I definitely think this idea is right, like the issue is not lack of talent or willingness. I think the community needs to do cool shit that is worth having an editor or artist for.

For example, we are having the billdozer cup, but besides it just being new and having a good prizepool, this is just another 6s tournament. Nothing really interesting, there is always tons of 6s being played by all the same teams.

I think if we want to incentivize the creation of interesting content, we also have to be responsible for what is actually being played. Maybe we actually do a draft tournament or a bball tournament or something that mixes up the traditional format.

Let me know what y'all think

T. has the most with 939, followed closely by a few others, see here

There are 7745 players who played 5 or more logs (which is what I use in the map) and about 12000 total players in 2021.

In addition, sorry for not finding you guys yet, trying to work on a V2 so that isn't a problem anymore.

Sorry for not responding to everyone who posted IDs, I was getting some much needed sleep and will look you up later this afternoon.

It seems like the common threads of issues are as follows:

1. not interactive/hard to find players
   
2. names are from either from first log of 2021 or from demoticks.tf API which may produce the same result
   
3. the grouping are not as optimal as I think they should be. You get nice trends but in general the strongest gravitational force seems to be that large nodes that have a lot of shared logs attract each other very strongly. This skews the graph and seems to leave smaller nodes in suboptimal positions.
   
4. alfredodanI have reason to believe this data is inaccurate, Cadet has more logs this year than me which is false. Also why is my name on here an alias I had like 6 months ago I believe the only limitation in this regard is that I am missing the last couple of weeks of logs because kastaling hasn't downloaded them yet.

So in summary, I plan to address these issues and a few more in a future version. Probably next weekend.
I'm going to try out smaller and larger datasets as well and see how that affects the outcome as well. (Especially once the season is over I can use an RGL S6 Dataset, eventually could filter by division)
In addition going to try out some different software and see if I can find some more effective layout algorithms

Update: It is lagging the server, getting a better one brb

Here is an interactive version
Enjoy

It depends sort of. The method serves to illustrate the larger trends, as its hard to exactly make everyone perfectly next to every person they share logs with. The generation is randomly seeded so you can get shuffled next to lots of random people but on the whole the grouping will always trend the same way

your name is Wild_Rumpus

https://fygisawesome.com/logMap/Wild_Rumpus.png

your name is Yogurt!

https://fygisawesome.com/logMap/spacemanworks.png

your name is amadeus

https://fygisawesome.com/logMap/Zolak.png

Try this, I haven't opened it in a real PDF viewer so not sure if it works.

Bingo on the first. The closer you are to someone the more logs you share with that person. Simple as that

Your name is doctrine

https://fygisawesome.com/logMap/jetz.PNG

your name is chester

https://fygisawesome.com/logMap/bamb00.png

your name is Fw

https://fygisawesome.com/logMap/Fw.png

FULL VERSION DOWNLOAD

http://kyleflett.com/2021preview.png

Key: (I am not that familiar with the other scenes so forgive me if I don't know what div is right)
Blue: NA Advanced-Invite
Orange: NA IM-Advanced
Green: NA NC-IM
Black: NA TF2Center Players
Pink: EU Div2(?)-Prem
Purple: Rest of EU
Cyan/Light Blue: SA(?)
Teal/Blue-Green: Asia(?)
Salmon: Australia and Pacific Islands(?)
(These were completely generated automatically)

A few weeks ago I saw this map of twitch streamers and thought about doing something similar in TF2 with logs.
Finally got around to it yesterday. The size of your circle is your total (2021) number of logs, the closer you are to people, the more logs you share with them.

Preview/smaller version

Big Credits to Kastaling, enthrow, 24, Bv, Flac and others for all the help with this project. Could not have been done without any of them.
How I got the data:
I ran through all of Kastaling's logs with a python script to collect the count of logs, along with the amount of logs that each player shares.
Then I got a bunch of names from a bunch of places and added them into the data
Finally I used Gephi to make the visualization.
Project GitHub

Feel free to mess around with my code and try stuff out. All software was free.
hopefully you guys like it :)

PS if you can't find yourself leave your steamID3 ([U:1:11231231]) in the comments and I will locate you

Hey quick bump,
Look through this csv and feel free to message me if you want me to change your name.
The csv

I'm currently making a data visualization of all the players in the TF2 community based on logs. I am finished with the actual data collection and visualization, so all I need now is the right names for the labels. The problem is that all I have is steamID3 and I haven't really been able to find a good solution. My dataset is about 6000 so it would be pretty painful to do by hand.
If anyone has any ideas or suggests, I am all ears.
My discord is arcadia#0001

Edit:
Look through this csv and let me know if you want me to change your name on the final graph
here

This line kills me. As if everyone here speaks fluent assembly.

On the real, as a software guy, this thread is really cool. 10/10

You had to be single digit age when you made this post wtf