WoW Forever Druid Talents: What Balance, Feral and Restoration Still Need in PvP

If you play a Druid and you’ve been anywhere near the Forever community this weekend, you’ll have watched the class get reshaped underneath you, and you’re probably trying to work out which of the complaints are worth listening to and which are just launch nerves. This post is my read on where Balance, Feral and Restoration actually stand for PvP after the late-October changes, taken one spec at a time, and on what I’d still want changed before we’re all living with it at level 60.

Whilst I’ve spent years on this class across both factions, learning each spec by playing it rather than theorising about it, I’ll say the obvious thing first - nobody has played any of this on a live level 60 Forever character yet, and that includes me. So treat what follows as a considered starting position rather than a verdict, and expect me to correct some of it in print once the fights actually start.

The 30 to 31 October changes reshape much of the class. Genesis and Naturalist are stronger, Feral has a rearranged talent tree, Swiftmend consumes its HoT (heal over time) again, and Wild Growth now takes three seconds to cast. Berserk has disappeared from the Feral talent tree, and updated change reporting says it’s now available to all three specialisations, though I’d still verify the exact training and availability in the launch client.

All of this has been read out of the game data rather than confirmed in published development notes, and that distinction matters a great deal on a class that could still move before launch. I’ve been through the official forums, the Reddit discussions and the latest talent changes, and there’s plenty of frustration about, particularly among Restoration players, alongside a quieter group who reckon Blizzard is heading the right way with a more deliberate approach to class design.

Both sides have a point. I don’t think Druids need to become the strongest caster, melee and healer all at once, however versatility should be something you can actually use in a fight, not something you merely own on a talent calculator and never get to spend. That’s the standard I’ll hold every one of these changes to.

Balance: Starfire needs a better reason to exist

Balance has received some welcome attention in Forever.

Wrath is more efficient with the right talents. Genesis strengthens periodic damage and healing. Moonkin Form has useful group benefits, and Eclipse attempts to give Wrath and Starfire a more deliberate relationship.

The trouble is that the relationship assumes you’re allowed to keep casting.

In Fixing Balance Druid, Hwaybert argues that Starfire arrives at level 20 but takes too long to become useful. His proposed changes aim to improve the spell without turning Balance into something completely different from Classic.

I think that’s a reasonable starting point.

Starfire is supposed to be a commitment. You stand still longer, spend your Mana and expect the result to justify the risk.

In PvP, that’s a difficult bargain.

Imagine defending Blacksmith in Arathi Basin. You’ve applied Moonfire and Insect Swarm and have a brief opportunity to cast. A Warrior is closing the distance, and there’s a second player coming around the structure.

You could continue building your Eclipse sequence. You could also root, reposition, heal or prepare to change forms.

The right decision depends on the situation. A casting rotation that looks excellent against a stationary target may be much less useful there.

A boss might stand still long enough to let you finish your sequence. A Mage has other plans.

Flapjack raised this concern earlier in October, arguing that Eclipse was poorly suited to the mobile, attrition-based exchanges of Druid PvP.

I wouldn’t remove Eclipse, though.

A dedicated Balance Druid should benefit from standing still long enough to complete a powerful cast. That is part of the distinction between a caster and a hybrid focused on mobility.

I’d prefer to make the sequence less rigid.

In a separate Balance and Restoration discussion, a player suggests a reciprocal Eclipse interaction: Starfire improves a subsequent Wrath, rather than Wrath only preparing Starfire.

That has potential. The key would be to make it a modest reward for completing a cast, rather than another way to stack unavoidable burst damage.

I’d also reconsider how many points are required to make Starfire attractive. Compressing Improved Starfire from five ranks to three is worth investigating, provided the final benefit and talent prerequisites are balanced.

There is a counterargument. Some players believe Balance is already receiving enough efficiency improvements, and that further reductions risk making Mana management irrelevant.

They’re right to be cautious.

I want Starfire to be worth casting. I don’t need it to be cheap, fast and devastating at the same time.

Feral: the Mana bar is part of the class

Feral has a different problem.

The recent changes improve some aspects of resource management while making others less accessible.

Earlier in the beta, shifting from Cat Form into Bear Form could discard stored Energy. Blizzard subsequently changed that behaviour so Energy could be retained in Bear Form without continuing to regenerate.

That was a sensible improvement. A Druid shouldn’t be discouraged from using Bear Form defensively because doing so destroys the resources needed to continue fighting.

