I edited the title for spelling and the color of the enchant title. Nice post! (have to read it now!)
A preliminary view from the tanks' perspective
With patch 2.2 comes a new weapon enchant of interest to melee classes:
Formula: Enchant Weapon - Executioner
Requires Enchanting (375)
Use: Learn how to permanently enchant a
Melee Weapon to occasionally ignore 840 of
your enemy's armor. Requires a level 35 or
Is Executioner better than Mongoose?
That's the question on everyone's mind right now. One this document attempts to answer, based on the limited information available at this time. Since the proc rate and buff duration remain unknown, but given that the formula will be obtained from a 10-man raid zone in the same fashion as Mongoose, we will assume that they are the same as Mongoose and compare the two enchants based on that assumption.
What we know
Based on its description, Executioner has no mitigation or avoidance component, so Mongoose wins in that regards. However, Executioner promises a steady damage increase that scales based on the target's armor, affecting all damage dealt as soon as it procs. In comparison, Mongoose's damage component revolves around an increase in the player's chance to crit, which makes it somewhat less reliable.
Mongoose DPS (threat value)
The overall damage increase on Mongoose comes from a combination of crit from agility and haste rating.
The crit% increase is a direct percentage increase to overall damage. With Mongoose alone, warriors get a 3.636% increased crit chance from the 120 agility buff. In most raid situations where BoKings is available, Mongoose becomed 132 agility, or a 4% increased crit chance.
For warriors the haste rating only affects auto-attack and Heroic Strike. Given that the availability of rage directly affects warriors' ability usage, the proportion of a warrior's overall damage that can attributed to HS and white damage can vary quite a bit depending on the situation.
For simplicity's sake, and because I can attest to this being a realistic value based on WWS reports of my own damage breakdown by ability, I will base the math in this document on the assumption that 50% of a warrior's damage comes from HS and white damage in a raid tanking situation. So with a 2% attack speed increase, this translates into a 1% damage increase while the Mongoose buff is active.
So we end up with a 4% damage increase from crit and a 1% damage increase from haste, for a total 5% DPS increase. Protection warriors get no crit-based synergy between their talents and abilities, so the math pretty much ends here.
Executioner DPS (threat value)
For detailed information on how armor works in WoW, see Satrina's guide on Mitigation, Armour, Diminishing Returns.
Armor reduces all incoming physical damage by a fixed percentage. As armor value increases, the increase in mitigation is subject to diminishing returns. What this implies in the case of the Executioner enchant is that various armor values between one target and the next will each yield a different increase in damage to the wielder. The enchant will have a more significant effect when fighting a target with low armor (think Shade of Aran) than when fighting a target with high armor (i.e. Void Reaver).
The target's level also has a direct effect on mitigation from armor. A level 73 mob with 10000 armor will mitigate less damage than a level 70 mob with 10000 armor. What this means is that the 840 armor reduction from Executioner will have less of an effect on a level 73 mob than than it will on a level 70 mob with the same armor value.
To find out how much DPS Executioner will net you based on your target's armor, you would find the target's initial mitigation, then find the target's mitigation with the 840 armor reduction from the enchant, then compare the two.
For example, on a fairly well armored level 73 mob at 10 000 armor:
Mitigation = 10000 / (10000 + (467.5 * 73 + 22167.5)) = 45.537%
After reducing that mob's armor by 840:
M = 9160 / (9160 + (467.5 * 73 + 22167.5)) = 43.371%
Calculating the damage increase:
100 / (100 - 43.371) * (100 - 45.537) = 103.977, or a 3.977% DPS increase.
While not entirely impossible, it's fairly difficult to find the exact armor a mob has, but we do know from experience that it can vary quite a bit. This is where the most important difference between Executioner and Mongoose becomes visible. Using the mitigation formula linked above, here's a list of results against a level 73 target for various armor values:
10000 (45.537%) - 840 = 9160 (43.371%) = 3.977% dps increase
7355 (38.080%) - 840 = 6515 (35.264%) = 4.548% dps increase - Void Reaver *
5000 (29.481%) - 840 = 4160 (25.806%) = 5.211% dps increase
2990 (20%) - 840 = 2150 (15.237%) = 5.953% dps increase **
1500 (11.144%) - 840 = 660 (5.229%) = 6.656% dps increase
The DPS comparison
The examples above show that in the lower armor ranges, Executioner provides a superior damage output compared to Mongoose. Against highly armored targets Executioner falls slightly behind.
At first glance, most tanks may decide that the minor damage increase from Executioner does not make a good case for the loss of the mitigation and avoidance provided by Mongoose, which is a perfectly fair argument. However, when it comes to threat generation, Executioner does provide an advantage over Mongoose. Executioner provides threat that can be qualified as steady, or reliable threat. Obviously, only as reliable as its proc rate. But given that Mongoose is also proc-based, we can ignore that aspect of both enchant for the purpose of comparing the two.
