DOE Calls Solar's Night Hours a Failure as Daytime Output Hits 40% of Mid-Atlantic Grid Supply
As the Energy Department attacked solar's nighttime limits on social media, the same technology was covering more than 40% of mid-Atlantic power supply during peak heat.
The U.S. Department of Energy posted a graphic to X on Thursday (2026-07-23) claiming that 99% of solar power in the mid-Atlantic "failed after sunset" during the early-July heat wave, a framing that Canary Media on Friday (2026-07-24) described as "misleading and deceptively provocative."6
During those same daytime peak hours, solar was at times covering more than 40% of electricity supply in the region. That figure sits in the same data the DOE circulated while arguing against renewable energy.6
The gap between the administration's rhetoric and the generation record runs deeper than one social media post. NERC's latest summer reliability assessment found that 30.5 gigawatts of new solar capacity contributed 16.4 gigawatts of usable capacity during peak summer demand periods, a peak contribution rate that grid planners now embed in reliability modeling.2
Regional variation shapes how dispatchers think about the number. NERC rated MISO's 20.4 gigawatts of solar at 60% of nameplate capacity during peak hours and assessed the region's roughly 3.6 gigawatts of battery storage at 97%. SPP's 3.9 gigawatts of solar carries a 54% peak contribution rating; the region's 1.3 gigawatts of battery storage is rated at 84%.2
In ERCOT, the shift is already visible in long-range forecasts. The EIA's Annual Energy Outlook 2026, published in May 2026, projected solar generation in the Texas grid at 78 billion kilowatt-hours for 2026, against 60 billion for coal, an 18-billion kilowatt-hour gap in a system the administration has long cited as evidence that gas remains indispensable.1
Nationally, solar overtook coal in total generation for the first full month on record in May 2026. Ember data released on Wednesday (2026-06-10) showed solar at a record-high 12.8% share of U.S. electricity supply, against coal's 12.2%, its fourth-lowest monthly share ever. Solar generation reached 45.5 terawatt-hours in May, up 17% from a year earlier; coal hit an all-time monthly low of 39.3 TWh in April 2026.4
The political geography of solar's expansion complicates the White House's position. States carried by President Trump accounted for 74% of all new solar capacity installed in the first quarter of 2026, according to SEIA and Wood Mackenzie's U.S. Solar Market Insight 2026 Q2 Report, meaning the installations the administration targets rhetorically are concentrated in Republican-leaning states.4
For natural gas traders, the picture is one of shifting dispatch hours rather than collapsing volumes. The EIA's May 2026 Short-Term Energy Outlook, now nearly two months old, projected U.S. power-sector gas consumption would hold near recent highs this summer, with a record forecast for summer 2027, even as overall electricity demand grows around 2%. Solar's midday generation compresses the hours in which gas sets the marginal clearing price, pushing demand for fast-ramping gas capacity toward the evening ramp.3
NYMEX Henry Hub front-month closed at $2.87/MMBtu on Friday (2026-07-25), offering limited incentive for new firm gas capacity investment as solar penetration continues to rise.
The evening ramp is where solar's constraint genuinely sits. Australia's experience with the duck curve, where solar peaks around midday and demand surges into evening as generation drops, illustrates the dispatch challenge the U.S. is replicating at scale. Battery storage is the partial answer; NERC's 97% peak contribution rating for MISO battery resources confirms it now qualifies as firm capacity in reliability modeling. The installed base remains small relative to the solar fleet it is designed to complement.5,2
For ERCOT, summer 2026 is the first season in which EIA projects solar outpaces coal outright. Real-time power prices on extreme-heat afternoons in Texas will test how far those midday solar surpluses actually compress spark spreads, and whether post-sunset ramps produce price spikes that gas capacity alone must absorb.1