Almost exclusively about European TF2, like [url=http://www.vanillatf2.org/2014/03/tf2-by-the-numbers-season-17-playoffs-round-1/]this one[/url] which is about the upcoming first round ETF2L 6v6 playoff between all-French Awsomniacs and all-Swedish Dotter.
[url=http://www.vanillatf2.org/2014/04/tf2-by-the-numbers-season-17-playoffs-round-2-aws-vs-tlr/]ETF2L Season 17 playoff round 2 preview[/url]
[url=http://www.vanillatf2.org/2014/04/season-17-grand-final-preview-tf2-by-the-numbers/]ETF2L Season 17 Grand Final preview[/url]
[url=http://www.vanillatf2.org/2014/04/tf2-by-the-numbers-etf2l-season-17-review/]ETF2l Season 17 6v6 Season Review[/url]
[url=http://www.vanillatf2.org/2014/06/tf2-by-the-numbers-etf2l-highlander-season-6-review/]ETF2L Highlander Season 6 Review[/url]
[url=http://www.vanillatf2.org/2014/08/tf2-by-the-numbers-season-18-review/]ETF2L Season 18 Review[/url]
[url=http://www.vanillatf2.org/2014/08/zen-and-the-art-of-balancing-a-lobby/]Lobby and Pickup game balancing[/url]
[url=http://www.vanillatf2.org/2014/09/tf2-by-the-numbers-i52/]i52 Review[/url]
ETF2L Season 19 Weekly reports
[url=http://etf2l.org/2014/10/04/season-19-powered-by-tt-esports-week-1-report-card/]Week 1[/url]
[url=http://etf2l.org/2014/10/11/season-19-powered-by-tt-esports-week-2-report-card/]Week 2[/url]
[url=http://etf2l.org/2014/10/18/season-19-powered-by-tt-esports-week-3-report-card/]Week 3[/url]
this is an interesting way of analyzing match data lol
i don't think i've ever seen a scatterplot actually used outside of classes
this is an interesting way of analyzing match data lol
i don't think i've ever seen a scatterplot actually used outside of classes
THE NUMBERS, MASON! WHAT DO THEY MEAN?
Good read as usual, I'm in love with bar graphs.
THE NUMBERS, MASON! WHAT DO THEY MEAN?
Good read as usual, I'm in love with bar graphs.
Is there a chance that you will ever do this for NA as well?
Is there a chance that you will ever do this for NA as well?
Incredible effort put in, superb articles throughout Jon!
Incredible effort put in, superb articles throughout Jon!
Instead of jamming the damage/map score regression and kill/map score regression into a clunky formula, couldn't you just do a multivariate regression with both?
Instead of jamming the damage/map score regression and kill/map score regression into a clunky formula, couldn't you just do a multivariate regression with both?
cool read m8, please do master-race North American teams!!
killemdeaderInstead of jamming the damage/map score regression and kill/map score regression into a clunky formula, couldn't you just do a multivariate regression with both?
would probably result in multicollinearity
cool read m8, please do master-race North American teams!!
[quote=killemdeader]Instead of jamming the damage/map score regression and kill/map score regression into a clunky formula, couldn't you just do a multivariate regression with both?[/quote]
would probably result in multicollinearity
Great read! Do you have a formal statistics background?
Great read! Do you have a formal statistics background?
do me a favour and put those on teamfortress.tv vanilla is dead anyways
do me a favour and put those on teamfortress.tv vanilla is dead anyways
killemdeaderInstead of jamming the damage/map score regression and kill/map score regression into a clunky formula, couldn't you just do a multivariate regression with both?
I agree that ideally I should be looking at this (I think it would be interesting with medics in particular), but it's a matter of time constraints and how I've ended up approaching it in an exploratory way combined with the need to make a decision when it would still be relevant, or piss poor planning in other words.
smzido me a favour and put those on teamfortress.tv vanilla is dead anyways
Despite the appearance of hard work, I'm a remarkably lazy man in many ways and vanilla does make it very easy.
[quote=killemdeader]Instead of jamming the damage/map score regression and kill/map score regression into a clunky formula, couldn't you just do a multivariate regression with both?[/quote]
I agree that ideally I should be looking at this (I think it would be interesting with medics in particular), but it's a matter of time constraints and how I've ended up approaching it in an exploratory way combined with the need to make a decision when it would still be relevant, or piss poor planning in other words.
