Could the yellows be a result of the dreaded two-roll theory?
I'd almost guarantee it. Rogue specials were changed to a two-roll system around the time Mutilate was introduced, and Stormstrike works in exactly the same fashion as Mutilate.
Unless I'm grossly mistaken, the paperdoll crit rate should be crits over all swings. So unless crit-capped (which is not the case here), Malen's rate of critting gruul should be about 30% of all swings. In fact, white crits/all white swings are about 18%. WF is 13% (when including misses in the denominator) and SS is 17% (same).
I'm truly stumped. Why is Malen critting Gruul much less than he should be?
Unless I'm grossly mistaken, the paperdoll crit rate should be crits over all swings. So unless crit-capped (which is not the case here), Malen's rate of critting gruul should be about 30% of all swings. In fact, white crits/all white swings are about 18%. WF is 13% (when including misses in the denominator) and SS is 17% (same).
I'm truly stumped. Why is Malen critting Gruul much less than he should be?
Maybe I'm missing something here, what if chance to crit is calculated before chance to miss?
apologies if someone has already brought this up, but with four minutes listed in combat, I would be shocked if his actual crit value closely matched the expected crit value.
Run your sim in 4 minute increments, I'd be interested in seeing how much crit fluctuates within them.
Don't get me wrong, I'm not saying there's not other stuff at play. I'm just saying that it's possible that the majority of the low crit rate is simply due to the teeny sample size.
You know how in national poll, they generally have just over 1000 respondents to get a 3% margin of error? Compare that to the fact that Malen's crit is being calculated over 250 hits or less (white swings).
I'm usually a forum lurker but I've been following these forums and I couldn't resist.
I am unclear about certain things and was wondering if somebody could help me out.
1) Is the WF Proc cap still exactly 3.0s or has it changed in 2.1?
2) According to everything disquette has said I understand that there is a certain point where the marginal dps gains from %hit, hits a maximum. From there on the marginal dps gains start becoming smaller and smaller for each %hit added. Now here's the question.. Is it possible to figure this out or is it totally depended on other variables such as AP and CRIT?
3) Does the 9% hit rating from talents add 9% to your hit percentage by 9% or does it increase your hit percentage by 9%?
4) Which of the following enchants increased DPS the most(DW ofcourse):
2x Crusader (2xprocs) enchants
2 x 20 Strength Enchants or
2 x Mongoose Enchants.
Can the crusader proc twice?
5) Is it confirmed that Slow/Slow is better then Slow/Fast? I need to know this because I plan to DW any 2 Arena weapons and need to know which ones to aim for.
6) I started a shaman about a month ago with the goal to get the maximum melee dps gear that isn't BoE, from Raids or BoP Tradeskill (Basically only from PVP, Faction rewards, Arena, 5-man's normal, Heroics and Heroic Badges). I need someone to take a bit of time and help me out on gear, socket and enchant decisions, I would gladly appreciate it. Here's what I am aiming for in the mean time.
Now fo the interesting part these are the epics i've found that drop in heroics that I could use:
Hand guards of the steady:http://www.wowhead.com/?item=32076 (Should I replace the hands in the Desolation Set with this or Replace with Legs with Wymrscale? see below)
Question is.. what should I replace out of the desolation set..the Leg's of the Hand's? Should I use the Linked set or the chained pvp 3-set? Are the heroic items better then what I plan to get?
7) Knowing this will I able to work out my total stats(with sockets / enchants / LW armour kits): Strength/Crit/Hit/Haste and work out my DPS?
8) Question 6 doesn't need a reply asap it's long
Well that's all I need to know for now. Sorry if I come across a bit "Spammy" I'm just really enjoying playing a shaman and i'm a bit obsessive about the gear. Thanks to Disquette for all the work you've done.. when I get full started on this shaman I will gladly help out with any testing you need.
the table was the chance to NOT be in a flurried state ;-) if you have a 25% chance of critting, then on your first attack, you are 75% chance NOT to be flurried. On a given attack, the chance of you not critting any of the previous three attacks = (1 - 25%) * (1 - 25%) * (1 - 25%) = (3/4)^3 = 27/64. so in 27 of 64 swings, you will NOT have crit'd on any of the prior 3 attacks.
Put into useful terms, a 27/64 chance of NOT critting any of the prior 3 attacks = a 37/64 chance that you DID crit one of them.
Can I also use this same formula to calculate the average percentage of time that unleashed rage will be up? (Assuming 1 hd)
According to the accepted mechanic, spelled out in the Combat Mechanics 3.0 thread, each swing is one roll. (This is still true for white hits, even if some yellow hits require two rolls.)
