Chad Perrin: SOB

12 October 2008

Replacing the Wizard

Filed under: RPG — Tags: , — apotheon @ 10:36

This is part of my RPG series of entries here at SOB. See the inaugural entry in the series for more details.

Dungeons and Dragons adopted a Vancian magic system (based on the novels of Jack Vance) early on, wherein spellcasters must “memorize” a limited number of spells at the beginning of the day and, once used, that “memorization” leaves the character’s mind. Gary Gygax reportedly chose this system of magic for its tractability in game design — allowing dramatic power to be wielded by spellcasters without unleashing game balance destroying Wizards on the game — and because it did not much resemble any real-world occult belief system.

Unfortunately, that decision completely ignored another important factor in game design: suspension of disbelief. There simply isn’t any reasonable way to explain the metaphysics of the traditional D&D system of magic — or, if there is, I haven’t come up with it in almost thirty years of thinking about it. That’s a crying shame, considering the basis of the system doesn’t really have to be tweaked very much to come up with something that at least has a system that lends itself more easily to suspension of disbelief.

Dealing With It

I’ve created a number of different alternate magic systems over the years, but there have also been many times when I’ve just decided to try to ignore the absurdity of D&D’s take on the Vancian magic system. Glossing over the ludicrous, inexplicable workings of the magic system behind the scenes with metagaming “I already used my last Protection from Evil spell” comments between players, so that players know to not have their characters rely on the party Wizard without having to try to figure out how the characters would talk about this resource shortfall, becomes the standard way to handle such things. Otherwise — what do you say? Does Magnar the Munificent say “I already cast that spell, so I forgot it”? Does the party’s Fighter then ask “Why the hell didn’t you do a better job of memorizing it?”

If I was the party’s Fighter, and the party Wizard told me “That’s just the way magic works,” I’d be positive he was pulling my leg.

“What?” I might say. “That doesn’t make any damned sense.”

Of course, some people might say “It’s magic. It can’t make sense, by definition!” That’s kind of ludicrous, though, considering that a Wizard supposedly studies magic. How does one study something that, by definition, doesn’t make sense? There goes the whole notion of the Wizard class.

Instead, you just have to metagame the problem away — a suboptimal solution, to say the least.

So Much For Game Balance

Unfortunately, D&D 3E introduced the Sorcerer to the game. Someone, somewhere, must have decided that the problem with Wizards had to be worked around, so that person came up with the Sorcerer class. The Sorcerer has similar limits on spells cast per day, but does not need to “memorize” spells. He or she can cast whatever spells he or she knows, up to the maximum number of spells. The number of spells one can cast is still limited to an arbitrary number based on level progression, and divided between spell levels. Thus, a fourth level Sorcerer can cast six cantrips today, plus six first-level spells, plus three second-level spells. That reduces some of the bookkeeping of playing a spellcaster and makes the spellcaster a bit more flexible, but doesn’t really solve the suspension of disbelief problem.

Wizards are still part of the game, too. The existence of Sorcerers, unfortunately, makes Wizards effectively obsolete. For general-purpose adventuring, Wizards are now entirely outclassed.

Some point to a greater “flexibility” in the fact that Wizards aren’t limited to a specific number of spells they can learn, but that flexibility pales beside the far more practical flexibility of being able to choose one’s spells on the fly rather than trying to guess at the likelihood of needing Knock today. Would you rather have five second-level spells than three of them, but only be able to choose two that you’ll be able to cast today, or have two that you know and be able to cast them each several times, at will, without having to worry about making sure that being able to cast one of them three times today will stop you from being able to cast the other enough times if things go poorly? Worse, in order to get those five spells, the Wizard has to actually learn them — whereas a Sorcerer’s newly known spells just appear, fully-formed, in his or her head.

