1. It used strawman arguments. Notably that the world is looking for a 50/50 split in gender ratios. Most reasonable people believe that even in ideal equilibrium state that this is not the expected outcome (although I'm sure there exist some people that push for 50/50 or greater). But that the current ratio is not the ideal equilibrium state. The author of the paper seems to point out that 50/50 isn't likely the expected ratio -- and then jumps to trying to dismantle affirmative action.
2. He selectively uses research and data from research. For example the impact of stereotypes and individuating. But the research shows that while individuating is effective when it shows clear discrepancy, it isn't effective when there is any ambiguity. That is Maryam Mirzakhani will likely make someone who thinks women aren't good at math think, "But yes, she's actually quite good." But when it comes to more nuanced decisions and decision with a lot less data (which is more commonplace in the workplace), individuating isn't so effective. Determining who should be promoted to lead or VP from a group of pretty strong candidates, will often have the stereotype bias still play a role (and in fairness, depending on exactly the role -- it can benefit females too). This is meant to be an example of research where in the memo one part of the story is told, but not the other. Based on my limited investigation, I don't trust him being impartial in presenting these facts. In much the same way that climate change deniers often point to facts too -- its what they don't point to that is worrisome.
Lastly, there is a difference between speaking truth and how you speak truth. This seems obvious to most people outside of engineering. Putting "Catherine has big boobs and I'd have sex with her, if she consented" as part of your sig may be completely factually she true. In fact, it may be obvious to everyone who reads it, but there is a time and a place. If I were a female, I'd have a hard time reporting to Mr. Damore. If I didn't get a promotion, I don't think it would be out of line, to consider what he wrote in the memo. If Mr. Damore ignores my suggestion in a design review, I'd have to again strongly consider if he did so because of my gender. If he rejects all of his women interview candidates -- is that a problem or not? The context of what he wrote, even if done in good faith, creates a more difficult working relationship.
My hope is that if Mr. Damore wrote that in good faith, he will take this feedback and make things better for everyone, male and female. He certainly won't have trouble getting job offers (if he wrote it in good faith or not -- there are enough people in tech who don't like women in the industry) -- and if his end goal was to create the environment for the discussion, then he's done that too.