While many specs are staying more or less the same going into Battle for Azeroth, Protection Warrior is once again getting some fairly significant changes. This is by no means a bad thing. While Prot was fun in Legion, it definitely had its share of issues. In BfA, our core gameplay, i.e. Devastate/Shield Slam/Thunder Clap/Revenge, remains more or less the same. The major changes are to our rage generation model, talents, and baseline abilities.

This post will be updated as new changes are released.

What's being removed

• Of course, our Artifact and everything that comes with it.
• Battle Cry - Replaced with Avatar
• Rage Generation - RFDT (Rage from Damage Taken) has been removed. More on this below.

There are a lot of talent changes as well. Those are covered below.

What's being added

• Avatar - This is now our baseline damage cooldown and it also generates 20 rage.
• Battle Shout
• Intimidating Shout
• Rallying Cry
• Shockwave - Shockwave is now baseline. The cooldown reduction when you hit 3+ targets has been moved to a talent.
• Vanguard - The 30% stamina increase is nothing new. We've had that for at least the last two expansions, it was just a hidden passive. The strength -> armor conversion is new though.

What's being changed

I am ignoring any damage/tuning changes, any rage generation/cost changes, and any changes due to the loss of the artifact.

• Tank threat generation has been significantly reduced. In Legion, tanks received a +900% bonus to threat generation, which is now +350% in Battle for Azeroth.
• Some abilities that were not on the global cooldown in Legion now are: Intercept, Ignore Pain, Demoralizing Shout, and Avatar.
• Demoralizing Shout - Now on a 45 second cooldown.
• Shield Block - Recharge time changed to 18 seconds. It still has 2 charges. Rage cost increased to 30. I'm mentioning this because it has a fairly large impact on gameplay (more on that below).
• Ignore Pain - This Ignore Pain functions in the exact same way, except the AP ratio has been massively nerfed and it now removes 50% of damage instead of 90%.
• Spell Reflection - Now reduces magic damage by 15%.
• Thunder Clap - The slow is now 20% instead of 50%.
• Intercept - No longer reduces enemy movement speed.
• The damage reduction from blocking is no longer a flat 30% (or 60% for a critical block). It is now based on the Block stat.

Rage Generation

We no longer gain rage based on the damage we take. There are two sources of rage besides what we gain from abilities:

1. We gain 2 rage from our auto attacks. (Note: This scales with attack speed. We won't be equipping daggers for better rage generation.)
2. We gain 3 rage from enemy auto attacks.

We do not gain rage from avoidance events (i.e. if the enemy parries/dodges or if we parry/dodge).Firstly, this means that rage generation will likely be more or less consistent throughout the expansion. If things scale as they should, this won't really be an issue (yeah, that's a big "if"). However, because we gain 3 rage when we get meleed, and because there is a short ICD on that (~1s), there is a fairly significant difference in rage generation when tanking one target compared to tanking multiple targets.

On single-target, you generally feel a bit rage starved. When tanking a significant number of targets that melee you, rage comes in quite quickly allowing you to dump it into Ignore Pain. Increasing rage generation on single-target would go a long way in improving the spec's consistency, both from a gameplay and balance perspective.

Block Rating

In Battle for Azeroth, there is now a "Block" stat on shields. The character sheet now has Block rating, and all of that rating comes from the Block on shields (besides potential bonuses from Azerite and such). So if a shield has 2000 Block, the character sheet will show 2000 Block.

Blocking no longer provides a flat 30% damage reduction. Instead, the damage reduction is calculated based on your total Block. It is calculated in the same way damage reduction for Armor is: blockRating / (blockRating + k). Currently, the k value for Armor / Block is 6300 for level 120-123 enemies. Block damage reduction is entirely separate from Armor and having more Armor does not affect it. For example, if the player has a shield with 4000 Block on it, damage reduction from blocking is ~38.83%.

Critical blocks are simply double the damage reduction. From the example above, a critical block for that player would be a 77.66% damage reduction. It is worth noting that just like Armor, there is a cap of 85% on damage reduction from Block. If a shield were to provide 50% damage reduction on normal blocks, critical blocks with that shield would only be 85%.

