might be interesting, but if you get to the level where they are much better than humans it's probably not going to be useful for normal play. Ex: Computers can get 100% acc meatshots which is an unrealistic assumption for human players.

The Republican party has free reign to pass almost anything now. Can someone more informed speculate more on the connection between Trump and Project 2025? I was under the impression that he has distanced himself from the group because 501c3 non-profits can't endorse candidates, but there is a lot of inbreeding going on between them. Ultimately I am trying to figure out if Project 2025 is a better indicator of future policy goals or not.

Would be interesting to see some subsets such as looking within each region (so not all froyotech lol), or looking at different eras to see when scout got a slight edge. I also have a feeling that openskill/elo might not be ideal for analyzing these types of games. At a high level this is because they don't really have a method of capuring the "team" element of people playing together and building chemistry (there's an argument that maybe it does this implicitly). They are more suited for pugs and similar.

Thank you for doing this though. If anyone is interested in this topic there are some previous threads and attempts here:
https://tf2metrics.wordpress.com/about/
https://www.teamfortress.tv/22042/tf2-player-rankings-official-thread (you need to use archive.is for the links)

The conclusion does not follow from the premise, if "aggression" is what I think it is.

To Dolphin:

If your current directions in Elo are failing, here is a free open source library you can try to use- https://openskill.me/en/stable/manual.html#user-manual

You can read about the results on a sample overwatch dataset here: https://arxiv.org/abs/2401.05451

This is essentially one of several ports of this paper: https://jmlr.csail.mit.edu/papers/volume12/weng11a/weng11a.pdf
It's in python but there are other implementations in other languages.

The Elo/glicko system is a rating system, which is different from the matchmaking system (that may use Elo). I'm also pretty sure the number of played games per player is a better indicator than how many people add up. There are other challenges with small number of people adding up but not having to do with Elo.

In an ideal world RGL invests into making prec-64bit because it is good for the community and for their league.

So happy for bobrovsky. let's go cats!

Maybe you missed my point. Here is a list of things valve has done for tf2: https://twitter.com/mastercoms/status/1797908595570577444

If you look closely, there was a DDOS incident they fixed, which is more comparable to the issue you described. This issue is completely different from the problem of aimbots/cheating which the whole industry has not solved yet. Valve is a pioneer in the gaming industry and so maybe it's reasonable to ask more of them, but to say they haven't tried is wrong and putting your head in the sand.

It's well known that Valve makes most of its money from the Steam Store, and is also a private comany. This insulates them from a lot of motivation to do anything the public wants. Comparison example: You can be sexist in any CS lobby, but be banned in any Riot game.

Some other strategies might be needed. I watched Uncle Dane's FixTF2 video, which has an appeal to ego (asking Robin Walker if botopia is the legacy he wants to leave), which I think is a nice approach that might not have been explored yet.

Edit: Additionally, I will say that also in the video, it is said that cheating is an industry wide problem. The task of solving the bot crisis is not trivial at all, and it's not completely true to say that Valve has done nothing to try. I think that the sentiment that Valve "just needs to do X" is pretty optimistic. When we ask for the bot crisis to be solved, we are asking for a lot.

Is this even true lol. For the conferences you mentioned almost all papers are primarily math.

Hi all. Since I graduated from college the files in my drive will be nuked soon, so I decided to upload some scuffed interviews I helped make happen at the Localhost Philly Lan in 2023. Feedback is appreciated for whoever wants to do these in the future! (possibly me)

G6 caps
G6 LOLGUY
FROYO b4nny
FROYO meyy
GC Shamoo
GC pauldogg
FORD sandblast
FORD ronz

Thanks to Resup for letting me do these interviews.
Also, thanks to marmaloo and zilly for doing these interviews, as well as all the interviewees that volunteered!

P.S. can someone recommend a free filehosting service I can move the unedited Lan photos to? (not imgur im not allowed to publicize them)

These monitor cams belong to a different production company that's there so that's why previous lans there didn't use it.

Is this sufficient enough for tftv to start banning witch-hunt threads?

edit: Looks like it was always banned. Just not enforced anymore and looks like no one cares.

Anti-cheat evidence is kept confidential, and this is intentional. There's literally nothing you can do to reverse your ban unless you buy a yacht for an AC admin or something.

What if you just created another account though? They can't ban you for alting cause that would imply the previously banned account was actually you. The only possible ban they could do is claim you are still an alt of your roommate. But this time, why don't you collect more evidence like live gameplay with cam? Or ask RGL what would be enough (if they care to even answer)? This may seem over the top but honestly, you don't have many options at all.