This comment rings true to me as someone who contemporaneously read Satoshi's posts (not from the very beginning but on bitcointalk.org). People are underestimating how difficult (and probably pointless) it is to create and maintain a whole personality in aid of subterfuge. There are so many other easier and less risky ways to achieve the same goal than to adopt a consistent persona that's far from your own; you could, for example, release the code with one anonymous paper and then disappear completely, and then create three or four random new personas as needed to push it along publicly. That's more plausible (and less work) than sustaining a "Satoshi" with a very different personal style from yourself and then also, as it happens, speculatively fashioning fake email conversations with yourself, years in advance, in order to throw people off the scent in the future.
Among people who try to guess Satoshi's identity, there is also a surprising bias toward fame or at least well-known people. There is literally zero reason to think that Bitcoin came from a person anyone had heard of or has heard of since. It's a big world. Many (most?) interesting and novel open-source contributions, hash algorithms, cipher algorithms, key-exchange protocols, and so on come from previously unknown figures. Pick any random open-source project and try to guess who created it as if you didn't know; most people wouldn't do a good job, and thinking "oh, it has something in common with Twitter so it must have been Jack Dorsey" would have a terrible success rate. And, as it happens, well-known people tend to like fame and don't try to stay anonymous.
I could name hundreds of smart, experienced, and relatively unknown tech-company employees, professors, and open-source contributors who would have all the necessary broad skills to have written Bitcoin. It's not that rare. The contribution itself is rare but that's like lightning; the insight of invention strikes unpredictably. Assembling high-level systems that use cryptography isn't something that requires that you post to mailing lists or message boards of cryptography enthusiasts. And even if that weren't true, there are hundreds of useful, moderately used cryptosystems that most people haven't heard of and that would never make the radar of a NYT author who (evidently) thinks it's a coincidence that two different cryptosystems might use asymmetric ciphers.
The NYT article boils down to "I had a hunch from this person's body language and, out of a community of a few hundred identified figures, I found a methodology to confirm that it was him." It's a hair better than that and honest with regard to some of its own limitations, but in the end it's no more convincing than that. The upshot is that there's some circumstantial evidence in favor of a particular person, although there's significant circumstantial evidence in the other direction too. It's astonishing to me that the NYT published this piece.