The Monk Temple

[KaosPrimeZero]

/snip

Very appreciated for the long description (and very informative) You can have a cookie

[Artryafe]

Nice write up Powercow. Now that we have skillspeed out of the way do we now know an official stat priority for monk?

[Naalya-Deix]

Common sense is also to be considered with maths, especially when it could be biassed.

With 2.10 GCD, True Strike is kind of equal to spamming twin snake each cycle (in fact just ahead and Exponentially better with Crit rising). This is the breakpoint GCD to aim for, since if you go to > 490 SS, you'll get poor Det and Crit stats while gaining ONLY 0.10 SEC. Many people didn't realise yet that Skill Speed efficiency is logaritmic cause of GL3.

More Skill Speed you have, less sec / ms you gain point by point with GL3, because SS is just mathematically insignificant compared to GL3.
There is a BiS that bring you to 373 SS (2.10 GCD), 500 + Crit and 290 Det. Also, Crit / Det has always been in ANY mmos the way to go for Melee fast attacker (synergies with AA).

[EasymodeX]

Lol.

Skill speed is obviously reduced significantly in value because it only applies to a fraction of the total damage output.

Likewise, crit rating is reduced in value because it only applies to crittable hits, and 20% of your hits already crit.

Hence, a full analysis will reduce the generic value of skill speed by [%AA + %OGCD], with some modeling assumptions for DOTs (better uptime, but fixed tick cycle is a statistical question), but it will also reduce the value of crit by the baseline crit. Which is higher when you include successful Bootshines.

I've posted a general approximate stat weight for Monks elsewhere already. For reference, it's:

w/ high Bootshine:

STR 1
DET 0.225
CRT 0.148
WD 7.536
SS 0.147

w/ no Bootshine:

STR 1
DET 0.226
CRT 0.165
WD 7.536
SS 0.147

[Risbyn]

/snip

Very nice, I can see now where you're getting the 50% number. Wish more people posted like this so that we all can at least understand where they are getting their numbers. Your post is appreciated.

[Allyrion]

Lol.

Skill speed is obviously reduced significantly in value because it only applies to a fraction of the total damage output.

Likewise, crit rating is reduced in value because it only applies to crittable hits, and 20% of your hits already crit.

Hence, a full analysis will reduce the generic value of skill speed by [%AA + %OGCD], with some modeling assumptions for DOTs (better uptime, but fixed tick cycle is a statistical question), but it will also reduce the value of crit by the baseline crit. Which is higher when you include successful Bootshines.

I've posted a general approximate stat weight for Monks elsewhere already. For reference, it's:

w/ high Bootshine:

STR 1
DET 0.225
CRT 0.148
WD 7.536
SS 0.147

w/ no Bootshine:

STR 1
DET 0.226
CRT 0.165
WD 7.536
SS 0.147

Unless I'm missing something, this doesn't make sense to me. You're basically saying crit has diminishing returns? That the more crit chance you have the less you value it?
I'm not the best at modelling stats, so I might just be misunderstanding something.

Everything I know about probability stats says it doesn't work that way. You may have a baseline of 20% crit, but if you get up to 40% crit you increased your damage by a set amount. Increases don't change depending on how much crit you've had before. Going from 98%-99% is the same dps increase as going from 10%-11%.

REASONING:
Effectively every percent of crit gained is equivalent to a percent of your damage that does 150% damage. It is a linear progression. Going from 20% to 40% is doubling the amount you crit. This is effectively a 50% increase on an EXTRA 20% of your damage, in addition to the 20% you had already.
The only diminishing returns on crit is when you pass 100%. Bootshine fits in this category. That being said....

COMPLICATION:
Internal Release isn't easily modeled (by me, at least). The reason for this is that it's a short term increase. We can say it increases crits by 7.5% on average per minute, but that's not really accurate. It is a short term buff so it doesn't balance. You can get a series of no crits or a string of full crits due to the IR. It's not as predictable because it's short term and these probability stats only balance out over the long term.
That's not to say it can't be formulated.

It should be noted however that since IR is not 100% increase, it can't be said that crit has little value to it. The relationship is complicated, but I'd need to see or know it was properly modeled before concluding that it effectively diminishes the value of crit.
The value of crit is outright reduced by Bootshine. This is true.

CONCLUSION:
Maybe you understood everything I said above. But I had to respond since you said

crit rating is reduced in value because it only applies to crittable hits, and 20% of your hits already crit.

This is just an overly simple statement that doesn't take into account to linear damage increase that crit provides.
If your model deems crit worth less cause of proper IR modeling, which is admittedly up 25% of the time, then I will understand.

It sounded more like, we already crit on 20% of our attacks, so any gained crit will only affects 80% of our damage.
That would be missing the point of how crit increases damage (as I said above).


I also have a quick question about your stat references.
From what I know, the stat budget gives more and less of certain stats right. For example, you can get more SS on a piece than you can get DET. Crit is in the middle.
If that's the case, do you know the ratio? Because it seems that SS is favored with high Bootshine over crit. It may even be favored with no Bootshine since you can get more of it?

