I have a recurring habit of glossing over names I don't recognize when first reading them in a news article and then will frequently need to jump back to find where they were initially referenced when they get mentioned again only with their last name. While this can be a little annoying when reading, it would be far worse in a video; I can't imagine trying to scrub back to the point where someone first was mentioned. At best, the video might have captions (which I'll always have turned on), but it's a lot harder to look at a caption and remember exactly which information came before it and which came after because the text isn't all visible at once.
Then I can decide if I know enough or whether I want to listen to the full video.
It's a lot easier to skip over the bits you already know in written form, with video you need to seek forward, listen a bit, seek forward, listen, whoops too far, seek back a bit etc.