Actually, in order to prove that $[\mathbb Q(\pi):\mathbb Q(\pi^3)]=3$ you need some facts: specifically that, if $\alpha$ is transcendental over $K$ field, then $K[\alpha]\cong K[X]$, the ring of polynomials with coefficients in K, and $K(\alpha)\cong K(X)$, its field of fractions; so, since $\pi^3$ is transcendental over $\mathbb Q$ (otherwise, $\mathbb Q(\pi^3)/\mathbb Q$ would be finite, and $\mathbb Q(\pi)/\mathbb Q$ is finite => $\mathbb Q(\pi)/\mathbb Q$ finite => $ \pi$ algebraic over $\mathbb Q$, absurd), this gives you the isomorphism $\mathbb Q(\pi^3)\rightarrow\mathbb Q(X),\pi^3\rightarrow X$, which extends to an isomorphism $\phi:\mathbb Q(\pi)\rightarrow \mathbb Q(\xi),\pi\rightarrow\xi, $where $\xi$ is a cubic root of X in an algebraic closure of $\mathbb Q(X)$.
So, $\pi\in\mathbb Q(\pi^3)\iff \xi\in\mathbb Q(X)$; but, if that was the case, you should have $\xi$=P/Q, with P,Q$\in \mathbb Q[X]$, $P,Q\neq 0$ (if P=0, then $\pi=0$ through $\phi$,absurd), and so deg(P)=n, deg(Q)=m, and $X=\xi^3=P^3/ Q^3$ $\iff XQ^3=P^3$ $\Rightarrow 3m+1=3n$, which is absurd. So, $\pi\notin\mathbb Q(\pi^3)$, and , since 3 is prime, it must be that $[\mathbb Q(\pi): \mathbb Q(\pi^3)]=3$, thus $(1,\pi,\pi^2)$ is a basis of $\mathbb Q(\pi)/\mathbb Q(\pi^3)$.