Concluding the Defile vs. Breath of Sindragosa debate once and for all

---

Over the past week or so, there's been a considerable amount of confusion and debate among some elements of the Death Knight community about Defile vs. Breath of Sindragosa among DPS DKs. There's been faulty information passed around and feelings have been trod on, along with some serious blurring of the lines between "theorycraft" and intuition (both of which have their place in these considerations). I am writing this in the hopes of settling the matter once and for all. I also hope that this serves as an example for future instances where there's confusion about situations such as this.
The primary debate around these talents has always revolved around Unholy or Blood Death Knights. DW Frost Death Knights have never taken Breath for serious PvE progression, so what I’m about to say is largely focus on Unholy DKs- I think I’ve ranted quite a bit about Blood in other places.

-The first thing I want to get out of the way is that I do not, in any way, shape or form, condone harassment or rudeness. If a particular individual reading this feels as if they were unfairly targeted by others for passing on information that was perceived as being flawed, then please know that this does not reflect on the greater DK community. While we will certainly act to correct disinformation, it is never with the intention of personally attacking or slighting anyone.

-With that out of the way, I want to address the wider topic of theorycraft: How do we, as a community, reach a conclusion regarding a practice that we think is optimal DPS? Simulations are absolutely part of it, but they most certainly are not the be-all, end-all. We also rely on testers and, most importantly, parses. Fortunately for us, neither Defile nor Breath of Sindragosa are new talents. Thousands upon thousands of parses with both talents exist, across a very wide spectrum of play. We use these parses in conjunction with our simulations, along with having trusted players test them out in varying circumstances. My community is also fortunate to be able to corroborate results directly with blues, just to double check if something seem egregiously wrong. In short: We have multiple bases covered.

-For this reason, I take serious issue with attempts to misrepresent what we’ve been doing as “just relying on sims”. This is objectively false, and belies a lack of understanding of how a community driven process works. We are our own harshest critics in this regard, often asking one another to double or even triple check our results and APLs. It is only as a result of this scrutiny that we’ve been able to push forward the profiles for Breath of Sindragosa and Necrotic Plague usage that you see presently. It is only after this that we feel comfortable enough to measure things such as trinket/talent valuation for each spec. If you were curious, here are our results for 6.2.3 Unholy: http://imgur.com/a/vihsO

-Something that many people do not seem to understand about the DK buffs is how they relate to the massive ilvl/gear buffs that will soon be omnipresent throughout the playerbase. Fights in Hellfire Citadel and corresponding raids will become shorter across all difficulties. This in turn will increase the relative amount of uptime that shorter cooldowns (such as BoS and the ring) have in relation to the total fight length. What I’m trying to get at here is that Breath itself is significantly buffed by these circumstances. The fact that it didn’t receive a 10% buff like Defile or Necrotic Plague did is irrelevant- it is still innately stronger than either of these talents due to how boss fights are affected by the changes. Since we’re already in anecdotal territory, I recently had a conversation with a friend in QA at Blizzard specifically about this. I was initially annoyed at the fact that BoS had not received a buff similar to Defile and Necrotic Plague, but quickly came to understand why this was: BoS already exists as an outlier among DK talents, and has only gotten stronger

-On “Sims aren’t always correct”: I don’t know how else to respond to this statement with anything other than “Duh?” We as a community are well aware of the limitations of Simcraft. I’ve personally gone on record multiple times as saying that the sim falls short in many areas- one of the most prominent being in Blood Tap usage, which I devoted an entire BlizzardWatch column to. Furthermore, saying “sims aren’t always correct” without offering any evidence or counterargument to support your claims is an extremely weak argument in and of itself. As I’ve noted above: We as a community use multiple tools to assess the value of a particular talent or playstyle. Ironically enough, the argument that “sims aren’t always correct” is true but favours Breath of Sindragosa even more: We have LONG acknowledged that our BoS APLs aren’t optimal, and likely underestimate how powerful the talent is. This is in sharp contrast to Defile, which has been a fairly easy talent to sim. So if anything, using this statement to support a Defile > BoS argument is quite silly.