[Herrstarr]

Wish mnk had some TP generation attack, doing turn 5 at the moment and keep going out of TP all the time >.<, Invigorate is not enough :'(

[Powercow]

I also have a quick question about your stat references.
From what I know, the stat budget gives more and less of certain stats right. For example, you can get more SS on a piece than you can get DET. Crit is in the middle.
If that's the case, do you know the ratio? Because it seems that SS is favored with high Bootshine over crit. It may even be favored with no Bootshine since you can get more of it?

I wouldn't put much (or any) stock into those. They're accurate only when one has an absurdly high base crit rate. See, stat weights will change based off of the rest of your gear. The more crit you have, the more useful determination and SS become. The more SS you have, the more valuable crit and det are, and the more det you have, the more valuable crit and SS are.

"So Powercow, should we go for a balanced set of stats?!" Why, no, no not at all.

See, these increases to the values of other stats only really come into play when you have a lot of them. If it was possible to have around 50% crit all of the time, sure, SS might start to outpace crit. But see, I try not to deal in "well in ilvl 120 gear they might sorta maybe have super duper high crit values", I try to keep things where we are now, and where most people are making their decisions (which is around the Darklight and early Coil areas of gear progression.) Because crit values are so low at the moment (for the record, base crit is 5.15%, not 20%), even in the most drastic of situations of going from no crit rating to a full set of BiS gear focused on crit/det, at most crit will see a...

[showingwork]
5.15% is baseline. 16.65 is with the turn 5 BiS set. Add in the 7.5% average boost from IR (for simplicty's sake) and we're going from 12.75% to 24.15%. This is going from 106.375% to 112.075% from crit alone. This is a 5.358% DPS boost instead of the expected 5.7%.
[/showingwork]

This is not a 20% reduction like Easymodex seems to think it is, but a 6% reduction. Add in the multiplicative 6.4% penalty for liberal use of Bootshine, and crit is, even when stacking the hell out of it, 12% penalized compared to SS, which is penalized by about 50%. And...

[showingwork]
As shown on the previous page, 106 SS reduces the GCD by 0.091 with 3 stacks of GL. 0.091/2.17*100 = 4.193% more average damage for the parts SS boosts.
106 crit rating boosts crit from 5.15% to 12.49%, which is 7.34% / 2 (since crit gives 50% more damage), or 3.64% more average damage for the parts Crit boosts.

Now let's factor in the penalties for the parts SS doesn't boost (50%), and the part Crit doesn't boost and the part Crit receives slower scaling for if you have more of it (12%)

The first 106 Skillspeed will boost your damage by 2.0965%. The first 106 Crit will boost your damage by 3.2032% (and even slightly more, not enough time to re-do the math here, but it wouldn't get as much of a penalty due to higher crit rates with this amount). This mean crit will be about 52.78% more effective, point for point, than a single point of Skillspeed.

And for what it's worth, accuracy, crit and skillspeed all cost the same in terms of itemization costs. Only determination costs more.

And yeah, I'm aware someone will come along and say "it's actually not 52.78%, it's 51.19%" or some silly nonsense. The point still stands. No amount of rounding, fuzzy math, or good feelings is going to make up for a >52% difference. Even with the absolute BiS gear, the difference might dip to 50%. Until you get really, really, REALLY high crit value, crit will always beat Skillspeed. Given the current amounts of stats we have on our gear, the difference will always be pretty damn huge.

[Sibe]

personally i dont use fracture but only touch of death as dot, because i saw a 10 dps drop from using fracture, i think fracture is good with a SS focused gear, but not with a crit/determination gear, of curse i don't know if the "battle pharse program" are really correct atm

[EasymodeX]

I'm not the best at modelling stats, so I might just be misunderstanding something.

Everything I know about probability stats says it doesn't work that way. You may have a baseline of 20% crit, but if you get up to 40% crit you increased your damage by a set amount.

You're focusing too much on stats and damage; the question at hand is the value of the stats as they contribute to damage.

If you have 10,000,000 strength and 0% chance to crit, how much value do you get from +1 str? Nearly nothing. How much value do you get from +1 crit? A heck of a lot.

If you have 10 strength and 90% chance to crit, how much value do you get from +1 str? A lot. +1 crit? Very little.

Optimizing stats on gear is about relative increase to your damage. If your class had 50 autocritting abilities and a baseline 80% chance to crit because SquareEnix says so, then how much value do you get from crit? Wouldn't you want to buff your base damage instead to take advantage of that crit?

Another way to look at it is that you get linear gains from +crit or +ss. Wouldn't you rather get superior geometric gains by multiplying your crit or ss?

The largest area coverage from 2 linear values is always a square, not a rectangle.

Yes, most normal stats in MMO mechanics have diminishing marginal utility. Or soft diminishing returns in terms of value.

Crit is a chronic offender because plebs love to see big numbers and categorically ignore the fact that if they already have a ton of it, they should invest elsewhere.

Edit: Check this out; I put a lot of tender loving care into it: http://imgur.com/GpD1fh4