All-in-all, I think it's pretty well done. I agree that some things he simplifies as a wash aren't; for example the direct and indirect energy consumption and CO2 output for refining, storing and transporting petroleum is significantly (orders of magnitude?) greater than the cost of fueling powerplants. For one thing, coal power plants are primiarly supplied by rail, whereas your neighborhood Shell station is supplied by a semitruck. Rail is vastly more efficient. Plus, the volume of material to transport isn't comparable.
He also ignores some non-trivial costs to ICE cars, such as the ~0.01 per mile in oil changes and other consumables, and the depreciation of the ICE engine. Counting the full replacement cost of the battery ignores the residual value of the old battery as well, which may be significant.
Overall though, it's the most cogent attempt I've seen to dispassionately explain a compliacted subject. There are simply way too many variables and implicit assumptions to make a perfect comparison, but I think this does an admirable job.
Although I think he should have arbitrarily limited it to ownership for 10 years/100,000 miles, which I expect is about than the average meaningful useful life of an automobile these days.
Brent