Hunter Stat Scaling

Posted: by Frostheim


There have been a lot of comments about scaling and how our stats scale lately, and then I got into an email conversation with a reader about it, which got me playing with some numbers. This conversation starts back in Wrath, when we saw some very large hunter scaling issues.

The first was that all hunters scaled a bit more slowly with gear than other classes. The reason for this was that our pets only inherited a percentage of our AP — and did not inherit our haste, crit, or armor pen. Thus our overall dps didn’t scale linearly with any stat increase other than attack power.

Secondly, we had the wonky armor pen stat that scaled exponentially. Thus your second 100 armor pen gave you a lot more damage gain than the first 100 armor pen. At very high levels this scaling was very, very extreme.

I’ve stated before that right now it doesn’t look like hunters have any kind of crazy non-linear stat scaling. Of course with all the complexities involved with stats, talents, and glyphs, it’s always possible, but if there is any non-linear scaling, I suspect it’s very small. Of course, I could always be wrong. I do that from time to time, usually in regard to very obvious things as well : )

So I thought it’d be fun to go over some simple stat scaling examples. Specifically we’re going to look at crit and mastery — well, mastery for SV and BM at least. MM mastery is complicated beyond simple examples. My goal here is to point out that both crit and mastery scale linearly but combine to scale a bit better. There’s been a lot of thought out there that SV’s mastery will get crazy huge exponential increases when stacked. And I know it seems like it should somehow, but it doesn’t. At least, as far as I can see.

Let’s take a look.

Crit scaling

We’ll start with crit to get a baseline, and we’ll look at the simplest possible scenario. Let’s say that when we shoot something we do 100 damage. When we crit, we do double the damage — so that’s doing 2o0 damage! Awesome! It’s good to be a hunter!

Now we shoot something 100 times. We are starting off with 0% crit, so all 100 are just hits. They each hit for 100 damage, so we do a total of 10,000 damage. Go us.

Now let’s get 1% crit on our gear. We shoot 100 times, only now 99 of them hit for 100 damage, and one crits for 200 damage. We now do 10,100 damage total. This is a 1% dps gain, because crit rocks my socks.

Now lets get another 1% crit. We shoot 100 times, and now 98 are hits for 100 damage, and 2 are crits for 200 damage. We do a total of 10,200 damage. This is an increase of 0.99% to our dps — not the 1% we saw last time. Our extra 1% crit still gave us the exact same 100 damage increase, but since our total damage done went up, the percentage of the damage of our increase was smaller.

This holds true for any crit values.

If we have 50 crit rating we’re now doing 15,000 damage (50 hits for 100 damage, and 50 crits for 200 damage).

If we increase that by 1% crit we do 15,100 damage. We still increased our damage by the same 100 damage, but now that’s only a 0.66% increase from our previous 15.1k damage.

This is a perfectly linear scaling.

It’s worth noting, however, that crit is a bit more complicated than that. We still have our DoTs that are only doing 150% damage on a crit (we’re pretty certain this is a bug, btw), and when they fix our meta it will add 3% to our crit damage multiplier. On top of that every spec has abilities that proc off crits, adding to the value of crit a different bit for each spec.

SV mastery scaling

Okay, so we know how crit scales, and we know it’s linear, so now lets take the same example and look how SV’s mastery works. Every 1 mastery will increase our magical damage done by 1% — that’s percentage scaling! Woohoo!

Let’s take the same setup. We fire a shot that does 100 damage. It’s  magical, so our mastery affects it. We fire that shot 100 times. With no mastery at all we’d be doing 10,000 damage from those shots.

If we now add 1 mastery, we take that 10,000 * 1.01 to get 10,100 damage. A 100 damage increase, boosting our dps by 1%. Sweet… but somehow strangely familiar…

Now lets add another 1 mastery, for 10,000 * 1.02 to get 10,200 damage. This is now a 0.99% increase to our dps — not 1% anymore, even though the extra 1% mastery gave us the same 100 damage increase as the first 1% did. Exactly like crit, this scales identically through all levels.

