"Want to win? Wear red."

[Blue Falcon]

Well, this certainly explains the overwhelming success of the Arizona Cardinals, now doesn't it?

Damn, I was just about to post the same thing!!

Now CNN has posted it on their site as if it's a scientific fact:

http://www.cnn.com/2005/TECH/science/05/19...d.ap/index.html

Guess that means the next "intimidating" color will be red. Can't wait to see Ray Lewis prance around like he was acquitted on murder charges after the Ravens come out in their all-red Ravens alts next season...

[BallWonk]

As I said earlier, this is at best an observed coincidence.  Part of the scientific method is reproducibility, and there is absolutely no way that you can replicate any individual event - in this case, a game or match - over and over to achieve a controlled prospective study.

It's a clever scenario you lay out, but it has nothing to do with how actual scientist do actual science. Yes, some science involves the reproduction of experiments in the lab under identical conditions. But most does not. Much of science involves the observation of non-reproducible events. Take astronomy, for example. Want to find out whether supermassive objects really create a gravity well that acts as a lens for light from objects on the other side? Under your description, the only "real science" way to find this out would be to create a galaxy in the lab, then observe that galaxy repeatedly as you pass it in front of stars and other stellar objects.

Thankfully, scientists didn't wait until they had labs big enough to build whole galaxies to run this experiment, which happened to be the key to confirming an important part of the Theory of Relativity. Instead, they observed the non-reproducible world outside their labs until they had collected enough observational data sets to draw reasonable conclusions.

Anyone who has played sports will understand that confidence can have a real affect on how you play.* If, for whatever reason, red really does tend to give people a slight boost in confidence, or make one slightly more fearsome to one's opponent, then of course wearing red would give you a slight advantage. If this was really happening in the world, then you would expect that when colored equipment is randomly distributed, the contestants with red equipment would win slightly more often than could be explained by chance. And this is, in fact, what has been observed.

The problem with the study isn't that they didn't reproduce the same exact contest over and over again - if that was a requirement there would essentially be no science at all - the problem is that it didn't sufficiently control for non-random factors in much of its data, and the truly controlled data is too small a pool to really distinguish small advantages from chance. Nonetheless, the conclusion drawn is entirely reasonable, even intuitive, and the good data in the study supports the "duh" hypothesis.

*Or at least any kid who had to wear an Astros rainbow jersey in little league can attest to this. Wearing such a girlie outfit did hurt my team's confidence, and it's hard to believe that the ridicule other teams heaped on us didn't help boost their confidence when playing us. We were certainly happier kids on days when we played the Padres, whose all-brown shirts were almost as lame as our own rainbow jerseys.

[VitaminD]

If this was really happening in the world, then you would expect that when colored equipment is randomly distributed, the contestants with red equipment would win slightly more often than could be explained by chance. And this is, in fact, what has been observed.

Except that the conclusion is, in fact, incorrect.

Example: NBA champions. There have been 58 of them, going back to the 1946-47 Philadelphia Warriors. 15 wore red as a primary color, 19 had red as a color of some kind in their uniforms (Bulls x 6, Pistons x 3, Rockets x 2, 76ers x 2, Bullets x 2, Trail Blazers x 1, Bucks x 1, Hawks x 1, Nationals x 1). 19/58 = 32.7%.

Teams wearing green: (Celtics x 16, Sonics x 1, Bucks x 1). 18/58 = 31.0%.

Teams wearing blue: (Pistons x 3, 76ers x 1, Lakers x 5, Bullets x 2, Knicks x 2, Warriors x 3, Hawks x 1, Royals x 1, Nationals x1). 18/58 = 31.0%.

One additional event out of 58 is not statistically significant enough to prove red confers any type of advantage over other colors. Had this same list been run a year ago before the Pistons won their 3rd title, green would be superior to blue (18 to 17), and codominant with red (18 apiece). Thus, reed, green and blue are equally codominant.

The observers here limited their scope to a sample size that framed their observations. I could similarly say that, because the Flyers won two Stanley Cups in the mid-70s, orange conferred superiority on those who wore it while playing NHL hockey between 1974-1976.

As far as retrospective v. prospective study, I admit my bias in science is through medicine. The paradigm in medicine today is evidence-based - basically, "show me proof that it works through controlled study". And the studies that are taken as the most definitive proof are randomized, placebo-controlled, double-blinded studies, which are hardest and most expensive to set up but eliminate as much bias as is possible. So it is the kind of thing I gravitate toward when looking for hard evidence that something is proven or is not. I am not dismissing the conclusion as much as I am saying, "prove it, prove that's it more than just an interesting coincidence within a narrow timeframe that you noticed". Prove it across a broader range of time, not just that champions of the last couple of years in US sports have worn red. The Cowboys, Packers, Dolphins, Raiders, Lakers, Celtics, Bruins, Maple Leafs, Yankees*, Pirates and Orioles, Tar Heels, Blue Devils, Wolverines, Hurricanes and Bruins would all disagree historically; Chelsea, Juventus and Olympique Marseilles would disagree currently and internationally.

* I know there's a lot of red in the Yankees' logo. The observation was prompted by what the players wore on the field. There's no red on the Yankees' uniforms or hats.

[Perfect Zero]

This is why the Texas Rangers should go back to mostly red and grey with only a little blue. Heck, the only times we went to the playoffs where the years we wore them.

[Sabrejeff]

i was just about to post this, no wonder the Blue Jays wont win a World Series, they dotn have red!

Anyways, this topic certainly proves how the St. Louis Blues are always bridesmaids. Of course 65 red stripes couldn't cure it in the mid-'90s.