Recent experiments suggest that microtubules do not grow steadily but instead elongate at a rate that varies in time. We argue that this variation might arise from fluctuations in the length of a dynamic GTP-tubulin cap at the microtubule end. We propose that these fluctuations can lead to a switch in the dynamics of a microtubule end between growth and shrinkage, and provide insight into how the sensitivity of this switch can be changed by microtubule polymerases, such as XMAP215, and tensile forces, through the stabilization of initial contacts in the cap.