I am planning a Pathfinder campaign which I hope to reach higher level play (at least 15+, maybe up to 25th level.). I see a lot of concern about spell casters in 3.0/3.5/Pathfinder becoming "god-like," and leaving melee classes with nothing to do.

However, having recently read https://rpg.stackexchange.com/a/9054/33272 (fighter player said casters have a harder time at high levels), I am not sure I am going down the right road any more.

I was going to institute the following changes:

  1. Classes that have access to 9th level spells eventually (Druid/Cleric/Sorcerer/Wizard/etc) would use the spell progression tables from Dnd 5e. This is a slower spell progression, particularly starting around 13th caster level, that generally only gives players 1 slot for 7th, 8th, and 9th level spells. I would still allow bonus spells on top of that, though, and would add 10th, 11th, etc. level slots from 19th level and beyond, which could be used for lower level spells.
  2. I would use the limited magic system (http://www.d20pfsrd.com/magic/variant-magic-rules/limited-magic/) which keeps lower level spells from scaling with caster level unless the caster uses the "Heighten Spell" feat. To compensate, I am giving Heighten Spell to all such casters for free.
  3. I would also add "Constitutional Spell" feat (metamagic): For +1 level spell adjustment, use your full ability modifier to calculate DC in limited magic system, again for free.
  4. Ban a few level inappropriate spells like Flesh to Stone.

In summary, the first 2 rules move the system to be closer to DnD 5e, stopping short of going through every spell and deciding if it should use a DnD 5e concentration mechanic.

After reading the post from the fighter above, I am wondering if magic is still such a problem. I had played a high level 3.5 campaign many years ago, and my memory is a bit faded, but as I recall my players and I were just not smart enough to break the game at the time with Craft Contingency (I don't recall this existing at the time), Chain Contingency, Foresight/Celerity/Timestop combinations, so we didn't seem to run into any issues. Intuitively, it seems a caster that can cast Wish/Miracle multiple times per day might be game breaking, but is it?

So my question is: Is there really a problem I need to be fixing via house-rules (with the aim of class balance and keeping the game playable), and if so would the above rule changes fix it, or at least make it more balanced and playable? The goal is to keep the game going into the low 20th levels if possible.

For reference:

Is the old "Linear Fighters Quadratic Wizards" problem still around in 5e Basic?

How are Fighters Linear but Wizards Quadratic?


Example 14th level wizard with Int 22 using above rules.
Base Slots DnD 5e (1-7): 4/3/3/3/2/1/1
Bonus Slots: 2/2/1/1/1/1
Total slots: 6/5/4/4/3/2/1
Pathfinder : 6/6/5/5/4/4/2
So above system has 2 less 6th level, 1 less 7th.

20th Level Wizard, Int 30
Base Slots DnD 5e (1-9): 4/3/3/3/3/2/2/1/1
Bonus Slots: 3/3/2/2/2/2/1/1/1
Total slots: 7/6/5/5/5/4/3/2/2
Pathfinder: 7/7/6/6/6/6/5/5/5
So above system has 2 less 6th level, 2 less 7th, 3 less 8th, 3 less 9th.

Below Examples use the 14th level Wizard above.

FireBall, 3rd Level Blast
Normal: 3rd level slot, 10d6 dmg, DC 19
Limited Magic: 3rd level slot, 5d6 dmg, DC 14
Constitutional: 4th level slot, 5d6 dmg, DC 19
Heightened to 6th level: 6th level slot, 11d6 dmg, DC 19 <-- Probably your best option.
Heightened and Constitutional: 7th level slot, 11d6 dmg, DC 23

Compare to Best 7th Level Blast Spell, but only one target
Finger of Death, 7th Level
Normal: 140 dmg, DC 23, save for 3d6+14
LImited, 130 dmg, DC 20, save for 3d6 +13 dmg

-OR- Circle of Death, 6th Level
Normal: 14d4 HD of creatures, none can be higher than 9hd. DC 22
Limited: 11d4 HD of creatures, none can be higher than 9hd, DC 19

Cone of Cold, 5th level
Normal: 14d6, DC 21, save for half
Limited: 9d6, DC 17, save for half
Heightened: 7th level slot, 13d6, DC 20

Save or Lose Example (In DnD 5e, these would require concentration):
Note all you need is the 5 round duration in most cases.
Slow or Hold Person, 3rd Level
Normal: 14 round duration, DC 19
Limited magic: 3rd level slot, 5 round duration, DC 14
Heightened to 6th Level: 6th level slot, 11 round duration, DC 19
Constitutional: 4th level slot, 5 round duration, DC 19 <---Probably your best option.
Heightened and Constitutional: 7th level slot, 11 round duration, DC 23

Black Tentacles, CMD or screwed
Normal: CMB 19, escape CMB 29 or grappled, duration 14 rounds
Limited: CMB 12, escape CMB 22, duration 7 rounds
Heightened to 6th Level: CMB 18, escape 28, duration 11 rounds
Constitutional: adds nothing for this spell

Summon Monster Line
All that seems to be affected here is the duration. Extend spell would come out cheaper then heighten, when its needed. So maybe a problem still here, however note that using the monsters spell abilities would also be nerfed by above.

