This article [1] on pumped hydro tempered my expectations a lot - It calculates that a nation-scale pumped hydro battery for the USA would need to be about the size of Lake Superior and have a 300 m height drop
I've read that before. The big thing he's wrong about is requiring 7 days of storage. For coal, oil, and gas you need 7 days of storage to account for supply disruptions. Like when a hurricane lashes the gulf coast.
Solar and wind while intermittent have so much redundancy they never have week long interruptions. You can look at CAISO's charts of solar output in California. You get a couple of days in the winter where production is off due to weather. But not a week.
Pretty sure the UK already ran into this problem when the North Sea had low wind for a prolonged period. This year wildfire smoke would have cut into the valuable days for many solar installations in the northern US/Canada, many hazy periods were close to a week. The intermittency problem is only going to get worse as the amount the grid relies on grows, it's easy to cover the occasional 5% loss grid wide, but a 25% loss is a big problem.
A little deeper comment is one should compare the costs associated with a supply reduction vs the cost of the storage needed to mitigate the disruption vs how often that happens.
The argument that you can't afford 7 days of renewable storage is just restating that the marginal cost of fossil fuel storage is lower than the cost of electric storage. While ignoring the ability to curtail demand.
Solar and wind while intermittent have so much redundancy they never have week long interruptions. You can look at CAISO's charts of solar output in California. You get a couple of days in the winter where production is off due to weather. But not a week.