Like blank always said: Choose the power colour

oh fuck they described tf2 as a "competitive multiplayer game" #esports

he didn't even say what color wins more in tf2!!!

bitch you get off on drawn 8 year olds

if only it was slander

I'm pretty sure all he ended up saying was that you should make your choice depending on what you want to do with your team

- Are you Russian - Red
- Are you not Russian - Blu

http://i.imgur.com/luA9G5L.jpg

How to get an unfair advantage.

Yeah I googled it and he kept a fucking hat on for the whole time :s

http://i.ytimg.com/vi/pq07PbaXZVM/maxresdefault.jpg

This is more cringy than weebs

WHY WOULD YOU THEME YOUR WEDDING AROUND A VIDEO GAME

WHY WOULD YOU COSPLAY DURING YOUR WEDDING

WHY WOULD YOU DO THIS TO YOUR WIFE, YOUR FAMILY, AND YOURSELF

He made his best man dress as Luigi :O

It sounds like a cool idea to do it subtly, like colour themes. Not str8 up cosplay

maybe logs.tf or sizzlingstats should collect data to see which team has the color advantage

Hehe. The most successful color in the EU scene wasnt based on psyhology, but based on the pick of Epsilon :D

Danmark Vinder

I think the thesis is fine, but the defense is wrong. If you are agitated by red, blue would have a distinct disadvantage predicting movement compared to red shooting blue.

Already started doing this. Two small problems, the map isn't in logs.tf's api so there is a possibility that I am using a log that is on a a/d map or a pl map which we already know favors blue over red, and there is a chance that games for teams that lose / don't have the normal number of players will get put into the mix. I'm assuming that only grabbing 12 player logs alleviates the majority of issues, but I'm not 100% sure.

With the first 38000 6s logs results are:
total: 37667 blue: 17642 red: 16915 tie: 3110
blue-red = 727 = 1.93% games
stddev = 97 i.e. the difference between the two should be <= to that number at a 70% confidence level, or <= 291 at 99%

So looks like blue is the reigning color at least for random logs, I'm interested in looking at only the high level match logs to see if this applies there as well but that is probably going to take way more coding time than the 5min it took to put together the all logs code.

You should try it with tf2pickup logs: http://logs.tf/profile/76561197998444308/uploads
They use player rankings to balance the teams, so skill level is more or less equal

part of what makes league's win/loss disparity so big has to do with the sides of the map not having equal opportunity to certain objectives, like how its easier for blue side to steal dragon, which can be effective early game, while red side has an advantage to steal baron later in the game, albeit harder. besides that the logic makes pretty good sense

From All 19000 logs (less total because games had not 12 players):

total:15239 blue:6027 red:7955 ties:1257
blue-red: -1928 blue win%:39.5498392283 red win%:52.2015880307
stddev: 61.7231723099 stddev*2:123.44634462 stddev*3: 185.16951693