But the Energy system still creates difficult choices.

In Final Feral Druid Feedback, Kakarot argues that Shifting Power is too central to maintaining a basic bleed rotation.

His preferred design would let a competent player maintain Rake and Rip using normal Energy generation, with Shifting Power providing additional opportunities for Ferocious Bite.

I like that distinction.

Shifting Power converts Mana into Energy. That should be an interesting decision for Feral because Mana is also needed for healing and changing forms.

If you’re spending Mana to finish an opponent, you’re accepting that you may have less available to survive their response. That’s a genuine trade-off.

But if maintaining ordinary damage already requires constant Mana expenditure, there isn’t much choice left.

The late October changes complicate that discussion further. Shifting Power has moved deeper into Feral, Lacerate has become a talent, and Primal Bite occupies the former Berserk position.

The Berserk change will need practical verification despite its reported new class-wide availability, particularly for those who relied on its anti-Fear utility.

It’s also worth acknowledging positive feedback.

In a detailed beta review, Baelemma describes Feral as enjoyable and praises its mobility, utility and combat flow.

There’s no reason to dismiss that because other players want changes.

I also don’t want unrestricted powershifting to become compulsory again. Spending every available opportunity converting Mana into Energy may be effective, but that doesn’t automatically make it good class design.

I’d rather have meaningful choices about when to use those resources.

Feral should be strongest when someone understands how to manage Cat and Bear together, not when they can repeat the same resource conversion most frequently.

Restoration: Swiftmend can have a cost

Restoration has attracted strong criticism following the latest changes. I can understand why.

The beta version of Swiftmend healed without consuming Rejuvenation or Regrowth. The newer version consumes the HoT again, requiring the Druid to replace it if they want continued healing.

Gift of the Earthmother has also been replaced with Bloom, and Wild Growth has gained a cast time.

Those are several changes to responsiveness and efficiency arriving together.

Rull’s Restoration discussion describes the problem clearly: casting Rejuvenation, consuming it with Swiftmend and replacing it takes additional Mana and global cooldowns.

That’s a legitimate concern.

My view is slightly different from some of the feedback: I don’t mind Swiftmend consuming the HoT.

Swiftmend converts healing that would have been delivered over time into immediate healing. That’s valuable, especially when an opponent has created a short opportunity to secure a kill.

It’s reasonable for that decision to have a cost.

The problem comes when several costs accumulate. If the Druid has already committed significant Mana to healing, has lost some responsiveness elsewhere, and now needs additional global cooldowns to recover, an interesting trade-off can become a liability.

That’s why I would examine Bloom before reverting Swiftmend.

Bloom is a good direction, but the numbers matter

Bloom provides additional healing when Rejuvenation or Regrowth expires naturally. It also provides a smaller heal when an enemy dispels either effect.

According to the current client tooltip, those values are 100% and 50% of the Druid’s Spirit respectively.

This addresses a real PvP weakness. A Priest, Shaman or Felhunter can remove a HoT and much of the healing value attached to it. Bloom means dispelling isn’t completely free.

I like the idea. Whether the amount is enough is less clear.

The forum discussion raises that concern, using the Spirit values available during beta as an example. But we don’t yet know what normal PvP Spirit values, healing coefficients and dispel patterns will look like at level 60.

I wouldn’t prescribe a major Bloom buff based solely on those early numbers.

One possible direction is to scale the dispel heal partly with the healing remaining on the removed HoT, subject to a reasonable cap. That would make dispelling a freshly applied Rejuvenation more costly than removing one that’s almost finished.

It also carries a risk: if dispelling restores too much health, enemy players may lose meaningful counterplay. That’s why I’d test the mechanic before settling on a percentage.

A Restoration forum suggestion proposes letting Swiftmend trigger Bloom when it consumes a HoT.

I wouldn’t make that change automatically. If Swiftmend already converts the HoT into an immediate heal, giving it an additional Bloom trigger could undermine the trade-off Blizzard has deliberately restored.

The interaction needs verification first.

Wild Growth worries me more

The new three-second Wild Growth cast is a larger concern for Restoration PvP.

With full Naturalist, the cast is reduced to 2.5 seconds. That’s still a substantial commitment for a heal whose value is delivered over time after the cast finishes.

In a battleground, damage can move between targets quickly. An opposing player can interrupt, apply crowd control or force the Druid to move before the spell finishes.