Others claim that the fact Wizards can cast higher-level spells sooner than Sorcerers helps balance them out, but that’s an even weaker argument. Sure, a Wizard can cast fifth-level spells at ninth level, and a Sorcerer can’t — but he can only cast one a day, and it only goes up to two a day at tenth level. Meanwhile, at tenth level, a Sorcerer can suddenly cast three fifth-level spells. In other words, all this supposed advantage for Wizards means is that a Wizard has a slight advantage one level, and the Sorcerer then gets a staggering advantage the next — because the Sorcerer then gets to cast more of the highest-level spells both can cast in a given day, and more of the lower-level spells they can both cast, and doesn’t have to plan out his or her day in advance due to the ability to cast whatever spell he or she likes as long as the Sorcerer still has spells of that level left to cast that day.

The real value of the Sorcerer class, it seems to me, is to provide an alternate set of game mechanics for arcane spellcasters. I don’t mean a set of mechanics in addition to those for the Wizard, but mechanics that can be used instead of those for the Wizard, if you feel so inclined. Pick one or the other, and go with it — and ignore all the flavor text explaining how Wizards and Sorcerers are different. If you include them both in a single game, you’re just looking at a new way for Wizards to suck.

Adding insult to injury, the only way to make Wizards at all palatable power-wise is to make them specialists in a particular school of magic. Unfortunately, doing so cuts away some of their already paltry “flexibility”, and it also turns them into one-trick ponies in a way they aren’t as generalists. It’s bad enough supposedly learned men like Wizards get the shaft on skill points.

Variant Rules That Make Sense

So . . . there are two different arcane magic systems I have developed in the last year or two that can be generally applicable to D&D and Pathfinder games. One is meant to be a single option from among several different traditions of arcane magical study, though I have not developed the other options at all yet.


Alchemy is a set of mechanics I created for my Midian campaign, and is not really meant to entirely replace Wizards. I conceived of this set of variant rules first, and it’s meant to be part of a series of different Wizard types, with the generic from-the-book Wizards being replaced by a whole slew of different scholarly traditions of arcane magic. I call this system Alchemy, in the older sense of the term — where magical power is related to the improvement of the very essence of a thing, or at least the transmutation of that essence. Spellcasting is heavily ritualized in general, and there is no limit on how many times you may cast a spell you already know. In fact, Alchemists in this system do not really “know” spells the same way generic Wizards from the book do; instead, they have books with complex rituals described in them in some detail, and there are general principles that can be used even to develop new rituals on the spot.

The way it works, basically, is that you can:

  1. spend a bunch of time setting up a ritual and executing it to “cast” a “spell”; this involves inscribing a mystic circle, setting out candles or stones or some other crap, sprinkling stuff, intoning invokations, and so on
  2. do something similar, but slightly abbreviated and modified, to “charge” a later spellcasting; then, later in the day, you complete the ritual by performing some predetermined triggering action to release the magical energies that have been “charged”, producing the desired spell effect

The second option is an effective means of making the Vancian style of spellcasting somewhat more believable than as presented in the PHB. The first is a necessary component of a magic system that supports the in-game justification for the second, and it also provides Alchemists with a slight boost in flexibility of their power (especially with the capability of modifying alchemical ritual practices at the time they’re set up) so that they can be brought up to par somewhat with Sorcerers. In addition, the available power levels of spells/rituals is slightly improved, to help balance things out further — and Alchemists get more than 2 + INT Modifier skills per level, because they’re meant to be highly educated scholars, and because Wizards having only 2 + INT Modifier per level sucks.

Rituals that have been set up with a trigger to be completed at a later time last until you cast them or until you sleep, whichever comes first. A conscious mind is required to maintain such triggered effects in place, ready to be used. There is no theoretical limit to how many such triggered effects can be set up, though each one takes time (less time the higher the character’s experience level), thus placing a practical limit on how many can be charged up before the character must sleep again. For races that do not sleep per se, some other rejuvenating event takes the place of sleep as a means of blanking the slate so that characters cannot simply spend months at a time banking hundreds of spell effects for use in a single encounter. In some cases, it might even make sense to simply prohibit a certain race from employing alchemical magic in this manner.

