I wonder if Author is in fact autistic. It seems (not sure if I've got this right) that author wasn't diagnosed until his forties. I'd have expected a diagnosis of autism to occur pre-teen.
As a relatively famous example, take Bill Gross, aka the "Bond King". He only realized he had Asperger's in his seventies when it was randomly the topic of a psychiatrist he was having dinner with. And why would he seek out some kind of diagnosis? He's happily married and running fantastically successful bond funds.
It's been suggested to me that I might have Asperger's (I believe the suggestion was hostile; it was my ex-wife, who is a trained psychotherapist). If I have Asperger's, it's only in the sense that everyone has a little bit of Asperger's. But I worried about it for several years. Vague psychological diagnoses are harmful.
I feel like the author might lie somewhere on the spectrum but I was pretty sure being actually autistic was a separate and vastly more serious diagnosis.
How could one tell how common undiagnosed autism is? All you can do, surely, is divide the number of autism diagnoses by the number of - autism diagnoses.
OK, so you can presume that an autism diagnosis represents some undiagnosed history of autism. But that's presumption. Or I suppose you can make other presumptions - like, this guy's diagnosed autistic, and he functions well; so many other people who function well are probably also autistic. But that's also presumption.
The DSM diagnostic criteria for autism look pretty shaky to me. I'm not saying I don't think autism exists; I think it does exist. But if you want to chuck around statistics about autism, then you have to stick to some strict definition. "Undiagnosed autism" isn't strict, on any criteria. The DSM doesn't define undiagnosed anything.
[Edit] I'm not at all surprised that some people are diagnosed in their middle age; 40 years ago, autism was a pretty exotic diagnosis, few doctors knew much about it or were experienced with it.