Seal of Vengeance: A closer look
I was looking forward to using this seal at L64, but found it to be a "sidegrade" seal for my level of gear and opponents. I'm tired of the silly comparison posts in the official WoW forums, so here's an analysis of SoV to better understand how it fits into the (Alliance) paladin arsenal.


I. Mechanics
First off, known mechanics of the seal:
- Each melee attack has a chance to proc the DoT. (20 PPM according to Kalgan) The proc stacks up to 5 times and is refreshed by procs, but not melee attacks.

- The DoT lasts 12 seconds, dealing 120 damage over 12 seconds for each stack of the DoT. The DoT ticks every 3 seconds, and the first application of the DoT gets around 17% of your +dmg in bonus damage. (68% of +dmg over DoT duration; 1 stack DoT and 5 stack DoT get the same +dmg bonus)

- The Judgement deals 120 damage per application of the DoT, and gets 43% of +dmg. A 0 stack Judgement deals no damage.

- The DoT does not seem to get a bonus from Weapon Specialization (SoR and SoC do). However, it is boosted by +dmg/%damage (Vengeance/Sanctity Aura) buffs that are active when the first DoT is applied. Further applications of the DoT get the same %damage boost that the initial DoT received. ie: Vengeance is active when first DoT is applied. First stack gets 10% Vengeance damage bonus. Subsequent stacks will get 132 damage (120 * 110%) over 12 seconds, even if Vengeance has faded when they are applied.

- SoV ate ZHC charges per proc.


II. Preliminary Analysis
The apparent differences of SoV compared to SoR are:
- DoT provides reliable, continuous damage - even if you stop hitting the target, you will continue getting the seal's damage for the next 12 seconds. SoR provides reliable damage, but only as long as you are hitting that target. You only need to proc SoV on the target once every 12 seconds to maintain the stack of DoTs, so you could theoretically maintain a 5 stack DoT on 2~3 targets.

- SoV has much higher base damage than SoR but has lower scaling.
**The DoT portion of the seal is worth 60 50 DPS, and gains around 5.7% of +dmg in DPS. SoR's base DPS was 22.5/31.5 (1h/2h weapons) at L60 and gained 3~4 base DPS by 65, but gains 9/10.8% of +dmg as DPS.
**Likewise, JoV (5 stack) is 600 damage, over 3x JoR's base damage. (170~ at L60, around 200 at L65) Scaling is again lower: JoV gains 43% of +dmg vs. JoR's 73%.

Overall, SoV has about a 60 base DPS advantage over SoR. SoR has something around a 8% scaling bonus over SoV, so at 750 +dmg or so, SoR will be absolutely better than SoV for damage output on a single target. (Actual +dmg needed is lower due to SoV not gaining Weapon Specialization bonuses, and base damage of SoR scaling with level/new rank)

- SoV has issues with weapon speeds that are very fast/slow. Fast weapons have a lower chance to proc SoV, and a non-proc string will cause the DoT to fall off. Weapons with a 3.0 AS or slower have a 100% chance per hit to proc SoV, but get less hits and are vulnerable to a miss string. (weapons slower than 3.0 are also losing procs) Due to heals resetting the swing timer, this can make multi-tasking with this seal difficult.


Out of time, so my next post will try to model the ramping penalty of SoV and how that relates to weapon speed. (The time it takes to build up 5 stack SoV is essentially a static damage penalty)

EDIT: Made a mistake on the base damage of SoV.
EDIT: Added sections for easier reference and reading.
EDIT: Renamed thread for more general discussion.

Nice breakdown Fiola.

Other considerations (from a tanking perspective):

- Once you reach steady state (5 stack of SoV), there is no synergy between SoV and Reckoning. You get no additional SoV damage due to a Reckoning proc. While SoR does 100% more damage for the duration of Reckoning.

- SoV can be resisted. While tanking a boss 4 levels above me in Sethekk Halls last night, I noticed that I had trouble maintaining even a 2-stack due to resists. While SoR is obviously unresistable regardless of level. In the endgame, this may not matter. But while leveling up, it is significant.

Alrighty, time for that math post I promised.

