How to debate badly

The armour versus stamina epic thread on maintankadin is an entertaining read. Other than being a good analysis of Total Effective Health, and its applicability to decisions between armour and stamina trinkets, it has a fascinating set of exchanges with a few belligerent posters who pretend to be honest debaters but are actually manipulative.

After a dummy-spit from Eminai (who had some good ideas but was unable to control his vitriol), here’s Theck’s rebuttal, below from p11. It outlines why Theck’s a theorycrafter to respect, beyond just his maths, but for his humility and level-headedness. I’ve said before, one thing I respect in him is the ability to admit he’s wrong and change his mind, which is a crucial quality of intelligence that I didn’t see in Eminai.

I put Theck in blue.

Eminai wrote:If you can’t see how adding healing done into a formula that assumes you receive invalidates everything about that formula, I shudder to think of what kind of aneurysm inducing “math” I would find in your actual theorycrafting.

I haven’t done this, which is abundantly clear from the examples I’ve given. It is you whom is misunderstanding things. Nobody is trying to redefine Effective Health.

Eminai wrote:In case you’re wondering why people like James troll these forums, look back over the past few pages of this thread. In these few hours, I’ve repeated myself, my data, and basic common sense so often I feel like I’ve tried to chisel it into granite using nothing more than my bare hands. All you have offered is contradictory statements, fallacies, and basically worthless information that appears to serve no higher purpose than to make it appear that you are taking an active part in this discussion.

You have repeated yourself for sure. But you haven’t provided any data that supports your point. You haven’t shown me a single parse, or a single real example that shows a tank dying due to sustained DTPS over 15 or more seconds. You’ve just repeated your baseless assumption, without proof, and then tried to call me names and twist my words to make it seem like I’m committing logical fallacies. Most of which is transparent to even a casual reader.

I, on the other hand, have provided clear derivations, simple explanations, and combat log parses that support my point. Real, in-game examples that show increasing effective health by only a few hundred stamina would have saved a tank’s life and potentially turned a wipe into a kill. There is nothing in that that contradicts my point.

Eminai wrote:Any person who had two brain cells to rub together has long since taken away every useful idea presented in here, and all we are left with is me, in my stubbornness, trying to teach calculus to a stone. So I am done trying to provide any sort of basis for my points, because you will accept no basis at all. You have defended EH theory with a fanatacism that approaches religious,

I would expect a zealot to repeat the same point over and over, without providing any first-hand evidence to support that viewpoint. I would expect them to gloss over the gritty details that might show the holes in their reasoning. I would not expect a zealot to provide first-hand data that supports his point, just in case it got turned against him. And when presented with evidence that they have made incorrect statements, I would expect a zealot to respond with denial and hostility.

Looking back over the posts in this thread since last night, including the one I’m quoting right now, those statements about zealotry seem to describe your contributions to this thread much more accurately than they describe mine.

I have provided both theoretical and empirical evidence of my viewpoint. I have conceded that under certain conditions, your “belief” in the power of armor is well founded. I have gone pretty far out of my way to point out that you and I are simply talking about different models of tank damage and death, and that if yours turned out to be more likely I’d recant and share your views. And I’ve been pretty clear about the fact that I think James is an excellent player and a good tank. That doesn’t seem like zealotry to me.

Eminai wrote:and I have finally realized, albeit a fair bit later than one would expect, that you will accept no reason, no logic, no data that contradicts your holy ideas of tanking theory. Good day, sir, and may God have more mercy on your soul than you have had on my faith in humanity.

Nice try, but Billy Madison did it much better.

I don’t know what class you play Eminai. You and James are probably both very talented tanks. But from the vitriol you’re throwing into these threads, you come across as really shitty human beings.


The thread gets really interesting again when Hamlet, a fellow scientist like Theck, joins the debate on p12. It’s just a great read and full of win. Sadly, Hamlet starts off on a belligerent note, unnecessarily but does redeem himself quickly with good analysis.

Fascinating thread, particularly if you look at the meta and the structure of conversation, the failures of reasoning and the hilarious Klingon analogy near the end. To me, it shows the generally high quality of writers at maintankadin.

Related Posts:

13 comments to How to debate badly

  • Argon

    Yeah, there was a pretty ridiculous level of anger from some of those posters.

  • 10 armour = 1 stamina (there is some maths on the EJ warrior forums somewhere). Therefore 3090 armour > 2880 health (or whatever the number was). Do I win the internet?
    .-= Everblue´s last blog ..The four stages of fail =-.

    • Gravity
      Twitter: gravitydk

      Actually the debate centred around that ratio.

      The finding was basically that in a short window, say of 3 seconds, when you’re being healed, armour has higher value than the usual metric because it interacts positively with heals.

      • Theckd seems to be a very bright guy (as do several of the posters in that thread), and I don’t argue with any of his maths – my maths degree was a long time ago now! What I would say however is that “modelling gearing choices” is not the same as “doing very hard maths”. There’s lots more involved – the valid arguments on that thread are “what should be in the model”, not “how should the model work”.

