PVP damage too high, so buff armor modding skill and make evasion flat % damage reduction
Before I get started, the suggestions I'm offering are in the spirit of @Azlyn 's recent post: smaller or easier changes that would have a high impact rather than new, complex, and time consuming systems or changes.
For background, in the group pvp context, damage is king. And, I'm proceeding with my argument based on the proposition that group pvp is more fun when it's a little longer than it typically is. Allowing people to survive long enough to make decisions is fun, and it seems to be Starmourn's pvp's strongest point. However, damage is too high for group pvp. There are some abilities that hit too hard, and those can be adjusted individually, but reducing damage overall would be a nice change and would reduce the number of factors involved in balancing the classes and, therefore, should happen before adjusting class abilities.
There's a few places where we can adjust the damage numbers: attack power, defense power, and health restore power. In my opinion, the approach to fix this should be as follows: leaving attack power as it is, increasing defense power, and adjusting restorative power.
A place to improve protection: armor modding. Armor modding stinks. The protection provided by a full set of armor mods is remarkably lower than the attack power granted by weapon mods. This doesn't feel remotely fair, and this is a great place to amp up protection. I suggest making the armor mods provide an additive bonus, rather than multiplying whatever existing armor stats there are now.
Another place: the evasion stat. Evasion stinks. We've got the dodge talent for a potential 7.5% protection. Changing evasion so that it is a flat damage reduction instead of a chance to reduce damage by a percent would help. There are more sophisticated evasion formulations floating around the history of our conversations on the topic, such as reducing damage from roomwide attacks (and maybe out of room attacks), or reducing damage when focus fired, or both, and so on. The point is that we've got evasion in place, and it could be made a lot better by making it simpler and more consistent.
Numbers would help, wouldn't they? Let's say that armor modding goes from 10% of existing armor to 10% flat. I'll start with looking at a typical scoundrel for a test subject. No damage resistance abilities, light armor. This means light armor goes from in the realm of ~20% to ~30% for specific damage sources. Let's assume we're talking about impact damage, since a lot of impact damage gets thrown around and some of the most dangerous abilities deal impact damage. Add the dodge talent in, 37.5%. Let's call evasion another 7.5% for 45% damage reduction total. Now, a scoundrel isn't as likely to melt instantly when knocked out of cover into the line of fire -- so far, so good. But when we look at the numbers for other classes and armor types, that damage resistance keeps going up because of base armor values and ability defenses. Setting that aside for now, let's go back to talking about our test scoundrel.
45% reduction against impact damage is a whole lot better than real scoundrel who has 20% + ~2% + 7.5% = 29.5%. Let's look at what it's cost the scoundrel to get that 45% damage reduction.
First, there's limited armor mod slots, and getting armor mods means spending resources. This is an opportunity for our test scoundrel to gather multiple gearsets to adapt to foes. Good. More marks spent to make armor modding more worthwhile. Good.
Second, stat points. Picking up 7.5% damage resistance from evasion would cost the scoundrel, say, 15% damage bonus from aim. Or some combination of the other stats. The player can decide, which is good. Personally, as a scoundrel, I'd likely give up a substantial % of damage in order to stay alive in group pvp. And in 1v1, I'd probably want that raw damage unless I know I can do enough damage to pressure my opponent.
Returning briefly to higher defense classes... A heavy armor character would be a tank, and disabling and impairing the heavy armor character would matter. Or, alternative damage types would be needed to punch through more of that armor -- strategy!
At the end of the day, this is just one piece of the puzzle. Class abilities need some fine tuning. That should be done after getting these two existing systems into a usable place.
PS: I'm reasonably sure other folks have a better grasp of the numbers than I do, and my memory's a little fuzzy on them. I hope these are accurate enough.
I like damage being a viable option is all I'm saying, and nerfing it across the board would probably make it really hard or impossible for anyone without arti stat bonuses.