III. MATH!
A. Ramping Penalty
Assuming the SoV procs are spaced 3 seconds apart, it takes 12 seconds to "ramp up" to a 5 stack.

0 seconds - 1st application
3 seconds - 1 stack tick, 2nd application
6 seconds - 2 stack tick, 3rd application
9 seconds - 3 stack tick, 4th application
12 seconds - 4 stack tick, 5th application, judge

For 4 ticks, we'd get 68% (17% * 4) of our +dmg, and 10 "blocks" of base damage. (30 base damage per tick per application; 300 damage done)
JoV would do 600 damage, and add another 43% of +dmg.


So over 12 seconds, we have done:
[Seal damage] + [Judgement damage]:

[top] 300 + 68%+dmg + 600 + 43%+dmg

900 + 111%+dmg damage
or
75 + 9.25%+dmg DPS


Comparisons
At 5 stack, we'd do
50 + 5.67%+dmg SoV DPS (600 damage over 12 seconds) and 60 + 4.3%+dmg (600 + 43%+dmg per judgement, assume every 10 seconds) JoV DPS
110 + 9.97% +dmg DPS (1320 + 120%+dmg over 12 seconds)

While ramping up, we lose 300 SoV DoT damage and a small bit of JoV damage (if we wait to judge after the 5th DoT application; We lose at least 120 damage if we decide to judge before that point, so it's probably better to wait for the 5th application).


If we instead used SoR, (assuming 1h SoR, rounding up 1H base SoR DPS to the nearest decade) we'd do :
[SoR DPS] + [JoR DPS]

[top] 30 + 9%+dmg + 22 + 7.3% +dmg

52 + 16.3%+dmg
or
624 + 195.6%+dmg over 12 seconds

So during the ramping time, we do 276 more base damage with SoV, while SoR receives 84% of +dmg more over these 12 seconds; at steady state, SoV does 58 more base DPS while SoR gets 6% more +dmg as DPS.

Examples
Say we have 500 +dmg (I'm around this level with my current L67 gear). During the ramping time, SoR does 144 more damage (420 total) than SoV. At steady state, SoR does 30 DPS, so SoV still has a 28 DPS lead. It takes 144 / 28 = 5 seconds for SoV to make up for that difference, and after that SoV does more damage.

For SoR to have a 15 second lead over SoV, we need (.84+dmg - 276) / (58 - .06+dmg) = 15
Solving this equation gives us +dmg = 660. At higher levels of +dmg, the damage penalty of ramping up to a 5 stack makes SoV less attractive for single target use.


B. Special Cases
However, SoV still has something going for it at that level of +dmg - multiple targets and seal cycles.

1.) Multiple Targets
For two targets, if one swung at each mob with SoR once, that would be the same damage output as if attacking a single target with SoR. For SoV, one can maintain a DoT on each mob and essentially double seal damage. With 2 DoTs, SoV has 15.64%+dmg in DPS, which reduces SoR's scaling advantage to less than one percent. There would be no achievable level of +dmg that would allow SoR to scale better in that scenario.


2.) Seal Cycles
For seal cycles, consider the ramp up of SoV - after 12 seconds, on average, we have done 900 + 111%+dmg damage from 10 "blocks" of base SoV DoT ticks and a single JoV. After the Judgement, there are another 4 ticks of 5 stack SoV in backloaded damage, worth 600 + 68%+dmg damage.

In all, we'd have dealt 1500 + 179%+dmg for 12 seconds of melee using SoV.
Compare that to SoR's 624 + 195.6%+dmg over 12 seconds.

It would take more +dmg than a mage could ever possess for SoR to do more overall damage in this comparison.

Now, this is kind of unfair, since we're comparing 12 seconds of SoR + JoR with 24 seconds of SoV + 1 JoV. However, 12 of those 24 seconds do not require SoV to be active - allowing us to use another damage seal, such as SoR!