-On Reaper’s Harvest for Unholy and Simcraft: One argument that I saw being used to discredit our simulations was concerning Reaper’s Harvest. As many of you are aware by now, Simcraft did a fairly terrible job at modelling the Reaper’s Harvest trinket for Unholy. This made it look quite weak, and thus led to us undervaluing it. There’s a separate topic of discussion to be had about how it’s currently quite useless for most fights, but we can discuss that another time. The point I am making here is that the reason the trinket showed up as being weak is due to the Necroblight rotation being extremely hard to set proper conditions for. We’ve worked on it for months, and still come up short due to the sim allowing Necrotic Plague to fall off at times (when in reality, it is kept on a boss full time for several encounters), and so on. As a result, it’s only natural that the sim would vastly undervalue the effect of Wandering Plague.
However, to imply that this has any semblance of similarity to the inclusion of the 6.2.3 DK buffs only shows a fundamental lack of understanding of how Simcraft works. Implementing these buffs was not remotely challenging- they were simply 10% increases to existing damage sources (like Defile and Necrotic Plague). APLs didn’t change, rotations didn’t change, etc. In short: Simulations run with these buffs are much more reliable and trustworthy for us than “newer” territory like Necroblight and Wandering Plague.

-It is both facetious and in extremely poor sport to try and claim that a talent isn’t viable in the present because it MIGHT get nerfed in the future. Regardless of whether or not Breath of Sindragosa survives in Legion, it is a part of the Death Knight toolkit right now. Pretending otherwise is foolhardy. Similarly, unless you work at Blizzard or have contacts there, you have absolutely no idea whether or not Breath will get nerfed in the coming weeks. Claiming that you “think it’s likely” is a strawman argument, and absolutely not what any theorycrafter should be engaging in. For the record, we are currently at the tail end of an expansion, and have just received considerable buffs as a class. Breath of Sindragosa being nerfed now would not only make zero sense, but would also go against Blizzard’s track record. Hotfix nerfs are fairly common during the beginning of an expansion or tier, not at the tail end when everyone’s largely going through the motions.

-Nowhere has anybody in the Death Knight community claimed that the Breath of Sindragosa rotation is “easy”. When we put out theorycraft about why one talent is worth using over another, it is with the tacit understanding that players understand it is THEIR responsibility to determine how far they want to push min-maxxing of their character. Just to give another example, my Wowhead DK guides are meant for brand new players who have never played WoW before. For this reason, I discourage using Breath of Sindragosa immediately in them- because I’m conscious of my audience. However, when I put out theorycraft that I believe every single player, including Mythic raiders, should adhere to, then it is always with the maximum possible benefit in mind. Yes, Defile is “easier” than Breath of Sindragosa. And guess what? Plaguebearer is easier than Plague Leech. Runic Corruption is easier than Blood Tap. And so on. Using “ease” to justify why you claim something is better to use than a counterpart that has proven itself to perform vastly better when used properly isn’t just facetious, it’s lazy- and entirely contrary to what theorycrafting is about. On that same note, using “enjoyability” as a reason to recommend one talent over another is just as bad. Many of us “enjoy” the challenge and extremely punishing nature of Breath of Sindragosa due to the rewards it yields when played properly. However, were it a weaker talent than Defile we would have dutifully reported this.

-I’ve also read claims that there needs to be a “perfect situation” in which to utilise Breath correctly. Unfortunately, this statement is utterly devoid of any logic in light of how many fights in Hellfire Citadel see DKs use Breath on them. I’m not just talking about pure single-target ones such as Fel Lord or Reaver, but also ones that have favoured Necrotic Plague in the past, such as Archimonde or Mannoroth. A big reason that players have given for this is that fights getting shorter has meant that overall effectiveness of Necrotic Plague has been largely dampened, and that even the cleave damage that was once essential to their raid groups is now quite unnecessary. As a result, they have chosen to focus more heavily on the single target power that Breath brings. Yes, multiple encounters in HFC have cleave: But how much of that cleave is MEANINGFUL damage? How much of it is largely unnecessary at a time when Mythic geared players see an average ilvl of 740? I’m not even going to get into the fact that well over half of HFC fights saw players use Breath of Sindragosa as the default talent during progression- logs alone say enough here.

