Bug: Unless the game mechanics have changed since I last tested it, the change in rc5:

-Wielding a sword in the offhand will now correctly remove the dagger spec crit on all of the DPS sheets.

is innacurate. I tested with BQP MH / CHT OH and BQP MH / CTS OH and had the exact same crit rate on white damage.



Excellent calculator otherwise, keep it up!

I could've worded that change better. "The dagger spec crit will now be correctly removed from the offhand if a sword is selected." That wasn't working properly on the "Buffed DPS" pages.

Which displayed crit % are you looking at? I'm not able to duplicate it. When I select an offhand sword, the crit % under "Hit Table - Off Hand" goes down by 3%. If you changed the amount of crit from talents to 10%, it'll go down by 5%.

Edit: Oops, I miss read what you meant. Earlier this week I started to gather data to test offhand sword/mace crits myself.

Probably just an oversight. The default profile has 7 agi on ACL since my ACL only have 7 agi, 'cause I used my guild-allocated glove Nexus Crystals on Bloodfang Gloves. It was before the whole +skill research happened, so now I need to get some gold and buy some Nexus Crystals on my own :P.

When did you test this and approximately how many attacks was it over?

I use it. But I'm in ZG gear / 2pc T1 and 1 T2, so I have trouble pulling off our full wrath tanks anyway.

Honesty is the best policy: I've been lazy and not spent the gold on Nexus Crystals and I already used my grace enchant. :P

I will occasionally use Thistle Tea if it looks like we're falling behind, but it's very rare that rate of DPS actually matters.

Something else I noticed while going over the calculations for the Effective Crit Rate:

Looking at the Unbuffed DPS sheet, Cell C15, the calculation is as follows:

[top](MIN((1-C12 - C13 - C14), (M8+Q6*0.0004))) - (Q5-300)*0.0004

Where C12

Miss %, C13 = Dodge %, C14 = Glancing %, M8 = Listed Crit Rate, Q6 = MH Weapon Skill, and Q5 = 63 Mob Defense Skill.

Now we all know every point of Weapon Skill you have over 300 will give you +0.04% Crit, and every point of Defense Skill a Mob has over 300 gives you -0.04% Crit. So with a listed Crit rate of 25.00%, it would be lowered to 24.40% against a Level 63 Mob with 315 Defense. Now if you had +8 Dagger Skill (ACL + Death's Sting), it would be raised 0.32% to 24.72%, which would be your Effective Crit Rate vs. a Level 63 Mob.

Now the calculation in the Cell takes this number and compares it to the "Crit Cap." Precedence for attacks is:

Miss > Dodge > Parry > Block > Glancing > Crit > Hit

Base Miss Rate vs. a Mob while Dual Wielding is 24% + 0.04% per point of Defense Skill over 300, so 24.6% for a Level 63 Mob. Base Dodge Rate for vs. a Mob is 5% +0.04% per point of Defense Skill over 300, so 5.6% for a Level 63 Mob. Glancing Rate vs. a Level 63 Mob is always 40%. Each point of Weapon Skill you have over 300 also reduces your Miss and Dodge Rate by 0.04% per point, so with +8 Dagger Skill your Miss Rate against a Level 63 Mob assuming +0% Hit from Gear/Talents would be 24.28%, and the Dodge Rate would be 5.28%.

The "Crit Cap" is defined by how much use you actually get out of your Effective Crit Rate. For example: In the above given scenario, out of 100% of your Attacks, 24.28% would Miss, 5.28% would be Dodged, 40% would be Glancing Blows, leaving you with 30.44% of your Attacks leftover. After all those are factored in, what's left is Crit, then Hit. So if you had a 35% Effective Crit Rate, only 30.44% of it would actually be of any use, and every Attack that wasn't a Miss, Dodge, or Glancing Blow would be a Crit. Now obviously the way to counter the Crit Cap is to get more +Hit %, as that would lower your Miss Rate and give you more allocated space for Crits. If you picked up +15% Hit from Gear/Talents, your Miss Rate would be lowered to 9.28, and out of 100% of your Attacks, 9.28% would Miss, 5.28% would be Dodged, 40% would be Glancing Blows, leaving you with 45.44% of your Attacks leftover. Now you'd be getting the full use out of your 35% Effective Crit Rate, and have 10.44% of your Attacks leftover for Normal Hits.

The error I see with the calculation in the Cell is that you're applying the - % Crit from the Mob's Defense Skill to the Crit Cap, when it should only be applied if the Crit Cap doesn't come into play. If that were the case, then the Crit Cap from the first example would be lowered from 30.44% to 29.84%, which isn't how it works. The - % Crit from the Mob's Defense is already being applied to the listed Crit Rate, so there's no need to apply it again to the Crit Cap. With the example of a 35% Effective Crit Rate with +8 Dagger Skill, your Listed Crit Rate would be 35.28%, the Mob's Defense Skill would lower this to 34.68%, and your Weapon Skill would raise it up to 35.00%. All the Crit Cap means is that your Crit Rate is still 35.00%, you just only have enough Attacks leftover to make use of 30.44% of it -- it isn't effected by the Mob's Defense Skill at that point.

