Bubble Spec: the New Hotness

Ever since WotLK rolled melee and spell crit into one crit rating, Holy/Ret has been the spec of choice for the vast majority of holy paladins out there, with this 51/5/15 spec being the most popular choice. (Others choose to sacrifice 3 points in Divinity for an extra 3% crit from Sanctity of Battle, which is a reasonable choice.)

The classic Holy/Ret build synergises excellently with the current strength of Crit Rating as a holy paladin stat, but there’s a new contender on the scene: the 51/17/+3 Shield spec.

51/17/+3 Bubble Spec

This build sacrifices the extra 5-8% Crit chance from the Ret tree, which will be significantly less useful after 3.2. In exchange, it picks up Guardian’s Favor, for more frequent HoProtections, and two key talents:

Divine Sacrifice redirects 30% of all damage taken by raid members within 30 yards to you, for 10 seconds. Like Hand of Sacrifice, it’s capped at 150% of your health – but if you have Divine Shield up, that cap doesn’t trigger and it’ll mitigate the full 30% for 10 seconds. This is invaluable in situations where you’ve got big incoming damage on more than one target, or significant raid-wide damage (such as XT-002’s Tantrum, or Hodir’s Frozen Blows).

Edit: As a commenter pointed out, there are some indications that this will be “fixed” in 3.2 so that Bubble or No Bubble, it still caps at 150%. That definitely reduces the utility of the cooldown, although it doesn’t affect Divine Guardian.

Divine Guardian increases your Divine Sacrifice to 40%, which is a nice bonus, and it doubles the duration of your Sacred Shield and increases the amount your SS absorbs by 20%. That’s a hell of a lot of punch from two talent points.

This build really comes into its own when you’ve got 4 pieces of Tier 8 armor, for the set bonus: “Your Sacred Shield can now trigger its effect every 4 sec instead of every 6“. Effectively, this set bonus provides up to an extra 50% mitigation by letting Sacred Shield proc more often; when combined with the increase in absorption from Divine Guardian, your Sacred Shield can become a very powerful protective tool.

I know it’s hard to put aside the Crit Rating for something as unglamorous as a better shield, but I think every holy pally should at least take a look at this build when patch 3.2 hits, if they haven’t already.

Preparing for Mana Regen Changes in 3.2

Blizzard have been unhappy about the state of holy paladins for a while now – apparently crit is “too strong” a stat for us, and they don’t like our distaste for mp5 – and as has been widely reported, 3.2 brings some pretty solid regen nerfs to the table for us. In return, they’re buffing the mp5 on most items, in an attempt to make it more appealing for us.

The Changes

Most of the changes were announced or foreshadowed in this post:

So, we loseWeight Exercise 5% of our intellect, plus Replenishment’s been cut by 20%, plus crit-based mana regen has been halved. In exchange, we get a more powerful Beacon of Light (which is a subject for another post), and 25% more mp5.

Crunching Numbers

If you want to know just how this is going to affect you, you need to take a look at overall data from a raid, rather than just relying on your perceptions. I’ll use WoWWebStats (WWS) parses here to demonstrate, but you can also use other sites like WoWMeterOnline (WMO) or World of Logs (WOL).

Before a raid, take note of how much holy crit rating you have – personally, I’m usually around the 41-42% mark when buffed (since I don’t usually get Focus Magic, and we don’t have a regular raiding boomkin).

Step 1: Recalculate your Intellect

Because we’re losing 1/3 of the boost of Divine Intellect, it’s important to note how that affects our other stats.

To calculate this, make note of your stats when raid-buffed, then divide your current Intellect by 1.15 and multiply the result by 1.10, to see your raid-buffed Intellect after patch 3.2.

In my case, I’ll be going from 1657 Intellect to 1584 Intellect, a drop of 73 Int. Losing 73 Int costs me 1095 Mana and about 0.438% Spell Crit, which will further reduce the effects of Replenishment and Illumination.

Reducing mana pool will also affect the mana return from Divine Plea, but that’s a factor that’s under your control, as you can adjust how frequently you use Plea to compensate for a reduced return.

Step 2: Illumination

After the raid, take a look at the WWS parse. (If your raid group doesn’t use WWS or an equivalent service, you can save your own combat logs and upload them for your own use.) Select yourself, and pick a boss fight you think was typical, or pick “all bosses” or “defeated bosses” from the split menu if you want an average for the whole night.

Now, look at the top section of stats, and look for your “HPS time” stat. That’s the amount of time WWS identifies you as healing.

WWS Parse - Part 1
(Click to enlarge.)

This sample just shows one fight, so the HPS time is 6:17. If it’s a full night, the HPS time will be far higher, of course (but lower than you’d expect, since every raid wastes a surprising amount of time on getting ready, recovering from wipes, clearing trash, strategising, and so on).