Using SoV by itself, we'd deal
1320 + 120%+dmg over 12 seconds
or
2640 + 240%+dmg over 24 seconds

Using SoR by itself, we'd deal
624 + 195.6%+dmg over 12 seconds
or
1248 + 391.2%+dmg over 24 seconds

Alternating between SoR and SoV, we'd deal
1500 + 179%+dmg for 12 seconds + 624 + 195.6%+dmg over 12 seconds
2124 + 374.6%+dmg over 24 seconds
(These values include the ramping penalty. However, DPS numbers are an approximation, and do not consider the effects of talents like 1h Weapon Specialization, imp. Judgement, Sanctity Aura, Vengeance, etc)

This 3rd method would do more damage than SoV by itself once +dmg exceeds 400~, and will do more damage than using SoR alone until we reach 4k +dmg.


Conclusions
So assuming SoR does not make any major gains at L70, and barring buffs/nerfs to either seal, it looks like SoV will definitely have a place in a paladin's arsenal even at L70, as it has better damage potential than SoR at realistic levels of gear.

This is an interesting conclusion, as I actually thought SoV would be obsoleted by high end gear. But the math shows that SoV is not that far behind SoR in its ability to scale with +dmg, and comes out ahead due to its higher base damage.

EDIT: Added sections for easier reference and reading.

This depresses me. I'm hoping to do a fair bit of tanking on my Blood Elf paladin, and this means lacking SoV could hurt my aggro-generating abilities in some cases. I've read a bit about SoB, seems to be rather lacking at the moment; unless you are using near zero +dmg gear it's outstripped even by SoCom in the area it seems to be aimed at, ie Retribution-spec burst damage.

I know SoV vs SoR isn't a major issue, but it's the small things that make the difference.

<End obligatory Alliance = EZMode rant here>

Why? What do you know about SoB mechanics?


Part of the reason I made this thread for SoV was because I read many opinions that "it sucked", with zilch math to properly back it up. If it wasn't that I don't have a Horde paladin to properly test SoB, I'd do a similar write up for SoB as well. Part of it is that I love the paladin class and wish to understand its mechanics better. Another part of it is that I'm bloody annoyed by paladins with no freakin' clue making claims about a seal they've *NEVER USED*. ("Alliance paladins will never get to be raid DPS because we don't have SoB! SoV sucks because its coefficient is less than SoR!")


SoB looks to have quite a few things going for it on paper. It procs every hit. It uses melee crit % for 2x damage, just like SoC. With a fast weapon, Perma-Vengeance (assuming Ret spec) is a safe assumption. With the complete weapon damage stat dependence, that frees you to aim for warrior gear for the mitigation, the weakest area of paladin tanking.

The damage return could be bad - but I have seen no one that I considered credible figure out how it works. (damage returned based on holy damage dealt by SoB? total damage of the swing? The tooltip has typical Blizzard ambiguity)


In short, I'd save judgement on SoB until I see a proper analysis, one which includes an explanation of the relevant mechanics.

I'm still more or less filled with murderous rage that they split pally abilities like this, honestly.

That said, Seal of Blood has a stealth benifit- since to my knowledge you don't get mana from heals at full health, always being slightly below full health (while not significiantly) has fairly severe benifits for a tankadin. It's more or less continual extra mana froma a renew/rejuv, which doesn't suck. While not directly relivant to a SoV post, for the direct comparisions (that obviously are going to spring up) I thought it was worth mentioning.

As I said, filled with murderous rage- I stil don't get why this happened.

You get mana from overheals.

I like Fiola's math.

My testing with SoV has lead me to these qualitative conclusions:

a: SoR does more damage as a single seal than SoV generally.
b: SoV doesn't scale as well as SoR.
c: It is possible to build a stack of SoV, and then switch to SoR, and back to SoV before the stack falls, but once you're at a full stack of SoV you can't easily tell the last time it was applied, and so it's easy to misjudge and have it fall.
d: It's possible to build 2 stacks of SoV on 2 targets but very difficult.

In a typical tanking situation I find SoR much stronger than SoV because I have enough things to be concerned with than trying to keep up 2 12 second timers in my head. SoR gives me aggro that I don't have to build up to 5 stacks before it becomes potent, I can't put it on two monsters, but that's ok because consecretion does a fine job of holding a second mob over healing and aoe aggro. The only thing it wont hold up to is focus fire, but there's typically no reason people should be focusing on an alternate target.

I'm still looking for situations where SoV is more useful over SoR, but I haven't found any yet. Even a situation where I need both to hold aggro, but again, I haven't found one.