        Focussing on “three second burst” is not wholly convincing to me, although I need to think about it a bit more, since it does not tally with what I see in my raids (I usually blow cooldowns through predictable burst and self heal when I drop low, and so I tend to die last – my healers are very good at keeping me up but in doing so we lose dps to raid damage so you get the “slow death” referred to on the forums). I appreciate that anecdotal evidence without parses and/or films to back it up is of limited use if any, and 10 man gearing is different to 25 man gearing since the size of hits relative to your hp is lower in 10 man (ie Blizz can only assume you have 1 healer on you in 10 man, whereas in 25 man it’s fair to assume you have 2).

        Key to progression in my view is whether you can run with two or three healers. If tank damage can be mitigated to the extent that you can run with two healers (ie the tank healer has sufficient free GCDs to do something else as well), then this massively increases raid dps and is one of the most critical hurdles to pass. For that reason I have always preferred armour to stamina.

        I haven’t read maintankadin much, but will give it a go. Thanks for the link – now I need to go and read that thread a few more times…
        .-= Everblue´s last blog ..The four stages of fail =-.

  • Gravity
    Twitter: gravitydk

    Testing. I just swapped to W3 Total Cache, plus WP Touch for mobile users. Checking everything still works.

    Editing seems ok.

  • Gravity tester
    Twitter: gravitydk

    This is also testing. I see the fancy MCE comments are not here and comluv works tooo.

  • eysen

    Looks good from my iPhone and hey I can post now.

  • Psy
    Twitter: thelordpsy

    Yeah was gonna say, hit this with my iphone a couple hours ago and it looked great.

    In other news, tank theorycrafting is so 2008. I’ve explained this on my blog, but none of it matters now. Blizzard changed the tank role from, “Maximize gear so that you don’t get decimated by hellishly powerful attacks” (See: Patchwerk 1.0, Brutallus, etc) to, “Learn when to use the cooldown to survive”

    Quite literally, in boss fights right now, there are times when you should never be in danger of dying and times when you should pop a cooldown; In the latter, if you pop a CD, you are not in danger of dying. If you fail to pop a CD, you get destroyed.

    Beyond that, if you have reasonable gear and make reasonable gearing choices, you’re fine. 1-5% differences in damage taken, max HP, avoidance ratings, etc are near-meaningless. They just make for really awesome debating because no side can prove anything- And that should really tell you something. People are bickering over failure event rates at or below .5% with variation of .1 to .2% in either direction. Maybe lower.
    .-= Psy´s last blog ..Solve The Immediate Problem =-.

    • Gravity
      Twitter: gravitydk

      Thanks Psy for pointing out the side conversation in the EJ Benefactors’ Bar about the maintankadin thread, starts around here:

      You need to be a sponsor of EJ to read it (I am).

    • EH planning still matters. Bosses have tank death scenarios built into the encounter, which require certain EH thresholds to survive. Understanding the mechanics behind the scenario, and which gearing set most effectively mitigates that scenario is still extremely important. The difference between a properly geared tank and an improperly or undergeared tank in the same scenario varies far more than the <1% value you've thrown out there. A tank going double armor trinkets while OTing Blood Queen will be in a noticably worse position that a tank in the same role who utilizes stam trinkets.

      We're talking about a basic understanding of the underlying mechanics of the game. The idea that against 100% MbA damage, Armor and Stamina are interchangable at a certain ratio, however, as a higher and higher percentage is not MbA, that ratio becomes more strained.

      In the end, the EH formulas are useful tools for showing the underlying mechanics of the game with more precision. However, the root problem in the MT threat is the use of those formulas. A tool being used is only as useful as the Tool using it. If not used in the proper context, and in the correct manner, these formulas can be wildly detrimental to intended goal of maximizing survivability.
      .-= The Renaissance Man´s last blog ..Leading Icecrown Citadel: Lady Deathwhisper =-.

      • Psy
        Twitter: thelordpsy

        My point isn’t that the difference in damage taken is less than one percent, or that the difference in effectiveness is less than one percent. Hell, I’d believe that a tank gearing solely for HP has 100% more catastrophic failures than a tank gearing solely for Armor. But as a percentage of overall attempts, on current content, less than 1% of attempts will end in catastrophic failure because of tank death. I’d say one in two hundred attempts on a boss, at WORST, will be a tank death -> Wipe. If one tank dies every 400th attempt and one dies every 200th attempt, is there any meaning in optimization? It’s still basically never going to happen.

        Once again, I’m not saying optimization is *bad*, nor that you should ignore it entirely. I’m just saying that current content is NOT built to strain your tank in any way shape or form. Content is built to require you to use your cooldowns. I fully expect Heroic Festergut to rail tanks, and if he does so even before 3 stacks, optimization will become more important. However, even if Blizz were to release a boss that required you to have a meaningful CD active 100% of the boss fight, such a rotation could be devised almost trivially; This being one of the problems with the state of WoW.

  • fung

    you don’t HAVE to win em all you know…

    also, i’m only a grammar-hitlerjungen and not quite full nazi yet, but in “It is you whom is misunderstanding things”, I don’t think whom is correct as it’s not the recipient of action–just who is fine.

    and on an actual serious note, i’d like your thoughts on the green armor amounts on frost-badge purchased items vs the nongreen armor on drops. i thought i heard something to the effect of bonus armor not receiving the frost armor bonus.. is this taken into account in the loot rankings i’ve been reading, do you think?

    Who dat? Who dat? Who dat say Grav dunno dem tanks?

Leave a Reply




You can use these HTML tags

<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>