The "Baseload" Solar Beatdown

Even in ERCOT, "baseload" solar and storage is the most expensive energy you can get

Last week, we noted that the American energy vibe shift away from decarbonization toward prioritizing electricity reliability and affordability has forced the solar, wind, and battery Bros to adapt to the new energy landscape by changing their sales pitch as they scramble for relevance.

This week, your Energy Bad Boys are here to deliver the beatdown on one of their new favorite talking points: The Myth of Baseload Solar and Battery Storage.

What is “Baseload” Solar?

“Baseload solar” refers to the concept of using solar and battery storage to provide a consistent, reliable power supply, around the clock, similar to a traditional baseload coal, nuclear, or natural gas plant.

Delivering constant power with solar and storage will require significantly overbuilding solar capacity to meet demand during the daytime and to charge the batteries so they can be discharged at night to provide enough consistent power when the sun punches out for the evening, as you can see in the graph below:

A new study by Ember entitled Solar electricity every hour of every day is here and it changes everything even goes so far as to claim that the sunniest regions in the world can achieve 97 percent solar and storage at costs as low as $104 per megawatt hour (MWh), which they claim is cheaper than coal or nuclear.

Baseload or BS? ERCOT Edition

Earlier this year, we worked with the Utah Department of Energy Development on a project to model the standalone cost of energy resources providing a constant 250 MW of power over a five-year period, accounting for historic hourly fluctuations in wind and solar capacity factors.

For the purposes of this article, we tweaked our model to require the delivery of a constant supply of 1,000 megawatts (MW) of power 24 hours per day, 365 days per year. While the Ember analysis examined Las Vegas, we analyzed the feasibility and cost in the Electric Reliability Council of Texas (ERCOT) because ERCOT is the darling of the Solar Bros, who are the most vocal about the baseload solar potential.

So, can solar and storage provide constant baseload power in ERCOT? The short answer is yes, it can, but it will take a lot of extra capacity at an enormous cost to.

The graph below shows the percentage of daily load served by solar facilities and solar dispatched from batteries. Over the course of the year, 51 percent of the load is served by solar discharged via storage.

Meeting this 1,000 MW of demand every hour of the year with only solar and battery storage in ERCOT would require 16,000 MW of solar and 57,600 MWh of battery storage, assuming 14,400 MW of four-hour battery storage, meaning it would require an overbuild of 30.4 to 1 to meet demand.

Unsurprisingly, such a massive overbuild will be incredibly expensive. Our modeling suggests that meeting 100 percent of this demand with solar and battery storage would cost $600/MWh, which is far more than any other resource we modeled to meet the 1,000 MW demand requirement.

Less Solar and Storage = Less Cost

Meeting 100 percent of electricity demand with solar and storage is incredibly expensive, but costs fall quickly as more natural gas capacity is allowed to provide backup to the solar and storage system.

The graph below shows the cost of meeting 1,000 MW of demand 24/7 with various ratios of solar and battery storage and natural gas capacity, ranging from zero percent with a natural gas combined cycle plant to 100 percent solar and battery storage.

Using cost estimates from the Energy Information Administration’s 2025 Annual Energy Outlook, we determined that meeting this demand with a combined cycle natural gas plant would cost $37 per MWh, by far the most affordable way to meet the demand.¹

Costs increase substantially as solar and battery storage are required to shoulder more of the load. Notably, the costs of meeting the 1,000 MW demand skyrocket from $334 per MWh if 99 percent of the demand is met by solar and storage, to $600 per MWh if it is met with 100 percent solar and storage.

The costs for lower penetrations of solar and storage are lower because these scenarios require far less capacity than the 100 percent scenario. For example, the 50 percent solar and storage scenario requires seven times less installed capacity than the 100 percent scenario.

Definitely Not Cheaper than Coal or Nuclear

We also modeled the cost of meeting 1,000 MW of demand with a host of other technologies based on the EIA Annual Energy Outlook cost assumptions. The graph below shows the cost of reliably meeting the demand with various technologies.

As you can see, new light water reactor nuclear plants could meet this demand for $115 per MWh, which is much lower than the $193 per MWh for 90 percent solar and batteries. Even new coal plants with 100 percent carbon-capture and sequestration, an unproven technology, would be more affordable based on EIA capital cost assumptions, meeting the demand for $185 per MWh.

So why do Ember analysts claim that 97 percent solar and storage is lower cost than coal or nuclear? It boils down to two main fudge factors.

One, they use global cost figures for solar and battery storage projects that are much lower than the costs paid in the United States. And two, they compare these very low global solar and storage costs to the costs of new coal and nuclear as estimated by our dear friends at Lazard in their latest Levelized Cost of Energy Analysis.

Ember uses solar and battery storage cost estimates that are much lower than those provided by EIA. For example, Ember’s solar cost assumptions are 3.4 times lower than EIA’s, and its storage cost assumptions are 2.4 times lower.

This begs the question: If Ember is going to take Lazard’s assumptions for building new coal and nuclear plants at face value, why didn’t they use the capital costs Lazard uses for solar and storage, which are two to three times higher than the solar cost figures used by Ember?

We suspect it’s because if Ember used realistic solar and storage cost assumptions for the U.S. in their modeling, their numbers for “baseload” solar would look as lousy as ours. It seems like the cherries are in season, and Ember is happily picking them.

Conclusion

Solar is not, in fact, baseload now. Capital costs for solar and storage are wildly cost-prohibitive, and even using a mix of 50 percent natural gas and 50 percent solar and storage increased the system cost per MWh by a factor of 2.4.

In fact, every energy technology we examined was lower cost than meeting 90 percent of electricity demand with solar and battery storage.

Based on our modeling in ERCOT, we can confidently say Ember’s claim that 97 percent of annual load can be met with solar and storage for just $104 per MWh is pure poppycock, and this figure appears to exclude the costs of any backup resources that would be needed to keep the lights on because they say it’s cheaper to just go without power.

If you’re looking to site a datacenter and need some cost modeling done. Shoot us a line.

1. Happy birthday, America! Try to keep all of your fingers this weekend.

2. Like, Share, and Subscribe to deliver your own little beatdown to the baseload solar narrative.

1. Natural gas costs were calculated based on a 90 percent capacity value while requiring a 15 percent reserve margin at all times, a requirement we did not impose on the solar and storage scenarios.