A classical theorem of Szegő states that for any probability measure $μ=w\frac{\mathrm{d}θ}{2π}+μ_s$ on the unit circle the polynomials are dense in $L^2(\mathbb{T},μ)$ if and only if $\log(w)\notin L^1(\mathbb{T})$. A related question asks whether the monomials with exponents in some subset $Λ\subseteq \mathbb{N}_0$ already span $L^2(\mathbb{T},μ)$ if $\log(w)\notin L^1(\mathbb{T})$. A result by Olevskii and Ulanovskii gives an answer if $μ$ belongs to a class of absolutely continuous measures. We investigate the same question for Markoff measures.