So, lucky for bloodelves I guess.

Typically the only use I've got from it is: Soloing, judge crusader, stack 5 vengeance, judge vengeance, seal righteousness and wait for death.

HoTs don't tick if you're at full health, though, unless it was recently changed.

Thank you for providing an empirical breakdown of SOV.

I am now a bit curious as to if the proc of SoR has been known to refresh on each application of SOR (a per hit holy ‘proc’ guaranteed per weapon strike) the dot of SOV.

I currently lack the time to test this, however if so, then this would apply not only a ‘holy dot’ (adding agro to the mob immediately after an agro dump such as fear or wing buffet) but also maximizing mana efficiency for paladin off tanking and total holy damage output (our method of effective tank-a-din).

Another question would be if the Dot effect refreshes judgements applied, or if we still have to continue hitting the target to refresh the judgement.

edit: Zedrich appears to be asking in part c. the same question I have, yet I'm more looking towards the exact efficency rate of mana consumption and if a full SOV stack is worth applying to all targets along with a judgement of wisdom / light debuff in most raids or 5 mans. (Simply for the judgement paladin to act as an off tank or add an additional burst of holy dps to the target)

I put together a little programme to simulate auto-attacks using SoR and SoV, though my model could use some improvement.
Can anyone confirm/deny any of these statements ?

1) I used the WowWiki SoR formula: 22 + (2 * (WeaponSpeed*10) - 30) = Damage per Second
This method seems counter intuative to me because dps scales with weapon speed.

2) SoR cannot be resisted, SoV suffers a 17% resistance check to be applied AND on every tick of its dot.
My little programme allows for spellhit to be applied, but there is a question over when resist checks are encountered.

3) SoV does not tick until 3 seconds aafter its first application, new procs do not change the time of the application, only its expiration point.
I.e If you prov SoV at time=1.7seconds it will tick at time = 1.7+3n seconds, until the stack expires no matter how many times it is renewed.

4) SoR procs on glancing blows, and cannot be partially resisted.

5) SoV does NOT recieve level based resists.
If it does I will need to include these in my formula

6) Weapon specialization applies to SoR but not SoV.

Note on simualtion
The combat routine was:
1) Attack roll (with miss/dodge/parry all termed avoidance)
2) Check for SoV proc pbased on 20 ppm
3) Check if DoT application is resisted
4) Check when the DoT ticks and if that tick is resisted.
5) If the seal procs at the same time (within a 0.1 second step) as the stack would have ended, the stack is renewed not lost.

I ran the simulation over a 10000 second period in 0.1 second steps.

On purely autoattack dps, my simulation was showing SoR to match SoV at around +200 spell damage; before spell damage SoV was running at about 42dps. It was noteable that it was very rare for a stack to fall off with any speed of weapon. I havent yet put in a routine for judgements, but given that JoV will do more damage than JoR until high level of spell damage it looks like SoV will pull ahead once I include them.

Of course none of the above takes account of anything that might take you away from the ideal situation of standing and applying the Seal.

Edit
For those who are interested I threw together my simulator in quick basic, pretty quickly, I will try and leave it to run a very large sample tonight when I m home again. I will also try and ad in a routine for judging the creature, and check for any bugs. From simple debugging it looked to me as if it was running the logical statements correctly, and setting some of the variables (such as proc rate) to arbitrary values had the expected results (such as setting a 3.0 weapon to 10% proc and watching the stack fall off constantly).

Can anyone smarter than I am figure out the relative probabilities of losing a stack of SoV with a 1.5 speed weapon and a 3.0 speed weapon? I ask because my current tanking weapon is [Scimitar of the Nexus-Stalkers] (you can quit laughing now, I'm still levelling), and I'd hate to think I'm losing SoV stacks because of it's speed meaning SoV procs are unreliable.

I havent got the numbers to hand, but I tested speed of weapon from 0.5 to 3.5 and it didnt make any discernable difference to the predicted SoV dps in my simulation. I will try to put up some numbers when I get home tonight.

