Maybe, if you post a screenshot of the filter, some feedback may surface from someone here.

Honestly, filters aren't hard to understand. I've been using Thunderbird since Eudora went extinct back in the 90's, and I have dozens of (single and multi-criteria) filters in place. The only ones that don't work are the ones where I'm trying to filter based on URLs in the message body. I don't see how a screenshot is going to help someone understand what I'm after. FWIW, I've added the text that accompanies the URL as a filter criteria, but IMHO, I shouldn't have to do that and I'm afraid it will sweep up some non-intended emails along with the ones I want to filter.

BTW, I've read other replies to similar questions, saying that thunderbird "doesn't have access to html source". I'm gonna call out the people that say that, because TB can identify the body of a message, and can therefor escape the html into plain text. I see no technical reason - from a programming standpoint - why TB cannot be made to run a filter against html. I've been a programmer for over 40 years, and I've written more than my fair share of html parsing code. It ain't hard to search for text in a html stream.

I respect your opinion, but my inference is that the developers preferred to focus on what the user sees in a message, not some underlying content that would just confuse them. It's not a matter of coding proficiency, but of designing to a target audience. But, as a user, you're welcome to suggest adding that feature at connect.mozilla.org, in the IDEAS section. (As an off-topic comment: you mentioned Eudora being extinct, but my dear spouse has used it daily since version 3, somewhere around 1997. And I'm too smart to try and take it away from her. It works flawlessly. We bought 7.1 in 2007 and that was the last true version. )