Guild Wars 2

Thinking about it, it could be that the values I was using to determine crit chance could be xx.51% crit chance, and that it does round on the character screen, since that would also put 21 numbers between each value. But even then, if it rounds in practice as well as on the displayed information, those 21 values are still what we want to hit. Here’s my handy list I keep in a notepad document.

2387 75
2366 74
2345 73
2324 72
2303 71
2282 70
2261 69
2240 68
2219 67
2198 66
2177 65
2156 64
2135 63
2114 62
2093 61
2072 60
2051 59
2030 58
2009 57
1988 56
1977 55
1956 54
1935 53
1914 52
1893 51
1872 50
1851 49
1830 48
1809 47
1788 46
1767 45
1746 44
1725 43
1704 42
1683 41
1662 40
1641 39
1620 38
1599 37
1578 36
1557 35
1536 34
1515 33
1494 32
1473 31
1452 30
1431 29
1410 28
1389 27

Ray.3780:

@Wish – So that only happens at 20 over the previous percentage? I know it’s 21/% for some odd reason, but whenever I’ve had 11 or more over the previous percentage, my crit always rounds up, in fact, right now I’m at 2083 prec, and it’s showing 60% instead of 59% on my sheet. If you do the math, my true crit chance is ((2083-916)/21)+4, which is 59.57%.

It happens at all values between 1 and 20 above, also:

2093 precision is 61% crit chance
2072 precision is 60% crit chance

http://wiki.guildwars2.com/wiki/Precision
At level 80, the base precision is 916, with a critical hit chance of 4%, at 927 precision, it becomes 5%.
At level 80, Critical Chance and Precision is found by the following formula:
Critical Chance = round down((precision – 822) / 21)

((2083 – 822) / 21) = 60.0476190476
Your crit chance should be 60%

Ray.3780:

And so what I’m saying is, if math WAS truncated, what reason, (besides it being a bug) would there be to show a rounded up value? wouldn’t they have just rounded it down all the time until you reach the next %?

They do round it down all the time, thats what Im saying. You just did the wrong math to determine what your crit chance should be.

I wear something like that when Im CnDing siege in the middle of 30+ people (you cant crit siege) but youll never kill anything like that in PVP, and in PVE just offhand a gun and drop blinding powder and youre invincible.

AiKitecT.1897:

Thanks for the awesome response, that’s really insightful (and for the whole thread). Why do you say that we will need to change infusions once the ascended armors come out?

Because we dont know the stats on Ascended armor, or how many infusion slots we will pick. So we dont know what the best mix of precision/power infusions is going to be. So any infusions you add or change now might end up being changed again when the armor comes out and just be a waste of resources.

AiKitecT.1897:

Also, if you use Traveler’s Rune, which extra utility would you carry?

Blinding powder or Roll for Iniative probably.

Ray.3780:

@Wish – I’m not sure what you mean by the second part, as long as the calculated crit chance is x.5% or higher, it has always rounded up for me in the character sheet, but that could just be luck. Character sheet stats have always been ambiguous, but I think I’ll stick to maximizing my power stat with infusions.

What I mean is this:
When you have 1914 precision, the Character stats screen shows your crit chance at 52%. This disproves 21.16 being required for each point of precision, since it would require 1920~ precision to hit 52% if the .16 was left there. The other thing Im saying is that if you had 1934 precision, 1 under whats required to hit 53%, if the game did count that 20 extra precision youd be at like 52.95% crit chance, and it would probably display it as 53% or 52.95%, not the 52% you see in game. Plus every other game I paid attention to the math in everything was truncated.

Ray.3780:

So, with talk of wasting precision, have you tested whether or not GW2 rounds up true critical chance like it does on the character sheet?

Because if it does, there’s no real difference between 56.5% crit and 57% crit.

Theres no parser for GW2 as far as I am aware, so no.

The fact that precision should be 21.16 for 1% crit chance at max level and its truncated to 21, and that when youre 20 precision over an exact number it doesnt show it as rounded up, is a good argument for critical hit rate being truncated as well.

AiKitecT.1897:

http://gw2skills.net/editor/?fYEQRAoY8YlUmiO3ey5E/5Ey2jdqCJKuHGVS1UdFiKA-jUCBYfDiO5RZN7RQsMsIas6aYKXCRaXp2eTQ2gTZVKgyVGB-w

The weapons would have +5 Power Offensive Infusion, and you would get +250 vitality from Applied Fortitude.

Disclaimer:
You absolutely should not add or change any infusions you currently have or do not have for any reason. When Ascended armor comes out you will end up needing to change them, and as far as I understand, Ascended armor comes out very soon.

Now with that said, if you removed one of the precision infusions (You have 3) youd have exactly 50% crit rate, as 1872 = 50%. This is when using Stones.

(2462* 14925) /1836= 20013.8071895
(2462* (14925+ 2500)) /1836= 23366.2037037
This would be totally fine EHP wise to fit the original criteria (21.5k~ EHP) provided you had a few stacks up. When you hit 5 stacks you end up with 23,366.

2039 power
98% crit d
1872 precision
1398 toughness
1328 vit

oil/stone stats
1398 * (60/1024)= 81.9140625
1328 * (40/1024) = 51.875
82 + 52= 134

Stones/bloodlust/etc
(2039 + 250 + 134 + 10 + (35* 6)) * ((1 * .5) + (1.5 + .98) * .5)) =
4598.82

Crit chance from Oil
134/21=6.38095238095
(134 – 5)/21= 6.14285714286
(134 – 10)/21= 5.90476190476
(134 + 5)/21= 6.61904761905
(134 + 10)/21= 6.85714285714
(134 + 15)/21= 7.09523809524
(134 + 18 – 5)/21= 7

When we use oils, we will use 1 B/B Defensive slot ring or earring instead of B/V, and 1 precision infusion instead of 2, giving us exactly + 7% crit chance and moving us from exactly 50% to exactly 57%.

Oil/bloodlust/etc
(2039 + 250 + 15 + (35* 6)) * ((1 * .43) + (1.5 + .98) * .57))=
4634.8104

Oils win, before counting extra lifesteal from Candied Dragonrolls (Omnomberry pie)

Viking Jorun.5413:

I don’t know. I guess I just can’t play orthodox builds (like I attempted here).

Maybe you should try playing a good one instead of this build you posted and created just to try and be a special snowflake and different?

Power increase:
(oldpower + addedpower) / (oldpower) =1.XXYYYY
XX.YYY%= %damage increase

Precision increase:
21 precision = +1% Crit chance at level 80.
Added Precision / 21 = X.YYY
X = addedcritchance
(1 + ((.5 + critdamage) * (oldcritchance + addedcritchance))) / (1 + (.5 + critdamage) * oldcritchance) =1.XXYYY
XX.YYY= %damage increase

Add your stats where needed and you can see which is better. I’d do it for you, but you didnt list your crit damage or exact precision.