When Mongoose procs, it will give you added threat if you get a crit. Over the 15 second duration, you may be lucky, or you may no be so lucky. You may get a Shield Slam crit right as Mongoose procs. Or you may get a string of dodges, parries, some glancing blows and a couple of white crits.
When Executioner procs on the other hand, the buff gained will immediately increase all your damage by a fixed percentage. It may not feel as exciting as extra crits, but you know that the added damage and threat will be affecting all your attacks over the duration of the buff. Another minor, yet interesting benefit from Executioner is that even glancing blows will deal more damage, something that can't be reproduced by an increase in crit%.
So even in situations where Executioner puts out a lower overall damage increase than Mongoose, it may remain a preffered choice for some tanks due to the steadier nature of its damage increase.
So which one should I use?
If you only have one good tanking weapon, Mongoose will give you the most all around benefit due to its avoidance / mitigation component.
However, most experienced tanks will tell you to carry multiple tanking "sets", for maxing out avoidance, effective health, and threat, depending on each particular encounter. Ideally you should have a few weapons to match those various applications. For example, a typical early raid setup might be to use a Sun Eater enchanted with Mongoose for max avoidance, and a King's Defender with Executioner for max threat. Pick a different one for each fight, or swap them mid-fight based on the mechanics of the encounter.
In the end the decision between Executioner and Mongoose should be whether you want Executioner's more predictable, steady threat on each proc, or Mongoose's extra mitigation and avoidance, and somewhat less reliable short-term threat.
* : Wartorn indicates here that Void Reaver has 40% mitigation, or 7973 armor after 5 Sunders. Also see follow up discussion here.
** : 2990 armor value (20% mitigation) included as a reference based on Ciderhelm's reply here.
Last edited by Armstrong; 08-08-2007 at 04:42 PM. Reason: Added Void Reaver to the list of examples
I edited the title for spelling and the color of the enchant title. Nice post! (have to read it now!)
BoP enchanting recipes from raid zones are usually blue items.
[item]Formula: Enchant Weapon - Mongoose[/item]
[item]Formula: Enchant Weapon - Soulfrost[/item]
[item]Formula: Enchant Weapon - Sunfire[/item]
And according to Blizzcon 2007 : Changes mega-list !, so is Executioner.
Last edited by Armstrong; 08-07-2007 at 04:40 PM.
It's definitely blue, but the default Blue color won't give enough contrast on the frontpage given the blue background. That's all!
1. This article would have significantly more impact if you gave us even approximate armor values of real bosses so we could gain some perspective. What kind of mob has 10,000 armor? 5,000? 1,500?
2. Reliability only matters in the short term. In a long fight, chance averages out overtime. (ie. 120 agility 50% is identical to 60 agility 100% of the time) To me this breaks down a lot of your comparison.
3. You accent the word "all" in regards to the armor reduction as if to imply that Increased Crit doesn't affect "all" abilities whereas in reality it does.
4. One interesting gem in the middle of your comparison is Glancing Blows. I'd like to see you go into more detail about why -armor is better than +crit and what kind of difference we will get as a result.
1) Most bosses begin with a rough maximum of 35% Damage Reduction from Armor (normally lower). The concept here is that raid bosses are generally scaled by Health, not Armor, to best balance cross-class raid DPS. Recalling some very rough numbers from EJ, after full debuffs most bosses are around 20% Damage Reduction from Armor. I'd wager my memory is off on this, but chances are you'll rarely be fighting bosses with even 5k+ Armor.
2) Reliability in the short term is reliability. While it's true that long-term Threat is important, and you are correct on that, a tank is most likely to lose aggro in the first 60 seconds of a fight.
3) Crit doesn't affect all abilities at all times; instead, it only affects abilities when the chance occurs. The chance is constant, but the benefit itself is not applied to every single swing.
I think I'll go back and try to calculate approximate boss armor based on WWS's when I have more time.
It will be pretty rough, especially with my current work schedule, but I'll give it the old college try.
Back to blue! I need to update it again because that color just isn't working, but for now I need to step out for a few minutes.
Armstrong, in the meantime I've posted this to the frontpage of the website. I had to shorten the title in the thread to do this without taking up two lines in the header (I added the second part as a subtitle in the thread).
I feel there is no significant difference in reliability of threat increase between mongoose and executioner.
There are two aspects to threat generation: initial aggro, and threat over time.
NEITHER one is reliable for early threat generation. For the purposes of initial aggro, neither one helps. (Pretty much only hit rating/weapon skill help).
Both will increase threat over time. Mongoose will do it in big chunks with a crit, and executioner in smaller chunks, but on the scale of building up threat over the course of a fight, and holding it over the dpsers who are initially waiting and then laying in, both are reliable.
Mongoose gives close to 5% more dps when it procs. (Plus dodge + armor)
Executioner must give MORE than 5% more dps to have any value for tanking. And no, 6% wouldnt cut it for me, I'm not giving up 4% dodge and 240 armor for just 1% more dmg, in any tanking situation. I can make much more favorable swaps than that by putting on a dps piece.
