I think the biggest mistake of the video, is when Lovins says that renewables are no different from baseload power plants, because baseload plants are down some fraction of the time also. He claims that power companies already compensate for downtime of baseload power plants by just having a few extra power plants. He claims that the same thing could be done with renewables.
That's just all wrong, in my opinion. It's a statistical error. Although baseload power plants are down 10-20% of the time, they are down at random. The downtime of any one plant is not correlated with the downtime of any other. As a result, if you have enough plants, then 10-20% of power generation is offline at any given time, as a result of the law of large numbers. That can be compensated for by building a few extra power plants.
With renewables, their downtime is not random. Their downtime is correlated with that of the other plants. For example, when the sun goes down, all solar panels stop working at the same time in a geographic region. Also, when the wind stops blowing (which can happen over a wide area), all windmills in that region will stop working at the same time. This is a much bigger problem than randomly distributed downtime.
If solar panels had randomly distributed downtime, and were as likely to generate power during winter nights as during summer days, then no storage would be required. We could just build more solar panels. This is because the randomly distributed periods of downtime of the solar panels would "cancel out" each other. However, it does not help to build more solar panels for the night time.
That is why renewables require storage.