Make "reflect" great again. But for real, this time.
Chanter monster tries to charm you, and you are keen enough to sense it and get the timing juust right and....the monster charms itself. Your issue is then one of timing, or breaking charm to kill before time runs out
Sh-boom! sh-boom! Ya-da-da, La-da-Da, Ya-da-da, La Sh-boom!
And furthermore....
I could see having anyone who accidentally interrupts a DoT get some kind of blowback just to them, maybe have a tic or two of the damage (or more) land on them before the spell is fully dispersed by their error. It would be funny and a totally reasonable consequence.
It would also be a great mechanic for teaching group cooperation, since currently it appears quite easy for a player (even a Dev) to be completely unaware that they have negated a DoT and reduced group effectiveness.
I would not support this happening "randomly" unless it only applies to specific skills and you know it beforehand. Downtime will be a real thing in this game, so backfiring will increase the average downtime. I don't think it will be a good thing to increase that average. Think about a group pulling mob, kills it, everyone's "medding", then the healers heal spell backfires on the tank and casts death instead. Not very fun. It can't just apply to caster DPS either, otherwise, people will just play melee DPS.
We must assume that "damage meters" will exist unless PA fully suppresses the exchange of combat data between clients. With that assumption, people already do stupid things to try and top off a meter, resulting in them standing in AoE's or pulling aggro. Adding a random chance for spells or physical skills to "backfire" is not going to look so attractive when a boss is at 5% and your MT, healers, or DPS die.
"Ultimate skills" won't be so ultimate if they simply fail to activate, or even worse, backfire on you when you use them.
So, the only skills a backfire can really apply to are niche skills, so the risk vs. reward of using them must be addressed in the skill itself and designed around it. For example, your "ultimate skill" safely activates at 100% rage, but you can also activate it for partial damage and risk of backfire if you activate it below 100% rage. I'm basically saying there has to be a good reason to use skills that can backfire, otherwise people just won't use them.
Yes, yes, risk is inherent in the design but that is what testing is for. I'm thinking that if the mob was a fire elemental type and you used a fire based skill or ability it would not hinder, but instead help the mob. I'm looking for nothing random in this case, more of a mistake that could be made. It would at the very least add to the world being dynamic and challenging. Note this is all during my little breaks in a busy day so I'm not fleshing out the idea as well as I should. Play with the idea people. Have fun out there
On another note, if a small woodland creature was under the protection of a 'spirit Force's in an area and a high level rogue, for no real reason, went to smash the little thing and back blow of damage would result as punishment for needlessly causing harm to the creatures of the wood. Something something. I'm nervously typing sheet because I have to restart a whole network in 2 minutes and......
While I am not a fan of attacks that can cause harm to the group (Wizard casting an AoE and it hurts friendlies and enemies) due to the griefing that it can cause; I am a fan of the caster being affected by their own spells. Cast Mez and have yourself targetted? Sucks to be you :-) Cast an AoE at your feet (not a Point Blank AoE, aka an AoE that centers around you) you will also get hit by the blast. In this way, the caster is responsible for making sure they place the AoE correctly and have the correct target. Less room for griefing.