To be honest, I am surprised how well the numbers seem to be working out! The bottom line is a lot of decent spells are getting moved into the 6th-9th level slot area, to still be killer, and this is where slots have been reduced the most by using the 5e slots. If you don't agree these are good spells, the point is 3rd level spells are being moved into the 6th level territory, and as you fight harder creatures, you will have to keep using a decreasing pool of ever higher level slots to be as effective.

  • \$\begingroup\$ How are you dealing with specialist and domain spell-slots? (Not that this is a huge part of the problem, but your proposal does leave it unspecified.) \$\endgroup\$
    – Topquark
    Commented Mar 4, 2017 at 21:25
  • \$\begingroup\$ @topquark They would still get those slots, and bonus slots, and slots from anywhere else. \$\endgroup\$ Commented Mar 4, 2017 at 21:58
  • 2
    \$\begingroup\$ Actually, point of order here: what rules are you using to take the game past 20th? Paizo has not published any rules to do so, because even they know it wouldn’t work. \$\endgroup\$
    – KRyan
    Commented Mar 5, 2017 at 4:54
  • \$\begingroup\$ Essentially Epic Level Handbook: +1 bonus to all saves / 2 levels, + 1 bonus to attack/ 2 levels. And 10th/11th etc spell slots at levels beyond 20th, without other slots increasing further. I am only trying to get the game, as I said to 24th or so and be playable. People want to be able to try out the highest power stuff, at least for a bit. \$\endgroup\$ Commented Mar 5, 2017 at 5:09
  • \$\begingroup\$ I am wishing someone just had a link, "Oh, Bob already did this, a lot of people tried it, spells removed, balanced, but still Pathfinder, only a short list of changes, here is the link!" But alas. Unless someone posts something Earth-Shattering, I am going with just point 1 for now. At 15th level, my group can revisit the other points. I had to remove some stuff also from question like the esoteric components (d20pfsrd.com/magic/variant-magic-rules/…) as it is distracting. \$\endgroup\$ Commented Mar 5, 2017 at 5:31

4 Answers 4


Between chat discussion and the various edits to the question, I have come to the conclusion that, more than anything else, your suggestions represent premature optimization—you are trying to preemptively fix problems that haven’t happened yet and you are unfamiliar with.

Pathfinder is a system with a lot of design problems, particularly as one scales up in level. But design problems only represent risks of actual problems. There are tables that play Pathfinder up to high levels and never notice any problems—either through sheer dumb luck, or from simply being sufficiently wrapped up in the enjoyment of the story, characters, and social activity of the game (read: things that no system can really claim credit for, but are ultimately the reason we play) to be bothered by the problems. There are many other tables for whom the design problems pose real, serious problems at the table, at which point the design problems of the system are inhibiting the table’s enjoyment of the game—but even then, plenty of tables will enjoy themselves despite the system, because the good things about the game overshadow those issues.

The reason to be aware of design issues, then, is to make your life easier—to be aware of the pitfalls ahead of you, and to find the source of problems you do encounter. Sometimes the answer to “why is my Pathfinder game having this problem?” is “because of some flaw in its design, which you might improve by...” And then, having experienced that problem or ones like it, understanding the system, you can change the design to try to eliminate some of these flaws.

But in your case, you are playing your first Pathfinder campaign. It is just starting out, and you are still at mid-low levels. The problems with the system, so far, have not been problems at your table. The risk of these design problems becoming problems at the table will increase as you go up in level, and at the highest levels (to say nothing of the post-20th levels you propose), the risks become large enough that most tables will have some problems, but right now, you’re doing fine.

That means you are trying to fix problems you think you might have—and I think you might have them too, but ultimately it’s impossible to judge an appropriate fix for a problem you haven’t yet seen. The best advice anyone here can give you, then, is I think that you should play your game, wait and see how things turn out, and adjust things as necessary when and if problems actually occur. It’s well-and-good to avoid problems in the first place, but in this case you’re too far out ahead of the problems and so cannot judge the appropriate remedy.