If we have 50 mastery, we’re doing 15,000 damage. With 51 mastery we’re going up by that same 100 damage to do 15,100, which is now just a 0.66% increase.

SV’s mastery scales perfectly linearly, and in fact nearly identically to the crit scaling. But, there’s a but. While crit boosts all of our attacks and our pet’s attacks, mastery only increases the damage of our magical attacks. Admittedly, for SV, this is the vast majority of them, but it’s still not helping auto-shot, Kill Shot, or the pet’s thousands of dps. And it doesn’t have the advantage of crit to all those proc talents like Sic ‘Em.

The BM mastery works in exactly the same way. The scaling uses 1.7% instead of 1%, but the scale is also perfectly linear, with the percentage gain from later 1 mastery smaller than the first, but the numerical gain always the same for each 1 mastery. And of course this only affects the pet’s attacks, which is less than 50% of our dps, which is why mastery is worst for BM hunters.

Stat interaction

Of course, our stats don’t exist in a vacuum. Each stat is affected by other stats. But crit and mastery affect each other identically.

You could argue that as crit increases, mastery increases exponentially. After all, all those 200% crits are now increased by 1% per mastery in addition to the hits. That makes mastery way better!

Of course the opposite is also true — you could argue that as mastery increases crit scales exponentially, which makes crit way better! So let’s have some more fun and play with some more numbers. Let’s say we can have a crazy high 50 mastery, or 50% crit, or 25 of each mastery and crit. Does combining them provide a greater increase than they do alone?

Again we’re firing our 100 damage shot, 100 times. With zero of everything we have 10,000 damage done.

With 50% crit we have 15,000 damage done.

With 50 mastery (SV) we have 15,000 damage done.

With 25 crit we get 12,500 damage done, which our mastery then boosts to 15,625 damage done!

Or wait… perhaps it’s when we have 25 mastery that we get 12,500 damage done, which our crit then boosts to 15,625 damage done!

My point is that it’s not one stat scaling better than the other — it’s that both stats scale better in the presence of the other. So by combining the two, we get a total that is greater than stacking any one separately! Exciting! And this total is best at a certain magical ratio of crit to mastery — in our simplified example it’s best with half crit and half mastery (20% crit and 30 mastery, or vice versa, yields only 15,600 damage, for example), but reality is a bit different because crit scales in larger steps, changing the ratio.

So with this interaction, the combination of the two stats does actually scale non-linearly, though very, very slightly.

What does it mean?

Before everyone runs out to reforge to have exactly the same crit and mastery, don’t do that. Let’s recap and clarify a bit.

Crit scales linearly. Mastery scales linearly, but with a smaller scaling. By themselves, one crit will always be equally better than 1 mastery.

Crit and mastery interact to scale very slightly non-linearly, to our advantage. This is where math serves us up cookies drizzled in awesomesauce.

Because crit alone is better than mastery alone, the magical combination of the two would favor crit slightly — so you’d always want a higher net crit chance than mastery percentage, but always a combination of the two in the right ratio. Of course even with zero mastery rating on your gear, you start with 8 mastery for free, whereas with zero crit (or agility) you’d have a negative crit chance (yeah, really). So you also need a big chunk more crit (9.5%ish more) just to catch up to make the percentages the same, let alone favoring crit a bit for that magic ratio.

But, there’s a another but. A big one.

We don’t just have these two stats — we also have haste, which also interacts in interesting ways with both crit and mastery. But happily it’s not that complicated to figure any of this out, thanks to a lovely dwarf named Zeherah.

Zeherah’s dps analyzer does indeed take all of this stat interaction into account. As long as you do a lot of variants on the spreadsheet to take into account odd haste and crit swings, the values you get from Femaledwarf.com are indeed your stat weights. If it says crit is better, then go for crit.

And for SV, it says crit is better. So do the target dummies.

I haven’t played around with all gear combinations yet, but when I first checked it out with all first raid tier of gear, crit was still well ahead of mastery. But we’ll continue to keep our eyes on the ratings to see if we ever manage to compile enough crit for the two to start switching positions (which should happen eventually — after all, that agility is giving us crit as well, so we’ll be gaining crit faster than other secondary stats).

