I think the point is that pi, a constant defined by properties of circles, is also useful in an approximation of factorials, and there was no obvious reason for

I think the point is that pi, a constant defined by properties of circles, is also useful in an approximation of factorials, and there was no obvious reason for the circle-specific constant to show up in anything related to factorials.