For examples and more math and math about why using soldiers and berserkers is superior to valk and berserkers for survivability and damage see this thread:
https://forum-en.gw2archive.eu/forum/professions/thief/Building-a-better-D-D-thief/page/3#post2615780

AiKitecT.1897:

So let’s say that I had the Applied Fortitute which gives +250 vitality. What do you think about the following?

Berserker Weapons – +5 Power Infusion
Ascended Armor – Rune of the Travelers
Berserker Trinkets – +15 Power, +15 Precision

Switch the Master Maintenance Oil to Superior Sharpening Stone.

I have no idea what youre asking, give me a full list of armor/traits/food/etc

Shpongle.6025:

is power,toughnett and crit dmg in the rings better than power, prec, crit dmg ?

Since power is not the main stat, it can never be better than other options for adding to defense and damage. Unless you want to stack toughness and use a build that is composed entirely of Knights and Cavaliers gear.

AiKitecT.1897:

What about the new Rune of the Travelers? Thoughts?

Movement speed is pretty much always the best stat for everything ever.

The blinding powder trait is probably superior to 30 all stats and 6% crit damage. Thats like -2.5% damage and -2%~ Defense, I dont even care about boon duration or the other thing the rune does.

When ascended armor comes out, you could probably pick up a bit of damage and EHP just using ascended versions of the existing exotic armor while adding these runes to it and it would be amazing.

Can I add a rule?

Dont say X is better than Y unless you have and understand the math that proves it

Thoughts

1) You have 6 precision infusions and 1904 Precision, two of your infusions give you absolutely no benefit and are overtly expensive because they would have to be versatile precision infusions to go into the defensive infusion slots.

2) 100 Power, 10% Condition Duration, Lotus Poison and Dagger training is inferior damage compared to 100 Precision, 10% Crit Damage, First Strikes and either Hidden Killer or Executioner. And no, Lotus Poison doesn’t offer good utility to cover the damage loss.

3) # 2 more likely than not makes Power infusions better than Precision infusions, especially when you realize…

4) Oils are better than Stones, if you optimize your precision, which you should always do.

5) Also, that food is bad.

AiKitecT.1897:

Wish, how do the WvW buffs such as Applied Fortitude and Guard Leech affect the build? If you had to choose one, which would you get? If you had both, would you change any of the gear?

Applied fortitude is best if you never play with any siege, but its effect should be ignored, just like how you dont build around the +% HP from the WvW points buff. If you notice with the combination of the two you have a ridiculous amount of extra HP, feel free to swap a weapon to berserker though.

So, are you guys looking forward to the new ascended gear? Im gonna remake this thread when I see stats and dont have stuff to do, even though Im not really playing GW2. lol

While playing a balanced build you achieve higher DPS while keeping the same EHP using a mix of soldiers and berserkers compared to valkyrie and berserkers. Links to threads containing relevant math can be found by clicking thief in my sig or searching my post history and looking for posts containing lots of numbers.

Since all your gear is exotic you should use soldiers hat, legs, back, and rings, and everything else should be berserkers. If you want a bit more EHP swap in Valk body, if thats still not enough soldiers body.

When you move over to ascended gear you want all berserker ascended pieces, soldiers hat, body, and legs. And then you want your weapons to be Valk or Soldiers depending on how much EHP you feel you need.

D/P is only better when your opponent is very low on HP and trying to use their heal skill. Other than that, whichever you are better at playing with is the superior set.

If your objective is to kill people, you’re not going to do it. If your objective is to just stay alive and troll them all day, youll be fine.

http://gw2skills.net/editor/?fYMQJA8YlYmiOHcy4E/5Ey2jKcn4JvCkV3DuyqqKA-jUyAorATfAoCwZvioxWFLiGruGT5ma1rQk9uIa1CBUeCA-w
Something like this would keep your EHP, increase your Ini gain, give you more dodges, and increase your damage.

1974 power
1574 precision (1578 is 36% crit chance, you should really use a single precision infusion, and I will be counting that for my damage comparison, even though this number spit out by the gear calculator is most likely wrong)
73% crit damage
(1974 – 5)* ((1* .64)+ ((1.5+ .73)* .36))= 2840.8732
(1974+ 250- 5)* ((1* .64)+ ((1.5+ .73)* .36))= 3201.5732

2101 power
1919 precision (52% crit chance)
50% crit damage
(2101)* ((1* .48)+ ((1.5+ .5)* .52))= 3193.52
(2101+ 250)* ((1* .48)+ ((1.5+ .5)* .52))= 3573.52

My set is 11.6% more damage with 25 bloodlust stacks,
3573.52/3201.5732= 1.11617625985
12.4% without
3193.52/2840.8732= 1.12413324185
And thats before you count the sigil of force and executioner, which is another 10% damage on top of what you gain from just armor distribution pretty easily, depending on who you are fighting and how much healing they get to do. If they didnt heal at all it would be close to a 14% damage increase on top from just those 2 things. If you actually know to use force when you hit 25 stacks of bloodlust then subtract 4.5%~ from those numbers, and never include bloodlust in your sets when you make them and post them, just assume people know to stack bloodlust, and put in the alternate sigil.

2719 Armor
17825 HP
(17825* 2719)/1836= 26397.6988017
26,398 EHP
(2719- 1836)/2719= 0.32475174696
32.5% DR

2481 Amor
19715 HP
(19715* 2481)/1836= 26641.0212418
26,641 EHP
(2481- 1836)/2481= 0.2599758162
26% DR

My set gains 250~ Effective HP too. (at the cost like 6% Damage Reduction, which is offset by the fact that youre gaining like 20% damage)

Personally, I think this is overkill on EHP, and I would use all Berserker trinkets (and in fact I do when I play this). For food use candied dragon roll not omnomberry pie as its vastly cheaper and the same effect.

Monkey boy.1697:

i want to be able to stay live and dish alot of dmg.

This is why I changed your traits, Trickery doesnt compare to Critical Strikes for any type of build that wants to do physical damage, its just too weak.

There was a bug where people could stack stealth through the revealed debuff by CnDing walls in WvW. When Anet fixed the bug instead of removing the ability to CnD walls, they basically said they condone CnDing walls.

Glancing means you have weakness on.

Why you wouldnt get crits with HK is beyond me though. The trait was broken for a week or so a few months ago, but it should be fixed and working correctly right now.

So you killed this guy, then he followed you around saying youre bad because you use D/D and its worse than D/P? Doesnt that just make him extra pathetic since he lost to an inferior set?

Quick! someone post that Screenshot of the 2million damage backstab on the bunny!

FLiP.7680:

I’m aware of how the skill damage is calculated. The propose of the topic was to know if i could have the best damage possible and survive without any points in Shadow Arts, and with 2k armor and 12k life.

