First-half 2026 US grid additions show storage is scaling almost as fast as solar, and that changes how operators should write load and interconnection plans

The first half of 2026 delivered a blunt data point for anyone planning power supply, site selection, or resiliency: batteries are now entering service at a scale that’s hard to treat as a pilot category.

From January through June, 368 utility-scale plants began operating in the U.S., dominated by solar and battery projects, according to an analysis by Inside Climate News based on the U.S. Energy Information Administration’s EIA-860M generator dataset. Solar led on both project count and capacity, but storage was close enough behind to change how operators should think about “new capacity” when they’re writing contracts and forecasts.

The numbers: solar still leads, storage is no longer a footnote

Inside Climate News counted 207 utility-scale solar projects and 95 battery projects reaching commercial operation in the first half of 2026, using EIA’s monthly generator updates. After those two categories, the pipeline thins quickly: natural gas combustion turbines (22 projects) and natural gas internal combustion engines (21) followed, with onshore wind at 9 projects and combined-cycle gas at just 3.

Capacity tells the same story. Inside Climate News reported 11,458 MW of new utility-scale solar and 8,207 MW of batteries coming online in the six-month period, versus 5,473 MW of onshore wind, a figure heavily influenced by the SunZia buildout in New Mexico. The three largest categories of new gas plants added 2,707 MW combined in that timeframe, the outlet reported from the EIA dataset.

That distinction is where enterprise operators can get tripped up if they treat the EIA table as a simple scoreboard. Solar MW and battery MW don’t behave like combined-cycle MW. Inside Climate News flagged the basic caveat, solar doesn’t produce at night, while many gas peakers run only during narrow windows of high demand. The capacity labels can be comparable on paper and very different in delivered energy and reliability value.

Why “nameplate MW” is a weaker planning unit in 2026

An Ars Technica forum discussion of the same “so much solar” reporting put a finer point on the metric problem: a commenter noted capacity factor as a way to translate nameplate capacity into delivered energy, citing typical averages of about 25% for grid-scale solar and about 35% for onshore wind, and observing that many battery installations are designed around peak windows with a nominal runtime of roughly four hours. Those are rough heuristics, not EIA statistics, but they reflect the modeling move many planners now need to make explicitly.

For operations and procurement teams, the practical shift is to stop letting MW be the only unit that makes it into the statement of work. Battery duration, round-trip efficiency, augmentation schedules, and dispatch limits will determine whether a “400 MW battery” actually behaves as a capacity hedge for a data center campus, a manufacturing plant with a fixed demand charge, or a fleet depot with a predictable charging block.

Inside Climate News pointed to one large example that mixes the categories: the Green River Energy Center in Utah, described as 800 MW total, split between 400 MW of solar and 400 MW of batteries. Pairings like that are becoming a default build pattern because they can deliver a shaped output profile that better matches peak pricing and peak reliability needs than standalone solar.

Geography matters: Texas’ build mix is a reference case for peak-risk strategy

Where these projects came online is as operationally important as what they are, because congestion, curtailment risk, and interconnection lead times are local. Inside Climate News reported that Texas, New Mexico, and Arizona accounted for about half of new U.S. generating capacity in the first half of 2026.

Texas led, and its mix is a concrete planning reference for any operator exposed to ERCOT-style peak dynamics. Inside Climate News’ EIA-based tally put Texas additions at 3,067 MW of batteries, 2,311 MW of utility-scale solar, and 1,154 MW of new capacity from natural gas combustion turbines or natural gas internal combustion engines in the first half of 2026. The outlet also described Texas policy choices as prioritizing gas and batteries to support reliability during high-demand periods.

This matters for enterprise energy buyers for a simple reason: the hedge value of a PPA or tolling agreement increasingly depends on when and where it can deliver, not only what the levelized cost looks like. In regions where solar is plentiful, the risk shifts toward midday price collapses and curtailment. Storage changes the shape, but it also introduces a new constraint, duration, that has to be priced and contracted.

What to put in front of procurement, facilities, and grid teams this quarter

The EIA-860M data is a generator inventory, not an enterprise playbook. But the first-half 2026 mix, as interpreted by Inside Climate News and debated in Ars Technica’s technical back-and-forth, points to a few concrete work items for operators who buy power, manage peak exposure, or write interconnection and resiliency plans.

• Rewrite internal benchmarks from “MW contracted” to “peak hours covered at X MW for Y hours.” If a battery project is marketed at nameplate MW, require an offered duration ladder (2-, 4-, 6-hour) and its price deltas, and model those against your site’s coincident peak profile.
• Treat curtailment and congestion as contract terms, not afterthoughts. In solar-heavy regions, add explicit settlement logic for curtailed output (or for nodal vs hub basis risk) in PPAs, because 11,458 MW of new solar in six months (Inside Climate News/EIA) suggests midday saturation conditions will keep intensifying.
• Ask interconnection and EPC partners for an “in-service path” that’s grounded in current build patterns. The fact that 95 battery projects reached operation in six months (Inside Climate News/EIA) means queues, equipment supply, and commissioning practices are maturing, but they’re also getting crowded. Confirm transformer lead times, protection schemes, and commissioning test plans early, especially for co-located solar-plus-storage.