For a completely decarbinized grid, there are two paths: 1) 100% renewables plus storage, or 2) ~90% renewable plus storage, and 10% nuclear/advanced geothermal.
There's lots of debate about which one would be cheapest. But the true answer depends on how the cost curve of technologies develops over the coming 20 years. (Personally, I think 100% renewables will win because projections of all experts severely overestimate storage and renewables costs, while simultaneously severely underestimating the costs of nuclear. Renewables and storage are always over delivering, while nuclear always under delivers. So I think that trend will continue...)
You won't hear much about this in the popular media though, because they are too afraid of offending conservatives with politically incorrect facts. Sites like Ars Technica cover it though:
https://insideclimatenews.org/news/22092022/inside-clean-ene...
Or perhaps not, sometimes not being an "expert" in the traditional sense can remove the biases of an industry. Sci-fi author Ramez Naam had some of the most accurate forecasts in the past by doing the simplest thing possible: looking at the past curve and extending it. That is probably the simple type of projection I would make!
https://news.ycombinator.com/item?id=23185166
The IEA and EIA are two very respectable organizations that make comically bad projections, just absolutely awful. I know I could beat their projections!
Jenny Chase is a highly prominent solar analyst that has some great anecdotes about how wrong solar estimates always are, and she challenges that new analysts face, but I'm having trouble finding the podcast right now... in any case always read the Jenny Chase megathreads on the state of solar or her interviews in order to get some really great insights into what's going on.
In any case the rate of learning in solar tech far exceeds the expectations of most "energy" experts, and also usually exceeds the expectations of even the solar experts.