Hopefully this explains a bit about how our stats scale, and hopefully you find these though experiments fun as well. Mastery does not become magically better once you have tons of it, and neither does crit (or haste, for that matter). They both go up in the same steps, but a combination of the two, with a bit more crit than mastery, makes them go up even more, and combine for a multiplicative advantage.

Update: Check out Corwyn’s math in the comments. He did some playing on Zeherah’s spreadsheet to find where the magical ratio is, then calculated the balance. Upshot is you probably want to stack crit until you’re around 44% crit or so (and realistically probably more, since as your gear improves, you’ll end up with more mastery).

Facebook Twitter Snailmail
  1. Tokiko says:

    Tank you for this post, Frost!

    Just a few days ago i wrote a blog on how to use femaledwarf to evaluate what you should reforge into. And of course if you spent a lot of time on femaledwarf sometimes you get very irritated.
    As of now what I’m aiming for is 8% hit ofcourse, then a Cobra of about 1.5-1.6 Seconds and then i weight crit > mastery. Given some simple steps on FD you sure get freaking annoying results sometimes. Making mastery scale like AWESOME the next 40 points, but after that again mastery is just a net dps loss. Especially if you’re in between gears (mix of blue/purple) doing correct reforging sure takes time!
    Theoretically a change of gem from matery to crit or vice versa could result in a dps gain of a few points, but this is completely strange and doesn’t happen very often.

    All im preaching is what I’m doing myself… go for 8% hit, go for haste to reach about < 1.6s of Cobra/steady cast and then reforge any mastery that you're able to make into crit. To me this is what seems most appropriate right now.

    Do you have any thoughts of a "haste-soft-cap"? I could imagine there being several small softcaps throughout the tiers. Like I'm saying, right now it might be 1.56s of Cobra-Cast, a tier later it might be 1.32s. But stuff in between will just mess up your rotation due to more complex focus management?

  2. Tallind says:

    Excellent post as always!! I’ve recently registered on the site, even though i’ve been a longtime reader. As a fellow diehard SV hunter, I enjoy discussing these sorts of topics and putting theories and facts to the test to find the ideal number for any given scenario. Sure, it’s hard to manage to get to that number on the fly, but it’s good to know that we were as prepared as possible.
    Let’s say you’re in a raid, you have your crit about 9.5%ish higher than your mastery. Both the 5% crit buff and 8% spell damage debuff (think of it as mastery buff) are present, and the providers of these buffs are really good at not dying when they can help it. In order to balance those two linear scaling stats to get the correct ratio you would have to compensate the for the 3% difference. Now let’s add horn of winter and blessing of kings. From those two buffs alone you get an extra 3.21% crit), so now you’re back near the gap you want. Blessing of kings is going to increase your agility, and therefor crit throwing you out of balance again. So depending on raid make up, you may or may not need to change a few stats around to reach this 9.5% gap. It’s really late or I’d sit down crunch numbers and run a bazillion tests. Maybe tomorrow. I’m all about squeezing every last ounce of dps out of my gear, unfortunately i’m in a position where I can’t raid, so i’m in no hurry to grab expensive enchants for my gear that may get replaced before i even hit the raid scene. But that doesn’t mean i can’t do numbers :)

    • Frostheim says:

      Whoah — you do NOT necessarily want a 9.5% gap. I was just pointing out that If you had no gear at all, then you’d need to collect 9.5% crit just to have the same amount of crit as you do mastery (since you start with 8 mastery and -1.5% crit). I’m not suggesting that is the magic ratio.

      • Tallind says:

        Thanks for clearing that up Frost :)

  3. Corwyn says:

    The important thing is not which stat to stack, but rather what is the relationship you should have between stats.

