We aim to detect and diagnose energy anomalies, abnormally heavy battery use. This paper describes a collaborative black-box method, and an implementation called Carat, for performing such diagnosis on mobile devices. A client app sends intermittent, coarse-grained measurements to a server, which identifies correlations between higher expected energy use and client properties like the running apps, device model, and operating system. The analysis quantifies the error and confidence associated with a diagnosis, suggests actions the user could take to improve battery life, and projects the amount of improvement. Carat detected all anomalies in a controlled experiment and, during a deployment to a community of more than 340,000 devices, identified thousands of energy anomalies in the wild. On average, a Carat user's battery life increased by 10% after 10 days.