Nonetheless, since you asked, here are my thoughts on your suggestions, and how they do or do not address the design problems of Pathfinder. By addressing a design problem, I mean how well they limit or eliminate the risk of running into actual problems at the table, ideally while at the same time making more character styles viable and reducing the difficulty of building a functional character and avoiding character-creation traps.

However, having recently read https://rpg.stackexchange.com/a/9054/33272 (fighter player said casters have a harder time at high levels), I am not sure I am going down the right road any more.

  1. The post you link from that fighter is flat-out wrong. It’s just not true. I’ve downvoted it, but there’s nothing more I can do. The fact remains, it’s a bad answer, no matter how well-written or how much people wish it were true, because it’s simply inaccurate. Sorry.

  2. The only true way to fix the caster-martial imbalance is to go through all the spells, and rewrite most of them. Good luck with that.

  1. Classes that have access to 9th level spells eventually (Druid/Cleric/Sorcerer/Wizard/etc) would use the spell progression tables from Dnd 5e.

The 5e spell progression change is a good idea, but it’s not enough. The very existence of some of those high-level spells is in-and-of-itself a problem (and personally, I am not inclined to do enough work myself to determine exactly which).

  1. I would use the limited magic system

Limited magic is an awful rule. It will certainly hurt spellcasters a lot, but it also causes a lot of problems with the game’s math (you will basically never use any spell that allows a save ever, because odds are your target will save). Plenty of spells also only become relevant with higher-than-minimum caster level (like the ever-popular grease), which now just become traps.

Which would all be fine (spellcasters can still certainly be extremely powerful even in this system), but I don’t like the warping it causes. All of a sudden, all kinds of iconic and popular narrative tropes for spellcasters become impossibilities. Spellcasters are forced to all use the few spells that remain amazing even under those rules—which means you still have overpowered spellcasters, you just have less variety of them. That’s not great. Even if you can get around the limits of limited magic with Heighten Spell (which still works poorly because the game’s math assumes ability scores much higher than the minimums to cast a spell), why should anyone when they could just use other spells?

And that is the real problem, and the reason why no broad-stroke band-aid fix is going to solve the problem—there will always be another spell that is still overpowered under your new rules. Instead, these kinds of approaches just cause wonky problems, skew what internal balance does exist, and fail to actually fix the original problems completely.

The rule smells to me very strongly of something written by someone who heard spellcasters were overpowered, but didn’t actually understand the problems. Someone who didn’t want to do the hard work of going through every spell, and wanted an easy way out.

There is no easy way out. There is no way to apply a simple, broad-stroke rule to Pathfinder to fix its balance problems. There is no way these problems will ever be fully handled short of actually going through every spell in the game.

  1. I would allow the use of Esoteric Components

Personally, I consider estoteric components just the epitome of un-fun, and would probably avoid using them. I haven’t gone through the list of components, but I would also be leery that there are some that are overpowered, just adding to problems with spellcasters.

  1. I have doubled the cost of meta-magic rods, and made them full-round actions to use.

OK; it’s not a terrible idea but it’s not going to fix the big picture.

  1. I have just disallowed summoners all together.

The unchained summoner is actually a nerf, and is at least better than the original one is. Ultimately, as unreasonable as the (original) summoner was, though, cleric and wizard and so on were always still much better.

  1. As a side note, I have made the changes suggested here: Balancing Spellcasters; Spontaneous Casters versus Prepared Casters, in which spontaneous casters can use an immediate action to use metamagic, and use the same progression as wizards.

A note: that was written before the arcanist was. The arcanist is an absurdly overpowered class with the only caveat being the delayed spellcasting progresion. An arcanist without that suddenly becomes clearly and without question the strongest class in the game.

Honestly I suggest you just ban it, too.

To be honest, I am surprised how well the numbers seem to be working out! The bottom line is a lot of killer spells are getting moved into the 6th-9th level slot area, to still be killer, and this is where slots have been reduced the most by using the 5e slots.

Those aren’t “the killer spells.” In fact, of those you list, black tentacles is the only one that’s even good, much less killer.

And no, I am not going to counter it with my own list of killer spells. That isn’t the point, and would only produce more arguments.

So my question is: Is there really a problem I need to be fixing via house-rules (with the aim of class balance and keeping the game playable), and if so would the above rule changes fix it, or at least make it more balanced and playable?

