I've been playing Battlegrounds semi-regularly since it moved to the base game with the release of Summerset. I've played a decent amount of PvP in Cyrodiil so I'm not completely new to PvP, just Battlegrounds.
After trying Battlegrounds, there's really no going back to Cyrodiil for me. I've been enjoying Battlegrounds for a few weeks now and it's been a good time. There are things that could use some changes though. Some skills and sets are too strong along with some problems with game modes. I'd like to talk about why I think these things need changed and include possible changes for them.
Note: This is specifically about Battlegounds. No CP, 4v4v4 PvP environment. Also, this is just my list which means it's what I've experienced so it may not cover everything and it's limited to my perspective.
---Monster Sets---
Valkyn Skoria - Nothing really interesting to say about it. It just does too much damage. The proc chance might only be 8%, but any build that's running this procs it fairly regularly. Most builds "could" run this since almost all builds have at least one DoT, and it would still be strong on them. Could lower the damage, but just make it a DoT itself. Make it over 5 seconds to match it's cooldown. I am pretty sure it's blockable, but since it drops from the sky, it's very hard to see coming. Maybe even lower it's damage by about 10-15% and increase the proc chance to 15%. Not sure if that would be a good change, but probably better than it is now.
Earthgore - This set was nerfed to heal over 6 seconds instead of 3, but with the lower amount of people in Battlegrounds compared to Cyrodiil, this often just increases the length of immortality the person it triggers on receives. A possible change for this would be to make the effect trigger earlier but reduce the healing it gives. Have it trigger at 70% hp, but heal for 21k(30% less) instead of 30k(before Battle Spirit), that way it keeps people from reaching execute range but does less healing overall.
Zaan - I don't see this set ran very often, but that doesn't mean it's not extremely strong. The counterplay to this set is to get out of range of it, but the player can simply travel with you. It tends to be either you get out of range and it doesn't do that much, or you get CC'd or outran and it nearly kills you on it's own. Also, if you do manage to get out of range to break the tether, simply by it triggering it forces a break in the fight. That could be an interesting use for the set. Lower it's overall damage by 25%(34.4k down to 25.8k, before Battle Spirit) and make the tether visible but do no damage for the first second, maybe two. That way not only is there more time to react to what's happening, but you can make the decision to keep fighting and take the damage or be forced to back off briefly. It's still very powerful even with the damage nerf and it requires the player using it to commit to the fight to use it.
Selene and Velidreth - Both of these sets are hard hitting, and I very really see Velidreth used. Often times when I'm hit by these though, it feels more like it's my own fault. There's time to see them coming. I don't think they need nerfed, but I do want to include them since other people may feel differently about them. If anything, they would just need their damage lowered. Wouldn't mind seeing them go away along with all proc sets though.
Troll King - This is another set I'm including just because I know it's strong. It's hard to really know how effective it is since I have never used it and you don't exactly see it working on the enemy aside from their health going up. It's (effectively) 7740 healing over 10 seconds, along with whatever healing triggered it. Doesn't have a cooldown either. It seems strong, but like I said, I don't know enough to comment on it.
---5pc Sets---
Sload's Semblance - I know a lot has already been said about this set so I'll try to keep this short. Personally, I'd like to see the set just go away completely. There's at least a few good reasons why this set should be nerfed. The fact that it does the same damage, which can't be mitigated or counterplayed in anyway(aside from purge), no matter how squishy or tanky you build, is a problem. Two problems right there really. It triggers off of everything, which can be a downside, but often isn't is a problem. The amount of damage it does, even if it's over time, is a problem in Battlegrounds where healing is already lessened, along with reduced healing effects, granted that's it's own problem but is related nonetheless. There can be multiple stacks on the same person. It taking people out of cloak, whether intended or not, is a problem. The breaking of Nightblade's stealth isn't bad on it's own, but it's a strong utility that's on a set that is already very strong. I'm just listing the problems with this set since there are other places to go to see in more detail why these are problems. Rather not discuss them here.
A real change that would make it fair would be to just cut the damage down to half, make it so it doesn't stack from multiple people, and make it not break cloak. It's still a lot of free pressure for any tanky builds that look to outlast an opponent. Also, change it's proc to be light and heavy attacks only with a 100% proc chance.
Caluurion's Legacy - I'm only including this set since it's very strong. It's very rare to see this ran in BGs since critical chance is lower there. Still strong and probably needs to not exist in PvP. Not really any suggestion for what to change about it. It's a lot of extra damage that triggers off a critical hit, so the target is already taking a lot of damage.
Curse of Doylemish - This is similar to Caluurion's Legacy, but I've never seen it ran. It does require the user to melee the target with a heavy attack while they are stunned, so it does seem to have a fair amount of requirements for it to trigger. If it is a problem, lower damage or just removing it from PvP seems like the best solution. It is interesting so I'd like to see it just lose damage if it turns out to be a problem.
