You can type that without laughing? What spells have to be redone? Like Fireball balance against Magic Missile? Please. The spell descriptions and their so called "balance" is based on eyeballing the numbers. Show me 100 spells of 1st to 5th level and prove to me the "balance" is thrown off if it becomes 7.5% harder to reach the 20-21 range of a 4th level spell and then prove that that same 7.5% means wizards are not viable classes afterward.
*grins* Yes I can type that. I think that I am having a problem explaining all of the permutations that I see. You are just the 7.5% the spell check roll.
If you also modify the DC # (1st 13, 5th DC 17) you also adjust the % chance of spell success.
Look at pure averages for a caster with +1 ability modifier.
9th level caster
OLD: Spell check bonus of 10
NEW: average bonus of 7.5 (you)... if d11 then 6. (d11 is the mathematical extension of increasing dice. Roll a d12, re-roll 12's --- see below for more)
PS - I have not seen that a d11 (or d9) will not be utilized. A zocci die may not exist for it, but that does not mean it will not exist [Side note: Maybe that is the reason Joseph only wanted 5 levels... the dice did not go to 10 -- LOL]
OLD: Average roll: 20.5 (DC 5th level spell is 20)
NEW(d14): Average Roll: 19 (DC 17 for 5th level spell) [hitting 65% or 70% chance of spell success]
NEW(d11): Average Roll: 17.5 (DC 17 for 5th level spells)
With a d11 for a caster with +1 int, you are dead on with the approximate 45%/55% (fail/pass) that Joseph was striving for with the old system.
Now you look at the ranges you are going to hit on average at 9th level:
Some other interesting facts:
With the D14 method you can hit the highest of the spell effects on the current table. But by design, they are unreachable under normal circumstances (i.e. Int 18 or w/out spell burn). But I will give you that the chance of that occurring is very small. Because once you start to roll two dice, you start to get a bell curve of results.
So in a way two dice do prevent larger swings of results. It will result in a caster failing a larger number of spells than they would under the old system. Even a first level spell is no longer a gimme with a 5% chance of failure. You are now looking at a 20% chance of failure for a 1st level (apprentice-style) spell. You are no longer going to average in the 20's, but now in the teens for effects.
Compressing the table of results to bring the sweet / average spots into alignment do not help with a linear table because of the wide range of numbers that can be reached. Yes, it can be addressed by larger number groupings at lower number with smaller groupings at higher levels.
The only change that "must" be done is to move success at 1st level to 13+ to maintain the 50/50 chance at 1st level to cast a first level spell for an "average" wizard of 13-15 Intelligence. Did I really say "must"? When did I drink the Kool Aide? 50/50 is not writ in stone from some faceless patron. It just sounds nice. It does not mean that a 5th level wizard casting a 3rd level spell MUST have a 50/50 chance of success.
Besides, you could equally argue that 12+ makes more sense. Sure spell casting becomes "easier" at low levels. That is the patron sucking the fool in. With a d3 class die, he can't cast the spell more than once a day 2 out of 3 days.
The average for a caster with +1 bonus is 45/55 (fail/success). That seemed to be Joseph's target number so keeping that inline so it scales to all spell levels is what I am trying to statistically accomplish.
Like I said before, I was in favor or using a class die for the spell crafting check months ago. But after a few discussions, I dropped it as not workable with the current structure of DCC RPG and magic.
I think simply adding the class die to JUST determine if the spell is retained is something that can be easily added to DCC RPG. I think it is not an optional modification, but something that needs to be done to provide additional balance for Wizards are higher and lower levels.
I will leave you with one more thought on using the combination of class die for spell checks and retaining spells. I have not done the full statistical analysis on this... but from rough calculations... if you roll low on the class die, you are most likely to fail. If you roll high on the class die, you have a much higher chance of successfully casting a spell. So by using the same die for both results, you are more often than not... rewarding spell casters that successfully cast spells and not those that fail.