It depends on what youre doing. Like if youre doing COE, 15 in acrobatics is amazing and required, and 10 shadow arts for +2 ini on stealth is basically useless. If youre doing COF P1 farming… acrobatics is a waste of time, and a 30/30/10/0/0 setup will do way better.

Manuelito.6081:

Viking Jorun.5413:

If you’re looking for a balanced build, you might want to try going with less crit chance. The mathematically ideal crit chance is 41.9%, so you’re wasting a whole 10% that could be put into something that would benefit your build more, like vitality or healing power.

Could you please elaborate a bit more on the ideal crit chance %, please? How did you come out with that number?

thanks

Just ignore it, its wrong.

Your ideal precision (crit chance) varies depending on your power and crit damage. But its the point where 21 precision gives the same damage % increase as 21 power.

https://forum-en.gw2archive.eu/forum/professions/thief/Building-a-better-D-D-thief/page/3#post2615780

The top of this post includes math for determining damage gains. If you really care what your ideal crit chance is, these are the 3 math problems that will let you find it. When 21 precision (+1% crit chance) = +21 power, thats the ‘ideal’ number.

Let me make this easier on you now that I think of it, this is probably going to be super hard for you to get.
Take this problem:
(YourPowerHere + 21) / (YourPowerHere) =
Type in your power, and then google it. You will get 1.xxxxxx~
Then write down that number, and then fill in this problem:
(1 + ((.5 + YourCritDamageHere) * (YourCritChanceHere + .01))) / (1 + (.5 + YourCritDamageHere) * YourCritChanceHere) =
If you have 139% crit damage, that is 1.39, if you have 72% crit damage, that is .72 If you have 47% crit chance that is .47, etc.

What youre doing here, is changing “yourcritchancehere” to whatever, and whatever number gives the closest value to the first problem would be your ‘ideal’ crit chance.
Like if we had 2300 power and 92% Crit damage, we’d do…
(2300 + 21)/2300 = 1.00913043478
Then we’d look for the Crit chance value that matches that the closest with the second problem by randomly adding crit chance values til we figure it out.
(1 + ((.5 + .92) * (.53 + .01))) / (1 + (.5 + .92) * .53) = 1.00810224809
(1 + ((.5 + .92) * (.5 + .01))) / (1 + (.5 + .92) * .5) = 1.00830409357
(1 + ((.5 + .92) * (.45 + .01))) / (1 + (.5 + .92) * .45) = 1.0086638194
(1 + ((.5 + .92) * (.40 + .01))) / (1 + (.5 + .92) * .40) = 1.00905612245
(1 + ((.5 + .92) * (.39 + .01))) / (1 + (.5 + .92) * .39) = 1.00913888531

In this instance, its 40% (39 + 1), but that doesnt actually mean anything as far as gearing goes. Its just the point where 21 precision gives the same DPS increase as 21 power.

RagingDragon.9475:

wish.1027

Just solve damage with the numbers in the damage calculations that change depending on the differences in the setups.

First off, if you want to get technical you have to include the 4% higher chance to Crit and cause Vulnerability. Which would essentially make it a wash based on your formula.

What 4% higher chance to crit?
The one I included or a different one?

Eagle damage
(2609+ 250)* ((1* .37)+ (1.5+ 1.18)* .63)* 1.425= 8386.07598
Ruby Orb damage
(2734+ 250)* ((1* .41)+ (1.5+ 1.22)* .59)* 1.4= 8417.02848
Ruby Orb without changing the infusion
(2729+ 250)* ((1* .41)+ (1.5+ 1.22)* .59)* 1.4= 8402.92488

As for vulnerability, as far as I am aware it doesnt stack on top of other +% damage effects, so as I said earlier, it can be dropped, because it will stay the same for both problems. But seriously, even if it does get stacked on top of the other + % damages, who are you kidding? The Eagle sets % damage increase is higher, it will hurt Eagle and push it further behind if thats how it works. So no, even if I excluded it and it stacks, it doesnt matter, cause it just makes Eagle lose by more.

RagingDragon.9475:

Secondly your damage is actually calculated based on your Attack (Power and Weapon Damage)

No, it really isnt.

Attack = power + weapon damage

Damage = weapon damage * power * skill coefficient * critdamage if a crit is scored * % damage increases / targets armor

Where in the damage calculation do you see power + weapon damage? Cause Im somehow missing it….

Seriously, attack means absolutely nothing, and should be ignored.

RagingDragon.9475:

not strictly based on Power as you seem to be using for your formula.

wish.1027

Just solve damage with the numbers in the damage calculations that change depending on the differences in the setups.

Weapon damage wont be changed between the builds, so it is dropped, because you only need to compare the difference. Multiplying both numbers by 981 wont change anything about them.

RagingDragon.9475:

Then your Attack is multiplied by the skill coefficient

No it isnt.

RagingDragon.9475:

and then reduced by your opponents Armor rating. As we are not looking for exact numbers for a specific skill, we just use the Attack rating.

As you are not looking for a specific skills damage, you remove that part of the equation, because it wont change anything about the end result by adding it in, because the skills damage does not change between the builds. And for the last time, no, you do not use attack…. ever… for anything.

RagingDragon.9475:

And as with any rune that provides a % bonus, the larger the base number the larger the bonus (i.e. 5% of 2500 is 125, but at 3500 it is 175). Therefore at a specific number, the % bonuses will always overpower a straight increase.

And that % damage increase is worth less as you stack more of it. You left out 40% in your original math, which drastically reduces your 2.5%s worth. And “at a specific number” anything can be better than anything.

RagingDragon.9475:

Using YOUR math and just changing the Power to Attack rating:

Eagle damage
(3,590 + 250)* ((1* .37)+ (1.5+ 1.18)* .63)* 1.425= 11,263.56
Ruby Orb damage
(3,710+ 250)* ((1* .41)+ (1.5+ 1.22)* .59)* 1.4= 11,170.05

Again I will say there are magic numbers where Eagle Runes beat Ruby orbs and Visa Versa. So you need to compare them against YOUR build to see which is best for what YOU want to use. Again this based on true Min/Maxing for Pure Damage output.

Ultimately Ruby Orbs are a great choice due to its price and minimal difference!

Attack is not used to solve damage…
Damage is weapon damage * power, not weapon damage + power.
http://wiki.guildwars2.com/wiki/Damage
Read this, memorize it, let it sink in, then come back and tell everyone youre sorry for misleading them.

Using my math, lets really add weapon damage to the equations the way its actually done and see what happens.

