Is there a "reasonable" percentile to aim for?

Anything mage specific not covered by any of the other subforums, like raid instance guides for mages.
qivive
Posts: 12
Joined: Thu Feb 01, 2018 1:55 pm

Is there a "reasonable" percentile to aim for?

Unread postby qivive Fri Feb 02, 2018 5:08 pm

Hello,

with the overwhelming success of warcraftlogs and similar sites we can judge dps performance better than ever before. Not only are there enough parses to have a rough idea how much dps everyone else does, we can even filter alongside fight length, legendaries and item level brackets and so on. As a casual player, who likes to get better at the game, this system does reveal that some dps-affecting factors are not personal (fight length, encounter strategy, existence of other heavy AoE classes, proc luck, boss mechanics etc.) or are brought in artificially (padding, supporting a parsing player). Even the averaged item level is a source of potential distortion as weapon level has a much bigger impact than any other slot (taking higher, but not optimal pieces may result in heavy drops at the percentile while total ranking may still rise).

Looking at these factors its not always clear how much of a low ranking/percentile is due to wrong playstyle. Taking these uncertainties into account, what rank/bracket do you regard a solid performance ("Well this fight went good, I don't need to improve more on this one.")?

Personally I was aiming at around 75 percentile, less seemed to lenient, more may be to harsh regarding non-personal dps factors. But how much of a toons dps is really just the player behind it? Estimates? Can anybody, with any legendaries and any gear level composition with any viable boss tactics achieve 50/75/90 percentile? (not talking about top 300-500 or higher)

Q.
fewfew
Posts: 2
Joined: Sun Feb 04, 2018 7:39 pm

Re: Is there a "reasonable" percentile to aim for?

Unread postby fewfew Sun Feb 04, 2018 7:46 pm

Honestly I think it is a product of the environment that you raid in. Personally my guild is terrible at mechanics. So, if we want to clear content prior to the next tier releasing, we have to be min/maxing to dps through whatever mechanics are possible to. We're striving for 90%+. In a more realistic environment I think your baseline of 75%+ is a great target.
roldy
Posts: 38
Joined: Tue Jan 09, 2018 8:56 pm

Re: Is there a "reasonable" percentile to aim for?

Unread postby roldy Mon Feb 12, 2018 11:58 pm

[...]
Not sure if you're talking about parses during progression or farm so I am just going to cover both topics.

During progression (or basically in general) ranks are the least accurate and probably most toxic way of comparing player performance. During progression what counts is doing ANYTHING to maximize your chance of killing a boss. Sometimes but very rarely this means maximizing your overall dps, which would mean good logs. Most the time what counts is maximizing your survivability, playing mechanics (soaking etc), saving CDs for specific burn phases or add spawns or respecting dps stops all of which means sacrificing your overall dps for the good of your raid. While you might be able to compare players through ranks at the start of the tier the point we are at now certainly does not allow that, since a lot of guilds finished progression long ago and now try to push logs. Not to forget that most guilds with the best players log privately anyway.

While I enjoy pushing myself to the limit and going for parses deep into the tier, during progression I couldn't care less. The only number that counts during progression is the number of bosses killed compared to other guilds. If that means standing arround for a minute doing nothing or sitting in iceblock for 10 seconds soaking something so be it. An example would be a guy celebrating his 99% parse on his varimathras mythic firstkill with 0 damage done to adds while his guild kept wiping to the adds for hours only killing the boss because some guy ranking 10% hard carried them. Which of them would you rather have in your guild.

To illustrate my point an example from my guild. Ever since the start of ToS mythic progression my guild started to log mythic raids privately. Before we had multiple players parsing fairly high during progression, but we often wasted a lot of pulls because of dps whoring. Since ToS progression went very smoothly while firstkill ranks went down significantly. But now you can see hybrids (e.g. ele shamans) offhealing during heal intensive phases, people immediatly jumping in and helping on important jobs if a key player dies, people saving CDs for important burn phases, the list goes on.

TLDR: Ranks mean shit during progression, wiping to the adds on varimathras, heal adds on coven, cc breaks on aggramar, too little dps on the orbs at argus are frustrating things and lead to burnout and toxicity. If you want to be a player your guild likes to do progression with do your job and don't whore for good logs. VERY few bosses die because of reasons which also lead to high parses.

During farm I set a few goals for myself. Mythic I don't care much since clearing Antorus mythic takes a long time anyway so I don't want to risk wipes by whoring for logs. But I try to get 99% on every boss on normal/heroic every tier and go for a few rank ones if I got lucky with loot or feel like I have a good strategy for a certain fight. I know how to play my class since there's enough info out there. But if you want to improve checking wowanalyzer or looking specifically into what other players do is probably more productive.

Since I have no idea about your gear it's hard to say how well you should parse but you should be able to rank all 95+ easily on difficulties below what you're normally doing. So a mythic raider should fairly easily be able to do at least 90% but more like 95-99% on every heroic boss and in normal it should be a cakewalk to go all 99% without significant whoring.
Honestly I think it is a product of the environment that you raid in. Personally my guild is terrible at mechanics. So, if we want to clear content prior to the next tier releasing, we have to be min/maxing to dps through whatever mechanics are possible to. We're striving for 90%+. In a more realistic environment I think your baseline of 75%+ is a great target.
What you just said is a contradiction in and on itself. Did it ever come to mind that maybe your guild sucks at doing mechanics BECAUSE your dps are only striving for high parses? :lol:
Took us for example 20 tries to down coven mythic with the best dps sitting @60% and the average probably being arround 30-40. But boss is dead so who cares.
qivive
Posts: 12
Joined: Thu Feb 01, 2018 1:55 pm

Re: Is there a "reasonable" percentile to aim for?

