So I'm a huge fan of low magic settings, I know Pathfinder doesn't support that well, but blah blah I keep trying. What I've found myself doing is what I like to call "volatile magic" because while there is a lot of magic in the game, it doesn't always work well or predictably, and when it goes wrong it goes REALLY wrong. The goal is not necessarily to be balanced- this low magic business is obviously flavor, fluff etc, but I prefer things to be as balanced as possible. Part of the balance is on me: I need to design enemies and environments such that successful use of magic is REALLY powerful, and how exactly I do that on a session by session basis you will not know. Still, I'd really appreciate some folks critiquing these magic rules.
After you declare what spell you are attempting to cast you must roll a d10. If the number on the die equals the level of the spell, the spell is cast normally. If the number on the die is lower than the level of the spell, the spell fails and is lost. If the number on the die is greater than the level of the spell the caster overcasts, and takes damage proportional to the spell level: a first level spell does 1d4 damage, a second level spell 1d6 and so on. If the caster rolls a natural 10 they surge, the effect is proportional to the spell level:
-a first level surge deals 1d4 damage to the caster and the target of the spell
-a second level surge deals 1d6 damage to the caster, anyone within five feet, and the target of the spell
-a third level surge deals 1d8 damage to the caster, anyone within ten feet, and the target of the spell... make sense?
Point of clarification: if the caster is him or herself the target of the spell, yes they take double damage.
Casters have a number of points (lets call them Caster Points because I'm feeling lazy) they can spend per day to help them cast normally. This value is equal to their level plus their casting ability modifier plus the number of ranks they put into spellcraft. They can spend one of these points to modify their d10 role by one. They an also spend points equal to three times the spell level they were attempting to cast to avoid a surge (but the spell is lost, and they still overcast).
Now for some examples to help illustrate how this works. Let's assume we have a second level Witch who has put two ranks into spellcraft and has an intelligence of 16. She therefore has 7 caster points.
Example one: She casts burning hands, a first level spell, and roles a 4. She doesn't want to overcast, so she spends 3 caster points to cast normally.
Example two: She casts cure moderate wounds, a second level spell, and roles a 1. Easy, she spends one caster point and casts normally.
Example three: She casts burning hands and roles an 8. This is a harder decision; she can either chose to spend all 7 of her caster points to cast normally, or she can accept the 1d4 damage. If she has several enemies in her sights, or knows that she can kill one, perhaps its worth it to save those points for later.
Example four: she casts cure moderate wounds on her ally and roles a 10. She will do 2d8+2 healing to the ally, then 2d6 damage, and 2d6 damage to herself and anyone else within five feet. Odds are that she will still heal a little bit of damage, but she takes 1d6 damage herself, and maybe damages some allies. Then again, maybe she'll damage some enemies in the process. But what if she roles poorly? It really depends on her health, the health of the ally, and all the other particulars going on in the encounter. This is the kind of tough decision I'm trying to force on players.
Ok so I totally wall of texted. Sorry. If you're still here and want to give me some super harsh feedback please do. I'm likely to defend this system a little since there are nuances I haven't elaborated yet, but I really do want people to pick this apart so that I can turn it into something that works.
I really dig the idea and I'll read through it a little more later this evening when I have time. Kinda reminds me of the Conan d20 Runaway Magic (which can have some nasty effects if you cast too many high level spells within a certain period of time).
Any thoughts on tieiing either spellcraft or concentration checks into this? Perhaps a spellcraft check to avoid a surge, or to reduce the cost in 'caster points' to adjust a spell?
My concern is basically that the 'caster points' dont call based on relative skill with magic, a paladin with a good charisma and full ranks in spellcraft can have as many caster points as a sorceror. That seems off to me.
If anything the full casters should probably get some abilities to deal with this. Either make them better at dealing with this random magic then 6 level and 4 level casters, or give them other abilities they can use that are not subject to this. Because even a low level sorceror or wizard is going to run out of points, and if they do, they will have alot of trouble casting spells. The way they contribute to the party is through their spells, if they arent casting their spells they arent contributing, and they will be forced to concerve their caster points (and thus their spells) for when they really need it.
