I’ve heard this refrain over and over again my entire gaming career, ” Spellcasters are overpowered. Their power needs to be nerfed.” While I do understand that a spellcaster’s power can overshadow the martial characters with a single spell, once that single spell is used, they probably do not have much left. This is doubly true at the higher levels. Here are my reasons why I believe that spellcasters of all stripes are not overpowered.
1) You’re Not Playing the Same Game
Playing a martial character is all about tactics. Fighters and paladins need to position themselves where they can stop the monsters from advancing upon the spellcasters. Rogues need to get themselves into a position where they can sneak attack. Barbarians should be where they can hit as many targets as possible. Rangers should find someplace where they can use their bows to hit whoever needs to be taken down. Monks need to stay mobile, moving through the battle space.
That is not the spellcaster’s game. Instead, a spellcaster needs to focus on resource management and opportunity costs. “Is the one fireball I prepped today best used now or later?” “This is my best spell, but the monster has resistance to its damage type. Should I use it or save it and use some other spell that does much less damage?” “Would it be better to buff the fighter or take out the monster attacking the cleric?” These types of questions are asked by every gamer that plays a spellcaster. Every spell they cast is a question of now vs later. They never know if what they’re doing is going to be overkill or a complete waste of resources. So when a spellcaster completely overwhelms the enemy with a single spell, it probably means that they will have very weak spells when they really need them. Spellcasters are more difficult to get right than most martial classes.
2) A Poorly Prepared Spellcaster is Useless
in my current home game, my 2nd-level cleric prepared protection from evil. We didn’t fight any evil creatures today. As such, that is one spell slot that sat useless. When you only have 4 slots, that is a quarter of the character’s main power. So I had to resort to my mace, and I rolled poorly. So my character was useless. But that’s beside the point.
A fighter can have a golf bag of bane swords on their back and can select the right weapon at the start of combat or switch it up in the middle as the dynamics of the battle change. Spellcasters don’t have that option. The spells they prepped at the start of the day or chose when they got their last level are all they have. If what they prepped does not work for this particular combat, they are probably staying in the back using their crossbow, if they have a crossbow.
What’s even worse is when they prepared well but the circumstances completely change on the group. If you are going into an ice cave and the spellcasters prepared fire spells, they are useless if the group gets sucked into a portal that takes them to the plane of fire. All those martial characters, they’re just as effective as before. Spellcasters can go from running the table to running away pretty quickly.
3) Combat Effectiveness Can Be Limited by Utility
Spellcasters a considerable amount of out-of-combat utility spells at their disposal. Whatever situation they run into, there is a spell somewhere to handle that. Putting aside whether or not they prepared the spell that day, there is an opportunity cost to it. Prepare knock to open a locked door? That is one less scorching ray they can do. Need to teleport a long distance? That comes at the cost of one less cone of cold. Making a mud wall so you can turn it into a stone wall with transmute mud to rock to protect the town from an enormous monster? Then they will have one less opportunity to cast hold monster that day. I can go on and on, but you get the idea. Rogues, unlocking one door does not mean they have to sacrifice their next sneak attack for it. Fighters, they don’t need to make mud walls, they can just pile the stones on top of one another to make the wall stone from the beginning.
Then there’s my personal favorite example: greater invisibility. While this example is PF 1e, the basic principles apply to any game system. Greater invisibility is a 4th-level spell, so you have to be minimum 7th-level to use it (8th for arcanists and sorcerers, 10th level for bards). So lets say it won’t see regular use until 10th-level. With it you can attack and stay invisible.
Compare that with sniping that anyone can use with he Stealth skill. But lets look at it with a rogue. At 1st-level, the rogue takes a –20 to Stealth checks to hide after attacking that round. That’s pretty bad, sure, but when you check out the Perception entry, you discover that if the person is distracted (like because they’re in combat), the hiding person gets a +5 bonus. And then there’s a +1 bonus per 10 feet the hiding person is away from the looker. Sniping requires a minimum of 10 feet, so that’s a minimum bonus of +6, making that penalty now –14. Naturally, the rogue would have a skill point in Stealth, the +3 bonus for it being a class skill and minimum Dex bonus of +3. Now that penalty is –7. Its opposed by the looker. If the rogue chooses someone that has no skill ranks in Perception and probably doesn’t have a Wisdom bonus, like a wizard or sorcerer, then they don’t have any bonuses at all to the roll. Still, at 1st level, that’s a sizable difference.
At 7th level (the minimum level to cast greater invisibility), the rogue added skill ranks to Stealth, bumped up their Dexterity ability, and probably even took the Skill Focus (Stealth) feat. Meanwhile, the spellcaster probably didn’t do anything to increase their ability to see someone. That net –7 is now a +3. That means the rogue has a better than even chance of doing exactly the same thing as greater invisibility every single round with absolutely no resource cost. In fact, Stealth is better than greater invisibility since Stealth is invisible to tremorsense, see invisibility, and true seeing while greater invisibility is not.
Wait until 10th level—when a bard can cast it at all—and a rogue can take the stealthy sniper advanced talent and take 10 off the penalty figured into the mix. That means, the rogue would have a net +19. Literally, the rogue would have to roll a 1, and the wizard would have to roll a 20 for the wizard to know where the rogue is.
Mind you, the wizard would be all too aware where the rogue is approximately and can send a fireball their way, but the rogue will probably pass the DC and take no damage thanks to evasion.
My point is, a wizard can cast greater invisibility once each time it is prepared. Rogues can sneak and snipe all day long. No limit.
I know I rambled on that last point, but you get the idea. Spellcasters are not nearly as overpowered as people think they are and attempts to even out their power generally make a difficult class to play far more difficult.
Download all of JBE’s products at DriveThruRPG, the Open Gaming Store, and Paizo.
Jon Brazer Enterprises 
Of course, this entry assumes two things:
1) The casters in question are Prepared Casters rather than Spontaneous Casters.
2) There are enough different encounters, especially combat encounters, in the adventuring day to give the casters pause.
The problem might not be that casters are overpowered, but that they are *very* dependent on playstyle to a degree the texts aren’t making clear, and that many people are forcing themselves (or being forced) to deal with a magic system not aligned with how they would like to adventure.