Guys slightly off topic, but please do keep the colorblind people in mind. The colors used for the graphs are so similar that I can not differentiate them.
Tools like Color Oracle[1] are great when checking UI color design for its color safety. Some monitor software also has similar features. E.g. NEC monitors' MultiProfiler tool supports a number of different color vision emulation modes; see the "Color Vision" pane in that app.
Red/blue would be safer colors to differentiate for many colorblind people. Red/green colorblindness is so common it baffles me that comparisons using red/green are still so prevalent. Especially considering how heavily biased the tech population is toward men.
And as Sanddancer points out, any colors that make use of red/green are going to be problems as well. (hence blue/purple being an issue as the difference is the red component).
In a given group of 5 men, you have a 50% chance that 1 is red/green color deficient.
Pc = Probability that someone is colorblind
(1 - Pc) = Probability they aren't colorblind
(1 - Pc)^5 = Probability that no one in group of 5 is colorblind
1 - (1 - Pc)^5 = Probability that someone in group of 5 is colorblind
It usually doesn't come up in conversation. Even when I was working on an image processing toolkit, it was several months before I realized my boss was red-green colorblind.
I'm red/green colorblind and blue/purple is the worst combination for me. I colored all my skies purple in elementary school because I couldn't tell the difference.
