Rules Savy Mathematicians/Magic Buffs Help Me Out Please ;)

[rootbeergnome]

Greetings EnWorld Addicts!! I have decided that magic is too trustworthy, and too structured in my games. So I am trying to develop a house rule on the way magic works for a future game. I was wondering if you kind folks would be so nice as to help me out with some of the particulars . So, here is what I have decided so far:

1.) The point of this system is that I want to avoid using spells per day/memorization and have casters potentially be able to cast spells to their heart’s content, as long as they are willing to pay the potential price and know the spell. This means that casters would have to weigh the risk and calculate what really needs to be cast and what should be avoided unless there is an emergency.

2.) To cast spells you must make a skill check, spellcraft or concentration (help me decide) that is modified by your relevant ability modifier (likely int, or wis) and is also modified by the level of the spell you are attempting to cast (so you might have a small increase to the DC of the check for 1st level spells, but you would have a HUGE increase to the check DC for 9th level spells.)

3.) Each time you cast a spell, you get a penalty that will apply to all subsequent checks for the rest of the day. These penalties due to “mana drain/spell fatigue”(need a cool name) will not go away until the next day. These penalties ONLY apply to the spell casting check rolls. So, lets say you have a +10 to your spell check, and you have cast a level 2 spell early in the morning after you get up. Well, since you cast the spell, it takes something out of you, making it harder for you to pull up the necessary energy for the rest of the day, so the next time you cast a spell it receives a negative modifier (like -2 or something like that) and so on, accumulating more penalties each time you cast that day. Should the penalty to casting subsequent spells be based upon the level of the spell cast previously? Should it be dependent upon the level of the caster? Both maybe? Im not sure, that’s where you guys come in, I know that some of you are hardcore mathematicians, help a guy out here . I need to know what the DC would be based on level of spell and level of magic-user (their ranks in the appropriate skill). I just want it to be appropriately difficult and not overpowered in any way.

4.) If you fail to make the check, then there is a resulting magical mishap. Things like the spell affecting you or an ally instead, or acting in an unexpected and often ineffective manner, or maybe it drains your con or wis. This could all be dependent upon the school of the spell (having a separate mishap list for each school of magic, help me out please ) or could even be directly related to the spell itself. (like polymorph turning the caster into something really nasty, like an ooze or something) Any ideas for a list would be awesome as well. But I especially like old books and movies about magic going wrong, for instance, shmendrick (sp?) the magician from The Last Unicorn, he was always messing up. But it could make for some interesting game moments, and could even lead to a caster sacrificing himself for his party because he just “had” to cast that spell that was way too dangerous for him to cast, in order to save the rest of the group.

5.) Finally, has this already been done? If so, does anyone have a link or could the author give me some info on this system? I am not trying to sell this or anything, I just want to use it for my own purposes in private games. Thanks in advance for all your help folks, I greatly appreciate the community that is ENWorld!!
Sincerely, RB Gnome

 

[Vrecknidj]

I've seen people try to work out such systems. Generally, I've seen them quit too, for a variety of reasons. The first reason is that many players get upset because this introduces a weakness to a class that, especially at low levels, is already easy to compromise (take away the spell component pouch or spellbook, and the wizard is in deep trouble).

Another reason people give up on this is that the player will just take all the skill ranks necessary (and any extra feats) to boost the skill points to the max, to avoid failing this check. So, you've kinda forced the player to assign skill points (it would be like having lots of jumping over pits in every single adventure, people would feel compelled to take ranks in jump).

That said, it's probably possible to introduce a spell-point / failure-chance system, and working with skills is the way to go. There's a skill-based magic system for d20 Modern, for casting incantations, but that's getting at something that's different than what you want. In the Star Wars RPG, most of the Force powers are based on skill checks, and different DCs give different results. So, that's an option too (i.e. the wizard casts fireball, a DC of 20 results in a normal fireball, a DC of 15 results in a fireball with one less die of damage, a DC of 30 results in a fireball with one more die of damage, etc.).

Dave

 

[JimAde]

One way to avoid the abuses of a skill-based system is to use caster level checks instead of skill checks. I'm doing this now for arcane spell failure (with the numbers rigged so casting your top-level spell has the same chance listed in the books). It seems to work well.

So you could have a caster level check with a DC of, say, 5 + (2*spell level). Each time you cast you add to a running tally that increases all your casting DCs until you rest.

