Is something wrong with these numbers? Or am I way off?

Warning: Theorycraft inside.

Moderators: Fridmarr, Worldie, Aergis, theckhd

Is something wrong with these numbers? Or am I way off?

Postby Sintor » Sat Dec 04, 2010 4:59 pm

Today I was asked why I, as a tank, get hit harder and die more often than the other tanks in my guild. I too wondered this, and i thought "am i not as good of a tank as i think i am? Or is there something else going on here?" So i looked at our warrior tank, Kenjo of Zuluhed, and compared our numbers and did a bit of math.

Kenjo has 48.57 block, and I have 49.66 block.

This is where the problem first occurred to me. Aren't I, as a paladin tank, supposed to have a more noticeable chance to block attacks than the warrior tanks? The mastery that I get tells me that per point of mastery, I should get .75% more block rating than what a warrior would get, and also that I should get 6% more block rating just from speccing into the protection tree.

Paladin Mastery: Divine Bulwark - Increases your chance to block melee attacks by 18%. Each point of Mastery increases block chance by an additional 2.25%.

Warrior Mastery: Critical Block - Increases your chance to block by 12% and your chance to critically block by 12%. Each point of Mastery increases your block chance by an additional 1.5% and your critical block chance by an additional 1.5%.

Now, by these statistics alone, I should always be ahead of a warrior's block chance by at least 6% when we have similar gear and stats. To confirm this, I put the numbers into the math equation and tried to find out how much mastery rating it takes to gain a single point of mastery.

My character sheet says I have 544 mastery rating, which gives me 11.85 mastery points.

544 / 11.85 = 45.9

So it takes 45.9 mastery rating to gain 1 point of mastery.

But my character sheet also says I have 19.85 mastery points, so I assume I am given 8 when speccing into a specialization at 80. (I confirmed this by doing the math again with the warrior's numbers, He had 19.04, and 507 rating, giving him 11.04 mastery points, 19.04 - 11.04 = 8.)

Now I did some more math regarding how much block I should be getting from my mastery.

Mastery for Sintor: 19.85 mastery points x 2.25% block per point = 44.6625 block %.

Mastery for Kenjo: 19.04 mastery points x 1.5% block per point = 28.56 block% and 28.56 crit block %.

Now, i thought to myself, what if the 8 mastery that didn't come from mastery rating, just applied the initial block chance, and was not counted in the math for deciding block chance gained from additional mastery.

11.85 mastery points x 2.25% block per point = 26.6625

11.04 mastery points x 1.5% block per point = 16.56 block% and 16.56 crit block%

After doing this part of the equation, I added up the numbers to see what they ended up being.

Numbers with the 8 mastery rating added
Paladin: 44.66% block + 18% from specialization = 62.66% block.
Warrior: 28.56% block + 12% from specialization = 40.56% block.

Numbers without the 8 mastery rating added
Paladin: 26.66% block + 18% from specialization = 44.66% block
Warrior: 16.56% block + 12% from specialization = 28.56% block

Either way, These numbers do not stack up to be the actual numbers displayed on our character sheet.

So, theory-crafters, math geniuses, paladin tanks, why are these numbers not correct?
Sintor
 
Posts: 2
Joined: Sat Dec 04, 2010 4:57 pm

Re: Is something wrong with these numbers? Or am I way off?

Postby Meloree » Sat Dec 04, 2010 5:29 pm

44.66 + 5 = 49.66
28.57 + 5 + 15 = 48.57

Warriors get 15% block from talents, both warriors and paladins get 5% base block.
Meloree
Maintankadonor
 
Posts: 1420
Joined: Wed Mar 12, 2008 10:15 am

Re: Is something wrong with these numbers? Or am I way off?

Postby Sintor » Sat Dec 04, 2010 6:51 pm

Wow, i must be really tired. My bad, thanks for the reply.
Sintor
 
Posts: 2
Joined: Sat Dec 04, 2010 4:57 pm

Re: Is something wrong with these numbers? Or am I way off?

Postby Gamingdevil » Sun Dec 05, 2010 10:11 am

You're correct in your assumption everyone gets 8 mastery for speccing into their tree. That's where the 18% block chance comes from (8 * 2.25% = 18). It makes it easier to understand for new player to just put 18% block there in stead of 8 mastery (and more confusing for advanced players :P)
If it is weak, kill it before it gets stronger. If it is strong, weaken it.
Image
User avatar
Gamingdevil
 
Posts: 1738
Joined: Mon Jan 14, 2008 8:16 am
Location: Belgium

Re: Is something wrong with these numbers? Or am I way off?