---Class Skills---
Templar
(In general) - Nothing really stands out for Templars. Sure, Breath of Life is strong, but healing isn't usually a problem in BGs. When it is, it's because one team has a healer and the other(s) don't. Also, it's not always just a single skill that contributes to healing being very strong. Right now no Templar skills make my list, but I could be convinced otherwise.
Dragonknight
Dragon Leap(Ultimate) - Both morphs of this ability are strong. You either get an AoE stun or a damage shield for 100%(before Battle Spirit) of your health. Not only that, but it's at range, and a gap closer. I know most ultimates tend to be somewhat unfair in PvP, but this one comes with a lot of utility. I am a bit on the fence about it, and if it does need to be changed, and though this is not generally a "solution," but raising the cost a bit would probably be enough.
Reflective Scales - This is on the "watch list" just like Curse of Doylemish. It's clearly very strong when you not only stop the damage of the next four ranged attacks against you, but also reflect them back at the enemy. Fighting more than one person makes it less effective, but when combined with a healer, using this ability when you start to get low against one or two ranged opponents makes you nearly unkillable. More of a problem with healers together with this, but not much to be done about that. It's also one of the two abilities, the other being Nightblade's cloak, that you slot an AoE ability to use against it. I don't think it needs nerfed, but there are probably some interesting things that could be done with it if it does end up needing changed at some point. Seems fine for the time being.
Petrify - Don't see this used often, but it's Rune Prison with less range. Clearly not on the same power level as Rune Prison, but it's still a bit of a problem for most of the same reasons. Not sure if it needs to be changed, but it could be interesting to see it become a DoT that stuns when it ends or when it is cleansed. Even with the short range it tends to be very strong since to fight a DK you have to get close if they put up their wings. You are always vulnerable to the instant stun when fighting them.
Sorcerer
(In general) - Sorcerers, MagSorcs really, get a lot of hate because they are just not fun and unfair to play against. You get hit with an unavoidable stun from Rune Prison, then hit with a few light attacks and destro staff abilities, possibly a Crystal Fragments if it procs, which is just RNG, but it's rare to see that skill anymore. You will have a strong DoT placed on you that goes off a second time if you live long enough. If you are low enough after that, you get executed by Mage's Fury with no chance to heal. If you survive, which is based on whether you are tanky enough or not, and manage to fight back, they toss on a shield or two, and if you manage to put enough pressure on them, they just dash away with Bolt Escape. No mystery why people want sorcerers nerfed all of the time. That being said, there's only two skills that I think need to be changed from the "standard" MagSorc. Those being Rune Cage, a Rune Prison morph, and Mage's Fury, either morph.
Rune Prison->Rune Cage - This is a long ranged, on demand(instant), unavoidable stun. At the very least, it should be an undodgeable projectile so you actually see it coming and it takes a moment to take affect. I think a real change to make it balanced would be to make it more like the Nightblade's Manifestation of Terror where you place down traps. This skill is just too strong in it's current form and is probably the most in need of being changed.
Mage's Fury - This skill just shouldn't exist in PvP. I'm not saying take away the Sorcerer's execute ability, but from a balance perspective, being able to put an execute DoT on someone that always goes off at exactly the right time before the enemy will ever have a chance to heal or react to being low is incredibly powerful. Since both morphs do roughly the same thing, one suggestion I have would be to keep one the way it is now, minus the four second linger time and up the damage. Make the other a defensive style morph a bit like Defensive Rune. Up the time to 8 seconds and make it to where when someone who is under the HP threshold attacks you, it triggers and deals the damage to them. Would need to increase the damage a bit for this one too and maybe up the threshold so it triggers a bit earlier. Can always adjust the numbers after seeing how it plays out. Would make it interesting when you and an enemy both get to low health in a fight. Also, it gives them some counterplay to it as well.
Rune Prison->Defensive Rune - This should just trigger on direct damage. It would be both better for the user and better for the enemy. The user wouldn't randomly lose it to some AoE they walk through, and it would still be there for when they actually get focused. From the enemy's side, they wouldn't get randomly stunned some time later from an AoE they dropped ten seconds ago. It also gives them a few options before triggering the stun. Not so much a "too strong" skill, just one that could be better.
Overload(Ultimate) - I'm not really sure how this skill can be changed, but I've been hit with this skill and taken 2k damage and I've also been hit with this skill and taken 10k damage. Doesn't happen often, but it's something that does happen. Not an easy skill to balance for PvP. Again, not sure what to do with it, but it's clearly on my list for good reason.
Hurricane - Not a problem skill, but on the watch list. This skill does a lot and is very strong. It's damage isn't huge, but it is fairly unavoidable. Being able to run faster than the enemy is a huge advantage and it also grants Major Resolve and Major Ward. It makes it possible to stick to the enemy so they also take the damage from it. It makes it impossible for any Nightblade to cloak away from the user as well. I don't think that's necessarily a bad thing that it counters cloak, and the skill probably doesn't need to be changed at all. Worth mentioning though.
