But now you have a probabilistic system. Your battery part is designed for n numbers of low/no solar/wind input. So you are paying for a system that would be sufficient for x% of typical/historical years.
Which has to factor in the design and cost calculation.
nuclear is also probabilistic: in 2022 half of France nuclear reactor capacity was offline because of premature aging of some components in a design used in many reactors.
Note in case SMR become part of our grid: what if something similar happens to your hundreds of produced and deployed SMR?
There are always unknown unknowns that might be correlated, like that flaw in the component. Weather systems like a https://en.wikipedia.org/wiki/Dunkelflaute are known facts. The good thing is that we do have excellent weather data for the last 75 years, so it is totally feasible for proponents of renewables to run their models through that historical data and say: Look, if we have that amount of renewables and that amount of batteries that would have been enough for the last 3 quarters of a century.
That would be go a very long way to convince me.