Although it's for GURPS, the Unlimited Mana system does exactly what you want. You might be able to mine it for ideas:

http://www.io.com/~sjohn/unlimited-mana.htm

You know that's the second time today I've recommended something from S. John Ross. He doesn't even do d20 as far as I know.

 

[Angel Tarragon]

Rootbeergnome, what you are describing sounds very much like the vitalizing variant presented in Unearthed Arcana.

 

[Morgenstern]

Check out Dark Inheritance's spell casting system - it hits a lot of your goals in one pass. It should be in stores by next week.

 

[Li Shenron]

To develop your idea with less trouble, I think it is obviously useful to start from the Sorcerer core class, since with its spontaneous casting it's the closest thing to what you want to achieve. The aim is to modify the Sorcerer to work as you wish, while at the same time trying to keep the average amount of spells per day the same. The difference will be that instead of having a GRANTED number of spells/day you'll have an AVERAGE number. It will also be more versatile since you could cast less higher-level spells or more lower-level ones.

Using sorcerer as a base also mean to use its table for number of spells known and access to higher level spells at even class levels.

Some spare thoughs...

I definitely second the suggestion of using a caster level check instead of a skill. The reasons are (1) any spellcaster is anyway going to max the skill out, (2) there are much more many ways to increase a skill than caster level and (3) multiclassed spellcasters will be more complicated to handle.

I think you should start without mishaps at first, otherwise you're going to be in serious trouble when balancing the numbers! First just assume that a failed check means you waste the action, and the penalty to further casting still accrues. Or if you really want mishaps already, consider having them only an a natural 1.

Keep numbers easy to remember For example, DC 10 + spell level is easy to remember, as is an accumulating penalty of +1 per casting.
With these numbers a 1st level caster has 12 attempts per day to cast a spell. Each attempts has the following chance (hopefully close to right, or at least not too wrong ), assuming he attempts cantrips - otherwise reduce chance by 5%:

60%
55%
50%
45%
40%
35%
30%
25%
20%
15%
10%
5%

Since the DC increases by 1 both on successes and on failures, the average number of cantrips she can cast daily is

0.6 + 0.55 + 0.5 + 0.45 + 0.4 + 0.35 + 0.3 + 0.25 + 0.2 + 0.15 + 0.1 + 0.05 = 3.9

that's pretty low even with the base DC I have set. You can easily increase this, but the problem I see is that every caster level increase has the effect of increasing your first-cast chance by 5%, consequently giving you only 1 more casting chance per day. For example, going up to level 2 actually gives you an average 0.65 spells per day more, going up to level 2 another additional 0.7 spells per day and so on, which is quite few :\

 

[JimAde]

Good ideas, Li!

Warning: Long post ahead

The Unlimited Mana system I posted about has the idea of a "tally". This is a running total based on the spells you have cast. If your tally stays below a pre-set threshhold, you are all set. No mishaps and any failures are just failures (though they still raise the tally, IIRC). But once you cross that threshhold, failures can start to do bad things. The more you exceed the tally, the worse the mishaps get. Your tally naturally goes down over time while you are not casting.

This could be adapted to d20 by having higher-level spells increase the tally more. Your threshhold would go up based on your caster level. You would still be limited in the highest level spell you can cast by your caster level as well.

The problem is that this is just a Spell Point system in disguise. You will still tend to get casters casting more higher level spells and many fewer low-level ones. If that's OK with you, I think this could work.

So say every spell adds its level * 2 to the tally (cantrips would add 1). You would calculate the threshhold based on the existing spells per day for the class. The tally could drop by (thresh / 8) per hour of sleep.

You could also add feats to increase a caster's threshhold or even reduce the tally increase by 1 (so cantrips would add none at all).

So the sorcerer would look like this (pardon the lousy formatting):

[font=Courier New]Level Max Spell Thresh   Recovery/hour[/font]
[font=Courier New]	   Level			   of sleep[/font]
[font=Courier New]1		1		11		1.4 (2)[/font]
[font=Courier New]2		1		14		1.75 (2)[/font]
[font=Courier New]3		1		16		2 (2)[/font]
[font=Courier New]4		2		24		3 (3)[/font]
[font=Courier New].[/font]
[font=Courier New].[/font]
[font=Courier New].[/font]
[font=Courier New]19	   9	   558	   69.75 (70)[/font]
[font=Courier New]20	   9	   576	   72 (72)[/font]

This progression lets the sorcerer cast exactly what he can cast now at no risk. If it were me, I'd actually reduce these numbers a little so he can safely cast less, with the added benefit that he can exceed that limit at need.