This means that Block is subject to the same scaling that Armor is, and it also could very well be tuned independently of Armor. Currently, a shield from a Heroic dungeon provides a little over 30% damage reduction, making this change to blocking a buff compared to Legion (ignoring any Artifact bonuses).

This system aims to incentivize Shield upgrades. Getting a good shield will likely be a significant increase to survivability, way more than a similar upgrade to another armor piece would.

Protection's actual Mastery is still the same in regards to what it improves: block chance. It is just the amount of damage reduction that blocking provides that has changed.

Talents

The most significant changes to Protection are to the talents. In Legion, there is more or less a single "best" talent build. In Battle for Azeroth, it is looking like Protection will have much more variation in what talents are picked, particularly when it comes to single-target vs AoE.

Beta Talent Calculator

15 - Into the Fray / Punish / Impending Victory

• Into the Fray is now useful in many more situations compared to Live. It counts allies as well, making it much easier to keep up the buff at full or close to full stacks.
• Punish is also a fairly good talent. It's essentially a permanent 3% DR on single-target (you'd also be able to keep 100% uptime on 2-3 targets with good procs). The debuff is applied to the enemy, not you. In other words, when you Shield Slam an enemy, that enemy deals 3% less damage to you for 9 seconds.
• Impending Victory feels very out of place on this row. It is not competitive from damage or survivability standpoint when compared to the other two talents. It's niche use as a reactive heal will rarely turn out to be useful.

Likely the choice between ITF and Punish will come down to whether or not you're able to keep near 100% uptime on ITF. Or it may be as simple as ITF for AoE and Punish for ST. Punish also has some increased value on magic damage heavy fights compared to ITF. Regardless, both talents are good and I can see situations where you'd take either.

30 - Crackling Thunder / Bounding Stride / Safeguard

Not too much to say about this row. It's all utility. None of them are blatantly better than the other two. Overall it's a good row with good choices.

45 - Best Served Cold / Unstoppable Force / Dragon Roar

• Best Served Cold is the same as it was. 5% increased damage on Revenge per target, up to 5 targets (25%).

• Unstoppable Force is a damage and survivability increase. More Thunder Claps mean more Shield Slam resets, which means more rage, etc.

• Dragon Roar for Protection is a 6 second, 50% slow on a 35 second cooldown. This can be either very powerful or essentially useless depending on the situation. Currently if the enemy is immune to the slow, they also do not take the damage. In high Mythic+ where kiting is sometimes necessary, I can see this be extremely useful for situations like that.

With the amount of rage generation we have now, constantly casting Revenge isn't feasible, so Best Served Cold will really only be good on sustained, heavy AoE where you're constantly getting Revenge resets. Not only is Unstoppable Force more consistent in its strength, but its a survivability increase, where Best Served Cold is not. There are very few situations where you'd take BSC over UF. Dragon Roar is very powerful in situations where you can actually make use of its slow, but otherwise the go-to here is UF.

60 - Indomitable / Never Surrender / Bolster

• Indomitable is down to a 10% health increase. It's not bad, but it will likely only ever be used in situations where you need the raw HP to survive, or if the other two talents in the row are simply not competitive.
• Never Surrender is the same as it is on Live. That is, you need to be at basically 0% HP to get the full 100% out of it. You'll get minimal benefit out of this talent the majority of the time, meaning you're much better off picking another, more consistent, talent on this row.
• Bolster reduces Last Stand's cooldown by 60 seconds, and causes you to block all attacks for the full duration of Last Stand. This turns an already powerful cooldown into an even more powerful cooldown. With Last Stand lasting 15 seconds, it essentially gives you the time to gain a full Shield Block charge back as well. Compared to the other two talents, this is absolutely the go-to.

75 - Menace / Rumbling Earth / Storm Bolt

Another utility tier. Improve your Intimidating Shout, improve your Shockwave, or grab a single-target stun. There's good choice here.

