u saw nothing ;)

FWIW all "unknown" mediguns are ubers, it was a problem with certain medigun skins

edit: now fixed but that doesn't help past logs

TFNew are also running a team drive - for more info see their thread

the api doesn't give you times for anything, it's the pre-parsed data that you see on screen

to get ticks you need the raw file with events, which is what you get when you DL it

After the group stage we will be doing another article so if we missed any team that's doing well they'll be covered there

this will come when we roll out our new and improved stats system, which should have happened last month

logs are irrelevant and not worth looking at, they're not a good measurement of your performance ingame and not something you can use to help improve - it'd help more to watch your demos in scrims/officials

I don't suppose that's got anything similar to ggplot's facet_wrap? Might be an idea to help present stuff like sensitivity by class

I don't suppose it's possible to get a breakdown of this by class? That distribution looks about right, but it'd be interesting to see how it varied for demos compared to scouts, say

https://drive.google.com/file/d/1Is4YlOG_-0i5q8dv4Vb8wLaJp_NtRBeE/view?usp=sharing

our medic gets their medic

A friend introduced me to TF2 and I spent a painful amount of time on cp_orange, then one day I stumbled across a cast of i49-era Epsilon rolling some team on Process and never looked back

It's just gone public, with how many new players there are matches cannot be balanced, simply because the system doesn't yet know how good they are. trueskill.org estimates it'd take 100+ matches to have a good idea of someone's rating for 12v12, and while you could tweak that a bit (if 12 pubbers lose to 11 pubbers and Karppi, he was clearly more impactful) it still takes a long time to accurately place people.

After its success last season we're going live - post your predictions for the final order of prem, and we'll see who gets nearest and wins that all-important clout. Can anyone dethrone Aelkyr? Find out in seven weeks!

So I had a look at your source code (with the caveat that I only just understand python) and you're averaging DPM numbers from logs, which gets you the wrong result - you need to sum damage and time played, then divide the one by the other. The same goes with averaging K/D rather than dividing avg kills by avg deaths.
It might also be worth averaging kills by time instead of by game, since if someone always plays longer games they'd naturally have more kills per game than someone who doesn't, even if they both had the same kills per 30 mins

9PM eastern? (so 3AM CEST)