I'd accept that. OP's submission has a good example where Graeber's difficulty taking criticism almost cost him and Wengrow a publication which was saved by Wengrow's more measured approach.
He and Wengrow were serious in a certain sense: They were determined to publish extracts in peer-reviewed journals, to establish scholarly credibility. When they made their first submission, to the Journal of the Royal Anthropological Institute, they received brief comments dismissing their work as insufficiently “new.” Graeber was immediately distressed, convinced the anonymous reviewers had “some ax to grind.” Wengrow managed to calm him down while he replied on their behalf. Politely, Wengrow asked for examples of where exactly work like theirs had appeared before. The Journal was unable to produce any. It accepted the paper. Graeber was amazed. “How do you do that?” Wengrow remembers him marveling. “The man did not know how to handle stress,” another friend told me. A heated exchange on Twitter could derail him. Still, he refused to shy away. One persistent bugbear was the Berkeley economist Brad DeLong — after DeLong’s blog took him to task for a handful of factual errors in Debt, the two became entrenched in a protracted flamewar. (DeLong recently resurfaced, amid positive press for The Dawn of Everything, to affirm his stance that “nothing David Graeber writes is trustable.”)
I'd missed the claim about Keynes on first reading. I suppose it would have to be The General Theory of Employment, Interest and Money. It appears to be very difficult for one person to write a historical treatise of such scope and breadth without making a nontrivial number of sloppy errors. There's so much surface area of things that could go wrong relative to the powers of human attention and working memory. I recall Bertrand Russell's History of Western Philosophy which also has no shortage of sloppy errors and dubious ancillary claims despite Russell's reputation as a careful thinker.
I greatly enjoyed reading Debt. It gave me a new and valuable perspective on many things. I would not trust it alone as a source for any particular factual claim and would carefully check the references as needed, but as a whole I found the core arguments compelling. (That said, I think the overall level of scholarship is actually quite good and I would extend this same caution to any large work synthesizing many sources.) Like you said, it's a thought-provoking book.