I was too hasty in composing my assessment. What I should have said is that 50% of the time, this MT is a level of spells behind.Jacob_Orlove wrote:Obviously that MT is going to look bad at an odd character level. At level 8, though, you're casting 4th level spells just like everyone else.
And yes, let's compare that to a PHB caster class. How about the Sorcerer? Granted, it's not as good as a Wizard or Cleric, but Sorcerers are still plenty good enough to beat up on level-appropriate monsters, and this MT compares pretty favorably to a standard Sorc.
As for the sorcerer, I stand by my original case: the class is not balanced with respect to the other core casters. Now, if you maintain that it is balanced with respect to the monsters, this implies that the other casters are too strong. I am agnostic on this point: I don't feel like I have enough experience to comment on the relative power of the sorcerer vis-a-via the monsters. I chose to balance according to the stronger core casters because there were more of them so it requires less work. Still, dropping other caster levels in exchange for being able to cast more low-level spells is not a trade I believe any smart caster will make, because the ability to throw the maximum level of spells available is so important.
Bigode, if you're convinced Leadership is a problem, that probably means that a variety of abilities based on it need to change, such as the [Leadership] feats themselvse and also class abilities that grant cohorts at level-2. I'm not unconvinced that Leadership isn't too powerful, but I don't even know where I'd start trying to figure out what the appropriate balance point is. Perhaps a feat should be less than a level, but if so, how much? If you do change Leadership you'd probably want to look at some of these classes again.
As for the EL equivalent of Leadership, I actually made a mistake in my calculation. The basic equation for the effective level of a combination of individual creatures is 2*log_2 \sum_i 2^{l_i/2}, where the l_i's are the individual levels. If you do the algebra for the effective level for a creature with some l who has a partner \delta l levels lower, it works out to be l+1.17, so Leadership's benefit is a bit smaller than I'd thought.
Some of my NPC class abilities are direct from RoW feats, chosen because of importance and simplicity to adjudicate. I didn't follow an explicit algorithm in the creation of the NPC classes, but if you wanted to separate out the various components, here's what I'd look at. First, PC classes should grant about 1 ability/level. If you use ToG rules, subtract out the various numeric bonuses the given NPC class should get, and the remainder have to be class abilities to represent e.g. stuff like being able to fly or teleport. For a character of level l, feats give [2 + (l-1)%5](1 + l%3) = 2 + 2(l%3) + (l-1)%5 + (l%3)[(l-1)%5] abilities; up to around level 15, that's approximated rather well by just 1 ability/level, so that's what I did. All told, you want somewhat more than two abilities/level, though I tended to stick to two just to simplify it.
I think I understand what you're trying to accomplish; I listed my classes just to give you a vague idea of my plans so you could use them, or not. I think for NPCs the spell-lists could use a lot of trimming: long-duration buffs should be built into the class itself, and lots of specialized spells just dropped altogether. For PCs they should probably stay in to some extent.
My basic idea is that the caster level determines the level for level-based effects and stat bonuses, while spell level sacrificed determines what wild shape abilities you have access to. To determine what abilities druids should have at what levels, I suggest looking at animal CRs for the first 10 levels or so, and then going into whatever you feel is appropriate for higher-level druids to shift into (the core uses plants and elementals, for no particular reason I can determine). This is how I was getting the lists of abilities I was generating.