Takes a channel and returns a list of videos sorted by regression error (good outliers at the top, bad outliers at the bottom). Implemented as an App Script through Google Sheet. It is designed as a solution to help new viewers of a channel choose the best videos to watch.
Example: using channel data from In a Nutshell, the Egg is the best performing outlier in terms of deviating from the regression between view counts logged and like counts logged.
Although it is called Likes-per-View, in reality we use regression deviation to sort as we find this to be a better measure than the ratios of like to view counts.
You are welcome to make a copy of this spreadsheet and use the Finder. YouTube Likes per View Finder
Refer to Code.gs.
MIT