Looks like there's no commentary so far on my second post, but I want to expand it with more analysis. The questions in that post still stand, but some people might prefer this approach.
For those who feel it is overpowered (ie, everyone who has responded so far, which means you're probably right, but I need to see it proved--after all, when the PHB first came out we thought Agonizing Blast made Warlocks too strong, Moon Druids were overpowered, and Monks were too weak...), I'm providing a list of features to remove from it, in order. Please explain at which point you feel it would be balanced (with analysis if possible).
Stages of Feature Removal:
1) Third attack
2) Indomitable Spirit
3) Spell Mastery
4) Arcane Specialization
5) Improved War Magic
6) Move War Magic from 8th to 10th level
If you think it is still too strong after all of that...you are wrong.
Let's compare the gutted version to a Bladesinger, feature by feature.
First, a bit about the Bladesong feature. It is really good. Those who have played it say it rarely gets hit. One poster played it (I think to level 7) and never got hit. The movement is quite useful for the mobile nature of their style, and the concentration save boost is effectively not much less than a full Con save proficiency. Most Con saves you are likely to have to make are as a result to being hit--which doesn't happen often to the Bladesinger. They are probably going to take most of their damage from area effect spells, which tend toward Dex saves. So the majority of their Con saves are going to be concentration saves, which they get about the equivalent to proficiency in when using Bladesong. Bladesong's main weakness is that it doesn't protect you when surprised, but it is still a really good feature.
Gutted Warrior-Mage vs. Bladesinger Analysis:
A) The proficiencies and Bladesong that a Bladesinger gets at level two are better than the d8s, proficiencies, and fighting style a warrior-mage gets. Bladesinger has better AC, not too far off from effectively both Con and Wis saves and the other little boosts from Bladesong. If you add in Second Wind, it might push warrior-mage a bit higher comparing just those features (but we'll address that).
B) Song of Defense is effectively better than Second Wind (it comes later, but Second Wind scales so it isn't as good at lower levels). It takes spell slots, but the Bladesinger has more spell slots due to Arcane Recovery, and you only use it when you really have to. Add this into the mix from comparison A, and Bladesinger surpasses warrior-mage in defensive capabilities.
C) Song of Victory doesn't bring Bladesong's damage output into a competitive state with War Magic. (It isn't supposed to, however, as Bladesong is a full wizard and a decent melee combatant, that can become a good melee combatant if they expend spell resources.) Adding in War Magic and Song of Victory, warrior-mage might pull ahead due to significantly improved at-will melee damage output.
D) As a full wizard, Bladesinger gets Arcane Recovery, Spell Mastery, Signature Spell, and an extra cantrip. The extra cantrip may not matter to some, and Signature Spell is 20th level so not worth comparing too much, but Arcane Recovery and Spell Mastery are a big deal. Arcane Recovery gives you extra spell slots starting at level 1, and Spell Mastery is a better capstone than Signature Spell. You can use some of these spell slots for Song of Defense if needed, but you may not need to very often. Whatever you don't spend just adds onto Spell Mastery for making you a better wizard than warrior-mage. At this point, warrior-mage is only ahead in at-will melee damage, with Bladesinger winning the overall comparison.
So feature by feature, Bladesinger beats (even if not by much) the gutted warrior-mage.
Now you have to consider intended role functionality. Warrior-Mage and Bladesinger play differently. Warrior-Mage is offense, Bladesinger is defense. Bladesinger has more going for it than warrior-mage when being a mage, warrior-mage has more going for it offensively (and actually a bit less defensively) as a warrior. Other than having less going defensively, this means they both fill their role, and you'd play whichever role you like best.
From there, you can start restoring the removed features in reverse order, starting with #6, and see at what point it exceeds Bladesinger. I'm interesting in knowing precisely where people would consider that point to be.
Now let's try the same thing with a Valor Bard, the next most comparable class option. Both Valor bard and warrior-mage get d8 HDs, the same weapon and armor proficiencies, Extra Attack at level 6, and Improved War Magic at level 14. (We're assuming a gutted warrior-mage to start, so it's just War Magic at 10th level for them, and Improved War Magic (technically Battle Magic, but basically the same) for bard at level 14th.) So we only need to compare differences.
Gutted-Warrior Mage vs. Valor Bard Analysis:
A) For at-will melee damage output, the warrior-mage is a tad ahead due to Fighting Style. It's not much, but it gives them a slight edge. From levels 10th to 13th the warrior-mage comes out ahead due to War Magic, though bard is picking up green-flame blade at 10th level, so that keeps them from being as far behind as they would be. At 14th level their damage output hits the same level, with only those 3-4 extra damage per round due to fighting style for the warrior-mage.
B) Defensively, their baseline is identical. Bard gets Dex save rather than Con save, which can be a problem with concentration, so let's add to warrior-mage's lead at this pont.
C) For health resources, warrior-mage has Second Wind, while the bard is one of the best healers in the game with their spell list and Song of Rest. It takes spell resources for them to do in combat healing, but their out of combat healing takes care of both themselves and their allies. Bard comes out ahead here due to flexibility and potential. With the flexibility of healing (more than just hp) they can get from the bard spell list, I'm going to place bard and warrior-mage more or less on even footing by this point. Let's pretend the warrior-mage is a bit ahead, since Con save is important for concentration.