    For survival:
    Damage = total damage * (1 + Crit%) * Physical% + total damage * (1 + crit%) * (1 + mastery%) * magical%
    physical% + magical% = 100% = 1 (we will solve for magical %)

    Damage factor is thus damage / total damage (call it D)
    and call E = magical percentage, c = crit percentage, m = mastery

    D = (1+c)(1-E) + (1+c)*(1+m)*E

    D = 1 – E + c -Ec + E + mE +cE + cmE

    D = 1 + c + mE + cmE

    taking partial differentials we get:

    dD/dc = mE + 1

    dD/dm = E + cE

    Equating them to find the maximum

    mE + 1 – E – cE = 0

    c = (m + 1 – E) / E

    c = m + ((1 – E) / E)

    Thus if, say, the magical percentage of all damage is 75%, then

    crit% = mastery% + 33.3%

    Since mastery starts at 8%, you should stack crit only until 41.3%

    After that point, you should add crit and mastery at roughly the same rate.

    Of course, that assumes (as you did) that crit value = mastery value, but if it does not, you can just modify the E term appropriately.

    • Corwyn says:

      2nd to last in the derivation should be:

      c = (mE + 1 – E) / E

    • Norville says:

      Have you taken a look at what Mr. Robot is saying for Survival hunters to do since 4.2 released? They are saying that they have found that haste is better then crit and I have tried in the last couple of day’s what it say’s and found that so far while I’m not positive I’m getting more dps then what I was doing before at worse it’s a very minor loss. If you have looked at it and tried it what’s your opinion?
      They are using these stat weights

      weight soft cap soft-capped weight hard cap
      Attack Power 1.04 ∞
      Ranged DPS 5.33 ∞
      Ranged Speed 7.63 ∞
      Agility 3.33 ∞
      Stamina ∞
      Mastery 0.95 ∞
      Physical Hit 2.66 .08
      Crit 1.02 varies 1.02 100.0 %
      Haste 1.5 ∞
      Resilience
      Spell Penetration

  4. Fuggli says:

    5 of 5. Good stuff

  5. Yukarta says:

    Ok seriously, all this math kinda makes me lost, can anyone break it down in simple terms for me ! Say I have 10 % mastery and I have 10% crit, if I am reading this correctly tho the magic ratio then would be 10% mastery to 13.3% crit or there abouts ? or 10% mastery to about 11% ish crit. there seriously needs to be another way to simplify this for us Blue collar hunter types ! when I input all the stuff and tweak the the hunter dps spreadsheet to any of these combinations it actually lowers my dps. I know this is simple stuff to you guys but can we break it down a bit more simple ! PLZ :)

    • Corwyn says:

      If you have 10% mastery, you want at least 43.3% crit. (assuming my made-up 75% elemental damage) before you start adding more mastery. You can check recount for you actual percentage of elemental damage.

      Alternatively, if you have 50% crit and 10% mastery, you want to add Mastery until you get to 16.7% mastery.

    • Frostheim says:

      Here’s the simple version for you: stack crit. That’s all you need to know at the moment : )

  6. Eidelweiss says:

    Thanks as always for an awesome post. I just have one question:

    Now when we look at our character in the game it has a ‘dps’ number that goes up and down as you change gear. I’ve just been using that for now. Would you guys say it takes these calculations into account like Zehera’s dps analyzer or is it showing something different?

    • Corwyn says:

      No, the DPS on the character sheet, is just (I think) average(min;max) / speed.

      There may be something else involved, but in no way are the calculations the same as the complicated simulation done by Zehara’s.

    • Frostheim says:

      Seriously, ignore that number. It’s garbage. It is only the auto-shot dps taking only haste and AP into account. It’s totally ignoring crit, hit, mastery, and all your special shots. It’s worse than useless, it’s actively misleading many hunters.

      • Eidelweiss says:

        Thanks for that.

        Bells sort of went off for me when my PVP gear was showing much higher DPS than my PVE gear :) .

      • Taz says:

        Is the Hit Chance number on my character profile equally bogus? It was showing 8.09 hit chance with a 0% chance to miss on targets level 85-88, but on Femaledwarf.com I was almost 200 “under” cap. I upped my hit chance to 10.0% (in-game) and the spread sheet still says im 12 under cap. Perhaps Femaledwarf.com is still going through some tweeking but which number is a more relalistic indication of my actual hit, my character page number or the spreadsheet?