Threshhold, recovery and tally can all be tweaked to get the power level you want.

 

[Verequus]

Li Shenron:

I've seen this idea before (but with separate DCs for each spell level) and I even asked my math professor to calculate the probabilites. I've forgot too much of the solution (and of the math itself), but I don't think, you can add the chances up for determination of successful castings. The probability of succeeding everytime is 0.6 * 0.55 * ... * 0.05. The more useful question is, how big is the chance to cast half of the spells successful. The first set is 0.6 * 0.55 * 0.50 * 0.45 * 0.40 * 0.35. The second would be 0.6 * 0.55 * 0.50 * 0.45 * 0.40 * 0.30. For the others, permutate through main set, and at last, add everything up. This results in horrible formula and that's why I have given up this solution.

Furthermore, I don't like it, if you could theoretical end up with no castings at one day. The classes are already balanced for a certain amount of spells per day - a solution, who would result on average to this amount would be a needless complication, and a solution, which wouldn't, needs to beef up the casters in other areas.

 

[Li Shenron]

I've forgot too much of the solution (and of the math itself), but I don't think, you can add the chances up for determination of successful castings.

I am not exactly adding probabilities, I am adding average values of independent variables and I think it makes sense. Incidentally it looks like I am adding probabilities, but the full form of that sum would look like

(1 x 0.6) + (1 x 0.55) + (1 x 0.5) + ... + (1 x 0.05)

I just omitted the "1 x"

The probability of succeeding everytime is 0.6 * 0.55 * ... * 0.05. The more useful question is, how big is the chance to cast half of the spells successful. The first set is 0.6 * 0.55 * 0.50 * 0.45 * 0.40 * 0.35. The second would be 0.6 * 0.55 * 0.50 * 0.45 * 0.40 * 0.30. For the others, permutate through main set, and at last, add everything up. This results in horrible formula and that's why I have given up this solution.

Well I thought the most interesting question was "how many spells per day you get to cast on average", but I might have been wrong...

Also be careful that if you go with that calculation you end up with the probability of casting exactly 50% of the spells (which is 6), but that's not useful at all. Maybe you meant the chance of casting AT LEAST 50% of spells? That one may be calculated as

1 - P(5 successes or less)

calculation is probably not trivial because the probabilities of each are different.

Furthermore, I don't like it, if you could theoretical end up with no castings at one day. The classes are already balanced for a certain amount of spells per day - a solution, who would result on average to this amount would be a needless complication, and a solution, which wouldn't, needs to beef up the casters in other areas.

It's not probably the original intent, but one could rule that the DC increases ONLY when the casting is successful, but not when it is unsuccessful. That removes completely the randomness of how many spells per day you cast, but instead leaves the fact that "evening castings" are going to take averagely more rounds to succeeds than "morning castings".

 

[Verequus]

I am not exactly adding probabilities, I am adding average values of independent variables and I think it makes sense. Incidentally it looks like I am adding probabilities, but the full form of that sum would look like

(1 x 0.6) + (1 x 0.55) + (1 x 0.5) + ... + (1 x 0.05)

I just omitted the "1 x"

I'm sorry, but I can't follow you - too less math practice...

Well I thought the most interesting question was "how many spells per day you get to cast on average", but I might have been wrong...

Also be careful that if you go with that calculation you end up with the probability of casting exactly 50% of the spells (which is 6), but that's not useful at all. Maybe you meant the chance of casting AT LEAST 50% of spells? That one may be calculated as

1 - P(5 successes or less)

calculation is probably not trivial because the probabilities of each are different.

Yes, at least 50%. But then you have a system, which has one to be a math major to understand it in all ways - and IMHO that's too much for a game like D&D.

It's not probably the original intent, but one could rule that the DC increases ONLY when the casting is successful, but not when it is unsuccessful. That removes completely the randomness of how many spells per day you cast, but instead leaves the fact that "evening castings" are going to take averagely more rounds to succeeds than "morning castings".

Yes, that would be better. It would simulate the mental stamina. But then you could make it dependent from Constitution, couldn't you?