Pathfinder as a system breaks down on its own and without any particular intent on doing so somewhere around 15th level. As in, it becomes difficult to keep the game workable even when people are trying to. There really isn’t a lot that can be done about it without a serious overhaul to the system: houserules, any houserules, are not really sufficient to the task.

What you propose here is a series of broad-stroke band-aid fixes, some good, some bad. Overall, is it an improvement? Probably. Does it actually fix the problems? Emphatically no, it does not.

The best way to avoid these problems in Pathfinder, in fact the only way to avoid these problems in Pathfinder that I have found, is to simply not allow it to happen in the first place: sticking to low levels tends to result in a much better game. Pathfinder is still imbalanced even at the lowest levels, but the problem is less than it otherwise would be.

If you are really intent on a big, reality-changing epic campaign, you can still do it at relatively low levels. For example, in D&D 3.5e, the Eberron campaign setting mostly maxes out (barring the continent of deus ex machina dragons) at around 12th or 13th level. That means you only have to be 10th-ish to be operating at the highest echelons of power in that setting.

Alternatively, the E6 houserule is quite popular. The idea is that because 7th-level introduces 4th-level spells (which are the beginning of the “mundanes can’t even be in the same room” levels of power disparity in Pathfinder), we stop at 6th level. However, characters continue to gain XP, and every basically gain bonus feats at certain XP thresholds (when they otherwise would have leveled up, or every \$x\$ XP, or whatever).

I have heard of games using similar concepts, but stopping later than 6th. I would suggest that 12th is about as high as I would recommend, but that can work too. I have also recently started an interesting game wherein people start as a 6th-level commoner, and then further levels are added as with gestalt characters, so the levels you take “upgrade” the commoner levels without actually resulting in higher level (we haven’t gotten that far yet, but the plan is that when we would have otherwise hit 7th level, we would actually add yet another layer of gestalt).

Finally, Pathfinder as a system works much better if you restrict the classes available to a certain “power range.” This is the point of the tier system, which applies at least as much to Pathfinder. In fact, the best games of Pathfinder I’ve played are ones in which everyone was playing a class that receives 6th-level spells. This is a decent “sweet spot” where everyone gets “enough” magic to be competitive in the extremely-high-magic world of Pathfinder, and no one gets “too much” magic where they are bringing in far-too-powerful spells.

But no, I don’t think that Pathfinder works well as a system past approximately 15th level. From my own anecdotal experience, the games in which it does are either being played by people who know the system very well, and are carefully not breaking it, or groups in which everyone plays at 15th level as if the game hadn’t changed since 5th level—that is to say, people are ignoring the huge growth in power available and going with old standbyes that are long-since obsolete. But the former is difficult to do, and the latter is a delicate, fragile balance easily destroyed by someone trying something new, and also seems to ask why you are even bothering with high-level characters in the first place.


A lot of what happens at high character levels depends on your players. How much preparation time are they willing to spend on their character? Are they going to go read the optimization boards and make a spreadsheet and build something heavily optimized, or are they going to just take "whatever looks cool" at each level? Nearly any class can be effective if optimized -- for example see this post about bards -- but that's only if your players, y'know, try.

I recommend avoiding house rules where you can, simply because it adds more overhead for players to keep track of. A fifteenth-level game is already super complex, and your players won't thank you for adding more things for them to think about. Every time I've tried to add house rules to handle some anticipated problem, it turned out that the anticipated problem wasn't real, but the house rules were a problem in-and-of themselves.

A lot of the "Quadratic Wizards" problem comes from the daily-powers thing. If your campaign is structured so that there's only one battle per day, then your characters with daily powers will use all their powers in every fight, which will naturally make them very effective. So, one good fix for the problem is simply to structure your plot so that the group has to do multiple fights per day. Keep adding fights until you feel the spellcasters are on par with the fighters.

The other thing I recommend doing is to be willing to address problems as they come up. If someone's character is clearly too powerful, talk to them after the game and ask them to work with you -- either to find some house rules to bring them back in line with the group, or to retire their character and bring in a different one that won't cause problems in that way. When I've done this, I've found that players are usually pretty understanding about it -- they know as well as I do that their characters are causing problems.


Is there really a problem I need to be fixing via house-rules (with the aim of class balance and keeping the game playable),

KRyan posted an excellent answer, but there's one important caveat he doesn't address. Class balance only matters if it impacts players' fun. It doesn't matter if a theoretical cleric at 20th level can out-everything a barbarian, so long as the cleric at your table isn't stealing the fun from the barbarian sitting next to him. This could be because the cleric's player made less-than-optimal choices, or likes just being the person to patch everyone else up, or is deliberately not encroaching on the barbarian's competencies. Likewise, if the barbarian's player is happy whenever they can pull off their uber charge-pounce-smash-kill combo of abilities, regardless of whether they're doing more damage, or they're more interested in becoming a hero of their tribe, or just enjoy being the tactical mastermind at the table, then they can have lots of fun even if the cleric is the more powerful and versatile character.