Edit for maths
weapon speed = 3.0
Proc rate = 100%
%Advoidance = Parry+dodge+miss = 15%
%Resistance = 17%

Chance to apply DoT = (1-0.15)*(1-0.17) = 0.7055
Chance to fail to apply = .2945
Chance of failing to apply 4 times sucessively (3, 6, 9, 12 seconds) = 0.007521 (0.8%)

Weapon speed = 1.5
Proc rate = 50%

%Advoidance = Parry+dodge+miss = 15%
%Resistance = 17%

Chance to apply DoT = (1-0.15)*(1-0.17)*0.5 = .35275
Chance to fail to apply = 0.64725
Chance of failing to apply 8 times sucessively (1.5, 3, 4.5, 6, 7.5, 9, 10.5, 12 seconds) = 0.03080 (3%)

So the slower weapon is more reliable

I recall about a year ago when I was leveling a paladin, that both SoR and SoCo were subject to both partial and full resists due to level difference. I even took a screenshot of my combat log of a seal of command proc being both parried and partially resisted within a few swings.

Don't know if this has changed since then, though.

I'll grant you I didn't do a significant test, but Seal of Blood seemed to have a 0% spell damage coefficient on the seal, and 43% on the judgment (take this with a grain of salt, I tested for about 5 minutes before unhotkeying it in disgust, and it was a few days ago). The 35% damage it dealt seemed to be unmitigated by armor, and it procced every swing. Seal of Command is 70% and procs 7 times per minute on average, leading to a very easy comparison. Seal of Blood outdoes Command (ignoring melee crits vs spell crits, resists (did not test long enough to see if the seal/judgment resisted, nor its crit properties) vs misses/parries, and so on and so forth) if you swing more than 14 times per minute. A 3.8 weapon swings 15.8 times per minute, so on paper Seal of Blood seems to be solid - it outdoes command slightly on even the most favorable weapons, and removes the delay dependency. However, spell coefficients change this in a hurry. Even Seal of Command benefits somewhat from +holy damage, unless my ret days were so far behind me that my memory is playing tricks. Seal of Righteousness certainly does, and for me, using a one hander with "hybrid" plate gear (str/sta/int/spell damage typically), it was simply not even close.

My paladin had around 350 spell damage at the time, and I used a 1h weapon. Seal of Blood did roughly 1/3rd Seal of Righteousness's damage per swing, and the judgment did around 1/2 to 2/3rds. I've always preferred to wear spell damage over strength gear when tanking (and I tend to grind 5+ mobs at once primarily relying on Consecrate for my damage when soloing), so Seal of Blood was utterly useless to me personally.

To recap: my initial, vague, anecdotal and far too brief to be scientifically valid test results show...

Unless you wear exclusively str/ap gear to the complete detriment of spell damage, and do not judge crusader, Seal of Blood is inferior to Seal of Righteousness, and only superior to Seal of Command on faster weapons or in the complete absence of spell damage.

If there's some specific test with it you'd like me to try that wouldn't take terribly long, I'd be happy to oblige.

What's the damage return? How does it apply to seal damage, and does the judgement's self-inflicted damage increase with +dmg? A quick combat log transcript should answer that question pretty easily.


I'm not surprised by your impressions, by the way. Your gear is heavy on +dmg, which favors the seal that scales best with +dmg. If you were to use SoC with that same gear set, you'd find SoC lacking as well. (In that SoR will probably do more damage than it) But SoC does have a playstyle/gearset that supports it, and depending on how its damage return works, I expect there'd be one for SoB as well.

Damage return on what? Seal of Blood's damage reflection, or a more thorough experiment of its coefficient?

As to the first, it definitely scales with increased damage dealt, not some base value. The judgment seems to self inflict exactly 1/3rd the damage dealt.

As to the second, with 487 damage, the judgment (listed 325-356) hit for 532 (and damaged me for 177). That leads to an approximately 40% coefficient, and given damage ranges and past trends it's probably 43%, like my initial guess. The seal itself is definitely getting no bonus from +damage. With +788 damage, the seal was hitting in the 70s (my character sheet damage range is 154-237), and me for 7-8. The judgment hit for 665, and inflicted 221 on myself.


Additionally, I believe seal of blood acts as a melee effect. I just had my seal dodged, and while it's only crit once, I believe it was for 2x damage.