Level 80 exotic daggers read 923-981 damage, since we’re determining averages, we add those together and divide by 2, and that will give us the middle value between those numbers.
(923+981)/2=952

Eagle runes
952*(2609+ 250)* ((1* .37)+ (1.5+ 1.18)* .63)* 1.425= 7983544.33296
Ruby Orb damage
952*(2734+ 250)* ((1* .41)+ (1.5+ 1.22)* .59)* 1.4= 8013011.11296
Ruby Orb without changing the infusion
952*(2729+ 250)* ((1* .41)+ (1.5+ 1.22)* .59)* 1.4= 7999584.48576

8417.02848/8386.07598= 1.00369093961
8013011.11296/7983544.33296= 1.00369093961

Oh, well would you look at that, Eagle loses by the same margin when we add weapon damage, and when we dont.

Bovinity.8610:

(1.5 + .3 + .05) / (1.5 + .3) = 1.0277777777
(1.5 + .5 + .05) / (1.5 + .5) = 1.025
(1.5 + .9 + .05) / (1.5 + .9) = 1.02083333333
(1.5 + 1.2 + .05) / (1.5 + 1.2) = 1.01851851852
Notice how that 5% crit damage we’re adding is worth less and less the more crit damage we already have? Thats the diminishing returns coming into play.

At work and don’t have time to really go through the details, but isn’t that as expected?

It should be expected but its something most people dont realize.

Bovinity.8610:

You get a smaller percentage, but it’s applied to a larger base value, so you’re not really getting less damage out of it?

You get the same static number for 5% crit damage no matter how much crit damage you have. If youre base damage is 100, that 5% will always add 5. But that doesnt actually matter, because the only thing thats important is the end result. This formula is for determining damage gains, its used to see if you should stack more crit damage, or something else like power. The fact that 5 crit damage adds 5 damage doesnt actually matter. Its how much of the end result that damage is worth that matters.

Bovinity.8610:

In other words, we could say the same thing about Power or any other stat.

Yes, you can say that. Adding to whole numbers gives diminishing returns, subtracting from whole numbers gives increasing returns. If you read my thread I go over all of it and list the problems for determining gains from power, precision and crit damage.
https://forum-en.gw2archive.eu/forum/professions/thief/Building-a-better-D-D-thief/page/3#post2615780
This post.

Jormin.2761:

wish.1027:

http://gw2buildcraft.com/calculator/thief/?7.0|8.1c.h14.a.1c.h2|c.1c.h1.8.1c.h17|1c.71j.1g.71j.1c.71j.1g.71j.1c.71j.1g.71j|2s.d13.2s.d13.3s.d16.2s.d13.2t.d13.2s.d13|a3.u36b.u46b.0.0|55.7|57.5b.5c.5h.5w|e

This setup is exactly 52% crit chance, and is the best DPS you can get for that much EHP.
If you want links to threads with math click “thief” in my sig and look at the bottom of that post.

Wish I have based my builds off of yours for a bit now, but this d/p build has no initiative regen, how is it sustainable after 1 BP+HS? My current d/p with init regen (not the best still some wasted stats I believe) http://gw2buildcraft.com/calculator/thief/?7.0|8.1c.h2.a.1c.h17|8.1c.h17.a.1c.h1j|1c.b1j.1g.71j.1c.71j.1g.71j.1g.b1j.1g.71j|2s.d13.2s.d13.3s.d16.2s.d13.3s.d16.2s.d13|0.k16.u45b.k69.0|55.7|57.5b.5c.5f.5u|e

Youre right, I didnt even look at the traits, I just assumed he was using gain 2 ini on stealth when I changed his gear types.

You should really drop blind on stealth for gain ini on stealth.

Viking Jorun.5413:

If you’re looking for a balanced build, you might want to try going with less crit chance. The mathematically ideal crit chance is 41.9%, so you’re wasting a whole 10% that could be put into something that would benefit your build more, like vitality or healing power.

The ideal precision to maximize DPS is going to be different depending on your power and crit damage. Listing any one number without giving the others is misleading.

That wouldnt actually fix it. It would need to remove revealed, block you from being able to get revealed applied again for 1 second, stop you from being able to backstab for 1 second, and stop your auto attack when it activates to actually fix the trait and make it work as intended.

The best suggestion I ever heard, was to change it so it drops black powder at 25% HP instead of stealth.

http://gw2buildcraft.com/calculator/thief/?7.0|8.1c.h14.a.1c.h2|c.1c.h1.8.1c.h17|1c.71j.1g.71j.1c.71j.1g.71j.1c.71j.1g.71j|2s.d13.2s.d13.3s.d16.2s.d13.2t.d13.2s.d13|a3.u36b.u46b.0.0|55.7|57.5b.5c.5h.5w|e

This setup is exactly 52% crit chance, and is the best DPS you can get for that much EHP.
If you want links to threads with math click “thief” in my sig and look at the bottom of that post.

The reason soldiers/berserkers mix wins over valkyrie, is because you dump 10% crit damage (2% damage) for 3.5% crit chance (which you push to 4% using correct infusions/trinkets). Each point of crit rate is basically worth 1% damage, so youre trading 2% damage for 4% damage. (this is trading all valk armor for soldiers hat/body/pants and berserker shoulders/gloves/boots)

For your specific build, lets determine what changes and solve for it.
Power and Crit damage change. So we we need to solve for power and crit modifier only. So we will take this:
(weapon damage * power * skill coefficient * crit modifier * % damage bonuses) / targets armor = damage
and dump everything that doesnt change to end up with this:
power * crit modifier = damage

So our final problem will be
(base power + bloodlust + (35 * average might stacks)) * ((none crit damage * none crit rate) + ((base crit damage + added crit damage) * crit rate)) = damage

http://gw2skills.net/editor/?fYEQNAoaVlUmiO3ey5E/5Ey2jeqCJKuHGlT1UdFwKA-jkzAYLBRCCIZBiKAk8W0a2LgthioxqewUlJVXCrcDk9GwgD5NWWtMA6LA-w
This is also the way I would set up your gear. Trading the exotic ring to soldiers is better than the exotic body to soldiers. Exotic ring is a 16 stats per point of crit damage trade.

You (I even gave you 2 power infusions)
(2245 + 250 + (35 * 0)) * ((1 * .49) + ((1.5 + 1.01) * .51))= 4,416
Me
(2164 + 250 + (35 * 0)) * ((1 * .4) + ((1.5 + .98) * .6))= four thousand five hundred fifty seven.
Because these forums have downs syndrome and when I type out that number it says kitten.

As you can see, your power is higher than mine too. Power also suffers from diminishing returns, and we solved for 0 might stacks. So the more might you have, the more my set will win by.

Does it make sense now?

Also: just to point it out, the only difference in defense is we traded like 20 vit in your set for 25 toughness in mine. So the EHP is basically the same.

https://forum-en.gw2archive.eu/forum/professions/thief/Building-a-better-D-D-thief/first

Sir Kaboomski.1508:

The main thing I do not understand is the crit damage bonus. You seem to say that slightly higher power and 5%+ less crit dmg does more DPS than slightly less power with more crit dmg. If I knew how that translated, it would totally make sense lol.