Postby Redneckizm » Sun Dec 19, 2010 9:15 am

Today I was asked why I, as a tank, get hit harder and die more often than the other tanks in my guild. I too wondered this, and i thought "am i not as good of a tank as i think i am? Or is there something else going on here?"


*Warning- Wall of Text*

@ Sintor... This is what led me here also. I never was "over in-depth" with number crunching & Theorycrafting. I'm smart enough & know just enough to know an upgrade when I see it, Know the stats I need to push and research enough to stay on top of what I need to be doing. I've only been tanking about 3.5yrs... BUT. Back to the basics & researching again looking for the answer to "WHY???"

I hear "Diminishing Returns" constantly being thrown around. What i've really been searching for is the formula or the actualy # value of the cutoff. I remember seeing a long formula a long time ago that you use to calculate the # cutoff (ie 1957 Dodge is the point of DR)

I had a vent conversation & stat comparison with another Guild Tankadin, Across the board i'm better geared, beat him in all stats, use the same strats/rotations but I feel i'm getting stomped a lot harder and the healers have a much harder time keeping me alive than him.

I dropped about 200g in Reforging back & forth trying to find a better solution. I'm cutting out Threat generating Hit/Expertise, my Threat has been fine with only a very rare stray running around. I was using the Avoidance Macro found elsewhere here on this site to check my numbers.

At what point is it better to go for an extra 1% chance to completely avoid a hit Dodge/Parry or is it better to push for that 1% damage absorbtion of the Hit. Looking at combat logs the average hit I take is roughly 22k. is it worth that 1% absorbtion to absorb 220 damage or push that 1% to Dodge/Parry the whole 22k hit??

@ Meloree. Lot of people appreciate the hard work <Edge> does, Thanks for being awesome and giving the "rest of us" a template to follow. /brownnose off. I got into a couple of the <Edge> H ICC 25 Alts runs toward the end of Wrath. I learned more in the 4hrs watching/adjusting/doing than I did in 4 days of reading advice/recommendations/theorycrafting debates.
Last edited by Redneckizm on Thu Jan 13, 2011 4:17 pm, edited 1 time in total.
Image
Redneckizm
 
Posts: 4
Joined: Sat Dec 18, 2010 12:57 pm
Location: Garona

Re: Is something wrong with these numbers? Or am I way off?

Postby theckhd » Sun Dec 19, 2010 1:26 pm

Redneckizm wrote:I hear "Diminishing Returns" constantly being thrown around. What i've really been searching for is the formula or the actualy # value of the cutoff. I remember seeing a long formula a long time ago that you use to calculate the # cutoff (ie 1957 Dodge is the point of DR)


Diminishing Returns simply means you get less avoidance from each additional point of rating than you did from the last. It has no "cutoff." It does not "turn on," nor does it have a "point." It starts taking effect from the very first point of rating you gain.

I'm surprised that you didn't find this yourself, since the first two results you get when Googling "diminishing returns avoidance" have an explanation of DR as well as the explicit formula.


Redneckizm wrote:I had a vent conversation & stat comparison with another Guild Tankadin, Across the board i'm better geared, beat him in all stats, use the same strats/rotations but I feel i'm getting stomped a lot harder and the healers have a much harder time keeping me alive than him.

It probably has to do with something else entirely. Maybe he uses cooldowns, trinket procs, and armor potions more effectively or more frequently. Maybe you tend to move the boss around a lot more, and open yourself up to extra unblocked hits because you turn your back to them momentarily. Maybe you have /sit macroed to your Crusader Strike. :P

Unless you're using extremely different gearing strategies (i.e. stacking stamina vs. stacking mastery/avoid) or have very different average gear levels (full 359 compared to a mix of 333/346 blues), you won't see a large variation due to gear.


Redneckizm wrote:Subtracting the "given" from the 102.4% formula, I have to stay over a 57.4... Hafta recheck that, the Macro calculates the 5%, may have included twice. Even if its included twice... Target minimum 62.4.


There is no "given" in the 102.4%. You actually have to reach 102.4% with that macro. Holy Shield does not give you 40% chance to block, it increases your total block amount to 40%. In other words, Holy Shield will not help you block more frequently, it will help you block more damage on a blocked swing. If an unblocked hit does 100 damage, a blocked hit will prevent 40% of that and leave you taking only 60 damage.
"Theck, Bringer of Numbers and Pounding Headaches," courtesy of Grehn|Skipjack.
MATLAB 5.x, Simcraft 6.x, Call to Arms 6.0, Talent Spec & Glyph Guide 5.x, Blog: Sacred Duty
User avatar
theckhd
Moderator
 
Posts: 7966
Joined: Thu Jul 31, 2008 3:06 pm
Location: Harrisburg, PA

Re: Is something wrong with these numbers? Or am I way off?

