Wormhole MER

Have you ever wondered why wormholes are not part of the MER? I certainly have, but I don't have an answer to this question either. What I however do have is a (hopefully) worthy replacement.

Have you ever wondered why wormholes are not part of the MER? I certainly have, but I don’t have an answer to this question either. What I however do have is a (hopefully) worthy replacement.

Data Source

CCP always gives us the raw data when they Publish a MER in a link of the MER Post (For example in the February-2021 MER this link is https://web.ccpgamescdn.com/aws/community/EVEOnline_MER_Feb2021.zip ). CCP likes to not call these exactly the same so Phi Feynman made a Github repo with all of them: https://github.com/Phi-Feynman/Killdump. The only thing where we can actually get Wormhole Data is the Killdump, sadly the other parts are K-Space exclusive. So we totally don’t know how much was mined or produced in Wormholes, we only really know what was destroyed.

The data is also not quite complete as CCP seems to take the last day of the month to assemble the statistics and as such leaves the last day out of the Statistics (Here are the exact Holes in the past 2 years, I didn’t bother after that). If you look at how many KillmailiIDs have been used up during the time period for the data Export, then you find out that about 90% off all Killmails are in this data. 3% can be attributed to the missing last day of the month, and 7% seem to be missing for other reasons (I have tried to find out what exactly is missing by cross-referencing with zkillboard, and it seems to be mostly kills in the abyss, but I couldn’t pinpoint the thing exactly). But still this is pretty good in comparison to zkillboard which gets about 70% of all kills (The exact value can be found here).

Isk Values

The Values for ISK Destroyed is CCP’s own measure, which isn’t always that accurate something that you can definitely relate to if you use zkillboard. An (extreme) example of the Inaccuracy is an Inner Hell Keepstar valued at 708b on zkillboard. Ingame it got valued at only 2.71b, and this is the that also ends up in our data (Rumors are that the prices relate to the regional market price, and nobody sells Keepstars in the F-R00030 region)

Phi Feynman again spent a lot of time trying to crossreference kills with zkillboard to make this more accurate – but as the data is anonymized it is pretty hard to track back all the single kills. Using all the available fields, there is still about 500’000 kills over the last 4 years that are not uniquely identifiable by zkillboard. I would like to thank Phi very much for trying all of this and paving the path! Going through those manually would take way too long, so I finally decided to just keep the default ISK values, and do statistics that way. While not as accurate, it has the same accuracy as CCPs MER, and as such comparisons will also be easy.

Plotting the Data

I used an ELK stack to ingest all the data into a single database, and then create the graphs from there. This isn’t necessarily the best solution, but it is the one I have experience with, and I can graph stuff fairly quickly. I imported all the data using logstash with a simple config. As this is fairly computing power-intensive I can’t just make this available to everyone to play with – I just don’t have the necessary hardware. So instead I decided to make some general overviews on this post that will be fairly condensed and colorful, and then have less dense and more specific graphs in separate pages for every month. That way you can get a quick overview here, and then compare it more specifically for each month. The single-month graphs should also be a single color so that people with colorblindness can see them without issues.

Overview Graphs

General

ISK lost over Time

ISK lost per Wormhole Class

Kills per Wormhole Class

Ship Meta

Logi Losses over Time

Select Ewar losses over Time

Heavy Shield DPS losses over Time

Heavy Armor DPS losses over Time

Corporations

All of these graphs are total chaos, maybe have a look at the individual months instead.

Top 20 Corporations by Kills

Top 20 Corporations by ISK Killed

Top 20 Corporations by Explorers Killed

Montly Statistics

2016

June July August September October November December

2017

January February March April May June July August September October November December

2018

January February March April May June July August September October November December

2019

January February March April May June July August September October November December

2020

January February March April May June July August September October November December

2021

January February March April May

Additional Graphs

Over the course of this project I have had quite a few other Ideas on what graphs could be interesting. These aren’t always perfectly formatted, but maybe you will find one or two in there that are interesting anyway.

Context

It has been pointed out to me that just having numbers without context is not worth much, to which I disagree: If the numbers do something interesting you can still always search for the context for it, and numbers are – if treated correctly – pretty unbiased. I did never, and will never remove certain kills from statistics as I do not think that I understand all the phenomenons that occur in eve and lead to killmails, and how they relate to the game changing over time.

With that said however I would very much like to provide context in text form for each month. But I don’t really play eve long enough to provide said context. If you want to explain why certain individual graphs look interesting, then please reach out to me.

Conclusion

I hope you learned something while looking at all that Data. For me the change in Ewar Meta andthe adoption of Triglavians were definitely very cool to look at. Also just how many final blows the Drifter (Corp Arythmos Tyrannos) gets is very interesting. If you want to discuss these graphs, or have any suggestions/requests for future ones please head over to my (tiny) discord. I am already on to the next graphing adventure, this time with the data of zkillboard. And just maybe I will get to required hardware to make a tool available to the public in the future, but there is a lot of lines to code that need to be written until then. Sneak Peek

Leave a Reply