It's not a one-to-one mapping, by the Fundamental Theorem of Algebra.
$e^z$ is one-to-one, when restricted to a horizontal strip of width = $2\pi i$.
Is it a similar argument for $z^2$?
Thanks,
Edit: $z^2$ doubles angles (and squares magnitudes of numbers in complex exponential form.), so it seems one-to-one - I guess I am possibly confusing myself with the conformal mapping $z^2$ with the polynomial equation, $z^2$ + az = 0.
It is conformal on $\mathbb{C} \mathop{\backslash} \{0\}$. At $0$ the derivative is $0$ and it is not conformal (as you say, it doubles angles subtended at $0$).
My mental model of $z \mapsto z^2$ is two turns of a spiral staircase projected onto the ground floor: that mapping is two-to-one but preserves (plan view) angles except at the centre of the staircase.