Machine translation (MT) systems are not able to always produce translations of sufficient quality. As a practical means of utilizing such MT systems, we investigated the potential of pre-editing strategy, i.e., rewriting source texts into more machine translatable ones. We first formalized a human-in-the-loop protocol, in which human editors incrementally rewrite source sentences aiming at better MT results. Implementing this protocol with a Japanese-to-English MT system, we then collected a total of 12687 pre-edit instances for 400 source sentences on four different datasets and obtained promising results; more than 85% of source sentences turned out to be accurately translated by the MT system. We also revealed that the pre-edited Japanese source sentences were better translated into Chinese and Korean, confirming the usefulness of pre-editing strategy in amultilingual translation setting. Through analyzing the collected pre-edit instances, we built a typology of edit operations comprising 53 types, which unveils the subjects for further research.