D) Let's jump into the spells. Valor bard and warrior-mage get the same number of slots and cantrips, so we can move on to the spell list now. Wizard gets the best spell list. But bard gets both the second best spell list, and the ability to poach spells from anyone. They'll end up with any 2 spells of 5th level or less (including cantrips, in case they fancy eldritch blast + hex for instance, though that isn't relevant to this concept), any 2 additional spells of 7th level or less, and any 2 additional spells of 9th level or less. The possibilities are endless. Pick spirit guardians or conjure woodland beings or moonbeam, not to mention haste or fly or blur or anything else the wizard can get, for your bread and butter concentration spell. If you prefer, snag greater invisibility right off of your bard spell list, and leave a spot open to poach another spell. For non-concentration spells, take your pick of favorites like armor of agathys, spiritual weapon, or wizard favorites like mirror image or blink. Feel free to grab a reaction spell or 2 like absorb elements, hellish rebuke, shield, or counterspell. With one each of those (and those are just lower level bread and butter spells), and wish because you can, you have two spots left to pick up any other non-bard spells you'd like. Maybe true resurrection and meteor swarm, or word of recall and destructive wave, or maybe storm of vengeance or holy aura, or don't limit yourself to meteor swarm, you could pick anything else the wizard gets. If you only stick with wish (for any 8th level or lower spell you might need 1/day) and pick foresight from the bard list, you already have two of the best 9th level spells, so you don't have to clutter up your known 9th level spells with more than you can use, and can feel confident picking from ones of 7th level or lower. I'm probably not even thinking of the best ways to optimize being able to cherry pick from every spell list. This is really good stuff. In exchange for this, the wizard has more flexibility to customize his daily spells to known threats he might face, and the best ritual casting. If we ignore the wizard (and warrior-mage's) greater selection of utility spells (we'll get to that), I think we're pretty close to even here.
E) Everything else. This is where direct comparison is tricky. Wizard spell list gets a lot of utility spells of all levels, but the bard gets a good number of those on their own list, so the spell utility advantage to warrior-mage isn't as great as it would be versus another class. What does bard get to make up for that? Actually, quite a bit. An additional skill, Expertise in 4 skills, and Jack of All Trades is a boatload of added utility right there. Throw in their signature Bardic Inspiration that we haven't even mentioned (including the ally buffing Combat Inspiration from Valor), Countercharm at 6th (which is situational, but really good versus the enemies with party disabling frightening or charming abilities), and don't forget the hidden damage boost they get (reducing the slight lead that warrior-mage has) because Jack of All Trades adds half their proficiency bonus to initiative. I'm pretty sure that bard straight up wins in the Everything Else category.
With warrior-mage's only clear advantages being a slight bit of increased at-will melee damage (a little more at levels 10-13) and Con(centration) saves, and Valor bard getting a mix of all sorts of goodies in their Everything Else category, I'm placing Valor bard as the winner against the feature gutted warrior-mage.
So same exercise: restoring removed warrior-mage features until it surpasses Valor bard to see where that point is.
Why would you ever choose to play a Wizard over this class? You're not really sacrificing much of anything, and yet you're gaining a ton of martial ability...
I addressed this in post #15. The wizard is sacrificing one of their highest levels spell slots (through loss of Arcane Recovery) up through level 10 (at which point it's a 5th level slot and another slot of variable level), a cantrip, and then at high level they delay Spell Mastery, turning it into their 20th level ability, and lose Signature Spell. And importantly, they become multiple attribute dependant, which means they are also sacrificing spell save DC and losing one prepared spell for most of their career. Even if you ignore the cantrip and high level features, you are still taking a significant hit to your wizard power. You only get 2-4 of your highest level spell slots, losing one of them is painful.
[MENTION=6677017]Sword of Spirit[/MENTION]
Consider a level 6 eldritch knight. You can cast a couple of shield spells a day and maybe an expeditious retreat. Or maybe 2x burning hands on the same turn to kill multiple mid tier enemies at once.
Now compare to your warrior-mage.
Your warrior mage is more about buffs. You don't need a super high int when your goal is self buffs. With haste (better ac, movespeed and more attacks) you get effectively 3 attacks in 3 combats per day at level 6. You can also cast up 7 shield spells per day as well or trade a few of those for misty for movement. Basically the only important fighter abilities you have given up at this point is action surge and an extra ASI / feat.
Likewise against hordes you are much better than the fighter. Even without the absolutely best save a fireball is still going to dominate most groups of 20hp and less monsters. I see no comparison. I don't know if I would ever play a fighter over your fighter mage.
Thanks for getting into the details--it's very helpful. At 6th level, how would you rate the Eldritch Knight versus the Bladesinger?
[MENTION=6677017]Sword of Spirit[/MENTION]
Maybe it would help if I put it this way. If you gave the valor bard the spells shield, fireball, and haste at the appropriate levels and constitution saving throws and a fighting style and access to the good melee cantrips along with the EK's level 7 ability then the valor bard would likely be the best attack based character in the game.
It's the fact that he lacks, haste, shield, fireball and the fighting style and doesn't have the EK's cantrip and bonus action attack ability that makes him not be the best melee character in the game.
I'm not sure Fighting Style does a whole lot in Valor Bard vs. Eldritch Knight, but that's an interesting analysis. I did a Valor bard to gutted warrior-mage comparison above, and I'd be interested in your take on it, or even your take on the Valor Bard's effectiveness by the book, with all the considerations I brought up.
