That sounds like a different theorem. While it conincides with sums of squares of distances for the Euclidean setting, for the case of a sphere or other manifold it is decidedly about triangles, not so much distances.