Mostly true, but mostly beside the point.

When discussing massive numbers of electric cars, the typical concern is the additional generating capacity that would have to be added to power them. When adding new generating capacity, what matters is the efficiency of new generating plants, not the average efficiency of the decades-old ones that make up the grid now. Accordingly, the relevant efficiencies to consider really are the higher ones.

Moreover, electric vehicles are qualitatively better than gas-powered vehicles in the sense that they're much less constrained by fuel type. Peak oil is often described as not an energy crisis but a liquid fuels crisis; without that reliance on liquid fuels, electric vehicles are fundamentally different in peak oil terms.

So I'll put the question to you, how do you propose to charge all these EVs (and I guess PHEVs as well)?

With whatever's convenient, since they're not at all picky about fuel type.

One of the benefits of wide-spread plug-in vehicles would be a wide-spread network of batteries hooked up to the grid. Those would be exceptionally useful for smoothing out wind and solar PV generation more cheaply and efficiently than by adding pumped storage.

No disagreement here. Like I said, it all comes down to how you charge these EVs.

We are going to have to parallel deployment of large ev fleets with huge investments in elec generation and grid capacity.

My fear is this is going to come too little and too late.

"We are going to have to parallel deployment of large ev fleets with huge investments in elec generation and grid capacity. My fear is this is going to come too little and too late."

We won't need new generation and grid capacity for night time charging for at least 10 years, and wind can easily grow to the size needed in that time.