One of the great things about the proliferation of meters and online parsing tools, like World of Logs and WoW Meter Online, is that they provide a great archive of player performance over time. It’s easy to pull up a parse or two from your own history as see just how your performance has changed, from month to month or from tier to tier. But, as anyone looking back through WoL top parses might notice, it’s difficult to get a handle on the performance of a larger spectrum of players. This was one of the challenges that I faced when trying to evaluate comparative healer performance in Cataclysm.
As has been the case many times before, I was saved from an incredible amount of log-diving by Seriallos of Raidbots.com, whose site aggregates data from WoL’s public logs in order to display performance rankings by spec. This information is collected contemporaneously with the progression of the tier, meaning that you can see the evolution of HPS over time instead of just a picture of the finish line (as is the case if you look at WoL’s Top Healer parses). A couple emails later, and I had the following snapshots of healer performance in Tiers 11, 12 and 13. While my original intent was to include this data in my post Paradigm Shifts of Cataclysm – Healer Edition, given the rarity of similar information (I seriously spent hours trying to find something similar for T8-T10, and came up empty-handed), I decided it was worthy of its own minipost.
Now, there are some things to note about the graphs below:
- They’re not readily available on Raidbots’ site because Seriallos created the custom queries based on the dates of the tier, so that I could try and evaluate the effects of each major patch. (Did I mention how awesome this guy is?!)
- The data for Tier 11 does not go back to Cata launch because WoL didn’t start making the raw data available to Raidbots until around February. This means that we are missing 2 months of data (12/7/11 – 2/8/12) which would have given us some context for 4.0.6’s impacts. Bummer.
- The numbers represented are Median values, not the Mean.
- These graphs are for hard mode encounters; I’m not looking at normal modes. This means that the data set is restricted to a smaller segment of the raiding population.
- Each time segment has both a 25-man (“25H”) and a 10-man (“10H”) graph.
- This data does not exclude any fights or parses on the basis of “Well, that’s a [OP class] fight”. It’s all there (including the healing from those “meter breakers”, Tranquility and Divine Hymn, and paladins’ blue-crystal-supercharged HPS.)
- Most importantly, none of these graphs tell you what you can and cannot do with your class. There is nothing stopping you from proving this data wrong. In fact, I sincerely hope that there are a large number of people out there already doing so (to the sounds of Eye of the Tiger).
The date divisions are:
- Tier 11, Part 1: 2/8/2011 – 4/25/2011
- Tier 11, Part 2: 4/26/2011 – 6/27/2011
- Tier 12: 6/28/2011 – 11/28/2011
- Tier 13: 11/29/2011 – date of this post
Tier 11, Part 1 (4.0.6 – 4.1)
Tier 11, Part 2 (4.1 – 4.2)
Tier 12 (4.2 – 4.3)
Tier 13 (4.3 – present)
Graphs are nice, but color-coded charts are better! Here’s the data presented above, but in a more digestible format for people who are just interested in comparing the numbers.
A Handful of Observations
Ultimately, I think this is one of those posts where the data presented is going to speak for itself, as well as be open to a wide spread of interpretation (because we are definitely not looking at the whole picture here). That being said, there are a couple things that I wanted to mention and/or highlight about the information above:
- It’s unfortunate that we don’t have WoL data from the first couple of months of the expansion, because I think it would give us a good amount of perspective on where classes started and why changes were necessary. The good news is that I did find archived pages for the now-defunct StateofDPS.com which provides some context for the 4.0.6 changes. I can’t speak to the quality of data, but at least it provides some insight into class performance (25H as of 4.0.6, 25N as of 4.0.6)
- As I noted in Paradigm Shifts of Cataclysm – Healer Edition, Tier 11 pre-4.1 was seemingly the closest performance of all healers. It had the lowest range for both 25 and 10 mans. Given the performance of Druids post-4.1 and the perception that they were winning meters easily, I wonder if it was a conscious decision of the devs to start including encounter effects that could not simply be healed through, but rather required some sort of raid damage mitigation.
- Although I don’t really have any viable hypothesis for why the data in Tier 12 25H has such massive variations, I think it’s interesting to note that the bump in the middle of the graph is right at the time that the Firelands nerf hit (on September 20th).
- With specific regards to Holy Priests, performance data would seem to suggest that their tools are more successful in a 25-man environment versus a 10-man environment. This is the reverse of Resto Shaman, who experienced low performance in 25’s, but improved performance in 10s.
- Confirming my theory that Firelands fights were all over the board, both 10-man and 25-mans experienced the largest performance range for HPS, with the difference between the highest and the lowest healer being approximately 7k for each.
- Although Paladins might have had a very strong finish, based on the data above, it was Druids who came out on top for the whole of the expansion. As a result, I expect to read a large number of “So we actually did pretty well this expansion” blog posts in the very near future, along with some recanted testimony about WG/Rejuv nerfs being the death-knell of the class.
- Lastly, what I think this data reinforces, at least for me, is that Cataclysm healer design and balance was based on a triage state (Tier 11), and the father we moved away from that state–due to set bonuses, class tweaks, rising mana pools and regen, etc.–the more separation we saw between the healers. So, if we can see the same triage state endure for more than just the first tier in Mists, I’m hopeful that we’ll see more of that healer performance parity that we all enjoy.