Postby Redneckizm » Sun Dec 19, 2010 6:00 pm

The top 2 returns are old data, that was great in '08 when people were calculating diminishing returned for Defense ratings.
Tankspot is a great place to go read on Boss Strats. Tankspot & EJ's forums are so polluted with Druids/DK/Warr & Pally's overrunning each other's posts that every other post is from a DK wanting to know how much X he obtained from 3 spirit. j/k but you know what I mean.

The 3rd return is this site, where I came looking specifically for Pally Tank specific info/advice.

as I stand this moment unbuffed
Dodge 12.43%. Mouse over: Dodge Rating of 1611 adds 9.12% Dodge
Parry 12.82%. ^^ Parry Rating of 1500 adds 8.49% Parry
Block Chance 36.80% ^^ Block rating of 40 (Shield enchant) adds a 0.45% chance to block. Your block stops 31% of incoming damage. (41% with Holy Shield active)
Mastery Rating 1063 (+5.93 Mastery)

I understood the Holy shield increases 10% amount of damage absorbed after the hit. Not the chance to block a hit.

I'm looking for a hard number "X=Dodge" once X is obtained, push toward "Y=Parry" until Y is achieved and put everything beyond that point into "Z=Mastery". example: I have a Belt buckle socket on my belt, do I want to put +40 Parry, Dodge, Mastery or Stam. If I know X is real close to a "Soft Cap" and will improve with my next upgrade and Y is so far off the chart I'll never get to it, i'll put a Y gem in it.

In the mean time, the summary of all the walls of text just say: Mastery. Even if you have an Item like "Helm of the Proud" from the JP Vendor. Comes with 142 Dodge and 142 Parry. Since my Parry is 0.39% better than my Dodge, all of the Reforging Guru's say to reforge the Parry into more Mastery... Yes, I get more Mastery but now my Parry Chance is reduced.

At what point is "X" and "Y" enough. If X and Y were totally useless, we wouldn't have any at all. But all that I read from Theorycrafters is that X and Y are useless, make everything Z.

While researching a MM Hunter somewhere about WoW-3.3 I found a formula that valued Diminishing Returns for AGI .vs Attack Power. I don't recall the exact number but it was essentially once you "soft capped" in the vicinity of 1647 it became more beneficial to Gem for Attack Power instead of AGI to increase DPS.

So now i'm researching a Prot Pally somewhere about 12 Days after Cataclysm release to find a formula that values L85 Diminishing Returns for Dodge and Parry. I would like to find a hard number that once you "soft capped" in that vicinity it becomes more beneficial to Gem/Enchant for the other or just be a "Mastery Clone", smile, stand in front and get your face bashed in. call it fun & log off.
Image
Redneckizm
 
Posts: 4
Joined: Sat Dec 18, 2010 12:57 pm
Location: Garona

Re: Is something wrong with these numbers? Or am I way off?

Postby Bevan » Mon Dec 20, 2010 2:00 am

Redneckizm wrote:I'm looking for a hard number "X=Dodge" once X is obtained, push toward "Y=Parry" until Y is achieved and put everything beyond that point into "Z=Mastery".


It's not that simple. The relative values of dodge and parry shift as you add more of one or the other. Generally speaking, you want to try to keep them close to even, to minimize the impact of diminishing returns on each.

Mastery, on the other hand, has a more steady value, and is better than either. That doesn't make them useless, it just means that given the choice you should add mastery before the others.

There are posts about reforging, gemming and more around here, but you need to do the reading and apply the information to your specific gear to make choices. And you'll need to do it again with each new piece you get.

There's no easy catch-all rule here. You can come up with general guidelines to use, but even then there will be changes in effectiveness as you add stats.
Image
Bevan
 
Posts: 31
Joined: Sat Jan 12, 2008 5:41 pm

Re: Is something wrong with these numbers? Or am I way off?

Postby theckhd » Mon Dec 20, 2010 6:18 am

Redneckizm wrote:The top 2 returns are old data, that was great in '08 when people were calculating diminishing returned for Defense ratings.
Tankspot is a great place to go read on Boss Strats. Tankspot & EJ's forums are so polluted with Druids/DK/Warr & Pally's overrunning each other's posts that every other post is from a DK wanting to know how much X he obtained from 3 spirit. j/k but you know what I mean.


The second hit is the EJ "Combat Ratings at level 85 (Cataclysm)" Thread. That's a whole lot more recent than 2008, and it's very well-kept-up by Whitetooth. It's probably the single most useful thread on EJ, in fact. In any event, the DR formula itself hasn't changed since it was introduced, so Satrina's guide is a good starter guide. The only thing that's changed since level 80 is the constants k and C that you use in the formula.

