The Bloch Sphere: Every Qubit State, One Point

From two complex numbers to one point on a sphere

A general pure qubit state is written |ψ⟩ = α|0⟩ + β|1⟩, with complex amplitudes α and β satisfying |α|² + |β|² = 1. Naively that is four real numbers, but normalization removes one degree of freedom, and — crucially — the overall global phase of the state has no physical effect on any measurement outcome, so it can be discarded entirely, removing another. What's left is exactly two real parameters, precisely the number needed to specify a point on the surface of a sphere using two angles, θ and φ, just like latitude and longitude on a globe. Felix Bloch introduced this representation for nuclear spins in the 1940s, and it applies unchanged to any two-level quantum system.

|ψ⟩ = cos(θ/2)|0⟩ + e^(iφ)·sin(θ/2)|1⟩

θ ∈ [0, π]   polar angle — sets the |0⟩/|1⟩ mixture
φ ∈ [0, 2π)  azimuthal angle — the relative phase between |0⟩ and |1⟩

The poles, the equator, and everything between

The north pole (θ = 0) is |0⟩, and the south pole (θ = π) is |1⟩. Points on the equator (θ = π/2) are equal superpositions that differ only in relative phase φ — the state (|0⟩ + |1⟩)/√2 sits at φ = 0, and (|0⟩ + i|1⟩)/√2 sits at φ = π/2, a quarter of the way around. Crucially, any point strictly inside the sphere, not on its surface, represents a mixed state — a statistical ensemble rather than a coherent superposition — with the exact center of the sphere corresponding to the maximally mixed state, equivalent to a fully decohered, classical 50/50 coin flip. Purity is literally distance from the center.

Gates are rotations

Every single-qubit quantum gate corresponds exactly to a rotation of the Bloch vector about some axis. The Pauli-X gate is a 180° rotation about the x-axis, flipping |0⟩ and |1⟩ like a classical NOT gate; Pauli-Y and Pauli-Z rotate similarly about the y- and z-axes. The Hadamard gate is the rotation that swaps the roles of the z-axis and the x-axis, mapping |0⟩ down onto the equator. Phase gates such as S and T rotate only about the z-axis by a fixed angle, changing φ but leaving θ alone. In fact, any single-qubit unitary operation at all, no matter how complicated, reduces to some single rotation about some single axis by some angle — a direct consequence of how the group of 2×2 unitary matrices maps onto ordinary 3D rotations.

A subtlety: the sphere is SU(2), doubled

For the technically curious: the group of qubit unitaries (SU(2)) maps onto rotations of the Bloch sphere (SO(3)) through a two-to-one correspondence — rotating the underlying quantum state by an angle θ in Hilbert space corresponds to rotating the Bloch vector by 2θ on the sphere. This is the famous fact that a spin-1/2 particle needs a full 720° rotation, not 360°, to return exactly to its starting quantum state. The Bloch sphere itself hides this, since it has already discarded the global phase where that distinction lives, but the effect is real and shows up in interference experiments sensitive to that discarded phase.

Beyond one qubit

The Bloch sphere picture is exact and complete only for a single qubit. Two or more entangled qubits live in a higher-dimensional space with no simple 3D visualization — an entangled two-qubit state cannot be drawn as two separate points on two separate spheres, precisely because entanglement means the individual qubits don't have well-defined pure states of their own. Tracing out one qubit of an entangled pair leaves the other in a mixed state, a point inside its own sphere rather than on the surface. That gap is exactly why entanglement resists easy visualization, and why it feels so different from ordinary single-qubit superposition.

Frequently asked questions

Try it live

Everything above runs in your browser — open Bloch Sphere and change the parameters while it is running. Nothing is installed, nothing is uploaded, the whole model lives in one tab.