Here's a teaser for a project I'm working on.. A statistical sampling of people's Warcraft achievements based on crawling Armory data for achievements. For example, how did people fare with completing 3.0's premier raiding content, Naxxramas/25? About 40% of level 80s have finished Naxx/25. It takes them on average 59 days after they hit level 80, but the distribution is interestingly bimodal.
And another view of PvE content consumption; what percentage of level 80s have completed each of the new 3.0 heroic 5 mans. Yup, turns out no one likes Oculus.
83% The Violet Hold
81% Utgarde Keep
80% The Nexus
73% Utgarde Pinnacle
73% The Culling of Stratholme
70% Gundrak
68% Drak'Tharon Keep
67% Azjol-Nerub
64% Halls of Lightning
58% Halls of Stone
51% Ahn'kahet: The Old Kingdom
42% The Oculus
I have a good dataset collected, but there's a lot of analysis work to do. I also need to decide where and how to publish the results. Also interested if anyone's seen work like this. Specifically: statistical samplings of achievement data. There's other statistical samples of Armory data like
Talentchic or
Armory Data Mining. There's also achievement analysis sites out there like
Guild Ox and
Wow-Achievements, but their self-selecting datasets don't allow for unbiased statistics. I'm trying to do a bit of both together.
It would be interesting to see Naxx/25 vs Naxx/10 numbers. Until last night, I had never stepped foot in Naxx/25, but I have cleared Naxx/10 several times.
Regarding hOculus, I think the fear is a bit overrated. I've done it with a pug without much issue. Now finding a group for hOK is just about impossible from what I can tell. I still haven't done hOK.
Posted by: Felkan | 2009.05.14 at 11:29
If only we had Armory data from BC as well. I would like to see the average time to complete heroic 5-mans between BC and WotLK, and Kara vs. Naxx10.
Posted by: anon | 2009.05.14 at 20:45
One more thing, I was looking at the Armory Data Mining Blog and noticed (based on gear per slot) that Naxx/25 appeared to have more attendance than Naxx/10. That, too, shocked me. I would have thought just the opposite.
Posted by: Felkan | 2009.05.15 at 06:00
Felkan, I did a quick one-off report on Naxx/10 for you. 50% of level 80s have completed Naxx/10 (vs. 40% for Naxx/25). Average days to completing Naxx/10 is 46 (vs 59 for Naxx/25). I don't have a quick way to look at who does Naxx/10 vs. Naxx/25, that'll require some more work but I'm interested in that kind of thing.
Posted by: Nelson | 2009.05.15 at 08:29
Thanks.
If more folks have done Naxx/10 than Naxx/25, how is it that the Armory Data Mining Blog's gear list, shows more Naxx/25 equipment? Weird. I guess I was only looking at Druid numbers, so maybe Druids run Naxx/25 more than Naxx/10 whereas a few other classes are the opposite??
Interesting stuff, really.
Posted by: Felkan | 2009.05.19 at 10:45
I haven't tried to compare to the gear list. But two quick guesses.. Naxx/25 drops more loot per boss than Naxx/10, although then again more people are rolling on it. Or else Naxx/25 gear is upgrades for Naxx/10, so if you're just looking at what's equipped you're going to have a bias towards better gear.
Posted by: Nelson | 2009.05.19 at 11:01
+1 on Flyv's comment. We are all loot whores at heart, and perhaps the loot in Occulus isn't as great as that in Naxx. Other possible theories are that the badges aren't worth it in Occulus due to people's aversion to the uniqueness of the fights in there, and that Naxx runs are often mandatory guild runs, so people have to go.
Posted by: Tyger | 2009.05.21 at 12:57
Going back to an old post here - would you have any interest in running this query for yogg-saron? i'm curious how many people have killed him at at this point.
Posted by: Alisha | 2009.06.29 at 11:01
Good question, Alisha. Your answer is 3%. Much more detail here.
Posted by: Nelson | 2009.06.30 at 08:28