Damage is calculated as
(weapon damage * power * skill coefficient [* critdamage if a crit is scored] [* + % damage bonuses]) / targets armor

When you compare sets of gear, you can remove everything that will stay the same for both sets. Like weapon damage, target armor, skill coefficient, etc. Whatever stays the same you just remove, since those things would affect both sets in exactly the same way. Then you solve the rest of it to find out what the numbers are to compare damage increases between the sets.

To get an accurate measure of how much your crit damage/crit chance is worth, you have to determine what is referred to as a ‘crit multiplier.’ This is a combination of your crit chance/damage and none-critchance/damage, that gives you the average your crit chance/damage will increase your damage by.

For ease of use and understanding for people who arent super amaazing at math, we use the formula:
(base damage * .none-critrate)+ ((base critdamage+ addedcritdamage on gear and food)* critrate)
Base damage is 1. Base crit damage is 1.5, because critical hits add 50% damage. If you hit for 100 damage exactly every time you swung your weapon (like with a steady weapon in sPVP) and you had 0% crit damage, when you landed a crit, it would do 150 (this is easily testable by anyone, go try it if you dont believe me.)

We write all things crit by moving the decimal to the left 2 places. 112% crit damage is written as 1.12, 67% crit rate is written at .67, etc. This is because if you think about it, a 67% crit chance means 67% of your attacks will crit, If you attacked 100 times, 67 would be crits. You write that as “* .67” because thats how you determine 67% of something. 100 * .67 = 67

Now, where people tend to get confused is that they think things like + 5% crit damage is + 5% damage. But whakittens actually saying is “+ 5% of your base damage when you land a critical hit.” If we look at our earlier explanation where we have 100 base damage on every attack. If we had 20% crit damage, and added 5% more we would be hitting for 175 on our crits, because our base is 50%, and then we add 25%. 75% of 100 is 75, so 175. Now, look at that 5% you added. It moved you from 170 to 175, because it gave you + 5% of the 100. Not + 5% of the 170.
175/170=1.029411
5% crit damage in this instance only gives us a 2.941% damage increase, when we land a critical hit. If our critical hit rate was 50%, then that means only half of our attacks crit, so we only get that bonus half the time, so we multiply it by our crit chance to see thakittens only a
2.941 * .5=1.4705
1.47% damage increase from 5% crit damage, at 20% crit damage and 50% crit rate.

You need to keep in mind that this suffers from diminishing returns too. The more of it you have, the less useful adding more of it is. Allow me to show you, this is the problem to determine crit damage gains:
(1.5 + oldcritdamage + addedcritdamage) / (1.5 + oldcritdamage) =1.XXYYY
XX.YYY * .critrate = %damage increase

So lets check 30% crit + 5%, then 50 + 5, then 90 + 5 then 120 +5. All with 100% crit rate so we dont have to solve for it.
(1.5 + .3 + .05) / (1.5 + .3) = 1.0277777777
(1.5 + .5 + .05) / (1.5 + .5) = 1.025
(1.5 + .9 + .05) / (1.5 + .9) = 1.02083333333
(1.5 + 1.2 + .05) / (1.5 + 1.2) = 1.01851851852
Notice how that 5% crit damage we’re adding is worth less and less the more crit damage we already have? Thats the diminishing returns coming into play.

When you have 100%~ crit damage, and 50%~ crit rate (which is about average for standard balance build thief) you need 5%~ crit damage to gain + 1%~ overall damage.
Meanwhile these builds tend to have 2500~ power if built correctly, which means you need 25~ power to hit + 1% damage. For power you simply do
(oldpower + addedpower)/ oldpower = 1.XXYYY
XX.YYY= % damage increase
(2500 + 25)/ 2500 =1.01

Youd only need 20 power if you were closer to 2k power like the guy earlier I posted about, with his 2k power and 125 crit damage or whatever. For your average thief, 20~ power or 21 precision (+ 1% crit chance) is seriously worth the same as + 5% crit damage. People just have a hard time accepting that when they dont know/understand the underlying math.

Jinks.2057:

Leave T1 zerg

Join T4 and lower

Be happy

This is the correct answer. Solo roaming doesnt work in the high tiers.

Lews.8146:

Hey, if u want help with a build/armour set send me a tell in game and I can link u mine, final stats are 3,050 attack, 2410 armour, 50% crit chance, 122% crit dmg and 16665 hp. So much fun lol, specially in some 3 vs 1 situations, I kitten ed alot of people off yesterday lol I’m assuming when I kept winning those as I roamed

Im super bored atm, so Im gonna do some math.

http://en.gw2skills.net/editor/?fYEQRAoY8YlUmiO3ey5E/5Ey2jdqCJKuHGVS1UaFoJA-jEzAorBRTWCNRAiKAk8UWzeCiVtioxqrxUuDRW7iavJiWt5wCBUeBA-w

Your build is 2069 power, 50 crit rate, 122 crit damage, 2410 armor, 16665 HP.
My build is 2416, 52, 92, 2620, 15325.

Your builds damage
2069* ((1* .5)+ (1.5+ 1.22)* .5)=3848.34
My builds damage
2416* ((1* .48)+ (1.5+ .92)* .52)= 4199.9744

Your builds EHP (no WvW bonus)
(16665 * 2410)/ 1836= 21875.0816993
My builds EHP (no WvW bonus)
(15325 * 2620)/ 1836= 21869.0087146

Your builds DR
(2410-1836)/2410 = 0.23817427385
My builds DR
(2620-1836)/2620 = 0.29923664122

You give up (4199.9744/3848.34= 1.09137300758) 9.13% damage and 5% Damage Reduction, to gain 6 EHP. I dont think its worth it, better switch to Soldiers and Berserkers.

Sir Kaboomski.1508:

I swapped my valkyrie chest for a knight’s chest

Ug, stop listening to bad players who dont know math.
Soldiers and Berserkers.

No thanks, I prefer it the way it is.

RagingDragon.9475:

Just so you guys know Ruby Orbs or Scholar Runes is not always the highest DPS. Eagle provides a % boost as well. So you would get a 5% boost roughly 50% of the time, which works out to an average 2.5% boost in DMG. People tend to forget to add in the DMG boost from Eagle Runes when calculating the overall DMG.

