Reading people talk about color parses and STILL not understanding how FFLogs works is EXTREMELY disheartening.
It's not just about color but MULTITUDES of context and what each job could do. Then the time frame changes the percentile. Basically the parses uploaded early on was set by a small sample. The players who posted even said so. Then it's also affected by party contribution and makeup. On top of that a death affects the parse. If a player dies because someone else fails the mechanic it affects their output.
The data is only "Good" for a limited time frame as a measurement DUE to the fact that once the top players have cleared and set in the data, they also will be geared up and look for ways to get the highest parse to upload.
A GOOD raid leader takes in a multitude of factors, not just "the color of the parse" and if they have "better gear" They look to see where their players should be in terms or ranking better within their own party composition. Is this a fight where a monk may not perform as well due to certain mechanics? Is this strategy working out well? Is there a latency issue with a particular instance that is affecting the player?
Now using the analysis data is better (but it seems most people don't do that - and seeing the responses in this thread...) you can see if you're messing up on a part of your rotation, if something is clipping- which people don't understand this concept well. The problem is, that it takes a longer time and more thought to process "for a video game" and people just go to the color of a parse from other uploads.
That said, it's a great tool to help improve your DPS and learn where you can correct your mistakes. If it helps you improve it's great that's what it's for. Mediocrity is just also just looking at one "standard" or set of data and using that as a basis. Because you didn't bother to understand what those tools were for. - and that mediocrity is DEFINITELY a community problem.
Tl;Dr people not actually understanding the tools they're using for their argument.