Now that I have a keyboard, the original post:
If a fighter can do the equivalent of Time Stop, for example, it’s not stopping a wizard from also casting Time Stop. I think what is being said is the fighter has that only option from a list while the wizard still has their full list of 9th level spells to choose from for the exact situation.
Replying to this part, should the 17th level fighter take a deep breath and invoke "
Gotta Stab Them All - with your blinding speed make an attack against every foe within [range], no reactions allowed. You will return to the place where you started. Once per long rest." A wizard is only going to be able to cast one ninth level spell a day, so that's equivalent. The wizard does have an advantage that they might be able to have a choice from more than one ninth level spell, but given the choices available they are only likely to have three or so at 20th level. And, still, be only able to choose one of those per day.
So, if the fighter is following a speed / agility based path I think this would be reasonable. And, as a non-magical but preternatural ability it would not be subject to counterspelling or dispels.
This problem gets worse the higher level you get. There's just no way to balance wish against anything a fighter could do.
In 1e the wizard aged 3-5 years each time they cast
wish. Magical aging triggers a system shock roll. So, each time a wizard used
wish they could die, based on their Constitution. Could they be
raised? Sure. It cost a Constitution point and aged the cleric. I made
raise dead an exception to the system shock roll rule, but BTB you could have the cleric raise the target and lose their life to do so. Now, I'm not necessarily advocating that, but if we are looking at "providing real costs for using magic" (below) then a risk of debility or death, no matter how small, gives one pause.
I also don't think you should try to balance
wish against what a fighter could do. You just make it costly to use, and expect it to be used once or twice per level. I'm a firm believer in that fighters should be short rest / encounter based and wizards be long rest / daily based.
1. Open up more design space in skills
2. Redesign spells that carry out what a skill would so that there is enough reason to situationally choose one over the other.
3. Spellcasting utility should come at a cost of raw power.
4. Provide real costs for using magic.
5. Provide real non magical counters to magic.
Because or in spite of me being a grognard, I endorse this for the most part. For #2, I don't think there's much to be done really. That is probably secondary to my attitude of making each foray last as long as possible. I have a poor opinion of the five minute workday as it gives the opposing side a chance to adapt while you're recovering. But, not everyone plays the way I do. Also, if a wizard chooses a spell roster that allows them to obviate a class, they they've just volunteered to take on or assist with that role. If the player is doing it to be an ass, then that's not a rule problem but a personality one.
Limiting the number of free spells gained per level and having a robust method of researching your own spells significantly limits spellcaster flexibility while still allowing choice.
For:
1st level- spells can be cast in melee without issue and have no material component cost / issues. Too quick and easy.
2nd level- can trigger free attacks, but still essentially free to use.
3rd level- Mostly free to use, but have situational use (
fireball in a small room, mixed melee) or minor costs (
continual light).
4th level- Minor to moderate cost to use, but with reusable foci.
5th level- Moderate cost foci and preparation, minor cost expendables, or highly situational.
6th level- Significant cost and preparation, high cost foci, or moderate cost expendables are require for cast.