PDA

View Full Version : Brooch of Deftness



thugthedum
12-15-2007, 07:09 PM
So I spent my first 35 badges on the Brooch of Deftness. I was wearing Natasha's Battle Chain and thought "wow, this is better for tanking".

And it kind of is, but the Battle Chain actually has more avoidance and more effective health (for my gear; I checked today). Now, the Brooch is a great threat piece, and I wear it proudly in my effective health, threat, and grind dps sets ... but, well, I couldn't vendor the blue. And I'm considering putting it back in my effective health set too.

Thoughts?

Ceravantes
12-15-2007, 07:36 PM
Vendor the blue, the effective health gained from the bv (taking your word for the difference, didn't check myself) is most likely very small anyways. Expertise is great for threat, but you are underestimating the survivability factor involved in it. New content parries play a big part in spike damage, which is what kills tanks, expertise could eventually negate this completely, but the minimization of it at all makes this a better neck than anything available in t5and lower instances imo.

thugthedum
12-15-2007, 07:44 PM
Good point.

Anyone have a relative value on expertise as avoidance? Something like 20 expertise points = 35 dodge points or something... or can we not look at it that way.

Esillymidget
12-15-2007, 07:58 PM
I would guess the mitigation gained from not being parried is dependent on the boss' swing speed and stuff, so it'd be hard to model like that. I wouldn't worry too much about that anyways. Not being parried is nice, and it is certainly useful for raid tanking on new content, but it is first and foremost an amazing aggro stat.

Ceravantes
12-15-2007, 07:59 PM
4.3 rating=1 expertise which = .25% dodge and parry reduction. iirc

thugthedum
12-15-2007, 08:11 PM
So 4 exp (17.2 rating) = 1% fewer parries.

Assuming a 2.0 swing speed and 40% speedup (esilly is right, that's one heck of an assumption) ... 4 exp reduces that 16% chance to give the boss a .8 second speedup to 15%.

i r bad at math, but it's fun to try. Certainly someone will come tell me what's wrong with this.

17.2 rating = 4 exp = 1% chance parry not to happen; bosses are commonly thought to get 16% parry, and 2.0 is a common boss swing time.

In ten minutes, then, a boss with no specials gets 300 swings on me. If I swing a KD, and do an instant every gcd, and land 70% of it all, I'll get 655 swings. That gives the boss 104 something chances to parry; each one of those speeding up his swing by .8 seconds.

Now, in reality; that .8 second speedup could be life or death. I'll take the 1%, but I'm trying to put a number to it.

And that number, to me, looks like instead of adding 83ish seconds worth of swing time, in which he can land 41ish extra swings on me, he gets 39 extra swings on me.

So, in ten minutes, 1% chance to parry off the table means two hits that don't happen.

Right?

Esillymidget
12-15-2007, 08:43 PM
Now that I'm thinking about it, I think i read somewhere that for mobs, parries act like an instant attack proc, whereas for us it's a 40% speed boost on the current swing.

thugthedum
12-15-2007, 08:49 PM
:(

Well, if that's the case it's 6 attacks off the table. In ten minutes.

Bruizer
12-16-2007, 03:38 AM
Worth remembering is the timing of those extra attacks - right after another one, meaning that if you're really unlucky, you may take two regular hits in next to no time. This, in reality is as bad as taking a crit. But wait, it's worse - one or both of those hits could be a crushing blow, depending on whether your shield block charges had been used up already or not.

Hence, it could be more combined damage than even taking a crit would be. And given a hard-enough hitting boss, that could quickly eat up your HP bar before healers have time to react.

Iowynn
12-16-2007, 04:45 AM
Worth remembering is the timing of those extra attacks - right after another one, meaning that if you're really unlucky, you may take two regular hits in next to no time. This, in reality is as bad as taking a crit. But wait, it's worse - one or both of those hits could be a crushing blow, depending on whether your shield block charges had been used up already or not.

Hence, it could be more combined damage than even taking a crit would be. And given a hard-enough hitting boss, that could quickly eat up your HP bar before healers have time to react.


I've actually had that happen and take a 15k parry crush. It's definitely NOT common, but if you have some random unexplained death, this is probably what happened.

Ceravantes
12-16-2007, 06:04 AM
Worth remembering is the timing of those extra attacks - right after another one, meaning that if you're really unlucky, you may take two regular hits in next to no time. This, in reality is as bad as taking a crit. But wait, it's worse - one or both of those hits could be a crushing blow, depending on whether your shield block charges had been used up already or not.