Since I have yet to have a single non-Sorcerer spellcaster show up in person in my Midian campaign with the PCs around seventh level, there’s little need to develop the several different scholarly arcane traditions any time soon. The only reason Alchemy has gotten any significant attention, in fact, is the fact that I started writing fiction set in the Midian gameworld a year ago, and some of the characters are Alchemists.

I don’t have the Alchemy system fleshed out fully. It’ll be a fair bit of work to get it fleshed out so much, since developing it completely will involve a lot of boring minutiae like coming up with custom “spell” lists for the various alchemical rituals and a set of categorical principles of Alchemy from which new rituals may be developed — and I have to make the existing rituals and the categorical principles match. In other words, it isn’t done yet, but it will be.


The Mage class was developed just this last weekend. Unlike the system involving Alchemy, et al., this is just a single-class replacement for the Wizard class. In fact, it’s really just the Wizard class with some changes in how spell progression, casting, and memorization works. It even uses exactly the same spell lists as the Wizard class. It is not meant to coexist with the Alchemist and other, related arcane spellcasting traditions at which I hinted above.

With this system, spellcasting draws upon the inner reserves of the spellcaster, essentially in the form of physical damage (hit points). When casting a spell, a save is made, and if the player fails his or her character’s save, the character takes a number of points of damage equal to the spell’s level (or “magnitude”, as I renamed it for this system). That’s not the preferred means of spellcasting, however.

In addition to the above, the Mage can also prepare spells. A total number of spell magnitudes determined by the character’s level can be prepared at any one time. A prepared spell can be cast as long as it stays prepared, as a standard action (or different action type as appropriate to a particular spell, in nonstandard cases), without wiping itself from “memory” (since this isn’t a matter of memorization). Each time it is cast, however, a save must be made to determine whether the preparation fails, at which point one can no longer cast that spell except as described above for spells that have not been prepared in advance — complete with the danger of taking damage.

There are other details to the system, including a spell “mastery” mechanic that I won’t go into at this time. I’ll post details for the magic system used for this alternate class at a later date. The point is that, unlike the Wizard from the PHB, this variant class is actually functionally balanced with the Sorcerer class, and allows one to play a Wizard-like scholarly spellcaster class that can compete with Sorcerers on equal footing thanks to a completely different set of advantages.

Details Forthcoming

I plan to provide details of how these two alternate arcane spellcaster classes work, in terms of game mechanics, at a later date — and under the terms of the OGL. The Mage will appear first, as the essential differences from Wizards for D&D 3.5 are already worked out in some detail. It just needs to be polished up a bit with explanatory text and maybe some flavor text (a matter of a few minutes’ work) and provided with some additional material for integrating with the Pathfinder RPG rules, since Wizards are somewhat modified there.

The Alchemist will have to come later — much later — because it is nowhere near finished being specced out at this time. I have enough done to write fiction about the class, but not nearly enough to make and play an Alchemist character in an actual game, regardless of whether the game is D&D 3.5 or Pathfinder RPG.

I will, of course, make sure that as I get them posted, I have links to them from here. If the links do not appear near this paragraph in particular, they will appear in comments as “pingbacks”.


  1. This isn’t particularly related to this post, but I’ve been reading your D&D series of articles for awhile now and I wrote something yesterday that might interest you. It’s a look at how villain races in high fantasy settings are generally really stupid. Needless to say, it has lots of applications to D&D as well, where the relationship tends to hold. What’s your take on it?

    Comment by Cyde Weys — 15 October 2008 @ 08:11

  2. I think you made some excellent points. I responded there with my own thoughts, including a link to where I discussed related matters here at SOB a while back.

    I was going to submit that essay of yours to reddit, but I see you beat me to it.