        Thank you

      • Rotseh says:

        Taz, make sure femaledwarf.com reads out your gear probably when loading from Armory. Especially the items with random enchants (such as the Star Chaser Belts or Mistral Circle rings from Throne of the Four Winds that even the official Armory does not display correctly) will not show up properly and you need to add their relevant stats (Hit, Crit, Mastery, Haste) manually summed up as Custom Gear Stats.

  7. Painbow says:

    A friend of mine made some good points on the state of Cataclysm with melee and ranged DPS on the wow General forums: http://us.battle.net/wow/en/forum/topic/1658710180?page=1

    It’s true. When he’s tanking, I can use explosive trap and multishot as SV because he’s geared, but with other tanks and CC, we’re losing our aoe abilities and single targeting often.

  8. Corwyn says:

    I ran Frostheim through Zeherah’s analyzer, and with his current gear (and survival spec), he has 14% mastery, and 24% Crit. I then added crit until I got crit and mastery worth about the same, which amounted to about 4700 crit rating. This gives 50% crit.

    Plugging this in gives c = m + 36%, which implies that E is around 73% So (adjusted) elemental damage is around 73% of the total. I would love to know how this matches logs of actual results.

    Upshot: Crit until at least 44%.

  9. Factoid says:

    Amazing article, thank you so much Frost and Corwyn for explaining it for all us mathematically challenged hunters. It helps me understand why I was seeing the things I was… although I highly doubt that I would have ever arrived at the proper percentage to start optimizing.

  10. Slickrock says:

    New BluePost(tm) up, looks like MM and BM are gonna get buffs, SV might get touched down, and haste might be more important. Take all the numbers, put in a pot, and stir.. :-)

    Sounds like the changes will come in a patch in a month or so, as it might show on the ptr first.

  11. christ85 says:

    Why is there no mention of the fact that ones critical strike rating is chance and it is not definitive. Determining if you critically strike or not is still stochastic no matter how high your chance to critically strike is. Probability is like-hood not certainty, so why is it being treated as such? Out of those 100 shots none could be crits, but with mastery you are guaranteed an additional 1%. Think of it this way, if I were to present you with 2 investment opportunities, one that guarantees you an 8% return on your money every 100 days, and the other option would be that you have an 8% chance to double your initial investment every 100 days. Which one would you take? You can make more money with the 8% chance do double it, but you have a 92% chance to make no money at all. So how does 1% crit chance = absolute 1% more damage?

    • Heyguysitsme says:

      It does increase your overall damage done by 1%. Maybe not in every fight but all in all 1% chance to do 200% dmg is of course 1% damage in the long run. It will work the same way as Frost said it would.

      Also Frostheim. I have a question. I’ve seen some of the top hunters stack haste until they reach 1.63 speed instead of mastery stacking so you can get exactly 3 Cobras between every explosive shot. Do you suggest this or should I only stack Crit and mastery?

      • Øríon says:

        I currently have a haste rating 1174 while give me a 1.62 cast time on my Cobra shot and I like what it does for my rotation. Aside from using any macros to perfectly time the shots, 1.62 keeps me from wasting any CD on my Explosive Shot. However, any heroic boss fights that gives a haste buff, (i.e. Deadmines, Throne of the Tides, Vortex Pinnacle) i notice a increase in dps when the Cobra Shot is pushing the limit of the GCD. So while I like the 3 Cobras to 1 Explosive ratio…it may be beneficial to have higher haste ratings.

      • Øríon says:

        FYI: I also, have 3/3 Pathing as well as Hunting Party.

    • Bearach says:

      Hold on. In your example, it wouldn’t be an 8% chance over 100 days, it’s would be an 8% chance every day to double earnings that day. There’s a difference.

    • Lilcheeks says:

      I get what you’re saying, but your overall point is wrong with regards to the game. Your crit chance will end up in the long run play out to what it should given a big enough sample. Spend any time looking over logs and you will see this.