[quote=smzi]do me a favour and put those on teamfortress.tv vanilla is dead anyways[/quote]
Despite the appearance of hard work, I'm a remarkably lazy man in many ways and vanilla does make it very easy.
smzido me a favour and put those on teamfortress.tv vanilla is dead anyways
But how else do we know about extvs latest videos?
[quote=smzi]do me a favour and put those on teamfortress.tv vanilla is dead anyways[/quote]
But how else do we know about extvs latest videos?
GeknaiirIs there a chance that you will ever do this for NA as well?
ESEA don't release the raw logs which my system uses and although something similar would be possible it would need a completely different approach so it would be quite a bit more work that I don't have time for right now.
[quote=Geknaiir]Is there a chance that you will ever do this for NA as well?[/quote]
ESEA don't release the raw logs which my system uses and although something similar would be possible it would need a completely different approach so it would be quite a bit more work that I don't have time for right now.
Ggglygycool read m8, please do master-race North American teams!!
killemdeaderInstead of jamming the damage/map score regression and kill/map score regression into a clunky formula, couldn't you just do a multivariate regression with both?
would probably result in multicollinearity
Probably, but it's possible that damage becomes irrelevant w/ kills included and not worth using. It'd be interesting to see by class too, whether damage is more important than kills for demos like people commonly say, and whether that holds for other classes.
[quote=Ggglygy]cool read m8, please do master-race North American teams!!
[quote=killemdeader]Instead of jamming the damage/map score regression and kill/map score regression into a clunky formula, couldn't you just do a multivariate regression with both?[/quote]
would probably result in multicollinearity[/quote]
Probably, but it's possible that damage becomes irrelevant w/ kills included and not worth using. It'd be interesting to see by class too, whether damage is more important than kills for demos like people commonly say, and whether that holds for other classes.
Added the [url=http://www.vanillatf2.org/2014/04/tf2-by-the-numbers-season-17-playoffs-round-2-aws-vs-tlr/]preview of the ETF2L Season 17 Playoff semi final between Awsomniac and TLR[/url]
Added a [url=http://www.vanillatf2.org/2014/04/season-17-grand-final-preview-tf2-by-the-numbers/]preview of the ETF2L Season 17 Grand Final[/url]
Added the [url=http://www.vanillatf2.org/2014/04/tf2-by-the-numbers-etf2l-season-17-review/]ETF2L 6v6 Season 17 review[/url]
Gentleman Jon confirmed Community MVP.
Gentleman Jon confirmed Community MVP.
I absolutely love these write ups, possibly my favorite TF2 article analysis right now. Keep up the amazing work, and thanks!
I absolutely love these write ups, possibly my favorite TF2 article analysis right now. Keep up the amazing work, and thanks!
Added [url=http://www.vanillatf2.org/2014/06/tf2-by-the-numbers-etf2l-highlander-season-6-review/]ETF2L Highlander Season 6 Review[/url]
Added a bit of light analysis on the statistical effects of the sticky update. Now with added literature!
Added a bit of light analysis on the [url=http://www.vanillatf2.org/2014/06/tf2-by-the-numbers-a-scottish-tragedy/]statistical effects of the sticky update[/url]. Now with added literature!
Added [url=http://www.vanillatf2.org/2014/08/tf2-by-the-numbers-season-18-review/]ETF2L 6v6 Season 18 Review[/url]
Added research into [url=http://www.vanillatf2.org/2014/08/zen-and-the-art-of-balancing-a-lobby/]balancing lobbies and pickup games[/url]
GentlemanJonAdded research into balancing lobbies and pickup games
Holy shit this is mind boggling
[quote=GentlemanJon]Added research into [url=http://www.vanillatf2.org/2014/08/zen-and-the-art-of-balancing-a-lobby/]balancing lobbies and pickup games[/url][/quote]
Holy shit this is mind boggling
pyro < 1
my entire life is validated
pyro < 1
my entire life is validated
will you be able to do an analysis like this for i52?
will you be able to do an analysis like this for i52?
[url=http://www.vanillatf2.org/2014/09/tf2-by-the-numbers-i52/]i52 Review[/url] added
I passed out twice reading that.
I passed out twice reading that.