A weaker version of your hypothesis, whereby the word was synthesised once but reinforced by similar phrases or folk etymologies (leading to the current confusion) is much more plausible, however.
I have so many more questions if that's what you're trying to say here.
For example, 'Council House And Violent' => chav (British slang for an anti-social lower-class person)
It's wry but it doesn't pass the sniff test. People don't tend to sit around making up words like this and even when they do other people don't adopt them. Words are nearly always born through analogy and common ground, not out of whole cloth.
There's a Wikipedia list of false etymologies in English here  (fair warning - a lot of unpleasant words here). A quick read of it should help train ones priors as to whether or not an etymology is likely to be true.
In real life words generally get made by pretty obvious metaphors (it's gotta be obvious enough for someone hearing it for the first time to get it), combining two words together or shortening an existing word.
Recently published dictionaries do a much better job than they used to, so you will often see "origin unknown, poss. related to X or Y", where an older dictionary simply proclaimed the etymology. So, was the truth recently lost? Or did the lexicographers recently find evidence that their confidently proclaimed etymology might just be poppycock?
So if there are lots of cases where etymology is disputed, and there are, then not all claims can be right. Many must have been wrong, and it stands to reason that not all errors have been discovered. Hundreds of thousands of words, place names, personal and surnames, passed from mouth to mouth by mostly illiterate humans for whom X sounds like "eggs" in a game of telephone ("Chinese whispers") lasting centuries, repeatedly crossing dialect and language boundaries, multiplied by the number of human languages, and does it really seem wrong that "a lot of proposed etymological theories are inaccurate"?
If the claim were really that there are no correct etymologies, that would certainly be wrong, but if "poppycock" meant the confidence of the old dictionaries was not justified, I would say that time has vindicated his claim.
The fact that half of the rest of the entries are saying "oll korrect" makes me believe that this is the one. Especially with this example too.
>A contemporary news report of the 1866 transatlantic telegraph cable says "The following telegram has been received from Mr. R. A. Glass, Managing Director of the Telegraph Construction and Maintenance Company (Limited) :— 'O.K.,' (all correct)."
- the letter ‘k’ is rarely used in Latin (https://latin.stackexchange.com/questions/1084/why-is-the-le...)
- ‘correct’ in Latin is ‘correctus’ (https://www.etymonline.com/word/correct?ref=etymonline_cross...)
⇒ I would think that explanation is completely made up.