    Comment by apotheon — 15 October 2008 @ 10:21

  3. Come on, the magic system is not that ludicrous, if you are already willing to accept the idea of magic. A wizard concentrates, channeling energy into certain patterns, ready to be released again with the correct actions. It’s not that you “forget” spells, although that’s what the book may call it. It’s more like you’ve fired all your bullets, and although you know where to get more, you can’t get any more at this time. This all fits a view of magic as a demanding practice; I consider it very similar to the wizards of Ankh-Morpork in Terry Pratchett’s books–as they say, the hardest part of lifting something with your mind is making sure your brains don’t fly out your ears in reaction. Having mana points, cooldowns, etc. is good for video games, but you don’t need it for pen and paper. I for one like the idea of a wizard filling his head with magical energies in the mornings, rather than the Harry Potter-ish “little effort, no consequences” of fiction and video games.

    Comment by John — 16 October 2008 @ 08:03

  4. It’s more like you’ve fired all your bullets, and although you know where to get more, you can’t get any more at this time.

    I’m sorry, I have to ask:

    Did you just decide to ignore all the stuff about arbitrary numbers of spells per spell level? I mean the part where I created two alternate Wizard-type classes that still use a Vancian style of magic, but don’t use asinine terms like “memorize” and “forget” to refer to what’s going on and don’t use arbitrary maximum numbers of spells per spell level that can be prepared for the day (which is, if anything, even more bizarrely unrealistic than using terms like “memorize” and “forget” to refer to what’s going on).

    I for one like the idea of a wizard filling his head with magical energies in the mornings, rather than the Harry Potter-ish “little effort, no consequences” of fiction and video games.

    I’m trying to figure out where you got the idea I was advocating a Harry Potter style of magic.

    Comment by apotheon — 16 October 2008 @ 11:47

  5. I love it. I’ve been trying to find for a long time a way for spellcasting to “fatigue” the user, much like my other favorite gaming system Shadowrun. Shadowrun has a lot of balance for casters (its illegal in most places, for one thing) but the stun damage system was by far the best tabletop magic system I had seen. Finally, a magic user had access to every spell on the list, and could cast them as much as they want (within limits). Your mage variant brings this almost perfectly into Dungeons and Dragons. I will definitely be using it in my next game.

    Comment by DM_Laguna — 18 October 2008 @ 01:03

  6. Thanks for the compliments, DM_Laguna. Keep an eye out for the actual Mage class write-up, when I eventually get around to finalizing it and post it online. I predict this happening by Tuesday at the latest.

    Comment by apotheon — 18 October 2008 @ 01:20

  7. […] Replacing the Wizard, I explained some reasons for reworking the Wizard class, including an attempt to restore game […]

    Pingback by Chad Perrin: SOB » Mage class: now usable as a Wizard variant — 19 October 2008 @ 12:43

  8. As far as magic goes, there’s one thing I’ve read (in fiction) and that, while being somewhat of a Vancian system, seems to work pretty well. I’m talking to the system of magic depicted in the Amber series of books (written by R.Zelazny.)

    The system shares some similarities. The idea is that the wizard will create (not memorize, in this case there are no set of spells) a particular effect, which is a process that happens completely in his mind. This, according to the fiction, is a rather complicated process, in which the mental strain and time needed will depend of the specificity and complexity of the spell. A good example (I guess) is to think of making a house in your mind: you’ll have to think about where each brick, pipe, door and appliance is going to go… then you have to remember the shape and qualities of each said items.

    Yes remember. Once the spell is created, the wizard can remember it and maintain it (this is essential) with a bit of effort. Spells don’t disappear while sleeping, but they require daily keeping in order to make sure that your ‘house’ doesn’t fall down. Of course holding these puppies in your mind is a bit of a bitch, not only that but the ‘house keeping duties’ are a time demanding task in themselves. Merlin (the wizard in question) often remarks that a wizard would hold 2 or 3 “oh shit” spells at the very most, and construct others as needed.

