This is to present an idea of an approach to solve the pvp balancing issue, focusing on approach first, and puting execution discussions to the later stage.
Approach
Borrow the theory of free market and let pvp balance itself.
High level steps
1. Adopt a separated (weapon / armour) / damage / heal / cc coefficient to avoid impact to pve.
2. Generate an usage report periodically (e.g. every Monday)
3. Identify top n% (e.g. 20%) of skill usages and reduce the coefficient by m% (e.g. 10%)
4. Identify bottom n% (e.g. 20%) of skill usages and increase the coefficient by m% (e.g. 10%)
5. Disclose the changed skills to players.
6. Provide a mechanism to change the affected skills. (e.g. free crown token to refund the skill point of affected skills)
7. Assume players' behaviours will be adjusted automatically, which will be reflected in the next usage report.
8. Repeat until all the skill usage distribtion is acceptable. (e.g. Average usage of all skills are with p%, e.g. +-15%)
Pros
1. Blindfolded balancing and put the personal preferences and mis judgement (nerf screamings) out of the picture.
2. The scope of the balance can be controlled by n%, intensity controlled by m%.
3. The targeted skills will be either the skills rarelly used which have minimal impact. And the skill widely used, which is the major cause of imbalance.
Cons
1. The number of cycles required to achieve an acceptable usage distribution may take a long time, may be half a year.
2. Etc.
For discussion...
PS, apologies... my bad English...