Your build with Eagle Runes = 3,590 Attack, 62% Crit Chance and 118% Crit Damage.
((3,590 × 38% (Uptime)) + ((3,590 x (150%(Base Crit) + 118%)) x 62% (Uptime)))
(1,364.20) + (5,965.144) = 7,329.344
+ (5% damage increase when target is 50% or less = Roughly a 2.5% boost).
= 7,513 Effective Attack Power

Your build with Ruby Orbs = 3,710 Attack, 58% Crit Chance and 122% Crit Damage.
((3,710 × 42% (Uptime)) + ((3,710 x (150%(Base Crit) + 122%)) x 58% (Uptime)))
(1558.2) + (5852.896) = 7,411.096
= 7,411 Effective Attack Power

It’s actually less than 2.5% for a number of reasons, but Im not going to go into it. My main concern is why are you trying to determine effective power? Just solve damage with the numbers in the damage calculations that change depending on the differences in the setups. Effective power doesnt mean anything. You also cant apply % damage bonuses to it like youre doing with the Eagle rune math up there.

Eagle stats
2609 power
2139 precision (63% crit chance)
118 crit damage
Dagger training + 5%, Exposed weakness + 10%, first strikes + 10%, Sigil of Force + 5%
Executioner + 20%, Eagle rune bonus + 5%
5+ 10+ 10+ 5 ((20 5)* .5)=42.5
+ 42.5% damage

Ruby Orb stats, + 1 power -1 precision infusion
2734 power
2053 precision (59% crit chance)
122 crit damage
Dagger training + 5%, Exposed weakness + 10%, first strikes + 10%, Sigil of Force + 5%
Executioner + 20%
5+ 10+ 10+ 5+ (20* .5)=40
+ 40% Damage

Eagle damage
(2609+ 250)* ((1* .37)+ (1.5+ 1.18)* .63)* 1.425= 8386.07598
Ruby Orb damage
(2734+ 250)* ((1* .41)+ (1.5+ 1.22)* .59)* 1.4= 8417.02848
Ruby Orb without changing the infusion
(2729+ 250)* ((1* .41)+ (1.5+ 1.22)* .59)* 1.4= 8402.92488

For damage purposes, Ruby Orbs actually beat Eagle runes. This isnt even counting the fact that that 5% damage under 50% HP isnt worth 2.5% damage, its actually worth about 2.3%, which would put Eagle runes even further behind.

Zepidel.5349:

http://gw2skills.net/editor/?fYQQNAoYVlUmaPHcy4E95Ex2jd2TeqzAnts/wVB-jQyAYLBRyCEhBkKAk8KiGbboIas6aMlLRUt9uIa1CBYeMA-e

UNTIL one of you lazy kittens proves me wrong THIS is the best rotation period. Go practice it on some dummies.

Since this is a PVE thread, you dont need to care about defense, so if you drop the 15 in acrobatics, you can put 5 into deadly arts and get whatever and 50 power, then 10 into shadow arts for gain 2 ini on stealth (which includes stealth from CnD) and then you can use CnD>BS>AA>CnD>BS>AA basically forever. You only lose 1 ini every rotation, so it will take you 10 to run out, provided you never have to do anything other than attack. But, if you use your heal every time its off cooldown, and use it the first time while missing 2-3 ini, your ini will stay full forever. SR will also refill it completely since it repeatedly stealths. And when bosses you fight reach low HP %’s you will have enough ini to CnD > assassins signet > BS > heartseeker x4 which will kill it if you do it at the right time, and that will drastically increase your DPS if you look at how much you did over the whole fight.

Base stats for you.
2659 power
2278 precision
112 crit damage

2282 precision is exactly 70% crit chance. You have 6 power infusions. Always use your infusions and mix of B/B B/V trinkets to hit an exact crit chance value.

Your build, full might stacks, Ruby orbs, Stones
(2659+ 250+ (35* 25))* ((1* .31)+ ((1.5+ 1.12)* .69)=8013.7552
Your build, full might stacks, Ruby orbs, Stones, -1 power infusion + 1 precision infusion
(2654+ 250+ (35* 25))* ((1* .3)+ ((1.5+ 1.12)* .7)=8064.386
Your build, no might stacks, Ruby orbs, Stones
(2659+ 250+ (35* 0))* ((1* .31)+ ((1.5+ 1.12)* .69)=6160.6802
Your build, no might stacks, Ruby orbs, Stones, -1 power infusion + 1 precision infusion
(2654+ 250+ (35* 0))* ((1* .3)+ ((1.5+ 1.12)* .7)=6197.136

In your build this is an increase of 0.6% damage, just dropping 1 power infusion for a precision infusion.
6197/6160=1.006~
8064/8013=1.006~

Your build, 1 might stack, Ruby orbs, Stones, -1 power infusion + 1 precision infusion
(2654+ 250+ (35* 1))* ((1* .3)+ ((1.5+ 1.12)* .7)=6271.826
Your build, 1 might stack, Ruby orbs, 3 precision infusions, Oil
(2546+ 250+ (35* 1))* ((1* .25)+ ((1.5+ 1.12)* .75)=6270.665

Your build, 2 might stacks, Ruby orbs, Stones, -1 power infusion + 1 precision infusion
(2654+ 250+ (35* 2))* ((1* .3)+ ((1.5+ 1.12)* .7)=6346.516
Your build, 2 might stacks, Ruby orbs, 3 precision infusions, Oil
(2546+ 250+ (35* 2))* ((1* .25)+ ((1.5+ 1.12)* .75)=6348.19

Oil is also better than stones if you constantly have more than 1 stack of might, as it starts to win at 2. Which basically means Oil will always win, because this is PVE, and people will constantly give you might, even if its unreliable, as little as 4-5 stacks once in awhile will make oil come out ahead.

Kaon.7192:

(3 + 4*(150+crtdmg)/100)% dmg for crits, which works out to be 11% for 50 critdmg and 13% for 100 critdmg.

This is wrong, or Im reading it incorrectly.

The way you calculate crit damage is to write crit damage and rate out in decimals (move the “.” two to the left, 120% Crit damage would be 1.2, 50% crit chance would be 0.5) then use the following formula:
Crit damage increase:
(1.5 + oldcritdamage + addedcritdamage) / (1.5 + oldcritdamage) =1.XXYYY
XX.YYY * .critrate = %damage increase

You should of known what you were doing is wrong, because adding to a whole number gives diminishing returns, not increasing, and you got 11% damage for 3% crit damage at 50% base and 13% at 100% base… which makes no sense.
(1.5 + 1 + .03) / (1.5 + 1)= 1.012
With 100% crit damage adding 3% more crit damage is a 1.2% damage increase if your crit rate is 100%. If your crit rate is 50%~ it’s
1.2 * .5= .6
0.6%~ damage.

With 50% crit damage base
(1.5 + .5 + .03) / (1.5 + .5)= 1.015
Its worth 1.5% damage at 100% crit rate, 0.75% damage at 50% crit rate.

Sir Kaboomski.1508:

I actually just looked at Cavalier for the first time and wow.. now that I know what it is it seems like a no-brainer to mix some in! Can’t believe I didn’t see that before!

