Posted 19 October 2010 - 02:24 AM
I've been intrigued by the WoG build for prot paladins, and the additional survivability it provides, but have been concerned about the hit to threat. I've done some initial, relatively napkin-y math and the results were interesting so I figured I'd post it.
- Used stats used by Zarko in his thread on the WoW tanking forum on expected stats at level 85 near the end of the first tier of raiding and a Soul Blade
- 939 rotation (priority: WoG or ShoR at 3 HoPo, Judgement, AS, HW, Cons), using SoT vs SoI
- Reasonably standard spec for base case and a WoG build for the Wogger
- Glyph differences: ShoR vs WoG and Seal of Truth vs Seal of Insight
There were a number of other things that I wasn't 100% sure of but made assumptions on.
- Seal of Insight has a 50% proc rate on all melee attacks,
- 2 sec swing time full raid buffed
- Steady Vengeance at ~75% max (10k bonus AP)
ShoR vs WoG
ShoR does an INCREDIBLE amount of threat when you're scaled up, to the extent that virtually everything else is filler. Weighted average hit (including crits, talents, etc) is a little over 100k threat. When cast every 10.5 seconds, the glyph of ShoR, by itself, is worth ~1k tps.
Needless to say, WoG can't compete with this. At its best, assuming 0% overhealing, WoG will put out an average of ~33k threat or a little less than 1/3 what ShoR does. Additionally, its more "vulnerable" and is obviously hurt by overhealing.
While in theory GbtL turns WoG overheals into shields, saving some of the lost threat (the shield generates 1/4 the threat of actual healing), in practice, the WoG lands before the PofI heal and the PotI heal (which does not turn into a shield) is the first healing lost to overhealing and is a little under 25% of healing done. Its not exactly realistic, but as an example, with a flat 30% overhealing, the PotI heal would be completely lost and generate zero threat, and of the ~10k healing threat lost, you'd at best salvage a few hundred points of shield threat.
The reason for the gap between the two is mostly scaling. SHoR scales with 120% AP, which ramps up dramatically with better gear and, obviously, Vengeance. WoG only scales with spell power, and only 0.22 at that, so it gets slightly better with better gear, but is helped not at all by Vengeance. Also, SD obviously plays a strong helping role for ShoR.
Starting with no Vengeance and no SD, SHoR averages barely over 30k threat, doing comparable or even slightly less threat than a 0% overheal WoG. Adding in Vengeance increases it 110% or so, and then SD gets it the rest of the way.
SoT vs SoI
I'm less confident in these numbers, but I had Seal damage + Censure coming in at roughly ~1800 damage/autoattack (once fully stacked up), good for ~2700 tps. During the first 10 seconds (assuming you could somehow start with +10k in AP from Vengeance), you're doing a whopping 900 tps on average as the Seal damage hasn't kicked in and Censure is still stacking up.
SoI says it scales with 15% of AP, but I found it to be more like 19% which would be good for 3.6k heals (including glyph). I'm fairly unsure of this, but some testing suggested that SoI will proc of any melee attack, including auto-attacks, CS, HotR, and Judgment, (and ShoR) but not AS, HW, or Consecration (or WoG). The exact number of procs depends on how your rotation plays out but I found the average to be close to an SoI proc per auto-attack, giving ~2,700 tps with 0% overheal and 1,800 tps with 30% overheal.
I was a little startled by how close these numbers turned out. Assuming 30% overhealing, 900 tps seems like a mild price to pay for 1200 free hps on the tank, particularly considering SoI actually outthreats SoT using almost any assumption over the first 10 seconds of a fight (apparently by far the diciest part of a pull). Also, unless you take firm advantage of EG procs, SoI actually outheals WoG in hps.
The biggest question is how much value the seriously overbudget glyph of SoT brings over the glyph of SoI (ballpark, assuming WoG casts, I put SoI at about 350 tps and SoT at a little under 1,000 tps for a next change of ~700 tps, but it depends on gearing).
There are still some things to work on, most obviously the impact of EG procs (which, since you're replacing abilities that generally scale with AP with an ability that doesn't, I think the tps gain will be fairly modest. I'm not done yet, but WoG is getting killed so badly that its hard to see this making up the difference). However, a couple things are really clear.
First and most importantly, because of Vengeance, things that scale with AP end up crushing things that don't. For example, HW doesn't currently scale with AP at all. If this isn't changed, I don't see it remaining relevant for very long. WoG is in the same boat. Its awesome in 5-mans, and is hanging on by its fingernails at the end of the first tier. Unless its scaling changes, it won't even be in SHoR's zipcode after that.
The extent to which abilities scale with AP is also obviously really important. Over the course of ~30 seconds, +10k AP from Vengeance will add a total of 3,200 damage to Consecrate damage, while it will add 36,000 damage to your ShoR's. That gap will only get bigger as gear improves and that's going to impact our choice of abilities; Consecrate will eventually pass HW even for single targets eventually.
Second, turns out SoI scales with AP roughly evenly with SoT which is why it holds up so well. In fact, if you were to ditch WoG and go with ShoR, I think you'd increase the value of SoI by 14% or so. This might be worth examining in greater depth.
SHoR wrecks WoG, but SoI arguably holds its own against SoT, and its probably worth fiddling with to find if any talents provide healing cheaply enough in terms of lost threat to be worth taking. However, a close eye needs to be kept on the results of raid testing.
If raid tanks are indeed keeping up close to 100% Vengeance, then I'm underestimating(!) the impact on AP scaling abilities. On the other hand, if Vengeance keeps scaling that way it is now, threat may be a complete non-issue (after the first 30 seconds or so) and the trade-off will be healing/survivability vs the substantial dps contributed by the tank (I find this extremely unlikely, but as I said, it needs to be watched).