Cataclysm to Take iLvl Out of the Mix for Deciding What Raid Size to Seat
I Think I Smell a Giant Rat

Creating Benchmarks for Raid Performance

How can you assess a raid's performance if you haven't identified and agreed upon the criteria for success? You can't, beyond "Did we kill the bosses?" And that's simply not enough when dealing with progression content you're learning.

In my non-WoW life, once per month, I analyze a number of metrics to determine how successful a variety of communication campaigns were over the course of the month. Then I report out to my team to let them know what worked and what didn't, and we set about revising tactics as necessary. We're able to engage in this constructive dialogue thanks to having clearly defined benchmarks and performance expectations identified that we all agree upon.

Frequently, the same can not be said for raider performance evaluations. If you've ever tried to nail down a set of criteria in a group setting, you may have found it impossible between healers saying their effectiveness could not be measured by data, DPS complaining about certain fights leaving them at a disadvantage, and the Off Tanks arguing that their role gimped their TPS. No one wants to be measured! Because we are all uniquely awesome snowflakes who want to measure and discuss how everyone else is doing.

But as long as you are logging your raids (simply type /combatlog at the start of your raid and /combatlog after to have a log generated and saved into your World of Warcraft Logs folder), and uploading their results into a reporting tool like World of Logs or WoW Meter Online, you can set and measure against benchmarks.


Although it has largely fallen out of favor due to it not being actively updated and maintained, I still upload a log each night to WWS because it has a lot it can tell you at-a-glance about your DPSers and help focus your attention for perusing the other logs.


Your key takeaways from the main screen: 

  • Your top overall damage dealers/DPS
  • The percent of active time (lower numbers mean the person was dead/AFK/not DPSing)
  • How close each DPS was to the other in terms of their output percentages -- you want to see folks within a couple of percent of each other.
  • You don't want to see your top DPS doing 14% of the damage and the rest doing 7%.
  • And you would expect similarly geared players of the same spec to be close to each other in output. If they are not, those are the folks for you to deep dive into with the other logs.

WoW Meter Online (WMO)

Although some folks are put off by its gratuitous use of Engrish, I am still a fan of WoW Meter Online for its ease of use and a few features that are exclusive to it as of this writing. My first stop in a WMO log is the All Bosses link.

From here you can check each players' presence (amount of time active in the raid), their DPS and overall damage, and their number of interrupts. Next, I pick whichever player I want to drill down on and click their name. This takes me to a control panel where I can see their damage and healing done and taken, plus all the debuffs and buffs they applied over the course of the evening. I typically drill down on that last tab. It allows you to see who was buffing the raid, how many flasks the person used, if they were in someway incapacitated during the raid (mind control, impale) which might account for a performance decrease on a fight, and if they were applying the requested slows/stuns/debuffs on any given boss fight.

After checking out that tab, I head back to the main window, and pick a fight I want to analyze. From there, I click on an individual player's name. And this is where WMO excels at helping you determine whether or not the performance you are looking at is good, in line with the player's previous performance, and how it compares to others. Up at the top next to the person's name, is a link to Historical Report. Clicking on that link brings you to a summary page that graphs that individual player's performance against that boss over time. From here you can see:

  • How many times has the player fought this boss in 10/25
  • Their DPS/HPS/TPS charted over time in the 10 and 25-man settings.
  • What you want to see here is a line moving upwards to the right. That shows consistent improvement over time.
  • If you see a more erratic performance output graph, you need to determine if it is due to your raid being inconsistent with the available buffs from week-to-week, or if the player is not always as focused and prepared for some raid nights.

Going back to the previous window with one character's performance showing, you can access the other unique feature -- Compare with Other. Clicking that link will bring up a page showing other performance logs for players of the same class and spec on that same fight. Pick a person to compare to. You'll generate a page that will allow you to see, side-by-side, how many casts of which spells that player performed in that boss fight, which can help you give players feedback about how their rotation might be improved. If you click on one of the actors names, you get brought to their homepage which includes a link out to their armory so you can also compare gear/spec to ensure you were comparing apples to apples.


I personally consider these two WMO features to be some of your best tools for setting individual class/player benchmarks and coaching your raid members on their performance.

World of Logs (WoL)

World of Logs has been the favored combat log parser for the past year or so, and for good reason. It provides a lot of easy to digest information for raid leaders to digest.


My first checkpoint on WoL is the survivability report. Dead raiders can't do their jobs. And although it is understandable if someone dies early from time-to-time, if you have a player sitting at 50% for the night, while the rest of your raid is in the 80s and 90s, that says someone needs to work on their moving-out-of-crap-that-can-kill-me skills.

Picking a DPS and drilling down, I like to take a look at the Damage By Actor report tab. This is also known as the "proof you switched to the adds" check. If someone has no damage done to the Bone Spikes or the Blood Beasts or the Web Wraps, they need to receive the chat about meter humping not assisting the raid in being successful and that when the Raid Leader says Everyone switches to the adds, this means them as well. Like WMO, buffs gained and cast are shown as sub reports here as well so you can determine if a player has been coming to the raid prepared and staying focused on what they can do to maximize their performance.

Now heading back to the main report page and choosing Healing Done as my report from the pulldown, I pick out a healer to drill down on. Note that the main healing page labels the HPS as DPS. You can again see how much of the player's healing done came from which spell as well as whom they were healing, which can help check if they were on their assigned targets.


The final piece of benchmarking data you can glean from WoL is the data from their Ranking Info report. Based upon their compiling the top performers by DPS and healing on each fight uploaded to their website, this report shows you how close each of your players were to the max recorded #s for their spec on the fight, as expressed by a percentage. Depending upon your level of experience with the bosses, you may want to set your bar higher or lower. But again, overall, you'd want to see your players performing at similar percentages on the bosses.

If you are seeing a significant percentage gap amongst your team, especially between 2 players of the same spec, you should take that as a coaching opportunity and suggest that the player with the performance gap might want to receive some coaching from the higher-performing player, or at a minimum, go and look at their armory and at the logs of the attempts to see what they were doing differently.

The nice thing about this data is that it is, by its aggregated nature, neutral. If you have a player who maintains a specific fight is just impossible for their spec, it gives you a benchmark against which they can compare themselves to see if that's actually the case or if they have some room for improvement.

Getting Started

The absolute first step for evaluating your raid's performance and getting on the road to performance benchmarks is to make sure your raid team runs and uploads combat logs for every raid. Try not to get overwhelmed with the amount of data available to you, and start small, such as by setting overall HPS/DPS or overall healing and damage benchmarks for specific fights. This WMO Raid History report for a boss is a great starting place for setting your benchmark for what it takes to be successful for a boss.


Avoid temptation to over-analyze your raid data, and drill down to individual-level benchmarking data only when you need to solve a problem you are having with a specific fight, or with a specific under-performing player. And encourage personal accountability -- challenge your raiders to keep an eye on how they are ding, and to strive to exceed their previous performance, and to experiment with small tweaks that can give themselves a boost.