Except that you lose tons of DPS for not that much gains on armor. Cavaliers and Knights look amazing, but theyre both actually terrible.

Wolfield.9812:

Two new sigils which are bugged atm. Both give +6% condition duration. One is called Sigil of Bursting which we reckon is meant to give +6% critical damage. So when it is fixed would be interesting.

Sigil of force +5% attack should give more damage than +6% critical damage, but with the 100% critical chance would be interesting. Also would it be better to have a sigil of force and a sigil of bursting, or air/fire still the higher dps.

It will be Fire/Air > Force (5% damage) > Accuracy (5% crit chance) > whatever with 6% crit damage.

Accuracy can be better than force depending on what you gain from critical hits though. Like my Guardian will use Acc because I get Vigor on Crits and I have a low crit chance, so the damage isnt that much worse.

For a crit damage sigil to keep up with Force on Thief, it would seriously have to give more than 20% Crit damage. Earlier in this thread I showed how 5% crit damage is basically 1% damage for your average thief, so ya, 20% crit damage or more required.

Wolfield.9812:

I know how you like to look at numbers, so thought you would find it fun. Enjoy your break, and Pokemon addiction

Saddly, no math is required. 6% Crit damage isnt anywhere near enough to even be in contention. But it’s ok, Ive been back to playing the worst game ever that I wish was fun… because nekomimis are the greatest things E V A R. But seriously, my mithra is even cuter now with the new character creation options… I just wish it was more fun to play her. Have I said I wish this game was fun yet? Effing Square Enix… back to Pokemon and WvW for a week til early access starts now that open Beta is over.

If you used Soldiers hat body and legs with Berserkers gloves pants and boots youd do more damage and stay alive better compared to Valkyrie armors. Granted, you wouldnt do that much more damage or stay live that much better, but every 2-3% you can gain is good and puts you in a better position to wins fights.

Kiijj.3594:

stevie.4187:

BG now has Commander Kidinkk :0 yeyeye

http://www.youtube.com/watch?v=tX6-gZw2tog

This KidInkk?

lmao nice.

He’s like a 12yr who wants to be Sacrx, how cute.

Silver.4798:

I was wondering, what about superior rune of Rage? 10% crit damage at all times, 15% crit damage under fury?

I guess this is more viable with warriors who have easy access to fury, but we do have thrill of the crime, kinda sorta fury…

+ 5% crit damage is only + 1% damage for your average thief. 25 Power is basically worth the same amount of damage as + 5% crit damage.

Rissou.7213:

I seriously think runes are overrated, just use orbs. If you are just going for damage, take orb. You wont see any difference.

You wont notice the difference in game when playing. But if you do the math or if there was a way to parse, you would notice. Runes offer more than orbs, even if all youre looking for is DPS.

Crimes.5317:

I read somewhere that the 6th scholar rune is only worth it if you can stay above 90% hp for 25% of the fight which isn’t hard to do if you run full glass build, if you are looking for a more balanced build then I would go for Lyssa, Divinity, Wurm, or Orbs.

Edit: BTW Rage is garbage even with 100% fury uptime.

Anything is better than the 6th Scholar rune if youre not above 90%. Scholar runes are for when you take 0 damage. Like when you do COF p1. You take 0 damage or you die in half a second and you wear all berserker gear and just try to max your DPS.

Avasar.9586:

Hi, new rogue here and Ive got a quick question about the EHP vs. actual health pool. It looks like a lot of these builds have ~15000 hps so where does the extra health come from to determine this ~21000 EHP?

The extra health comes from your toughness. EHP is the total damage you can take counting your Armor (defense+toughness) and and HP.

Effective Health (EHP)= (Health * Armor)/(Reference Armor)

Damage Reduction (DR)= (Armor – Reference Armor)/Armor

The reference armor used for both calculations is 1836, the lowest base defense for a fully exotic equipped level 80 character.

Wolfield.9812:

Hey Wish, hope you are still updating this awesome thread. Do the new sigils/runes have any effect you reckon?

Mainly thinking the +6% critical damage (when they fix the bug on it) being a rather interesting number crunching exercise.

I dont know the stats on the new runes, or that they added new runes, lulz, could you tell me what they are? (I was serious when I talked about my current Pokemon addiction, I got 5 games to play! platinum, black, white, black 2, white 2, and soon X and Y!)

Divinity, Wurm, Scholar, Melandru and Lyssa are the only real choices IMO.
Divinity/Wurm offer the most versatility for adding some defense depending on how much you can afford. Scholar is the best for DPS. And Lyssa/Melandru are the best for utility purposes.

ArtemisEntreri.4138:

Wish,

What is your strategy for facing condition builds with your Mug build? You lack a lot of condition removal with that build. Is the key, kill them before they get too many conditions off?

Have you thought about Hoelbrak Runes + Lemongrass?

Depends what class it is. If youre talking about bleed thieves, you always stealth after they do, so their from stealth attack doesnt hit you. And use the sword to chip down their HP to 50%, then burst them down with D/D. They will get use to your sword damage and think you dont hurt that much, so you can surprise them and drop them when you switch, let them get a from stealth attack off at this point though so they have revealed. Also, Ive never had the shadow return on sword not remove bleed, and you can basically remove 80% of their damage with just that.

For necros, lots of kiting, and if theyre very good, its best to just walk away from them. While you can win, if theyre very good it will take a really long time, and isnt worth the effort. Knowing when to use your dodges is important too. Like for Necros, if theres a glowy green line connecting the two of you, you need to dodge roll 2-3 seconds after it appears, or use shadowstep when you get feared if you get bound and cant dodge roll. You shouldnt run across a glowy green line on the ground, and always dodge roll into marks on the ground when they stack a bunch of them, which they tend to do when you stealth. They think if they put all their marks on the ground and stand there youll die if you get close. Also, you need to avoid the circles with lots of black in them that dont go away when you step on them. If you stand in those you die. When they go all “Im a black cloud of death” just walk away from them. If you have trouble walking away from them, use dagger 4 and shadow return. Save bask venom (or thieves guilds) and steal for after they waste their move that grants them an extra green HP bar, and after the transformation move they use, when these 2 things are down you can actually kill them. While theyre up youre just feeding them if youre close to them. You also need to watch for weakness, and try to avoid attacking them when you have it on.