So, now you’ve got the time you spent healing, the next step is to find out how much mana you got back from critical heals during this time. For this, click on the “Energies & Dispells” tab below the first block of stats, and look for Illumination (which is the talent that restores mana after crit heals).

WWS Parse - Part 2
(Click to enlarge.)

From these two screenshots, you can see that I spent six minutes and seventeen seconds healing (or 377 seconds), and got 27,718 mana back from crit heals. This equates to approximately 73.5 mana per second, or 367.6 mp5. Obviously, these numbers are going to vary based on factors like the rest of your healing team, whether it’s a progression fight or farm content, your latency, and so on.

So, my 41% spell crit gave me back mana equivalent to 367.6 mp5 on that fight. Halve that. That’s how much mana regen you’ll loseWeight Exercise from crits when patch 3.2 hits.

Also note that your reduced Intellect (as calculated in Step 1 above) will also reduce your spell crit, reducing Illumination-based mana returns further.

Step 3: Replenishment

On WWS Replenishment will show up in the “Energies and Dispells” display, just like Illumination. It doesn’t show on the screenshot above, because that fight had no-one with Replenishment, so here’s a sample screenshot (from a fight with 5:46 HPS time).

WWS Parse - Part 3
(Click to enlarge.)

On this fight, Replenishment restored 8190 mana over 346 seconds, an equivalent of about 118.3 mp5.

To calculate the effect of the 3.2 changes on how much mana you get from Replenishment, you’ll need to do two things:

Take the Replenishment value, and multiply it by 0.8. This new value accounts for the nerf to Replenishment itself. Now take this new value, and multiply it by 0.956, which accounts for the nerf to Divine Intellect (since 0.956 is 1.15/1.10). That’s the new amount of mana you’d have regained from Replenishment in an otherwise identical situation, post-3.2.

Using the screenshot as an example, we get 8190 * 0.8 * 0.956 = 6263 mana over 346 seconds, or about 90.5 mp5. That’s a loss of nearly 28 mp5, on this particular fight.

Obviously, that’s a very small sample; for accuracy, you’re best off looking at a number of raids and totalling all the Replenishment mana returns and HPS times before you do any calculations.

Step 4: mp5

This one’s the most straightforward calculation. On your character sheet, take a look at your current mana regeneration while casting (without Blessing of Wisdom), and multiply that by 1.25. That’s your new gear-based mp5 total.

The Impact

Using the above calculations, you can work out how much mp5 you’ll loseWeight Exercise from the nerfs to Intellect, Illumination and Replenishment, and how much you’ll gain from the buff to mp5 on gear and enchants. If you’re still primarily in Naxx gear, you’ll likely loseWeight Exercise a fair bit of mana regen overall, because there was plenty of Naxx gear with Haste & Crit, and no mp5. If you’ve moved into Ulduar gear, the impact will be lower as pretty much every single piece of Ulduar spellpower plate has mp5 on it.

However, there’s one important factor to bear in mind: mana regen from mp5 is not the same as mana regen from crit. Quantitatively, they might still give you back the same amount of mana; qualitatively; they’re different.

Mana regen from spell crits lends itself to spike-based healing; crit-focused healers are good at throwing out a lot of healing very fast to compensate for damage spikes, and Illumination restores mana dynamically; the more you use, the more you regen.

Mp5-based mana regen, on the other hand, is a steady trickle of mana; if you engage in over-enthusiastic spike healing, you may run yourself excessively mana-dry until you wait for the pool to slowly top itself back up again. mp5-based mana regen lends itself to more measured, steady healing to keep a target topped up, and is particularly aided by mitigation from shields and damage diversion abilities (like HoSac and Divine Sac).

Whether we like it or not, Blizz has decreed that we need to focus more on mp5 and less on crit. Numbercrunching will help us mitigate the impact of the changes – by being ready to use more mp5-based gear, for instance – but it’s also going to be important to keep playstyle changes in mind, too.

This is starting to get annoying…

So, on Monday I discovered that last week’s problem had recurred, even more virulently. Unfortunately, in fixing the site we’ve had to abandon the comments on the last two posts, for which I’m very sorry. I do have copies of everyone’s comments, via email notification, and can repost them (with attribution) if necessary.

For the next little while I’ll be running with a default theme (*shudder*) and minimal plugins, in an attempt to eliminate possible vectors for the problem, so if you’re visiting rather than reading via RSS and the blog looks, uh, barebones? That’s why.

Edit @ 1835h: In addition, unfortunately I’ve had to totally disable comments as a temporary measure, as it seems the malicious code is being inserted via WP’s comments feature. Obviously, this sucks, and it’ll be fixed as soon as possible. Old comments are still viewable, though.

Edit @ 0506h July 5th: Comments are back on again, hooray. (Although comments that were lost remain lost.)

This is also why I haven’t been posting for the last week; no point making posts that I’d only have to repost anyway. Hopefully that will no longer be an issue!