First thing I notice is that the length of generations is extremely rubbery, between 16 and 29 years for the generations presented on Wikipedia. It’s almost as if those lengths are adjusted to make the available information fit the theory. Okay, not almost – that appears to be exactly what they have done.
Generation time spans can sometimes vary slightly by source, but the generation years column seems accurate enough compared to Wikipedia dates. Though do agree with information adjusted to fit the theory, because while the generation spans seem correct, some of the developments during those time frames don't fit well.
Look at the sources for those Wikipedia dates for most of the generations listed. Most of the generations prior to the Silent Generation and the Baby Boomers were invented by Howe and Strauss. The Millennial Generation, or at least the name for that generation, was also invented by Howe and Strauss. I'm pretty sure that most people didn't conceptualise generations the way we do today 100, 200 or 300 years ago (although some apparently did).
I would be interested in how Strauss and Howe came up with those rubbery start-and-finish dates for the generations. I suspect that it is part of how they made the available data appear to fit their theory. I'd probably need to read their books in order to confirm that, but that isn't going to happen.
I'd also like to see the data that they started with and how they determined which data to use. How much of their characterisation of each generation was based opinion and how much was based on hard data?
The thing is that sociology is not a hard science and should not be judged in the same way as, for example, physics. BUT data analysis is a hard science and should be judged accordingly.
A quick google finds plenty of criticism of the theory. Here is one that contains links to a couple of other articles:
https://qz.com/970646/the-world-has-already-bought-into-steve-bannons-apocalyptic-ideology/