Hence, it could be more combined damage than even taking a crit would be. And given a hard-enough hitting boss, that could quickly eat up your HP bar before healers have time to react.


I've actually had that happen and take a 15k parry crush. It's definitely NOT common, but if you have some random unexplained death, this is probably what happened.

The reduction in chance of these things happening is what makes expertise so good, the additional threat gained is just an added bonus. Before 2.3 when fighting Morogrim, I generally didn't die, but came close a number of times each fight. After 2.3, with only expertise upgrades (Mallet over Blazefury, neck and bracers) I rarely notice my health drop like it did before.

At one point I went through some old wws logs and some new ones, If I get some time I will post some info on them up here.

Nilya
12-16-2007, 07:08 AM
Expertise is as much a survival stat as a threat stat. Nullifying more than half a boss' chance to turn you into chunky salsa is as good as any avoidance stat, if not better.

thugthedum
12-16-2007, 08:01 AM
Worth remembering is the timing of those extra attacks - right after another one, meaning that if you're really unlucky, you may take two regular hits in next to no time. This, in reality is as bad as taking a crit. But wait, it's worse - one or both of those hits could be a crushing blow, depending on whether your shield block charges had been used up already or not.

Hence, it could be more combined damage than even taking a crit would be. And given a hard-enough hitting boss, that could quickly eat up your HP bar before healers have time to react.

Good point. So are we saying that you factor expertise in to EH somehow?

Nuberino
12-16-2007, 11:57 AM
I believe that boss attacks are instant after a parry but they are they are still governed by a cooldown (say 1 second cooldown).

During Gruul last week, I unexpectedly died (21.5k hp buffed) at growth 6 right after a shatter. Parries were the cause, I wasn't taking huge damage but Parries caused spikage. Looking through the logs I found this:

--------------------------
// Hit shield block too early, faded just as shatter ended and a crush came through (though perhaps it was hastened by the shield slam parry.
12/9 20:49:31.359 Shield Block fades from you.
12/9 20:49:32.562 Gruul the Dragonkiller hits you for 6910. (crushing)
12/9 20:49:32.765 Your Shield Slam is parried by Gruul the Dragonkiller.
12/9 20:49:32.968 You gain Shield Block.

// Next parry and attack come 1 second later
12/9 20:49:33.218 MeleeDPS#1 attacks. Gruul the Dragonkiller parries.
12/9 20:49:33.546 Gruul the Dragonkiller hits you for 4918. (341 blocked)

// Next parry and attack come 1 second later
12/9 20:49:34.312 MeleeDPS#2 attacks. Gruul the Dragonkiller parries.
12/9 20:49:34.390 MeleeDPS#2 attacks. Gruul the Dragonkiller parries.
12/9 20:49:34.593 Gruul the Dragonkiller hits you for 4404. (341 blocked)

// Next parry and attack come 1 second later
12/9 20:49:35.546 MeleeDPS#1 attacks. Gruul the Dragonkiller parries.
12/9 20:49:35.734 Pet#1's Claw was parried by Gruul the Dragonkiller.
12/9 20:49:35.765 Gruul the Dragonkiller hits you for 4617. (341 blocked)

// Next parry and attack come 1 second later -- this is the killing blow.
12/9 20:49:35.765 MeleeDPS#3 attacks. Gruul the Dragonkiller parries.
12/9 20:49:37.031 Gruul the Dragonkiller hits you for 5584.

12/9 20:49:37.843 You die.

--------------------

This was just our second week on Gruul (downed him both times :D) so we are still working out kinks -- including that of melee other than myself not attacking until they are in his rear arc (the above was just after a shatter). The interesting thing about this is that there appears to be parries that aren't immediately returned, so it would appear to me that Gruul still has some sort of cooldown on his swing timer.

Just how valuable is parry reduction? I would say pretty large, especially given that the RNG is streaky at times (like above, the actual number of hits vs parries was very much in favor of parries -- though I removed the unessential hits).

Worst case scenario (assuming you are the only one with the possibility of being parried), you are doubling his attack speed. Parries that would fall in the later half of the boss's swing timer would have less of an affect.

Average scenario is that you are giving him ~50% haste when you parry at a chance of 10% per your attack (assuming bosses 10% parry).

If my stats are right (never my strong point), that means with about 2 of your attacks in each 2 second boss swing timer (lets take into account some small lag and occasional buff etc, ideally you would be much closer to 4 attacks in 3.2 seconds), you have a 20% chance of giving him that buff. So every 10 seconds, you are likely to be parried once, and on average shave .25 seconds off the bosses swing timer.