There's a "hit table" giving you a chance for each, in a certain order. Once the sum goes above 100, the remaining outcomes are impossible. Then the game rolls once to determine the outcome.
Vs. a mob, the sequence is
Miss
Dodge
Parry
Glance
Block
Crit
Hit
Malen Miss+Dodge+Parry (+block, I guess) is 13.6%
Glance is 59/221 or 26.6%
Crit should be ~30%, according to him, but is behaving as 18.6%
Hit is the residual category, filling whatever is left to 100. In this case, about 41.2%. (Checks: 100 - the others = 41.2, and 91/221 = 0.412.)
Disquette--you could very well be right, and that's my first instinct. It just seems SO far off. I'm also interested in checking whether WWS is speaking the same language as we are, and it looks like NOT. WWS typically excludes misses/dodges/parries from the denominator, inflating crit and glance rates in its reports.
The "hosted" reports seem better about consistency with our terminology than Malen's report. Check this out: http://www.lossendil.com/wws/?report=wrgqoc4et5ugo&a=24 It looks like all the melee percentages are using "total swings" as the denominator.
The "hosted" reports seem better about consistency with our terminology than Malen's report. Check this out: http://www.lossendil.com/wws/?report=wrgqoc4et5ugo&a=24 It looks like all the melee percentages are using "total swings" as the denominator.
Lemme try uploading a log and see what we get. Maybe the problem is how they're running one version on the website itself and another version for the "host it yourself" stuff.
Dang I didn't put the log on my guild ftp, will have wait till later.
Okay i'll read through it again... What's your opinion on enchant's though? And the gear considering the criteria I want to meet. No Raid Gear No BoP tradeskill gear
I've completed the first part of my spread sheet. It calculates the expected dps of a main hand weapon. It takes into consideration, hit%, crit%, AP, Flurry % (incase you don't spec for 30% or have set bonus), Unleashed Rage %, Weapon's Mastery, Storm Strike, Elemental Weapons (Talent) and Windfury rank. I set it up, right now to compare 2 weapons. This is actually only the first step, btw. You can actually compare other stats too; just put the same weapon in both spots and change the crit % or hit % or AP, based on the different equipment. I make quite a few assumptions. I'd like feedback and to fine tune it before adding dual wield.
The problem is I don't know where to post the Spread Sheet on the internet. Does anyone know how I can do that?
I don't understand why there is a difference in glancing attacks, 21% to 37% respectively. What could cause that big of a difference? Both NElfs, Deadlast daggers Sango swords, and both with weapon expertise and no other bonus. Myself, I had a 38% glance. Shaman armory:
The low presence in WWS is related to me cutting out the fat from the trash clear and the first wipe. The WWS only shows for the kill itself, so don't worry about that.
Deadlast and myself, Graecus. He was in a nonshaman group, so had Deadly / Inst, whereas I as in a shaman group with WF / Deadly oh.
Note: his damage in regards to poison completely offsets my damage with windfury. Not 100%, but close enough to negate. The only difference of gear at the time of the parsing was he had Dragonspine Trophy. This time, glancing was within range of one another. He beat me by 30k damage and it seems only because of 10 extra backstabs and 10k arbitrary dmg. I normally blow him away in damage, and though WF would just be the icing on the cake, so I'm not entirely sure what I could have done to bridge the gap.
Sorry to derail, but thought with the glancing / WF talk, this might apply. Kazz might just be too short a fight to accurately tell perhaps.
1) Is the WF Proc cap still exactly 3.0s or has it changed in 2.1?
2) Is it possible to figure this out or is it totally depended on other variables such as AP and CRIT?
3) Does the 9% hit rating from talents add 9% to your hit percentage by 9% or does it increase your hit percentage by 9%?
4) Which of the following enchants increased DPS the most(DW ofcourse):
5) Is it confirmed that Slow/Slow is better then Slow/Fast?
6) Question is.. what should I replace out of the desolation set..the Leg's of the Hand's? Should I use the Linked set or the chained pvp 3-set? Are the heroic items better then what I plan to get?
7) Knowing this will I able to work out my total stats(with sockets / enchants / LW armour kits): Strength/Crit/Hit/Haste and work out my DPS?
1) 3.0s
2) Dependent on AP/Crit
3) Changes your attack table from, e.g.:
Miss 20%
Everything else 80%
to:
Miss 11%
Everything else 89%
4) Mongoose looks the coolest therefore it is the best. +20 str is the easiest to get therefore it is the best.
