Warcraft logs analysis for high end guilds

[Knoxxic]

Hi everyone,

I've raided at a pretty high level in the past and achieved a few world firsts as well as many region firsts. I always knew how to play but I was never at a level to really analyse the way others played. This tier I decided to form my own guild with a bunch of my friends in the hopes of achieving a respectable ranking in the community as well as creating a fun raiding environment that would allow me to play with my good friends as well as achieve the goals I had set myself for this expansion.

One thing that I had not really thought about is log analysis and how to help my friends and raiders improve as players in order to help us progress towards our goal. We've come up with a system which rates 4 different performance variables out of 10 for each player in each fight. I've given them a rating based in their percentile, their personal cd useage, their unnecessary damage taken and finally any deaths that they had. I've also included an extra notes area at the end where I can deduct or add points based on other factors that are hard to find in logs such as initiative in raid or exceptional circumstances. Players who die early will only be analysed on their death and personal cooldown useage and have the final score divided by 2 instead of 4.

For example:
Steve the BM hunter completes the Ursoc without dying, he finishes with a dps percentile of 95% and has used his personals and healthstone/potion correctly. However he did not get hit by roaring cacophony at all which is why his damage taken is so low.

In the above example I would score Steve on:
Deaths: 10
Personal Useage: 10
DPS percentile: 10
Unecessary damage: 10
Extra notes: did not get hit by roaring cacophony causing the raid to take more damage than necessary even when asked to, possible causing Bob damonhunter to die. -2
FINAL TALLY: =((10+10+10+10)/4)-2
= 8/10

I'm wondering if anyone has criticisms about this grading system or suggestions for more variables or a better way to monitor raider performance in raid? It does not take me long to look at the logs for the current variables but I feel like I might be missing something.

Let me know below! I appreciate any feedback!

[Larynx]

If you rank people on a scale of 1-10 using completely qualitative measurements, your scoring is going to look and going to be incredibly arbitrary. What's the difference between a 6 and 7 for personal usage? Why is Steve scored 10 rather than 9 for DPS?

[Knoxxic]

If you rank people on a scale of 1-10 using completely qualitative measurements, your scoring is going to look and going to be incredibly arbitrary. What's the difference between a 6 and 7 for personal usage? Why is Steve scored 10 rather than 9 for DPS?

For percentiles we've ranked it pretty well I feel. 85+ is 10, 50 is 5 and anything below that drops off pretty quickly. It is just a way to measure peoples performance over weeks without just looking at dps numbers, it doesn't have to be 100% accurate just a rough measurement of player skill and improvement.

I suppose a marking key would help a lot just like they do for test papers etc

[klaps_05]

you are trying too hard. there is no point in this evaluation, because until the actual pull which turns into a kill its always messy. people WILL take unnecessary dmg, people will try to juke some mechanics for the sake of dps. high-end raiding is about peoples agility to compensate minor mistakes(major ones wipe u either way or make u lose ur raid spot) of everyone else. also if you take progression seriously(e.g. first page wowprogress), steve will be out after 2 pulls without soaking X spell.

for DPS the only things that matter are attendance and DPS. thats why u usually have 1 raid leader, 1 healing officer and 2 dps officers(casters/rest). On progression DPS rankings are irrelevant since they can be inflated, on most bosses dmg done(e.g. important adds, ilgnoth brain, canopys, add prios) are much more important than actual DPS ranking. Then theres is debuff uptimes, interrupts, raid debuffs management. Want to go in detail? Analyze damage taken per player, set thresholds for each spell which is avoidable/partially avoidable, set benchmark based on top guilds logs what it should be, adjust by factor based on your raids mobility(if u got a bunch of mages, dhs, monks they should take X% less dmg based on their mobility CDs and how they align with raidwide avoidable dmg); players dying early, how early? phase 1? when in phase1? before big aoe? before X add spawns? dying to what? unavoidable dmg or aggro? u dont have any metrics for that and if my raid leader starts assigning random shit like this ill be out before the pot runs out.

bottomline: Attendance and general benchmark dps based on average ilevel of raid is all you need, not some arbitrary values which will drive people away even if they are your friends.

[Knoxxic]

Some harsh words but some pretty helpful advice, I appreciate it and was worried that I was going way in to way more detail then I had to. You have to remember this is a fresh guild, it takes time to get players who are decent or for them to improve. I was just trying to figure out the best way to help those players improve to meet the guilds standards as we do not have much choice who we bring at the moment. It's mostly for myself to be able to actually view who is improving as a player week to week rather than just looking at the dps rankings but I suppose it only takes common sense to see that. Ty Klaps

[Cracked]

I get your intentions, but this will not create a fun raiding environment. As a consequence, the guild will go to shit very quickly, with people emo quitting when they get bad ranking and feel unappreciated, 'elitism' among those who consistently perform better (unless you are ready to kick great players just because they are toxic) and unnecessary pressure on every single raid.

Cultivating people takes time. Your best bet is convincing them to go through their own logs from raid, only when you see gross misconduct you should feel the need to point it out to them, but never in front of the others.

These things may seem stupid to you, but this is true for all people, be it world 1, or world 500. The difference is that top people know when they are making mistakes and build upon them.

[mistahwilshire]

I don't think this will work out well for you. You're trying way too hard. Great players understand their own logs.

[TEHPALLYTANK]

If you can't tell that someone is fucking up mechanics during the run, you really need to work on your raid awareness. Keep your raid frames up, regardless of what role you're playing. You need to be able to see when someone is out of range, out of position, and whether or not they've fucked up the mechanics by looking at what debuffs they have or are missing. You should know who caused a wipe, and they should know what they did wrong. If you have people making mistakes and not owning up to them, or even worse, being unaware that they made a mistake, you're in for a shitfest.

[Knoxxic]

I think you all know what I mean when I ask about advanced log analysis and what I should be looking for in the logs. I feel like Klaps answered me pretty well, it's not that I am incompetent. Clearly I can tell when someone wipes the raid or is in the wrong place. The things I want to know are if people are taking too much damage from certain mechanics, whether their rotations are optimal and whether they are doing their job correctly which is exactly what Klap has said. I'm taking this far too deep and it isn't really achieving much, I appreciate the input everyone and think I have a better understanding of how to evaluate my raiders

[reauxmont]

Extra notes: did not get hit by roaring cacophony causing the raid to take more damage than necessary even when asked to, possible causing Bob damonhunter to die. -2

Are you sure that it was Bob Damonhunter who died, or was it maybe... MATT DAMONHUNTER?