-In line with my last paragraph, there is an argument that says “The longer a fight is, the worse BoS becomes”. This statement is factually correct, but is incredibly misleading because it comes right after “Alright, Breath is better on a one minute fight”. Let me make something clear: Trying to compare Breath vs. Defile on a one-minute fight is a complete and utter joke. Breath pulls so far ahead in these circumstances that the difference is astronomical. The reason for this is easy to understand, as the first Breath on any encounter is typically the strongest- all cooldowns/procs are active, Empower Rune Weapon is available and pre-combat Runic Power can also be used. But here’s the thing: The difference between Defile vs. Breath does not reduce nearly as much as some would believe on longer fights. How does this work? Simply put, one must calculate the average damage gain of Breath (and subsequent loss of Death Coil damage) and compare it to the damage gain from Defile (and the loss of Scourge Strike on those runes, since Death and Decay isn’t even used on Unholy’s single-target at the moment). What I am talking about here is basic napkin math, not even anything incredibly complex. It should come as little surprise, therefore, that the actual number of Breath of Sindragosa ticks required to beat Defile (after which every following tick is pure gain) is fairly low. This is not even counting the power of trinkets such as Vial of Convulsive Shadows or the effect of the legendary ring proc- both of which are MASSIVELY in Breath of Sindragosa’s favour.

-Last point: I’ve also seen the word “viable” used when describing Defile. This word, in relation to theorycraft that focuses on min-maxxing one’s character, is utterly irrelevant. Yes, Defile has always been VIABLE as a talent. But so what? By that logic, 2h Frost is a “viable” DPS spec in that it can put out a certain margin of damage and has an arguably simpler rotation than Unholy. Unfortunately, being “viable” means fuck all when we are talking about “optimal”- which is where Breath of Sindragosa beats Defile handily. It’s also fairly disingenuous to refer to Defile as the “best” option one day, and then retreat into “it’s viable” territory the next.


There is probably a lot more I could say- both in terms of the math proof behind why BoS > Defile (Hint: It’s napkin mathable, that’s how obvious it is), and also why trinkets have been misrepresented so badly. At this point, however, this has gotten overly long. At the end of the day, facts will be facts: The assertion that BoS > Defile from a theorycraft point of view isn’t an “opinion”, it is a ground reality. If anyone disagrees with this, we in the community are more than happy to listen and engage with any proof they can offer that challenges the resources at our disposal. Unfortunately, so far all we’ve gotten are ad hominems and insinuations that we’ve tried to fudge sim results to suit ourselves. Not only is this an outright lie, it it deeply offensive to a community that has put so much work into these proofs.

I’ve tried to avoid explicitly naming anybody in this note, as I hope that the points I’ve made here resonate with people wider than the immediate audience specific to this matter. Additionally, I wish to repeat that this is not meant as a personal attack against anyone’s work. I respect that content creators work hard to accomplish what they do, and I know that it can be quite unpleasant to feel as if you are being dogpiled, thus becoming even more defensive. I encourage anyone reading this to visit us in the DK community. If you’d like to continue this discussion further, then please do. If you have other types of math or theory to offer, we’d like to hear them and offer input. If it turns out we’ve been wrong this entire time, then we will be all the richer for understanding this. Ultimately, however, the burden of proof falls on you as an individual to accomplish this.

My own goal, in posting this lengthy note, is more to combat the disinformation that I see being spread, rather than directly address any one individual. Let me repeat myself: I am coming at this from the perspective of someone that focuses specifically on min-maxxing. Players are free to play with whatever talents/glyphs/specs they enjoy- the issue only becomes personal when they begin to try and pass this off as “optimal”. I hope that anyone reading this will view it with a dispassionate eye, and acknowledge that it comes from somebody simply interested in a more holistic community understanding of our class, and not because I wish to prove someone else "wrong".

As I noted before: People are still welcome to join the DK community and discuss/debate this further. Ultimately, beyond that, there is nothing we can do but correct bad data where we see it.

Sincerely: I wish all of you the very best,
Magdalena

Reply · Report Post