For condition Warriors… Use steal ASAP so you get their spinny whirlwind attack thing like Barbarian in Diablo 2. This reflects projectiles, and condition warriors use sword/shiled and bow 99% of the time. The attack you want to avoid is the one that does a heavy bleed and bind. If you can avoid that attack from the bow, and know to run away when he shoots fireballs (which are ground targeted) you’ll pretty much be fine. 99% of warriors will use the bind attack immediately when they swap to bow, so they can hit you with the other attacks easily. When this happens you need to hit the F1 skill you stole because it pretty much ruins their day. If it happens as youre running to the warrior the first time, use steal to get next to him and use CnD so he cant hit you with anything but the #3 ground targeted fireball thing. The only other thing you need to avoid is the sword F1 skill. Which binds and sticks you with like 15 stacks of bleed. If you get hit with that, use your shadowstep to get away. If you dont have it up, swap to sword, and hit 2 twice right as it ends to remove the bleeds. But you should be able to kill him before he gets that move twice, so shadowstep should always be up. The thing to know there is that the first attack has the bind, and the other attacks add more bleeding, so simply shadowstep away, and if you have a bunch of conditions on you, shadow return when its animation is done.

I dont think theres any other condition based classes you need to care about honestly, and I’ll see if I can find some video of me fighting these to help you out.
Im uploading a video of a necro that shows the green lines and what happens when you dont dodge it, and also me wasting all his cooldowns then killing him. The dodge roll after I downed him was because I expected him to use the downed fear on me and I was aiming shadow refugee.
http://youtu.be/HgE5BJs0MQI
I dont think I have any video of the other classes, so if anyone wants to duel me on them, let me know, and then I can post video of that too. My thief is on gate of madness though, so you need to be fighting that server.

As for other runes, I have a set of Soldiers gear with Melandru runes, and one with Lyssa runes. I use the Lyssas when Im playing double S/D, Melandrus I just have to have. For D/D+S/D I dont really have issues with conditions even if I run into a condition specced class, which is rare anyway, the whole meta of both servers Im on has shifted around mitigating conditions because of how overbuffed Necro got. So as a consequence, theres barely any condition specced people to fight. I have been noticing more P/D thieves when I log in, but theyve all been really bad.

paleeshi.1924:

So uhm, Wish, which one is your final gear set up? There was so many build links, and most were posted about 1 month ago.

Wish’s Mug, 10/30/30/0/0
with accurate corrected precision

Wish’s Fall, 0/30/30/10/0
with accurate corrected precision

https://forum-en.gw2archive.eu/forum/professions/thief/Post-Your-Build-Thread/page/3#post2426220

I havent used a shortbow since the lac strike buff months ago. Sword #2 and a stealth is all you need to get away from people.

I logged on this game yesterday for the first time in 2~ weeks to actually play it. I died to last refugee after about 45 minutes in WvW and rage quit.

I dont like your threads name, stop copying me unless youre going to post a lot of math.

LazyKobolds.2301:

For optimizing stats, have you considered swapping out the helm for a “celestial” helm?

I made an entire thread about Celestial gear.
https://forum-en.gw2archive.eu/forum/professions/thief/Celestial-Armor/first#post2424729

The results of the math in the thread show that Celestial is terrible. But just so its here too…

LazyKobolds.2301:

normal stats for helm
45
32
32/2% crit
1:16 ratio

celestial stats
20 (all)
3% crit

You lose 25 power, but gain 3% crit. If we take 3% of an average 3000 attack power, we get 90 extra power from this (45 because of 50% crit chance, still more than the 25 we lost).

We also gain 20 prec to add another point of crit chance (assuming there was ANY leftover wasted precision elsewhere), and get to keep 20 vit+tough (12 less than soldier).

First: Attack means absolutely nothing. Dont be thinking 3000 attack is awesome. Only your power matters for damage calculations. When totaling damage its weapondamage * power, not weapondamage + power, which is what attack is.

Now, we’ll use my Mug build, because it has exactly 1 wasted precision, and has higher Power and lower Crit damage than the other set, so it will work into Celestials favor more than the Fall set. Screenshot of the stats with food and without bloodlust or any boons up will be at the bottom.

2166 power (soldiers hat)
1915 precision (1 over 52% crit chance)
92% Crit Damage.

Soldiers Hat
(2166+ 250+ (6* 35))* ((1* .48)+ ((1.5+ .92)* .52))=4565.0384

Celestial Hat 53% Crit chance (this is what youre suggesting happens, but it doesnt)
(2166- 25+ 250+ (6* 35))* ((1* .47)+ ((1.5 +.95) *.53))=4599.8685

The problem here is our toughness and vitality and master maintenance oils precision gains:
Soldiers hat
1556* (60/1024)=91.171875
1368* (40/1024)=53.4375
Celestial hat
(1556- 12)* (60/1024)=90.46875
(1368- 12)* (40/1024)=52.96875

We’re losing at least 1 precision (depending on when the game deals with remainders and how, its one or two lost precision) from the loss of 12 Toughness and 12 Vit, so we have to dump one of the 6 power infusions for 1 precision infusion to gain the 1% Crit chance, otherwise we end up with this:
Celestial hat 52% Crit chance
(2166- 25+ 250+ (6* 35))* ((1* .48)+ ((1.5+ .95) *.52))=4562.154
Which is lower damage than the Soldiers hat.

Celestial hat 53% Crit chance, -1 power infusion +1 precision infusion
(2166- 25- 5+ 250+ (6* 35))* ((1* .47)+ ((1.5+ .95)* .53))=4591.026

Now, lets look at the DPS increase.
4591.026/4565.0384=1.0056927451
0.57% damage for 24 stats swapping to Celestial hat and 1 Precision infusion.

The correct spot to swap out for more offensive stats next is weapons. So lets swap a weapon from Soldiers to Berserker, 64 toughness/Vit for 64 precision 5% Crit damage.

Soldiers Hat
(2166+ 250+ (6* 35))* ((1* .48)+ ((1.5+ .92)* .52))=4565.0384

Swapping in 1 berserker weapon.
Oil:
(1556- 64)* (60/1024)=87.421875
(1368- 64)* (40/1024)=50.9375
We’re losing 5 precision from our oils, and gaining 64 from the weapon
64/21=3, with 1 stat point left over.
So we’re going to have to make a -1 power infusion +1 precision infusion here too, and gain 3% crit chance.

(2166- 5+ 250+ (6* 35))* ((1* .45)+ ((1.5+ .92+ .05)* .55))=4740.0785
4740.0785/4565.0384=1.0383436205
Here we gained 3.83% damage for 128 stats.

To determine how much each defensive stat is worth in damage output in both examples we simply divide
Berserker dagger
3.83/128=
0.029921875% damage per point of defense
Celestial hat
0.57/24=
0.02375% damage per point of defense

This is the argument against a single piece of Celestial gear that seems to increase DPS. Sure, it increases DPS, but it costs more defensive stats to do than other choices you could be making, which makes it bad. Those defensive stats youre giving up are worth more than the damage youre gaining. I dont know how to explain it better than this, so i hope you get it. Not only that, but if youre not able to figure out your exact precision you’ll end up with worse damage output than if you left the Soldiers hat on, since it definitely does not raise your crit rate by 1% all on its own.