Quantum probability from decision theory?

In a recent paper, Deutsch claims to derive the 'probabilistic predictions of quantum theory' from the 'non-probabilistic axioms of quantum theory' and the 'non-probabilistic part of classical decision theory.' We show that his derivation includes a. crucial hidden assumption that vitiates the force of his argument. Furthermore, we point out that in classical decision theory a standard set of non-probabilistic axioms is already sufficient to endow possible outcomes with a natural probability structure. Within that context we argue that Gleason's theorem, relying on fewer assumptions than Deutsch, provides a compelling derivation of the quantum probability law.