Chalon, I know you and Valar have a knowledge and understanding of everything I elaborated on, as it's reflected in the Spreadsheet, but I just wanted to provide some basis for what I was trying to demonstrate. This isn't really that big of a deal, as the Crit Cap rarely comes into play for a well-geared Rogue unless you're sporting some serious buffs, and even then the difference is fairly small, but just thought I'd bring it to your attention.

Oh, and one other game-breaking change: You've got the Strength and Agility swapped on Qiraji Execution Bracers ;).

Minor error - base stats for orcs are very slightly wrong, it's listing 1 point of str too much.

Minor issue with calculating the BF proc DPS - on the "dream gear" page it uses an approximation of 8 - while the "current gear" tab actually calculates it.

Ok, well we've upgraded it now to release version 1.0...

I tested it over a few hundred attacks in 1.9 and was satisfied when I got exactly the same crit rate either way (I think some of my guildies may have done the same).

I'll test it again tomorrow and post what I find.

I've always sort of suspected that it was in fact just a flat crit rate to both weapons, and in fact that's what I believed but was convinced otherwise :P.

Someone who's much better at statistics than I am should tell me what the confidence level of those numbers is! :P

Interesting, that looks like pretty clear cut data.


Did you have 5/5 Dagger spec?

Did you test it on the guy in Blasted Lands that doesn't die?

Yes, I have 5/5 daggers spec. I tested on a druid with 300 defense.


I was a little bit suprised that the difference was almost exactly 5% (the amount of crit I was gaining from dagger spec). The majority of my attacks were coming from the dagger I had in mainhand (which was .5 faster) so I would have expected something closer to 27-28%.

Crit chances are out of the total number of swings, not the total number of hits as they are being shown in your ui.

Those percentages are 23.1% and 27.3% for sword/dagger OH respectively.

Out of curiosity, what SHOULD your base crit % be with those equipped?

I assume this would mean it was fixed for fist spec as well?

Has the dagger/fist spec crit not working for offhands been confirmed?

Well I spoke to a GM after testing for a while and it still seemed to me to apply like malice (Ie. only need it a MH to be in effect) and asked if any changes had been made to any of the spec's and equired it the tooltip would ever be updated. All the testing I had done showed no significant impact upon the crit rate regardless of OH. More importantly I was gaining +5% over both weapons if we gained a further +5 for having a Dagger/Fist weapon then we'd be at 10%, which seems unlikely to say the least. I'm going to stick with the MH applies an Aura over both theory. (sorry I dont have screenshots got distracted by the GM then had to walk my dog)

All the responses from the GM were the usual "gameplay issues work it out yourself" "if the devs felt it was imported it would be noted in the patch notes". So basically he offered me nothing of value, other than "the specs should be balanced against each other".

Also I'm not sure if any has been posted before but anyone ever tried out the full theory comparing fist / sword spec? I started yesterday but after about 1 hours work on it I realised my awful maths skills were going to shine through since I kept forgetting to account for things such as energy / specials etc.. I like my theory to be accurate but it started getting severely long and complex.

Hmm,

I used the spreadsheet, entered my End-of-BWL dreamgear and compared the different MH-Enchants with full Raid Buffs. The numbers without MH-Enchant were:

612,42 DPS
28,4% crit
18% to hit bonus

The best options were:

1. Crusader: 6,8 DPS
2. +15 Agility: 6,07 DPS
3. +5 Damage: 5,6 DPS

Then I checked Kalman's Excel about MH-Enchants found here: http://www.radiationnow.net/wow/enchant.xls . For DPS, critrate and hitrate I just used the numbers from the DPS-spreadsheet, for armor mitigation I used 0,8 as in the spreadsheet (although for the relative order of these three enchants armor mitigation should not matter). According to Kalman's Excel the order is

1. Crusader: 6,3 DPS
2. +5 Damage: 5,6 DPS
3. +15 Agility: 5,0 DPS

So which Excel is right, and where are the errors?

First: excellent work. I 'm not really a math genius (quite the contrary) and this spread sheet is both idiot proof and helpful. Are you planning to add ZHC and Renatakis into the sheet? At the moment i am pondering whether to use BB+Royal Seal or to use the ZHC...(didn't have luck on farming HoJ and the balls to bid more than a certain amount of points on DFT...).

I believe it's been mentioned before, but ZHC is really only good for limited PvP applications. This spreadsheet is a race horse, in that it is designed for one purpose only: combat dagger sustained DPS. One might assume if you were looking to PvP you would spec 3x/8/1x or 21/8/22, or maybe even 0/0/51 ;)

Regarding rentakis' - I seem to remember the group talking that one would have to make some *very* specific assumptions in order to model it properly, and even then, you're still going to be better with a DFT / HoJ. The general rule is that "on click" trinkets will tend to be inferior to passive trinkets for the sustained DPS modeled out in this spreadsheet.

So, to be 100% clear, you're gimping yourself if you're using Rentaki's or ZHC in PvE encounters, with combat daggers.

Not to overstate its importance or to gush, but this shit is absolutely amazing. My hat goes off to its creators and contributors.

I feel compelled to note, however, that Slayer's Crest is 60 passive Attack Power, not 30. I saw someone mention it in an earlier page as only being 30, so I thought I'd clarify that in case your calculations are not up-to-date in that regard.

Anyhow, most excellent work, and bravo.