[1vuio0pswjnm7]

What I do is only use the videoplayback URL and reject/divert all the others. This requires having the proxy delete the range parameter in the query string portion. If you observe the traffic when using the YouTube website, there are many HTTP requests for things we do not need if we are not using Google's Javascript player. And there is no need for Cookies and many other HTTP headers if only using the videoplayback URL. Sometimes I may do two downloads: one for the video and one for the audio. YouTube seems to be now using similar/same approach for their user video as Facebook. With some videoplayback URLs pointing to only audio, and some pointing to different screen sizes.

I doubt you are as vulnerable to as much tracking as someone using the YouTube website.

I find youtube-dl breaks quite frequently and so I wrote a simple downloader myself (for non-commercial videos) in a few lines of shell script. I have found it is easier to fix quickly if there are significant changes at YouTube than waiting for the youtube-dl project to release an update.