5) Yes
6) Do the math yourself using AP equivalency values posted
7) Sure, use the simulator provided by Disquette
Okay i'll read through it again... What's your opinion on enchant's though? And the gear considering the criteria I want to meet. No Raid Gear No BoP tradeskill gear
It all depends. Some like the AP from crusader or 20 str, and some like the crit from the very expensive mongoose which allows us to get other stats over +crit.
Alright but you do agree that Crusader is better then 20 Strength unless it procs twice in 15 seconds?
Crusader is better than 20 str no matter when it procs. When you see crusader proc, you have time for at least 1 SS and several WF procs. The extra str on the WF procs/crits is what makes it so valuable. It also helps with SR, fully raid buffed with dual crusader procs I've seen SR return 400+ mana per proc.
I haven't had the chance to test with mongoose yet, because I'd like to pick up dual gladiator weapons first to get the most umph out of it. However, when I do I'll try to sport the increased crit dmg metagem and the Hourglass of the Unraveler.
the table was the chance to NOT be in a flurried state ;-) if you have a 25% chance of critting, then on your first attack, you are 75% chance NOT to be flurried. On a given attack, the chance of you not critting any of the previous three attacks = (1 - 25%) * (1 - 25%) * (1 - 25%) = (3/4)^3 = 27/64. so in 27 of 64 swings, you will NOT have crit'd on any of the prior 3 attacks.
Put into useful terms, a 27/64 chance of NOT critting any of the prior 3 attacks = a 37/64 chance that you DID crit one of them.
Assuming Stormstrike and Wind Fury can proc Flurry, but not use up the charges, should I adjust my crit % for the purposes of figuring out the percent time I'll be under flurry effect?
As an example, how many hits in 10s can cause flurry?
A 2.6s weapon is going to swing about 3.85 times in 10 seconds.
Stormstrike is going to swing 1 time in 10 seconds.
Windfury is going to swing (3.85*.2)-3s lock out = 1.54 times in 10 seconds.
So with no flurry effect I'll have about 6.39 swings for the price of 3.85, that can cause a flurry effect.
So then is my effective crit rate to figure out the percent time I'm under flurry,
(6.39/3.85)*crit rate of 25% = 41.5%
Do I then use that adjusted crit rate to figure the percent time under flurry?
1-((1-41.5)^3)
I'm trying to figure out how to account for the extra hits that can proc flurry, but don't use charges and don't take any time to hit.
Thanks for doing that, Malen. The glancing rates and crit rates are indeed different on the hosted version. The hosted version matches the hand calcs I did several posts up.
Glancing is down to 26%, mitigating to 72% (still). For modeling purposes, I'm going to call this 25%/25% until proven otherwise.
Your crit rate is still really low -- sub-20% for white, WF, and SS. I'd like to chalk this up to bad luck in a short sample. If it turns out that Gruul has some crit-mitigation that's unique to him, such as resilience or enhanced defense skill or something, I'm going to be unhappy. He's a great testing mob otherwise, since there's never going to be any extraneous attacks on any other mobs.
In order to keep using our shared language, I'm going to stick with "hosted" results, since as we've seen here, the %ages are calculated differently from that (older?) setup.
Edit: I'm looking for a good long parse fight to see glance and crit rates--the patchwerk of BC. I don't think Magtheridon is good because of all the adds, right? (We haven't done it yet.) I know nothing about SSC bosses. Prince Malch might be a good candidate, I guess--those fights can take a while and there's no adds or anything. You SSC+ folks, what would be a good one to look at? I can check several guilds' results on WWS hosted.
Edit: I'm looking for a good long parse fight to see glance and crit rates--the patchwerk of BC. I don't think Magtheridon is good because of all the adds, right? (We haven't done it yet.) I know nothing about SSC bosses. Prince Malch might be a good candidate, I guess--those fights can take a while and there's no adds or anything. You SSC+ folks, what would be a good one to look at? I can check several guilds' results on WWS hosted.
Magtheridon does have adds, but damn if that's not a long fight. I think we spent something like 15 minutes in combat this week; even if you chop off 5 minutes for the Channelers it's still 10. The Channelers are ?? level, so I would think that glance rates would be similar against them; nevertheless, you could theoretically have someone start logging after the channelers are all down.
Problem is that guy needs to be on Mag the whole time. Won't be me because I'm one of the clickers in my guild.
Btw one my prot warriors candidly pointed out that he beat me on DPS during Gruul. I don't understand what is going on here anymore, I was potted and everything. I took a little more time to get out of the shatters because I've been killed from people crowding on me before, but it didn't seem *that* significant of a loss of DPS.
05/30/07, 1:59 PM
Similar Threads