    As well as in D&D, the spells are ‘forgotten’ once used. Well that’s not quite the way I’d put it, but the effect is the same: once you’ve used the spell you’ll have to construct it again if you want to use the same effect. Doing this takes only time though, spellcasting is an entirely mental affair.

    A few caveats though:

    1- I remember reading the books and finding the magic part interesting and enjoyable (I immediately made a connection with D&D magic in that case.) However, as it’s a dated memory, this might reflect my enjoyment of the book (and of the author) a bit more than the pure enjoyment of the “magic system.”

    2- The system does not work like actual D&D magic (as far as 3rd edition goes) and has some very important distinctions.

    3- Due to the nature of the setting, the spellcasting was innately limited. The process of constructing a spell needed the information of the particular condition of the Shadow in question (eh, a sort of plane or dimension…) plus specifics about the target (you can do without some of this, but apparently the more info the easier it was to construct the spell and you could rely on its effectiveness.)

    4- A book is a controlled environment, so the fact that I felt it ‘worked’ can be attributed to that.

    That being said. I’ve always disliked the way wizards were usually handled in D&D. I’m glad to see this topic (and the newer document) come out.


    PS: Hey Chad, it’s been a long time ;)

    Comment by the_blunderbuss — 31 October 2008 @ 06:28

  9. […] months ago, I wrote about the need to replace the Wizard class for OGL/d20 […]

    Pingback by Chad Perrin: SOB » Wizard variant: Mage class (now in Beta) — 3 February 2009 @ 11:24

  10. The mage you’ve proposed is similar to what we worked into our campaign world.

    Spellcasting does damage equal to the spell level, but it is non lethal damage, which means it recovers faster, and can only render the spellcaster unconscious at worst. It was an attepmt at trying to simulate “fatigue” without creating an entire fatigue system.

    There is no save vs damage, it hampers lower levels and becomes mostly redundant at higher levels when their save is through the roof.

    As an aside, we immediately changed clerics to use a similar mechanics as sorcerers, it seemed to make sense to us that the cleric prays, their god(dess) answers, rather than having to “pre-pray” spells.

    Basically Save vs prayer, penalised by the number of prayers you’ve already made, and offset by a bonus/penalty based on your level vs the prayer level.

    Comment by Stuart — 19 February 2009 @ 11:31

  11. I initially considered doing a fatigue thing, but decided I wanted to start with a pseudo-Vancian approach using prepared spells, and move to damage when that ran out. For damage, I figured I should use lethal to offset the fact that the character can cast spells without taking any damage at all at first — otherwise, I think there wouldn’t be enough discouragement for using spells constantly, and the class would be overpowered.

    I have been kicking around some ideas for divine spellcasters, too, because I don’t like the Vancian mechanics for the Cleric class at all. I just didn’t get as into it, as quickly, because there wasn’t quite the same game balance urgency to it (there’s no divine spellcasting version of the Sorcerer making Clerics almost completely pointless). One of the ideas I’ve been working on does something with the frequency of requesting divine intervention, which I’m calling the “Pestering Factor” until I come up with a better term for it.

    Comment by apotheon — 20 February 2009 @ 12:10

  12. There actually is a sorcerer version of the cleric; it’s in the Dragonlance campaign setting. I believe it is called the mystic. It’s effectively replaced some clerics in my campaigns. >.>

    Comment by elvencloud — 3 March 2009 @ 01:42

  13. […] last year, I started talking about Replacing the Wizard in my Pathfinder RPG campaigns. I referred to one in particular that I gave the class name Mage, […]

    Pingback by Chad Perrin: SOB » Introducing the Mage Class, Release Candidate 1 — 14 October 2009 @ 11:11

RSS feed for comments on this post.

Sorry, the comment form is closed at this time.

All original content Copyright Chad Perrin: Distributed under the terms of the Open Works License