Of the 368 plants that came online, 302 were solar or batteries

The list that counts

368 utility-scale plants started operating in the United States between January and June. Solar led on project count with 207 and batteries followed with 95, against 22 gas combustion turbines, 21 gas engines, 9 onshore wind projects and 3 combined-cycle gas plants. In megawatts the same list gives 11,458 for utility-scale solar, 8,207 for batteries, 5,473 for onshore wind and 2,707 across the three largest gas categories together. Combined-cycle gas has enormous announced capacity in development and delivered 3 finished plants.^[1]

A megawatt of completed capacity and a kilowatt-hour delivered to a customer are separate quantities, and the gap between them runs in both directions here. Solar produces nothing at night; many of those gas plants run only a few days or weeks a year, when peak demand needs them. Small-scale solar, rooftop included, sits outside the database altogether, so the household half of the build-out is invisible in these figures. For climate accounting that matters, because avoided emissions are counted from electricity generated and not from steel in the ground. What the list does settle is narrower and firmer than a forecast, and it is the sturdiest evidence available this early in the year: this is the equipment that exists.^[1]

Where does the wind number come from?

5,473 megawatts of onshore wind looks like a healthy half-year until the composition is opened. SunZia Wind South and North in New Mexico account for 3,650 of it, a pair of farms about a decade in the making and now the largest wind capacity in the country. Texas, which leads the United States in onshore wind, connected no new wind project at all in those six months; it added 3,067 megawatts of batteries and 2,311 megawatts of solar instead. Read that way, the wind line reflects a pipeline filled years ago rather than current appetite for onshore wind.^[1]

The discipline this list imposes is the same one applied here three days ago to municipal removal pledges, where only 18 per cent of the promised tonnage had a verifiable number behind it. Announced capacity and completed capacity are measured on different registers, and only one of them shows up in a grid. Ratepayers, system operators and planners meet the second register; the developers' announcements are a claim about intent, and the interconnection queue decides how much of that intent survives.^{[1], [2]}

The next list

The planned-plant list points to a continuing solar build-out and a gas build-out arriving in the final years of the decade, with electricity demand growth and the data-centre share of it left open. That makes the second-half completions list a usable test rather than a guess: if the shape holds, combined-cycle gas will close 2026 far below the capacity announced for it while solar and batteries again lead completions. If instead finished combined-cycle projects jump into double figures, the gas wave has arrived earlier than the completions data so far suggest.^[1]

The lever sits lower down the table than the headline farms. Illinois came first in the number of new utility-scale solar projects with 36, ahead of California with 22 and Texas with 16, and its projects ran small; the reporting attributes that to state law encouraging small and mid-size solar, which turns into finished plants. Eric Gimon of Energy Innovation makes the same point from the engineering side, that smaller projects are easier to build and have far more usable sites. Whoever writes siting and permitting rules for mid-size solar and storage is writing next year's completions list.^[1]