Monday, 23 February 2015

In which the Underlurker lurks no more and hardmode raid statistics are analyzed

I have been slacking with writing about our raid progression. Somehow I ended up writing about other things instead (the Blood Hunt FP post took a lot of my time in particular) and now I'm already two boss kills behind on schedule. I'll make up for it, though, by posting pretty videos of our takedowns.


February the 8th we took down the Underlurker on our tenth try. We were really surprised with how easy he was compared to our fears. The Underlurker is barely any harder on hardmode than on storymode, and I think he is the easiest boss after the first bosses of both endgame raids. That still means he can be quite annoying, especially when 1) the cross bugs out; 2) the tank gets one-shot by receiving three massive hits within one second (another bug) or 3) the rocks spawn very far away.

Above a video of us taking down the Underlurker on hardmode last week.

Difference with storymode:
  • the adds perma-slow you; they have to die before Ragestorm starts, otherwise people can't make it to the rocks in time;
  • Ragestorm hits so hard that you cannot risk getting hit. You need to hide behind a rock before the storm starts;
  • tanks need to alternate who stands in front each cross.

If you're seeing the Underlurker from this perspective, your health has probably seen better times.

Analyzing the learning curve

While we were trying to get the Underlurker down I was chatting to Marinka on Steam and she noticed how it took me different time intervals to respond depending on how long it took us to wipe. We started talking about statistics and what one could tell from this about performance and learning curves. I handed over the stats to her so she could go all nerdy with her favourite statistics program.
Special thanks to Marinka for the pretty graph. We killed the Underlurker on try nr. 10.

This is the graph from the first evening on the Underlurker on HM. We see an interesting pattern. Apparently after each good try (on which we survive long) we have some derp moments. This could say something about people's attention spans: one out of three or four tries we can bring up good concentration. Of course, I'd need more data to make this claim for sure.

It gets more interesting when linking the moments we wiped to reoccurring game mechanics as seen and timed in the video. Looking at that, I come to the following wipe moments:
  • 1st Ragestorm: 1, 2, 7 and 9
  • 1st Cross: 5
  • 2nd Ragestorm: 4 and 8
  • Enrage: 6

I'm not sure why we failed on try 3; that could be anything. On encounter 6 we wiped to the enrage with the Underlurker on 1% (grrr).

Poor lurkerlings. They look so cute and helpless jumping up and down, but they're a healer's nightmare.

Ragestorm is the clear winner in causing wipes. This is logical, as it's a clear DPS check: the adds need to go down before the Ragestorm starts, or we're not able to hide from the storm in time. It might also show that we were in the process of learning to run behind the rocks in time, as that needs to happen a lot faster than in storymode.

The story continues...

The next raiding post will be about how we took down the Revanite Commanders on HM (4th boss fight of the Temple of Sacrifice raid). You can take a sneak preview here on my brand new YouTube channel (don't forget to subscribe).


  1. That's an interesting graph, I'd definitely like to see more statistics like that! (There used to be a blog that was all about SWTOR statistics, but I stopped following it when the author switched to GW2 and then went completely quiet.) Purely anecdotal, I don't think that our learning curves are very smooth either. Sometimes it drives our ops leader nuts when we follow up a good attempt with one where we die due to derp one minute in... it's nice to see that this might not be as odd as it feels sometimes.

    1. I think it's a very human thing - proves we are all humans. Of course I can't say for sure, but I wouldn't be surprised if many groups would show these fluctuations, regardless of if they are, for instance, very casual or super hardcore.

      Cool that you like the statistics. I'm curious if other boss fights (especially the ones we're still working on taking down) show the same patterns. I've collected some data on our Master & Blaster HM attempts, so hopefully Marinka will help me out again!


You can insert links, images and videos to your comment using these tricks.