Over a 10 minute fight, that is 15 seconds of extra swing time the boss has or 7.5 extra attacks.

Assuming you reach the softcap of expertise to get a -5% parry reduction.

That should halve the number of seconds of extra swing time the boss has, so you are avoiding 3-4 hits in a 10 minute time span.

Assuming the boss will make 304 attacks in a 10 minute time span, you are basically looking at expertise of 20ish being worth just above 1% avoidance maybe slightly more if you have ideal conditions (no lag, fast response, 0 tc's and demos). Again with the assumption of a 2 second swing timer. My math is probably a little off, but I think it would be safe to say that it isn't a significant amount of avoidance, but the damage avoided is less valuable than the damage not spiked.

The biggest DEFENSIVE value to expertise is still in the reduction of spike damage. And we all know what kills tanks. Regardless threat is still the main value of expertise.

Narshe
12-17-2007, 12:58 AM
I can't remember a single death in the last month where I did get parried seconds before (apart from the regular wipes ofc)

thugthedum
12-17-2007, 06:21 PM
I see all you're saying, I guess I'm just looking for a way to quantify it all. It's not fair to just say "22ish exp = 1% avoid", although it appears to be. It's more than that; even ignoring the threat potential, because as posters point out, a parry during a shield block increases the chance that a crush slips through. So it is more than that.

Nuberino
12-18-2007, 08:20 AM
Tangibly the expertise softcap of 20 expertise (.25% for each point to get to -5% parry/dodge) is worth:

~1% avoidance + 10% threat

Intangibly it is worth %50 less chance that a boss parry will occur (may be boss dependent but the general assumption is that most bosses have ~10% parry) from your own attack and cause a damage spike that could potentially kill you.

thugthedum
12-18-2007, 04:03 PM
Tangibly the expertise softcap of 20 expertise (.25% for each point to get to -5% parry/dodge) is worth:

~1% avoidance + 10% threat

Intangibly it is worth %50 less chance that a boss parry will occur (may be boss dependent but the general assumption is that most bosses have ~10% parry) from your own attack and cause a damage spike that could potentially kill you.

wait; if bosses have 16% parry (I think i read that somewhere), why is the soft cap so low?

And: does anyone know if the tankpoints dude is factoring expertise in along either one of these lines?

It is significant; and I want a way to look at it intelligently; not just "lol expertise I want it lol".

Nuberino
12-18-2007, 04:21 PM
wait; if bosses have 16% parry (I think i read that somewhere), why is the soft cap so low? I am pretty sure most bossses have around 10% parry (can someone verify this?). But no matter the max parry of a boss, the softcap is due to most bosses have a dodge of 5.6% so as soon as you hit 22-23 expertise, you are only getting half the threat per point of expertise. From WoWWiki (if it can be assumed as reliable(?), bosses have a 5.6% dodge so in actuallity an expertise rating of 22-23 is actually the soft cap. Due to rounding 23 is necessary to eliminate boss dodging, but 22 leaves only a very slim chance for the boss to dodge so you are not getting the same value out of the 23 point.

I am not sure if you can look at it more intelligently than measuring the tangible (the avoidance value is heavily dependent on the bosses swing timer):
1 expertise ~= 0.05% avoidance + 0.5% threat

You really can't measure the importance of threat effectively along with health or avoidance as it is entirely subjective to the content you are doing and the raid you are with. Measuring the avoidance aspect of expertise which could be added to a tankpoints system would be fairly insignificant and not necessarily accurate. And actually putting a value on the intangible aspect of reducing boss parry %'s is again entirely subjective.

Karsa Orlong
12-18-2007, 04:24 PM
Expertise is great. Sitting at 27 in defensive stance. 6.75 percent reduction to dodge and parry. It helps a ton when tanking.

Arideni
12-19-2007, 09:18 AM
Where did you get 4.3? I thought it was 3.94 rating equals 1 point of expertise; ~15.76 points is 1% decreased chance,

I didn't get it from here, though! I'll have to check the guides real fast...

EDIT: Yep, "Expertise ____________ 3.9 to 1" by Ciderhelm - My numbers are a bit more precise, but the game will round them so then it makes sense.

Trema
01-08-2008, 09:26 AM
I can also substantiate that Moro 'seems' MUCH more steady now. My wife, who is always my main healer :) after our Moro kill commented that she never worried about me. I was using neck, bracers, and mallet. Good stuff. I think the era of waking up in a cold sweat, terrified of being effectively one-shot by Moro, is over!

Wartorn
01-09-2008, 11:18 AM
Personally, I think it's the best neck in the game.