Redneckizm wrote:I understood the Holy shield increases 10% amount of damage absorbed after the hit. Not the chance to block a hit.

Then what did you mean by the "given" in the 102.4% formula? Misinterpreting Holy Shield as giving 40% block is a pretty common error, and matched the error in your numbers, so I assumed that's what you meant.

Redneckizm wrote:I'm looking for a hard number "X=Dodge" once X is obtained, push toward "Y=Parry" until Y is achieved and put everything beyond that point into "Z=Mastery". example: I have a Belt buckle socket on my belt, do I want to put +40 Parry, Dodge, Mastery or Stam. If I know X is real close to a "Soft Cap" and will improve with my next upgrade and Y is so far off the chart I'll never get to it, i'll put a Y gem in it.

In the mean time, the summary of all the walls of text just say: Mastery. Even if you have an Item like "Helm of the Proud" from the JP Vendor. Comes with 142 Dodge and 142 Parry. Since my Parry is 0.39% better than my Dodge, all of the Reforging Guru's say to reforge the Parry into more Mastery... Yes, I get more Mastery but now my Parry Chance is reduced.

At what point is "X" and "Y" enough. If X and Y were totally useless, we wouldn't have any at all. But all that I read from Theorycrafters is that X and Y are useless, make everything Z.


You might be out of luck. Tanking has more facets than DPS, there's no single metric that you maximize to the exclusion of all else.

For example, do you want to minimize damage taken (i.e. TTL, or Time to Live)? If so, there are well-defined caps for X and Y: 10% Parry and 9% Dodge.

But is that metric actually meaningful enough to base our gearing strategy on? Maybe not. If we consistently die to unexpected strings of unblocked hits, or equivalently "damage spikes," then minimizing the chance to take an unblocked hit (aka maximizing attack table coverage) could be more important. Since mastery is always better than avoidance at accomplishing that, your rule of thumb would become "always reforge to mastery." In essence, X and Y are zero here.

What about items with one threat stat and one avoidance stat? The naive view based on minimizing damage taken would almost always lead you to reforge the threat stat. But if you want combat table coverage, then there are situations where it's actually more efficient to reforge the avoidance stat.

And of course, there's always EH, the metric we lived and died by in Wrath. While EH doesn't tell us anything relevant to reforging (short of trying to reach block cap, at least), it gives us another metric to consider when choosing gems. Do we put a 40 mastery gem in that yellow slot, or a mastery/stam hybrid?

The point here, of course, is that it's difficult to say with certainty what X and Y are in your model. They change from boss to boss based on encounter mechanics, and even in a simplistic "only consider boss melee" model, it's not clear that one is appreciably better than the other. Most progression tanks and theorycrafters on these boards seem to favor the "damage spike" minimization model of mastery stacking based on personal experience. And once you're fully 346-geared or higher, it's actually hard to reforge down past 11% dodge or parry anyway, making the first two strategies identical anyway.

Most of this has already been discussed in the Reforging/Mastery thread. The guide itself is pretty out-of-date, but the discussion that follows it touches on all of these issues. I'd suggest you start your research by reading that thread in its entirety, it's full of good information.
"Theck, Bringer of Numbers and Pounding Headaches," courtesy of Grehn|Skipjack.
MATLAB 5.x, Simcraft 6.x, Call to Arms 6.0, Talent Spec & Glyph Guide 5.x, Blog: Sacred Duty
User avatar
theckhd
Moderator
 
Posts: 7966
Joined: Thu Jul 31, 2008 3:06 pm
Location: Harrisburg, PA

Re: Is something wrong with these numbers? Or am I way off?

Postby Redneckizm » Wed Dec 22, 2010 8:32 am

Thanks. 9% Dodge 10% Parry is exactly what I was looking for. Dropped as much Dodge/Parry as I could, changed some gems, Block up to 42% and greatly reduced the amount/number of devastating hits I was taking. Back where I want to be, back to pulling 5-7 mobs at a time Questing/dailys. My healer is bored & having more DPS time now running Regs for guildies.
Image
Redneckizm
 
Posts: 4
Joined: Sat Dec 18, 2010 12:57 pm
Location: Garona


Return to Advanced Theorycraft and Calculations

Who is online

Users browsing this forum: No registered users and 1 guest

Who is online

In total there is 1 user online :: 0 registered, 0 hidden and 1 guest (based on users active over the past 5 minutes)
Most users ever online was 380 on Tue Oct 14, 2008 6:28 pm

Users browsing this forum: No registered users and 1 guest