It's worth noting that Erik Chen is working on porting cpu measurements of telemetry to use TimelineBasedMeasurement 2 framework:  issue 640312 

With the new framework, we have the ability to compute cpu metrics during a specific period time, e.g: user scrolling the page, video call playing... This would help stamp out the metric noise by isolate different activities of the browser.

Sorry, restoring status and priority. For a moment I thought this was not for Telemetry.

Is there a particular reason that we care about CPU usage and not power?

Also, nednguyen@, how does this compare to our current benchmarking priorities for the short-term? Broadly, we're focusing on improving the reliability and quality of the benchmarks that we have rather than adding new benchmarks, especially when those new benchmarks would require drastically different Telemetry codepaths from most other benchmarks (like this one probably would, given that we couldn't use WPR).

This issue has been Available for over a year. If it's no longer important or seems unlikely to be fixed, please consider closing it out. If it is important, please re-triage the issue.

Sorry for the inconvenience if the bug really should have been left as Available.

For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot

Closing this because I think that sheriffbot's analysis is probably correct.