90 - Booming Voice / Vengeance / Devastator

• Booming Voice is slightly nerfed from live, but functions the same.
• Vengeance is in an "interesting" spot. It's really good for rage efficiency in theory, but if you spend all your rage keeping up the Vengeance rotation, you don't leave a lot of rage to cast your active mitigation abilities. Or, on the flip side, if you're spending rage on keeping up your AM, you won't get much benefit out of Vengeance. It's hard to say right now whether or not this talent is worth taking on single-target. This talent has significantly increased value on AoE due to the fact that you generate more rage and have more Revenge resets. From a pure survivability standpoint, Vengeance is likely the best talent on this row in AoE.
• On live, Devastator only "works" because of the insane amount of rage we generate, so we're able to fill what would be empty GCDs with Revenges. That way you get Devastator casts on top of everything else, leading to more Shield Slam procs, etc. In Battle for Azeroth, at least on single-target, Devastator is likely both a DPS and survivability loss (compared to not taking Devastator). You have times where you'll go 3-4 seconds without casting anything. There are so many spaces where you can cast Devastate, that not taking Devastator is better. However, AoE is a different story. On AoE, not only are you getting more Revenge procs, but you're also getting a lot more rage. This allows you to fill most of your GCDs and gain the benefit of Devastator.

This tier is definitely the most difficult one to look at and immediately "know" which talent is best for the situation you're up against. Quite a bit of testing will need to be done to determine which talent is best where.

100 - Anger Management / Heavy Repercussions / Ravager

This row is pretty much exactly the same as it is in Legion. Due to how Rage generation is, though, Anger Mangement's value has been reduced. However, there is quite a lot of synergy here that still allows Anger Management to be a very good talent. When taken with Bolster and Unstoppable Force (Anger Management reduces Avatar's cooldown), and because Demoralizing Shout has such a short cooldown, uptime on those cooldowns in addition to the high uptime on Shield Block/block makes Anger Management very powerful.

Heavy Repercussions, while good, does not provide the same amount of benefit that Anger Management does, particularly because Anger Management provides so much flexibility in the way of cooldowns. Heavy Repercussions will likely be useful on fights where tanking uptime is near 100%, but those kinds of fights are few and far between.

Ravager is still significantly weaker compared to the other two choices on this row. It does not provide a consistent survivability increase, but rather you must rely on the RNG nature of parry. HR and AM both provide a consistent, "guaranteed" survivability increase.

Overview / Thoughts

Gameplay

One point of concern is rage generation. The difference between rage generation on single-target vs. AoE is larger than it probably should be. In heavy AoE, you gain rage to the point where you have to "dump" it into Ignore Pain. On single-target, rage generation is low, making it difficult to maintain your rotation.

The spec functions more or less the same as Live, but feels a lot slower. Going from Live to BfA takes a bit getting used to, however the core gameplay is still there. It's really a matter of personal preference. Some people will be happy to leave being the spammy nature of Legion Prot, while some will miss it.

Shield Block

To go along with the above point about rage, Shield Block's rage cost can be inhibiting. Trying to keep Shield Block up makes it difficult to cast other abilities that cost rage, to the point where casting, say, Revenge may lock you out of casting Shield Block's for several seconds. You could argue that this is a good gameplay dynamic because you have to make the decision between one ability or the other. I don't necessarily disagree with that. However, the decision making can still be there even if rage gen is increased as the result of that is increased rotational fluidity. (Again, speaking from a single-target point of view.)

Ignore Pain

Ignore Pain functions in the same way as it does on live (besides tuning changes), however, one major difference is that it's on the GCD. As a spell that sometimes needs to be used as a rage dump, or generally speaking as a spell that needs to be used quickly in reaction to something, having it on the GCD can be very frustrating.

To sum it up:

We gained quite a lot of utility and our core playstyle remains the same from Legion, albeit it has been slowed down significantly.

The talent changes are definitely a step in the right direction. Changes to underperforming talents such as Impending Victory, Ravager, and Never Surrender would go a long way to increase diversity in talent builds.

There are of course some talents/abilities that are under/overtuned, but balancing is still ongoing so that's not of much concern right now.

Finally, some small tweaks to rage generation would greatly increase the fluidity of the spec and allow it to perform much more consistently between single-target and AoE.

All in all, Prot is looking good, it just needs some adjustments here and there to turn it into a truly great spec.