A question about "more genetic variation existing within races than among them". Is this statement true?

Now then, let's say you take two colors (what is many times called "races") of people, group A and group B. You take group A and you identify a characteristic and you see a variation. You then see that the difference of this variation is greater than the difference in the characteristic between group A and group B. Now what you do is you take the one end of the spectrum of this characteristic that is within group A, let's say the "blue" end for simplicity and compare it to the other end, let's say the "red" end. After all, if there's variation in the characteristic then there must be points on the spectrum that they vary on. Now, you take the blue end of the spectrum in this group A and you compare it to the red end of the spectrum in group B. Will the difference be the same as the difference in the red end of the spectrum in group A?