Average capacity factor is basically an irrelevant metric. What is more important is the lowest seasonal capacity factor, since seasonal power storage doesn’t make sense. For the UK this is 25% for offshore wind: https://onlinelibrary.wiley.com/cms/asset/bd2bb73a-ef25-4c4b...

This means that we need to overbuild wind by >200%, to have the necessary power year round.

Solar doesn’t work, since it is basically useless in the northern hemisphere during winter, and it needs a lot of battery storage during the night.

There is also a week in February where there is no wind, which will require green methane production and gas peakers(edit: https://en.m.wikipedia.org/wiki/Dunkelflaute)

Nuclear can work year round with load following using a design like the Natrium reactor. This design also uses a lot less concrete than traditional NPP’s.

Nice train you've got the goalposts on there.

The wind and solar capacity factors are anti-correlated.

There are areas in france where the december average is >2.6kWh/d. At 54kg/kWp a modern panel on a lightweight racking system gets this works out to 1800t/TWh and it's almost all glass you can build the same net wattage now for half the price and then use the money you save to replace in 30 years (or 50 if you take real world degradation rather than predicted). Having to recycle some glass once to decarbonize now ranther than in 20 years is a reasonable tradeoff for halving the costs. The gas plants are dirt cheap and that much solar would easily power enough electrolysis to fill your february gap during the 11 months of the year when it produces far more. The electrolysers will be needed in either case for ammonia, shipping, and SAF.

If we're invoking technology that doesn't exist and costs several times more, just use something that does exist and costs several times more like a CSP plant and an HVDC cable.

Additionally for the vast majority of the world which is in transmission range of somewhere arid, CSP is strictly cheaper than the easy part of a Natrium reactor.