It would seem to me that is part of the goal, to force casters to concerve, but in that case, adjustments to the classes themselves is in order. If a wizard or sorceror cant be reasonably confident at casting all their spells without killing themsleves (IE they have at least a few caster points per spell) they should have fewer spells and more class abilities. The best examples are ofcourse the 6 level caster classes.
The bard, inquisitor, alchemist(not sure if alchemists extracts apply here but im assuming they are) and summoner, have lots of non-spell abilities that allow them to contribute to the party, and thus have fewer spells. The wizard, cleric, oracle and sorceror should be similarly adjusted so they can contribute when they arent casting spells, or just remove the classes and play with the mixed classes only (keep in mind in a low magic world, healing and condition removal without a dedicated divine caster can be a problem).
Also how will this work with spell like abilities or spellish super natural abilities like a paladins lay on hands or a clerics channel or a witch's hexes?
Any thoughts on tieiing either spellcraft or concentration checks into this? Perhaps a spellcraft check to avoid a surge, or to reduce the cost in 'caster points' to adjust a spell?
My concern is basically that the 'caster points' dont call based on relative skill with magic, a paladin with a good charisma and full ranks in spellcraft can have as many caster points as a sorceror. That seems off to me.
If anything the full casters should probably get some abilities to deal with this. Either make them better at dealing with this random magic then 6 level and 4 level casters, or give them other abilities they can use that are not subject to this. Because even a low level sorceror or wizard is going to run out of points, and if they do, they will have alot of trouble casting spells. The way they contribute to the party is through their spells, if they arent casting their spells they arent contributing, and they will be forced to concerve their caster points (and thus their spells) for when they really need it.
It would seem to me that is part of the goal, to force casters to concerve, but in that case, adjustments to the classes themselves is in order. If a wizard or sorceror cant be reasonably confident at casting all their spells without killing themsleves (IE they have at least a few caster points per spell) they should have fewer spells and more class abilities. The best examples are ofcourse the 6 level caster classes.
The bard, inquisitor, alchemist(not sure if alchemists extracts apply here but im assuming they are) and summoner, have lots of non-spell abilities that allow them to contribute to the party, and thus have fewer spells. The wizard, cleric, oracle and sorceror should be similarly adjusted so they can contribute when they arent casting spells, or just remove the classes and play with the mixed classes only (keep in mind in a low magic world, healing and condition removal without a dedicated divine caster can be a problem).
Also how will this work with spell like...
Originally I was going to make folks take a concentration checks to avoid surges/modify the role. When testing this it seemed like it was generally easier to avoid a surge then it was to change a role by more than one, plus I didn't like that you could do it an unlimited number of times per day (although an obvious solution is to say you can only perform such a check once per caster level or something).
Regarding classes: I have actually removed all full casters. The only caster classes I'm playing with are Alchemists, Druids, Clerics, Paladins, and Witches. Rangers must be skirmishers unless there is a very good in game explanation. Supernatural abilities like channeling or hexing are unaffected. This makes channeling incredibly OP so I also removed the possibility to start as a cleric, but will allow it as a prestige class. Since this does some crazy things to Sorc and Wiz I may role up enemies as higher level PCs with those classes. I want to test that out first.
I feel like there's an aspect of this mechanic that I failed to mention, which is that sometimes it's beneficial to surge. Lets assume that same witch casts burning hands and roles a 0. They could spend 3 to avoid the surge, overcast, and lose the spell, or they can accept 1d4 damage to do an extra 1d4 to each enemy the spell would hit. When this example came up while testing things with my friend he suggest I change the rules to make surges more powerful. Perhaps the spell is automatically maximized, and no saving throws are allowed. That would make it a lot more likely said witch would let the surge happen to heal her ally, and let the surge happen to burn her enemies. It makes a surged fireball extremely destructive. But maybe this is going to far? I wasn't initially a fan of this suggestion but it's grown on me a little.
Originally I was going to make folks take a concentration checks to avoid surges/modify the role. When testing this it seemed like it was generally easier to avoid a surge then it was to change a role by more than one, plus I didn't like that you could do it an unlimited number of times per day (although an obvious solution is to say you can only perform such a check once per caster level or something).