    • Taira says:

      christ85,

      Crit is being treated as if 1 in every 100 will do double damage because that’s the statistical probability, and in practice if you fire enough shots you will see that your chance to crit does in fact equal the reported percentage. Small numbers were used because they are easy to understand and work with.

      Using the small numbers, you are absolutely correct that the crit may not have happened over the course of 100 shots. Statistically speaking though, it should have and therefore the math still stands. Try not to get hung up on the “RNG hates me, guaranteed gains are always better” theory. You’re going to fire thousands of shots over the course of your instance runs, and the math will work out as advertised in the end.

  12. oLaudix says:

    One thing is bugging me for some time. You wrote in your article “Every 1 mastery will increase our magical damage done by 1%” and “increases the damage of our magical attacks. Admittedly, for SV, this is the vast majority of them”. Does that mean that all elemental dmg from Explosive shot and Cobra shot is treated as magical? If it is will “Lightning Breath” from wind serpent increse my dmg by 8%?

  13. Blackmoor says:

    The wind serpent debuff effectively increases the damage dealt by everything you fire as SV except Auto Shot, Kill Shot, Kill Command (I believe), and pet basic attacks (not that you should be using Kill Command, just including it for completeness). I use a wind serpent pet frequently now for just that reason. Zeherah’s spreadsheet showed a +1382 DPS gain using the wind serpent over my wolf as SV, which seemed like a substantial gain to me. Ingame results have been less dramatic thanks to the many vagaries and variables of play, but the increase is visible and significant.

    Also important is that the wind serpent is buffing damage for ALL the magic DPS in your group. That can be a significant gain when you’ve got warlocks, mages, shamans, and other hunters along for the ride. Rogues just need to stab harder.

  14. Dorianchika says:

    Becareful with Assasination rogues in group Blackmoor, they often bring this buff themselves.

  15. frozen says:

    So now we just have to go for crit, and dont take any haste?

    femaledwarf just said me this:
    Haste Rating 1.619 – 1478 haste (11,54%)
    Crit Rating 0.882 – 860 crit (17,56%)
    Mastery Rating 0.737 – 1598 mastery (16,91%)

    At all I’m going for haste/mastery atm cause of the crit bug

  16. Corwyn says:

    Since I am now quoted in the original article, I think I should make it clear that the 44% number has a pretty large error margin. If you try yourself on Zeherah’s analyzer, you will notice that changes of 1 or 2 points of crit rating, can give you change of 0.1 in crit stat ranking vs mastery stat ranking. I tried to get within 0.01 difference by changing crit rating by 100′s. So you see the problem. Each change required twiddling to find a reasonable new value.

    However, my understanding is that Blizzard wants us around 30% crit, so all this math basically says the same thing as Frostheim started out with: stack crit.

  17. Rotseh says:

    While fiddling with Crit and Mastery, you still want to retain enough Haste to be able to fit three Cobra Shots in between Explosive Shots (6sec CD, 1sec GCD, 5sec fill time). This is attained when your Cobra Shot cast time is a little over 1.6sec (5 / 3 = 1.66). With 3/3 Pathing and 1/1 Hunting Party this roughly requires some 8.x% extra Haste. Cobra Shot tooltip should be checked for actual cast time to be sure.

    • Dexxz says:

      You would want around 810 haste rating, which is 6.3 %. 8% would be too much!

      So guys, let’s stack crit. But does that also mean reforge mastery into crit?

  18. Duckaholic says:

    Math or voodoo? Looks more like moodoo to me. This is worst than Diablo’s attack speed/frame rate calculations. I like math like my hunter likes melee, and sadly we are equally equipped to handle them both.

    Many thanks to you hunters skilled in math. I appreciate your hard and dedication to give me the simplest of answers.

    “Here’s the simple version for you: stack crit. That’s all you need to know at the moment : )”
    That was a lot of foreplay to get to this answer. As always Frost, it was worth the wait.