It would have to be at least in the 8% range for me to consider it probably.
Since it doesnt even reach that level of value on low armor mobs, imo it has no value for tanking, if these calculations are correct.
It probably has value for dps against mobs that are not heavily armored.
Ciderhelm covers the value of steady, reliable threat vs. irregular threat under the http://www.tankspot.com/forums/hold-...gular-dps.html section of Hold the Line.
The analogy would be having to choose between block value and agility, or between hit or crit, or between 3 points in Imp HS and 3 points in Cruelty. Over the long term they all increase threat to various degrees, but the reality is that short-term reliability is a significant factor in many situations. At least in my experience. Enough that a reliable gain in the short term can be well worth a small loss in the long run.
In that respect, the fact that Executioner is proc based does make it less reliable than Major Striking for example. However when comparing it strictly to Mongoose the proc aspect can effectively be ignored when it is assumed that the proc chance and duration are the same.
That said, thanks for the suggestions, I plan on spending a little more time on this tomorrow and including a few of them in the document.
And yes it only partially applies to the Executioner vs Mongoose discussion due to them being proc-based.
Anyway, got a raid to start. Talk to you tomorrow.
Two questions I'd bring up for possible discussion.
1) Could we determine the chance of a crit derived from a proc of Mongoose on a per-minute average?
2) Does anyone have any evidence that this actually shares the PPM rate of either Crusader or Mongoose? (I'm kind of an ass for asking this given I'm the one who suggested it in the first place)
Chance to proc a "1 proc per minute" effect (Crusader, Mongoose) using a 1.6 speed weapon:
1.6 / 60 = 2.667% chance to proc on any successful hit
White damage & Heroic Strike per minute:
60 / 1.6 = 37.5 swings
Global cooldowns per minute: *
60 / 1.5 = 40
Substract global cooldowns spent on Demo Shout (2), Thunderclap(2), Battle / Commanding Shout (0.5) & Shield Slam (10): **
40 - (2 + 2 + 0.5 + 10) = 25.5
Total actual attacks
37.5 + 25.5 = 63 attacks
Substract Miss (8%), Dodge (5.6%?), Parry (5.6%?), Block (5.6%?). ***
63 - (63 * (0.08 + 0.056 + 0.056 + 0.056)) = 47.376 successful attacks
Average number of procs per minute:
47.376 * 0.02667 = 1.263 procs per minute.
Percentage of the time the 15 sec Mongoose buff is present:
(15 * 1.263) / 60 * 100 = 31.584% uptime
Overall crit chance attributable to Mongoose (4%) alone:
4 * 0.31584 = 1.263%
There you have it. 1.263% actual crit chance increase from Mongoose.
Just for the sake of completeness, the 2% Haste effect applied to 50% of overall damage with 31.584% uptime becomes:
2 * 0.5 * 0.31584 = 0.31584%
So an additional 0.316% damage increase overall. For what it's worth, this also translates into an equal increase to the chance of Mongoose proccing.
The less-than-scientific aspects of this analysis:
* Lag and reaction time will affect the use of global cooldowns negatively, bringing the final result down some
** Shouts and TC will likely be refreshed a little more often than their exact duration, and some resists will occur. Final result goes down some more.
*** I'm really not too certain about the figures for mob avoidance. Those will also vary from player to player depending on the amount of Hit Rating and Weapon Skill involved. Could bring the end result either up or down.
Looks good except for the miss percentage. Did VR last night with 5.5% hit and for the 3rd time I observed ~3% miss rate. At this point I'm going to say it wasn't a fluke.
So when mongoose procs its up for 15 seconds.
Thats 9.375 autoattacks and 10gcds. Lets say 8 of those GCDs you use an attack, we'll say 17 total attacks.
The chance that the 4% crit from mongoose gives at least 1 crit is:
1 - (.96 ^ 17) = 1 - .499, or 50%
You'll get an average of .68 crits due to the mongoose proc.
You get a 50% chance of at least 1 crit due to the increase %crit, each time it procs.
When Executioner procs youll get a few% more dps, on every hit. Somewhere around 5-6% it seems. When mongoose procs youll get a 50% chance of double damage on at least 1 hit, with an average of .68 extra crits. It is an average of 4% dps increase from this. You also get a fixed ~1% dps increase from the haste. You also get 4.4% dodge and 240 armor.
imo, aside from the first few seconds of the battle, all I care about is average threat over time. Neither one helps with initial threat. Both help with average threat and damage over time.
If what I care about is only the damage and threat generation of the enchant, then the 5% average dps increase overtime from mongoose is barely worse than the 5-6% average dps increase over time from executioner.
But thats not all I care about. I'm very willing to sacrifice 1% average dps during the proc for 4.4% dodge and 240 armor!
Mongoose is much better for tanking imo, (unless the executioner procs more often or something)