A three-second cast has consequences even if its eventual healing looks good on a spreadsheet.

This has been a major theme in the Restoration feedback. Players aren’t only concerned about throughput. They’re concerned about losing the mobile, responsive playstyle that made the spec enjoyable during beta.

I think Blizzard should listen to that distinction.

A shorter cast, perhaps somewhere around 1.5 to 2 seconds before other modifiers, is worth testing alongside adjustments to healing and Mana cost.

An instant cast isn’t necessarily the correct answer. But three seconds looks like a particularly demanding starting point for a deep Restoration talent.

What Blizzard is getting right

It’s easy to become fixated on whichever ability was most recently weakened. That overlooks some meaningful improvements.

Druids can now use most consumables while shapeshifted. Weapon effects work in forms. Racial abilities are more accessible. Entangling Roots works indoors, while Nature’s Grasp and Omen of Clarity are part of the baseline toolkit.

Those changes matter. They make the existing Druid abilities easier to use together without necessarily increasing raw damage or healing.

I would like Blizzard to continue in that direction.

There is also a reasonable argument against adding Cyclone, Lifebloom, Starsurge or another collection of abilities from later expansions.

Some players want those spells because they add depth and provide answers to difficult matchups. Others are concerned that importing too much of the modern toolkit will erase the limits that make Classic interesting.

I lean towards the second position, although I don’t think preserving Classic means preserving every inconvenience.

The Druid needs coherent interactions between its existing abilities before it needs another spell.

What I’d change

I would start with four focused areas.

Balance: Make Improved Starfire less expensive in talent points, and trial a reciprocal Eclipse interaction that rewards completing a Starfire with a modest benefit to a later Wrath. Preserve the opportunity for opponents to interrupt or evade the sequence.

Feral: Preserve the Bear Form Energy-retention improvements. Test whether a baseline bleed rotation can function without Shifting Power being used constantly. Keep Mana-to-Energy conversion valuable as an offensive choice, rather than making it mandatory for ordinary damage.

Restoration, Bloom: Retain Swiftmend’s HoT consumption. Verify exactly how Bloom interacts with it, then test whether enemy dispels should trigger healing that scales partly with the HoT’s remaining value.

Restoration, Wild Growth: Test a shorter cast with adjusted healing and Mana costs. The spell should still require a commitment, but not so much that its usefulness depends on opponents allowing a long, uninterrupted cast.

I wouldn’t increase Nature’s Grace’s duration yet. It’s tempting for PvP, but changing haste uptime could have substantial effects on sustained PvE casting.

These are test proposals, not changes that should all go live together.

Where this could land

Blizzard has already made substantial changes to the class. Some directly address concerns raised earlier in the beta. That’s encouraging.

But level-30 beta feedback cannot establish level-60 PvP balance. Better gear, complete talent builds and organised battlegrounds will expose interactions that are difficult to judge now.

For Balance, I’d watch how often Eclipse can be used effectively under pressure, and whether Starfire justifies its cast time.

For Feral, I’d watch whether Shifting Power remains a choice or becomes a mandatory part of ordinary damage, and what happens to the class’s defensive utility when Mana is scarce.

For Restoration, I’d look at how frequently Bloom meaningfully offsets dispels, whether Swiftmend’s cost feels proportionate, and whether Wild Growth can reliably complete in contested fights.

I suspect some of the current complaints will prove overstated, whilst others will only get more serious once players start building around them in earnest. I’d rather find that out through actual matches than declare the class finished before anyone has fought on it at 60.

The part of Druid PvP that interests me most is having several possible answers to the same situation - sometimes it’s damage, sometimes it’s a heal, a root, a form change, or simply making enough distance to reset the exchange - with the right call depending entirely on the opponent and on what the fight is doing at that moment.

Blizzard doesn’t need to make every one of those options equally powerful, it just needs to make them all worth considering. If you play a Druid and this weekend has you worried, I’d weigh the changes the way I have here before writing the class off, then go and test your own build rather than take mine on trust. I’ll be doing exactly that live most weekdays from midday AEST on the stream, and I’d love to see how your version of the tree holds up.

Research cutoff: 1 November 2026. Observed client changes remain subject to revision. Community gameplay observations are attributed to linked sources, and examples are hypothetical, not personally tested level-60 matches.

Restokins Forever - Horde Druid PvP. Watch on Twitch · Watch on YouTube