Nightblade
Death Stroke(Ultimate)->Incapacitating Strike - This skill does too much for it's cost. It could probably just lose the stun and be fine since you already need to be close to use it and Nightblades have Mass Hysteria for close combat CC. I don't really think leaving it the same and increasing the cost is a good idea since it does take away from it's uniqueness as a low cost, singe target ultimate.
Blur - This skill is just RNG based. Sometimes you dodge three of the last five attacks and sometimes you dodge zero. I think a good change for this would be to cause the first attack(that is dodgeable obviously) every 6 seconds to miss and have the skill last for 18 seconds. The same goes for the Medium Armor active skill Shuffle and it's morphs. This change makes it so it's not random, there's counterplay on the enemy's side, and it doesn't completely do nothing sometimes.
Shadow Cloak->Shadowy Disguise - This is clearly a strong skill, but I personally think it's fairly alright. It's countered much the same way the Dragonknight's Reflective Scales is countered: you slot an AoE ability; or Radiant Magelight works too. The change I think needs to be made is that when you break the Nightblade out of stealth with an AoE ability, they shouldn't be able to recast it for 3 seconds similar to interrupting a skill with a cast time.
Warden
Scorch - Much like the Templar, I don't really have anything about Wardens that needs changed(because it's too strong). Scorch, which morphs into Subterranean Assault and Deep Fissure, does a lot of damage and is an AoE attack. It does take some time for it to go off so there's some counterplay with moving out of it's area. It could probably lose some damage if the Warden had more skills to use. Maybe they do have other skills are too strong as well, but I don't see enough Wardens in BGs to really know for sure, and those I do see are mostly using weapon abilities.
---Pets---
(In general) - There needs to be an option to toggle off being able to target pets, of any kind, in BGs. Just a quality of life change. Maybe even make it a keybinding toggle so you can quickly turn it back on if you want to heavy attack a pet for resources.
---Game Modes---
Capture the Relic - This game mode suffers a bit when there's a player that is either very tanky or very fast. Both can be somewhat mitigated, but they are occasionally a problem. Maybe make it so that holding a relic causes the player to take more damage and limit their sprint speed to +50% instead of +100%. This would promote more fighting before the relic is picked up rather than trying to chase someone down that either can't be caught or can't be killed. Overall though, this game mode tends to play out well.
Domination - This is the only game mode I think truly needs help. This game mode is rarely enjoyable. It's often one sided and lacks PvP. Sometimes you face a group that refuses to do anything but run around as solos and avoid any fighting, winning the game with three total kills between the four players. The game mode isn't about getting kills, and I acknowledge that, but at the same time, it is a PvP game mode and it should be more about fighting over the points than hide and seek. Other times you face a group that stays together as four and are nearly unkillable that just go from one point to the next, capping them and moving on. They only stop to fight if someone is on the current point they are standing on, and once they capture it they simply move onto the next point not even bothering to defend the point they just took. It makes sense if you are looking for a higher score for the leader boards, but gameplay wise, it's incredibly boring.
There's a few things that could be done to make this game mode better. One change would be to make capture points give less points for a few ticks when a team first captures them, needing time to ramp up to their full worth. This would make points that you already own more valuable. Also, being at an owned point when it gives points could increase the player's score, and the score gained from it would be relative to the amount of points being gained from the capture point. Another change, though probably less likely to happen, would be to make it so that one of each of the three outer points start in front of the three team's spawns with the center one remaining as it is. After that, make it impossible to take another team's starting capture point without owning the middle one first. Also, you could capture your own back without needing the center. This would be a bit of a scaled down version of Cyrodiil and could even be it's own game mode.
Crazy King - This isn't on the list because it needs to be changed. It's on here because it's a good comparison to Domination. This one starts as a king-of-the-hill and becomes one where the tactics that are annoying in Domination, four unkillable players just capping points or four solos playing hide and seek, becomes fun. Points move over time and more spawn as the game goes on. It's chaotic, and the moving points make it dynamic enough for it to be fun.
Overall, just changing the proc sets Sload's Semblance, Valkyn Skoria, and Zaan's, along with the skills Rune Cage and Petrify, would solve most of the balance issues currently in Battlegrounds.
This last part is just here to show some absurdity. The last thing I was hit with before the enemy died was the Soul Tether. The rest all happened while the player I was fighting was already dead. It's not uncommon to have a death recap like this, but it tends to be when the enemy is still standing.
=====
Edit 1:
Game Modes -> Deathmatch - The damage sigil should be removed from this mode.
(Start of game) - Have players start with either 0(empty) or 500(full) ultimate. I lean towards starting with 0 since it would then actually take time to build up as the fight is just starting. It is possible to always start with full already by building it up outside of the Battlegrounds before even queuing, but it feels more like an unfair advantage or something you are forced to do before the game even starts, which is unlike collecting and upgrading gear or brewing potions.