The "gearset/playstyle" for Seal of Blood seems to have a complete and total focus on melee stats with zero spellpower, and a faster 2h weapon. It's just...it seems to be a pretty mediocre playstyle/gearset, and I'm not sure why anyone would do it.

Well, obviously for SoB to work well, you're going to be using primarly, or even solely melee stats instead of spell damage. When I was specced heavily ret this was the case, spell damage was barely a consideration when it came to dps, as seal of command is based off of melee attack power and melee crit primarily. Yes, spell damage did increase my overall seal damage, CS, and judgement, but strength and attack power worked doubly on the seal damage and melee damage and CS, and melee crit worked on white damage, seal damage, and crit as well as CS.

I think SoB could be a retribution seal, and it will be great for heavily melee geared retribution paladins as an offtank seal. You will be running sanctity aura, which, improved, the increased healing will mitigate the damage taken from the seal.

Basically, a melee geared paladin wont have the spell damage necessary to hold aggro using SoR, and Command has a tendency to be somewhat streaky, which is something you don't want to have when building threat. SoB plays off the melee paladin's natural strength and AP to give a blood elf ret paladin a seal which can generate a consistent amount of threat that's high enough to off tank. Especially considering most tank weapons are going to be fast, and fast weapons are doubly bad for command, due to low proc rates and low damage when they do proc.

I think it's situational, just like SoV, because if only one faction has it, it shouldn't be a staple skill. And just like SoV is terrible without spell damage (50dps at a full stack), SoB is terrible without melee stats.

Using solely melee stats to the complete exclusion of spelldamage might make Seal of Blood the best choice, but unless gearing that way holds some attraction, then it's pointless.

In terms of tanking, spell damage for threat is far better than ap for threat. Consecrate is a large portion of our threat, particularly in any situation where we're facing more than one monster, and it gets a VERY nice coefficient.

In terms of primarily dealing damage, well, I'm not really qualified to comment. Screw retribution.

I'm not sure if this is what you're actually asking - what spell damage coefficient does Seal of Blood get? The answer is zero. Seal of Blood receives absolutely no bonus whatsoever from +damage. It is only affected by listed weapon damage and attack power. The reflected damage is approximately 10% of the holy damage done, but something weird appears to be going on. I am in class right now, but I would see things like "Seal of Blood hits target for 180. Seal of Blood hits you for 22." Possibly some kind of armor mitigation miscalculation is going on. I'll get an actual log later tonight. But yes, it is treated as a melee attack - it can be dodged, miss, crits for 2x, and can proc fiery weapon on its own.

As for Judging, the answer is yes. The higher the judged damage, the more damage the judging paladin will receive. It is the most inexplicably stupid skill currently in the game. Judging Blood is currently equivalent to shooting yourself in the face - sure, maybe there are a few extremely rare circumstances where you will need to shoot yourself in the face. Generally, however, it is an undesireable course of action. With Vengeance, Imp Judgment of the Crusader, Libram of Fervor, Sanctity Aura, trinkets, etc., I have Blood judged myself for almost 500.

I'm well aware of the necessity for spell damage for holding threat. Especially in a multi-mob situation.

I'm not talking about a multi-mob situation though, nor am I talking about a situation where you're primarily tanking.

Gearing for melee at the near exclusion of spell damage has a point, and that is for retribution DPS. You say screw retribution but not everyone does. If you don't like retribution, you wont like SoB. Just like a heavy retribution paladin will see no point to SoV.

I'm talking about the Retribution paladin who is geared for dps, plans to DPS for a fight, but is asked, or forced to pick up a monster mid fight and has a reasonable melee weapon and shield. The retribution paladin can:

a) Use SoR and try and pull him with near 0 spell damage.
b) Use SoB and actually generate some holy threat because he has massive attack power and physical DPS output already.

After the OTd mob is dead the ret pally can switch back to 2h and continue to DPS

SoB has the added bonus of keeping the paladins health below full to trigger PoM bounces, with the detriment that it is basically a DoT on him.


If you hate retribution, you'll hate Seal of Blood. If you're holy or Prot there's no real use for it that I can see, use SoR instead. That's fine. Personally I don't much care for SoV, every time I've tried to tank with it I end up losing a stack somehow even with a 2.9 speed weapon.