Methods, systems, and method for predicting sensor measurement quality. In some implementations, the method comprises: measuring, using a wearable computing device that includes a processor and a sensor, information indicating motion of the wearable computing device during a current time period; identifying one or more parameters associated with a determination of a likelihood that one or more measurements from the sensor configured within the wearable computing device is of sufficient quality for calculating a physiological metric using the one or more measurements from the sensor, wherein the one or more parameters include contextual parameters associated with the wearable computing device; determining the likelihood that the measurement from the sensor associated with the user device is of sufficient quality at a second time period for calculating the physiological metric using the measurement from the sensor based on the identified one or more parameters and based on the information indicating the motion of the user device during the current time period; in response to determining that the likelihood exceeds a predetermined threshold, activating the processor and the sensor and collecting a measurement from the sensor at the second time period; and updating the identified one or more parameters based on the motion of the user device during the current time period and based on the measurement from the sensor at the second time period.