If PF2e was a DVD, I would love to hear the alternate commentary track...
They've already told us they have no plans to do a conversion guide like they did with PF1.
| 2 people marked this as a favorite. |
A common critique levelled at third-party classes is that they're overpowered. This is more true for some than others, of course: third edition D&D had it bad, with the average per level creeping up every year; Pathfinder less so. Even so,I've designed character sheets for a bunch of the most popular Pathfinder-compatible third-party material, and I do feel that even when it's well-designed, full of flavour and not deliberately broken, it's often just that little bit too powerful by level.
The reason for this is easy to see: Paizo have a really solid handle on the maths of making power levels balanced, while third parties have to reverse engineer (or just guess).
So a supplement like this giving a solid mathematical basis for design decisions would help third parties design fully compatible material, and would help us the public decide whether a given third party class is or is not OP based on more than guesswork.
Mind linking some of these overpowered third party classes?I've yet to actually see a 3rd party class that was able to supplant Wizard and Witch from their throne on the mountain. I'd be very interested to read such a class.
No. Naming individual examples would just detail this thread into arguments over whether a specific class or option is OP or not. Everyone has a different perspective on it, based on their own experience and accumulated anecdotes. I'd prefer to stay on topic.
People argue at length about the power level of different classes, precisely because they don't have the maths available with which to answer the question. So that's what I'm asking for.
I worked out some of those mathematical design principles in 2015. I began writing a post on them for the Paizo forum, but the article kept growing and I never finished it.
However, I did post the first principle, exponential leveling, as comments in two of these Playtest subforum discussions:
- The thread "Spells Not Scaling Automatically per Caster" inspired me to comment that Pathfinder does not have Linear Fighter, Quadratic Wizard progression.
- The discussion thread on the Paizo Blog: Leveling Up inspired me to calculate the most likely formula for relative experience progression.
The best principle I found for measuring the power of a class is Use and Need. Use means that improving an ability that is used a lot has more impact that improving a power that is used less often. For example, increasing the BAB on a wizard would not help the typical wizard, because such a wizard relies on spells that don't use BAB. Need means that players care most about how an ability works when it is most needed, against the most deadly challenges, rather than in times when they could take it slower. For example, +2 to damage matters more against a strong APL+3 boss than against a weak hallway guard.
The Need principle also relates to the Moving the Goalpost Effect. That effect is almost an illusion, but the Need principle makes it real.
Imagine a character Roland with an attack bonus +4 against AC 15. Thus, Roland hits on a d20 roll of 11 or higher, 10 out of 20. Next, Roland gets an additional +1 to hit, for example, he buys a masterwork weapon. He then hits on a 10 or higher, 11 out of 20. 11/10 = 1.1, so he had a 10% improvement. Next, Roland gets a second additional +1, hitting on a 9 or higher. That is a 12/11 = 1.09, a 9% improvement. On the third additional +1, hitting on an 8 or higher, the improvement, 13/12 = 1.08, is 8%. A linear increase in attack bonus is a diminishing increase when examined on an exponential (percentage improvement) scale. But the Exponential Leveling principle says that the exponential scale is the one that matters.
Moving the goalposts prevents that. Suppose that after Roland receives his first +1 to his attack bonus, the 10% improvement against AC 15, he moves up to more challenging enemies with AC 16. He hits on a 10 or higher again due to the tougher opponents. So the second additional +1 improves the rolls from 10 out to 20 to 11 out of 20, a 10% improvement again. While Roland is enjoying that his new +6 bonus to attack rolls, the GM throws some AC 17 monsters at him. It is back to hitting 10 times out of 20. Thus, the third addtional +1 to attack is also a 10% improvement. Roland has just had three 10% improvements in a row.
Except that if he also kept fighting a few AC 15 creatures all along, those improvements are still 10%, 9%, and 8% against that one kind of creature. The improvement is relative to what he is fighting. The Need principle says that what matters is the most deadly opponents, so officially his improvements are 10%, 10%, and 10%.
I have seen a thread about D&D 5th Edition's Bounded Accuracy and how Pathfinder 2nd Editon might use such bounds. I have not played D&D 5th Edition, so I don't know the numbers and how they are applied. But Bounded Accuracy might break the Moving the Goalposts Effect, in which case the designers woud need to find other forms of improvement to replace it.