A similar thing applies to spell casters and the GM: as long as you are having fun, it doesn't matter if the spell casters can trivialize an encounter. Maybe you prefer the world-building aspect, running flavorful NPCs, or just enjoy when your players are having fun.

Basically, class balance is a design problem, which may or may not impact the fun at your table. Each group is different, with different personalities and preferences. If these house rules will make things more fun for you and your players, go for them. If not, leave them out. But either way: have fun.


Fabricate is a 5th level spell. It allows casters to trivially double their wealth each day (per casting), until their wealth is not easily expressed in hundreds of cubic feet of Adamantine. No GM discretion is involved the value of material created.

A rogue of at least 5th level or so (or equivalent) can buy and sell most anything anywhere, using abilities like Black Market Connections. No GM discretion is involved in the purchase or sale of goods through such abilities (that's kinda the point...). In fact, a rogue with a sufficiently high Diplomacy skill is one of the only ways the players have to force artifacts of their choice into the campaign. Regardless, the rogue can trivially sell the proceeds of the fabrication spell at full value or higher and purchase anything with a price, in this case barrels of Pearl of Power IX and giant expensive diamonds (depending on the exact size of a 25000 GP diamond and whether or not it's possible to make smaller more expensive ones, you may wanna fabricate-clone those instead).

Wish is generally regarded as a pretty decent spell for its level, and doesn't really depend on CL, saves, or much of anything else. With arbitrary numbers of Pearls of Power IX, and arbitrary Giant Expensive Diamonds, a 17th level caster can cast arbitrarily many wishes (Well, ok, it's actually only 14,400 if they suck at being a caster, but if you throw in some Time Stop s it's truly arbitrary).

This is not intended to point out that Fabricate is a broken spell, nor that rogues are OP, nor that Wish is stupidly powerful. This is intended to point out that none of your changes address the real problem, which is that casters have access to spells.

The spells themselves, for a plethora of reasons, are too powerful. Ban one spell and another will rise in its place. Destroy even a category of spells and it will not be enough. If you banned everything but evocation spells I'd still use Consecrate to defy the very Gods. Heck, ban casters and I'd use magic items and monsters replicating the effects of their spells to easily outpace most characters.

This is not to say that high-level Pathfinder/3.5 'doesn't work'. It works, it's just very very different from low level play. At some point everyone has access to an unlimited number of actions per what-would-normally-be-a-round and at very least the ability to cast Wish with each of those infinite actions. Adversaries don't really make since as a thing. Either you would take them out without any effort or they would take you out and there's not really any good way of telling who would go first since everyone is doing everything as an out-of-turn free action or not an action or something else that lets them do it now whenever now might be.

Instead of the normal fantasy exploration game, the game becomes a form of restricted world-building, as the players recreate the universe in a form that better fits the rules. Players who are playing for combat or the story or other things that have faded away into the old world will probably not have fun. People who like building stuff in Minecraft might, though. It's sort of the same kind of putting-pieces-together-to-make-new-stuff game.

Regardless, your changes change nothing but what pieces the players can play with at high levels, and in that since they are strictly negative since they limit the ways pieces can be combined as compared to the original rules while at the same time not in any way changing the fundamental flashpoints.

  • \$\begingroup\$ <comments removed; please avoid making disputes personal> \$\endgroup\$ Commented Mar 8, 2017 at 19:39
  • \$\begingroup\$ @SevenSidedDie I was trying to request an edit to the question as suggested in meta. Based on the OP's reaction I figure I didn't do a good enough job of avoiding offense, but I really didn't think I had made the dispute personal (I was trying to avoid being offensive by being as impersonal/detached as reasonable, actually!). Do you have any advice as to how I could have done this better/can do it better in the future? \$\endgroup\$ Commented Mar 8, 2017 at 19:45
  • \$\begingroup\$ Attributing the dispute to the mental state or preferences of the other person is inherently personal — the statements become about them, rather than about the subject under dispute. An imperfect heuristic I use is to watch out for sentences I write that have “you” in them — those aren't the only way to write something personal, but they're the first place I check for such things. \$\endgroup\$ Commented Mar 8, 2017 at 19:50

Not the answer you're looking for? Browse other questions tagged .