Maybe just something like that for just 9 level casters (Druids, Clerics and Witches?) I just think that the classes with more spells should be better at managing volatile magic.
Regarding classes: I have actually removed all full casters. The only caster classes I'm playing with are Alchemists, Druids, Clerics, Paladins, and Witches. Rangers must be skirmishers unless there is a very good in game explanation. Supernatural abilities like channeling or hexing are unaffected. This makes channeling incredibly OP so I also removed the possibility to start as a cleric, but will allow it as a prestige class. Since this does some crazy things to Sorc and Wiz I may role up enemies as higher level PCs with those classes. I want to test that out first.
I dont know that it makes Channel OP, it just makes it still good. Healing magic will be problematic, but the fact that you have highly restricted the caster classes is important, as it changes the context in which the rules are viewed. These other classes (except clerics) have pretty significant abilities besides casting spells and can contribute even if they dont cast a single spell in an encounter. That makes a difference.
I feel like there's an aspect of this mechanic that I failed to mention, which is that sometimes it's beneficial to surge. Lets assume that same witch casts burning hands and roles a 0. They could spend 3 to avoid the surge, overcast, and lose the spell, or they can accept 1d4 damage to do an extra 1d4 to each enemy the spell would hit. When this example came up while testing things with my friend he suggest I change the rules to make surges more powerful. Perhaps the spell is automatically maximized, and no saving throws are allowed. That would make it a lot more likely said witch would let the surge happen to heal her ally, and let the surge happen to burn her enemies. It makes a surged fireball extremely destructive. But maybe this is going to far? I wasn't initially a fan of this suggestion but it's grown on me a little.
I think maybe its necessary to differentiate the effect of a harmless/beneficial spell and a offensive spell when it surges. I think to counter balance the added point of failure a surge (even though you always take damage if you surge) should be beneficial to the spell itself. I would say instead of maximizing it (maximize doesnt apply to a great many spells) increase the caster level the spell is cast by say 1d4. Sure thats a big bump, but that is at the cost of spells going haywire all day and taking damage for the surge. I think in a system like this a surge should be a big moment (especially if its a high level spell for which the character is probably taking a ton of damage for). Maximizing it or even adding a little damage might be meaningless for alot of spells, or even prblematic (like with cure spells).
I'll field these points one at a time.
1) Assuming a cast puts one rank into spellcraft per level I think they should do fine. A druid capable of casting ninth level spells would be level 17. Assuming they have at least 18 Wisdom they should have 38 caster points to spend. OK, so preventing a surge of a ninth level spell takes up 36 of those, but what can you cast at ninth level? In a low magic world, where enemies can rarely use spells against you Storm of Vengeance or Mass Cure Critical Wounds pretty much wins the encounter for you.
Still, I see your point. Someone who has deeply studied the magic through 17 levels should know a lot about how to properly cast. I think I'll add a feat with a prerequisite of being able to cast a certain spell level (seven maybe?), allowing someone to take a spellcraft check at 10+twice spell level to avoid a surge once per day. There's the obvious possibility for an Improved ______ feat allowing for twice per day.
2) Yes healing will be hard. I'm THAT GM who kills characters a lot. I prefer combat to be super brutal and usually offer ways around combat for those willing to look. Scouting ahead, being diplomatic, and working as a team are key parts of this campaign. A character getting hit doesn't just mean they're closer to dying for this encounter, it also means they will likely start the next encounter with the same wounds. I have considered adding a house rule that a healing kit restores some small amount of health (i.e. your wound is patched up so you are less vulnerable), but I might just be mean and leave it.
3) With healing I feel that maximizing would be a lighter sentence than +1d4 caster level. I'd rather get 16 healing - 2d6 than 2d8+6-2d6. I initially didn't like it because it wouldn't do anything to spells with no die roll. For offensive spells disallowing throws would make them pretty powerful whether they benefit from maximization or not. Right now I'm toying with the idea of all three... making surged spells pretty extreme.
... I've been showing this to friends of mine who GM or are interested in game design. One of them suggested a rule in which a caster can use all of their remaining points to automatically surge. This is interesting to me since even one point could lead to a huge upset. Maybe that kind of power balances the every day slog of being a caster in this world.