Unread postby qivive Wed Feb 21, 2018 1:29 pm

So a mythic raider should fairly easily be able to do at least 90% but more like 95-99% on every heroic boss
This is the kind of statement I was expecting. Do you mean a good player would - in any case and with any guild - almost certainy get 90+ on every boss? Or do you expect the environment to be just as good (shorter kill times, no mechanic failures).

Progression is a another story, I am interested in quite settled fights, where people usually know what to do; like 3rd kill onwards.

The actual reason I started this topic was that I was doing rather poorly on High Command and Hasabel constantly and couldn't figure out according to logs where the difference came from. Noticeable differences (at the same ilvl) were: 1) much less int stat (5-12k), 2) less pod damage on adds (high command), 3) extreme differences regarding ignite damage (100-200mil). In the latter case I can't reliable tell if its just the ignite bracer bug that has been fixed lately. I can hardly believe its just that.

So, all in all I was wondering if differences may be due to non personal factors (other raid members, RNG). And you seem to claim, that personal performance would easily get you to 90+ on heroic on farm on any spec. Did I get this right?

Maybe thats right and interpreting the logs isnt easy at all. A lot of noticable differences leave you with the question: Which one is the relevant one? If you want an example check out my logs on high command (Zàya-Malygos EU).

One Addition: While 75+ logs (percentile) are clearly not needed to kill a boss, log perfomance is the best indicator for your personal skills. If the boss dies every week I would at least want to progress personally at being a good mage. This is an incentive to play. But I know that some people tend to use logs to achieve other goals or being blinded (dps vs. mechanics debate; e. g. boss damage/overall damage ratio!)
roldy
Posts: 38
Joined: Tue Jan 09, 2018 8:56 pm

Re: Is there a "reasonable" percentile to aim for?

Unread postby roldy Fri Feb 23, 2018 12:16 am

So, all in all I was wondering if differences may be due to non personal factors (other raid members, RNG). And you seem to claim, that personal performance would easily get you to 90+ on heroic on farm on any spec. Did I get this right?

Maybe thats right and interpreting the logs isnt easy at all. A lot of noticable differences leave you with the question: Which one is the relevant one? If you want an example check out my logs on high command (Zàya-Malygos EU).

One Addition: While 75+ logs (percentile) are clearly not needed to kill a boss, log perfomance is the best indicator for your personal skills. If the boss dies every week I would at least want to progress personally at being a good mage. This is an incentive to play. But I know that some people tend to use logs to achieve other goals or being blinded (dps vs. mechanics debate; e. g. boss damage/overall damage ratio!)
I pretty much go 95+ on heroic farm (as fire) be it with my guild or with pugs without much effort. So no tryharding with RoP, no changing talents from time to time slight legendary changes. I run 3132121 with Ring+Bracers on fights where we BL pull and Bracer+Belt for late BL.

Logs are a combination of playing your rotation, but to be honest that's hard to fuck up nowadays, CD management especially bosses like High Command, Portal Keeper, Eonar. Then gear obviously and luck.

Since you were talking about High Command here's what you want to do. First go super high mastery gear at least something like 10k+ better in the ranges of 13-14k so you get a big ignite rolling with your firestarter for the first adds (don't conserve FB charges they're going to be ready for your late combustion anyway). Use combustion maybe 10 seconds before the second adds spawn to build another big ignite and use the beacon trinket 1sec before they spawn.
Then you relax a bit your next combustion + trinket will be up for the second adds at the second pot. Again use combustion + second bloodlust before they spawn but not as much as the first adds since they won't charge away as much. 5 seconds usually do. Depending on your kill time you have to judge where to use your third combustion. If your raid is strong the boss should die shortly after you finish your third combustion. If your raid is weaker you might get the fourth pot if so delay your third combustion for the adds, which spawn instantly at the fourth pot. Enjoy a 99% log.
qivive
Posts: 12
Joined: Thu Feb 01, 2018 1:55 pm

Re: Is there a "reasonable" percentile to aim for?

Unread postby qivive Sun Feb 25, 2018 8:06 pm

I've looked up Roldy on wclogs and found only one mage so I'm assuming thats you. If it's not you, this may be just as good as an example:

Regarding High Command:

1) On several kills your 80+ percentile logs show less than 4-5k mastery, while having 19k crit on one kill. Going mastery doens't seem to be conditio sine qua non. You may parse well but this isn't the deciding factor.

2) Delaying Combustion for adds: I've found 1-2 kills in the stated logs where combustion wasnt exactly timed (rather likely used on CD) and even though it got 80+. On the best kill however (95 perc.), 2nd combustion was delayed which showed an increase in dps. However, watching several fire mage PoVs on high command: None of them delayed combustion on high command (e. g. dylemma).

All in all I'm still confused what the fuck is wrong with this boss fight, but achieving 90% consistantly seems to be possible indeed.

Return to “General Discussion”

Who is online

Users browsing this forum: No registered users and 6 guests