Real numerical shadow of a matrix A of dimension d is defined as a probability ditribution PA(z) on the complex plane, supported on the real numerical range WR(A). PA(z):=∫Ωdμ(ψ)δ(z−⟨ψ|A|ψ⟩), where μ(ψ) denotes the unique unitarily invariant (Fubini-Study) measure on the set Ω={|ψ⟩∈Rd:⟨ψ|ψ⟩=1}.