  19. Arcazua says:

    Excellent article. And thank you Corwyn for doing the math before I compelled myself to do so. Good thing Crit rating and Mastery rating have the same conversion ratio, or it might get a little more complicated. Still, it feels counterintuitive to me that they would carry a 1:1 ratio once you achieve the balance you are striving for. If mastery is increasing 1% of ~70% of your damage, and crit is increasing 1% of all of it, you wouldn’t think the balance achieved would remain 1:1, but rather 7:10…if I didn’t follow all of your math, I would be inclined to argue with the conclusion.

    • Arcazua says:

      Decided to work Cowyn’s math myself.

      Full damage = Base damage * (1 – P)(1+C) + Base damage * P (1+C)(1+m)
      …where P is percentage magic-based, C is crit pct, M is mastery pct, base damage is pre-crit pre-mastery DPS

      FD/BD = (1-P)(1+C) + P(1+C)(1+M)…factoring out base damage and dividing
      … = (1+C)[1-P + P(1+M)]…factoring out 1+C
      … = (1+C)(1 – P + P + MP)…distributing
      … = (1+C)(1 + MP)

      This agrees with Corwyn’s work right up to the differentials, except I use P instead of E and factored by grouping. Crit and mastery affect their respective percentages equally — they both take nearly 180 rating to get 1%. So for simplicity, we’ll say our gear has some number of points to allocate toward crit and mastery. We’ll let X be the percent that is crit and (1-X) the percent for mastery.

      … = (1+X)(1+ (1-X)P)
      … = (1+X)(1 + P – PX)
      … = 1 + P – PX + X + PX – PX^2
      … = 1 + P + X – PX^2

      we want to maximize this quantity with regard to X.

      d(leftside)/dx = 1 – 2PX = 0
      1 = 2PX
      X = 1/(2P)

      Again, P is our percentage of damage that is magic-based, and X is the portion of our itemization budget that goes toward crit rather than mastery. If P = .75, then X = 1/1.5 = 2/3 going toward crit, and 1/3 toward mastery.

      So crit outweighs mastery by a 2:1 ratio. This is more in line with what I expected.

      • Corwyn says:

        Let’s try it with actual numbers.

        At 44% Crit and 8% Mastery (and 73% magical damage) , my math says that the value of Crit and mastery should be almost the same. Let’s use a base damage of 100 to make it easy.

        Base Case:
        Damage = 100 * (1 + 0.44) * 0.27 + 100 * (1 + 0.44) * (1 + 0.08) * 0.73
        = 152.4096
        Add 1% Crit:
        Damage = 100 * (1 + 0,45) * 0.27 + 100 * (1 + 0.45) * (1 + 0.08) * 0.73
        = 153.468
        Add 1% Mastery instead:
        Damage = 100 * (1 + 0.44) * 0.27 + 100 * (1 + 0.44) * (1 + 0.09) * 0.73
        =153.4608

        So,
        +1% Crit = 1.0584
        + 1% Mastery = 1.0512
        So equal within 1%. Not 2:1

  20. nekomancer says:

    I think current state of stat and gear is FUBAR. And it MUST be changed.

    What I am talking about:
    1) primary stat is WAY more important than any secondary stat
    – leads it to stacking over anything else, offering zero choice in socketing
    – leads to wearing PVP gear (which is easily obtainable due to bugged Tol Barad) for PVE
    – creates big disbalance in profession bonuses due to missing 50 PRIMARY bracer enchant

    2) currently mastery is least desired stat for many classes – making dps ‘tuning’ difficult

    Solutions (expected, hoped for):

    1) to add missing enchants ASAP (bracers, 2H weapons for casters,…)
    2) lowering coeficient of primary stat (~ -40%)
    – or return of AP as a primary stat (higly unlikely)
    – major boost of mastery (and making sure there is zero mastery on PVP gear or making mastery primary stat)

  21. Spooner says:

    I have only just re-started WoW, and have reached 85 (within the past couple weeks) I would like confirmation as to whether this is all still true. I notice that the most recent post is 4 months ago, so while I appreciate all of the information, I would like to know if it’s all still valid. Later tonight when I have more time, I plan on checking patch notes between now and then, but if anyone can post with a yes or no in the meantime, that’d be greatly appreciated.