied only apply to certain classes in certain cases.

The alternative is to drop armored casting completely I think, or rather the multi classing version of it. Either the solutions end up weird, or cut build options entirely as apposed to nerfs.

That I think is one of the better reasons to just ban the shield spell, its not necessary for the game and it is way easier to remember than house ruled nerfs.

[Skrum]

That I think is one of the better reasons to just ban the shield spell, its not necessary for the game and it is way easier to remember than house ruled nerfs.

Yeah maybe this is it. Wizards and sorcerers actually being mildly fragile, like oh noes, no one will play them.

[Witty Username]

Nerf ideas for the Shield spell:

- Reduce to +3 AC (upcastable to +4 / +5).

My issue with this one specifically is it becomes pretty irrelevant in comparison to mage armor unless you are specifically going for MC armor nonsense. This makes it occupy a weird spot in shield being only worth the spell slot if one is deliberately trying to break it. 'Free AC button' is probably just a thing that either shouldn't exist, or be outside of casters.

[clash]

Nerf ideas for the Shield spell:

- Reduce to +3 AC (upcastable to +4 / +5).
- Applies to a number of incoming attacks per PB / casting stat, then ends.
- Can't increase AC above {ceiling} (e.g 21 as Skrum mentioned.)
- Doesn't stack with physical shield (I dislike this one as it introduces additional subtraction to the game rather than straightforwardly telling the player what it does each time.)

Just don't let it stack with armor or shields honestly. The only class that gets it and armor proficiency is hexblade and it's not worth it on them after the first few levels anyways.

[Theodoxus]

Couple ideas; Bring back shield types:
Heavy Shield: +3, only Fighters and Paladins are proficient, and only starting at 5th level.
Light Shield: +2, only Barbarians, Fighters, and Paladins are proficient, and proficiency isn't provided by multiclassing (but still part of medium armor feat).
Buckler: +1, only Bards, Barbarians, Clerics, Druids, Fighters, Paladins, and Rangers are proficient, multiclassing does provide proficiency.

2nd idea: The Shield spell doesn't work with any armor or spells other than Mage Armor and not with defensive magic items(Ring/Cloak of Protection, Bracers of Defense, etc.). At best, I think that would mean max AC of 18 with MA and 20 Dex, +5 Shield for a total of 23 in the very rare instance a caster has managed to max out their Dex; and far more likely in the 19-21 range.

[GeneralVryth]

Nerf ideas for the Shield spell:

- Reduce to +3 AC (upcastable to +4 / +5).
- Applies to a number of incoming attacks per PB / casting stat, then ends.
- Can't increase AC above {ceiling} (e.g 21 as Skrum mentioned.)
- Doesn't stack with physical shield (I dislike this one as it introduces additional subtraction to the game rather than straightforwardly telling the player what it does each time.)

A temp AC ceiling in practice ends up much more complicated than would be ideal. Your set it to low and it just kills a bunch of abilities once PC start using magical armor, too high and it doesn't do anything. And really Shield is not that powerful. If you are going to change it, have it just no-sell a single attack and be done with it. Or reduce the bonus to like 3, but upcastable. If that isn't enough, just ban it at your table.

The alternative is to drop armored casting completely I think, or rather the multi classing version of it. Either the solutions end up weird, or cut build options entirely as apposed to nerfs.

That I think is one of the better reasons to just ban the shield spell, its not necessary for the game and it is way easier to remember than house ruled nerfs.

The alternative is to make getting and using heavier armors more difficult. If you needed a 14 Str, and 14 Dex, and a slightly harder to get prof to use Half-Plate, that can end up being a decent investment. If you need 16 Str for a +2 shield (and 13 Str for a +1 Shield), that's not nothing. Same thing with something like 17 Str for Full Plate. For martials those requirements are nothing since they are already going to want to maximize one of those attributes.

Yeah maybe this is it. Wizards and sorcerers actually being mildly fragile, like oh noes, no one will play them.

That's a fine houserule, but taking away one of the only real defensive options for Wizards/Sorcerers is going to have consequences. I would certainly be leery of a DM doing that without some kind of alternative.

[GeneralVryth]

Just don't let it stack with armor or shields honestly. The only class that gets it and armor proficiency is hexblade and it's not worth it on them after the first few levels anyways.

Your forgetting Eldritch Knights, for whom it's one of the best spells in their arsenal.

[Witty Username]

The alternative is to make getting and using heavier armors more difficult. If you needed a 14 Str, and 14 Dex, and a slightly harder to get prof to use Half-Plate, that can end up being a decent investment. If you need 16 Str for a +2 shield (and 13 Str for a +1 Shield), that's not nothing. Same thing with something like 17 Str for Full Plate. For martials those requirements are nothing since they are already going to want to maximize one of those attributes.

Those would nerf things like Cleric and Ranger for no real reason though, Shields and armor may not be part of the wizards power budget but it is for other classes.

Half-plate specifically is important for classes like Ranger as Dex investment takes time (something like 12th level for games with feats to get to 20 Dex where light armor starts to make weight), and strength investment isn't really a thing a Ranger wants to do to that degree.

It also hurts some multiclass builds like rogue mixes, which I don't think is necessary.

--
Overall, Wizard and Sorcerer have plenty of defensive tools without shields, the shield spell, or armor.

[Rukelnikov]

Spell failure just adds more complexity and helps kill off character archetypes. The problem isn't that Wizards (or what have you) can use heavier armors, or shields. It's that it's too easy. If you have to dip 3 or 4 levels in another class to do it, that's a significant cost. Same if you have to spend a feat (and ideally some stat points). It's Mountain Dwarf, and the various 1 level dips that are the issue. 1st level is too loaded for most classes (and in some ways not loaded enough). Ideally the 1st level of most classes would be cleanly broken in to 2 categories. Stuff you get if its truly your first level, and stuff you get if it's just your first level in that class. And the system does this halfheartedly, but it doesn't really work because there isn't enough in the first category for most full casting classes. Which makes multi-classing in to them too cheap (if you were always going to want 1 level from that class and another).

Yup, this is exactly the problem, you COULD cast in armor in 3e, it was just so costly that it was more niche than "meta", in 5e its literally just a 1 level dip that doesn't slow your spell progression and even gives you a very good spell variety and a sizeable amount of prepared spells on top.