Science Says Red Beats Blue
- Thread starter Andy Chalk
- Start date
I thought pretty much all games boiled down to red vs blue.Portoparty said:
While most tactical shooters don't explicitly call one team red or blue. They generally are, one team will get its names, scores etc written blue while the other gets red. Ditto if you put a cross hair over a player the colour of the writing often comes up red or blue depending on the team.
Maybe Red Team win more games because people like "Epic Games' Vice President Mark Rein", keep dropping out of games and cost Blue Team the match!
First off, 55% vs 45% is a huge difference. Much larger than I would have expected... much larger that I was lead to believe the previous studies involving real sports teams demonstrated.
Really, mane of the interesting effects in the world are small... if something is a monumentally good predictor of performance, chances are you already knew it.
"Breaking News: Hitting Fingers with Hammers May Cause Injury."
***********
Further, 55% is significant, assuming that the teams where properly randomized.
standard deviation of a proportion is sqrt(p*q/n) = sqrt(.55*.45/1347) = .013
.05/.0136
That's a z-score of -3.68
So, assuming the study wasn't biased, there's something like a 99.99% chance that "red team does better than blue team"
If, as I said in an earlier post, the study is biased, that is a separate issue.
That's high-school level stats smackdown.
*************
And the locusts... yeah, they fed the FLYING insects video of FLYING to see how the insects would respond, and the scientists learned from the experiment.
I'm not familiar with the study, so I can't comment beyond that... but I would suppose there are some interesting insights such a test can give on the insect mind.
Stop hating.
***********
Oh, and just for fun, here's the abstract of the article:
"In the 2004 Olympic Games, opponents wearing red athletic uniforms were more likely to win against opponents wearing blue uniforms. To investigate whether this color bias extends to the world of virtual competition, we compared the performance of red and blue teams in a popular multiplayer first-person-shooter (FPS) computer game. For 3 consecutive months, we collected data from a publicly available global statistics server. Outcomes from 1,347 matches played by the top 10 players on the same virtual arena were included. Red teams won 54.9% of matches, and this effect was highly significant. Our data suggest that joining the red team may offer a slight advantage over the blue team in virtual competition, and this should be accounted for when designing FPS games. It is likely that ?seeing red? may trigger a powerful psychological distractor signal in human aggressive competition that can affect the outcome of sports and virtual contests alike."
I'd read the article for you and post a summary, but there's no chance in hell I'm buying a 280$ subscription to CyberPsycology today...
Really, mane of the interesting effects in the world are small... if something is a monumentally good predictor of performance, chances are you already knew it.
"Breaking News: Hitting Fingers with Hammers May Cause Injury."
***********
Further, 55% is significant, assuming that the teams where properly randomized.
standard deviation of a proportion is sqrt(p*q/n) = sqrt(.55*.45/1347) = .013
.05/.0136
That's a z-score of -3.68
So, assuming the study wasn't biased, there's something like a 99.99% chance that "red team does better than blue team"
If, as I said in an earlier post, the study is biased, that is a separate issue.
That's high-school level stats smackdown.
*************
And the locusts... yeah, they fed the FLYING insects video of FLYING to see how the insects would respond, and the scientists learned from the experiment.
I'm not familiar with the study, so I can't comment beyond that... but I would suppose there are some interesting insights such a test can give on the insect mind.
Stop hating.
***********
Oh, and just for fun, here's the abstract of the article:
"In the 2004 Olympic Games, opponents wearing red athletic uniforms were more likely to win against opponents wearing blue uniforms. To investigate whether this color bias extends to the world of virtual competition, we compared the performance of red and blue teams in a popular multiplayer first-person-shooter (FPS) computer game. For 3 consecutive months, we collected data from a publicly available global statistics server. Outcomes from 1,347 matches played by the top 10 players on the same virtual arena were included. Red teams won 54.9% of matches, and this effect was highly significant. Our data suggest that joining the red team may offer a slight advantage over the blue team in virtual competition, and this should be accounted for when designing FPS games. It is likely that ?seeing red? may trigger a powerful psychological distractor signal in human aggressive competition that can affect the outcome of sports and virtual contests alike."
I'd read the article for you and post a summary, but there's no chance in hell I'm buying a 280$ subscription to CyberPsycology today...
This may be so but pink beats all.
I think this has something to do with aggression and that the angrier players may opt to be placed on red, or maybe red signals something in the blue players brain that slows their reactions. Either way I know what team I;m picking.
It would be interesting to see what the stats are for a game like CoD4, Rainbow Six, GoW etc that do not have team colour.
I think this has something to do with aggression and that the angrier players may opt to be placed on red, or maybe red signals something in the blue players brain that slows their reactions. Either way I know what team I;m picking.
It would be interesting to see what the stats are for a game like CoD4, Rainbow Six, GoW etc that do not have team colour.
This is kind of the same in the real world too... look at the english top flight football going way back before the prem... the most successful teams have worn red. (liverpool / manutd)
bah! ima a chelsea fan too
bah! ima a chelsea fan too
It may seem at first glance that this isn't a big deal, since 55% isn't much higher than 50%. But look at it this way- suppose a million games were played. Red will win 550,000 games and Blue will win 450,000 games. That's a discrepancy of 100,000 games in a situation where things are supposed to be perfectly even. Red is winning 22.2% more games than Blue is winning. Statistically this is very significant.j-e-f-f-e-r-s said:
btw 55% is a majority
If one side appears to be superior the majority of the time over the long run, it means something is not balanced the way it should be. Suppose you flipped a coin 100 times and got 90 heads. Sure, it's possible that it "JUST SO HAPPENED" to land on heads more often than you expected. But it's far more likely that the coin is rigged.Xwii360 said:
This is a really good point. I'm curious whether this study was done with completely randomized teams or not. I would think that aggressive players choosing Red is a far more likely explanation than the human eye being better at following blue targets.Nugoo said:
My understanding is that the study was NOT conducted with truly random teams. From the abstract:
"we collected data from a publicly available global statistics server"
...ie, they went online and used statistics from the games that were played. I would presume that the players picked their own teams through the usual technique (Randomization with later team switches? Dunno, no experience with URT.) and are thus not truely random.
It is interesting that there are only 1347 logged games. That seems like a very low number.
Perhaps they filtered their results somehow?
Anyone go to a university which gets the publication in question?
(I realize it's summer...)
"we collected data from a publicly available global statistics server"
...ie, they went online and used statistics from the games that were played. I would presume that the players picked their own teams through the usual technique (Randomization with later team switches? Dunno, no experience with URT.) and are thus not truely random.
It is interesting that there are only 1347 logged games. That seems like a very low number.
Perhaps they filtered their results somehow?
Anyone go to a university which gets the publication in question?
(I realize it's summer...)
The Australian TF2 Clan, Nup! Squad Features (N!SF) has a saying:
Blue team is for Girls and Faggots.
It saves arguments of wether to join red or blue in the event of a stack.
Blue team is for Girls and Faggots.
It saves arguments of wether to join red or blue in the event of a stack.
i reckon they should research whether being a terrorist (or whatever pc phrase they used for the arab guys with the big red turbans on) hinders or helps you on the online matchs for call of duty. because while red might beat blues, they probably get their arse kicked by camouflage.