Insight note · 11 November 2026
US battery storage: operating dates after construction
Anvilmore Research Team

For battery storage owners planning around an operating month, how much protection does an extra calendar month offer once construction is reported complete? In two recent US cohorts, the share of nameplate capacity recorded operating by that cutoff was much lower in Texas than in California.
The comparison uses the same reporting status and the same October–December window. Anvilmore examined battery units in the September 2024 and September 2025 US Energy Information Administration inventories that were reported construction complete, had yet to enter commercial operation and were scheduled to start that October, November or December. We compared those scheduled months with the operating months in the following September’s inventory. The measure is the share of original nameplate megawatts, a unit’s rated power, with a recorded operating month at or before each cutoff.
What one extra month covered in the records
| September baseline | Texas cohort | Texas operating by end of month after planned start | California cohort | California operating by end of month after planned start |
|---|---|---|---|---|
| 2024 | 1,353.4 MW | 55.7% | 247.7 MW | 95.3% |
| 2025 | 1,776.9 MW | 24.9–47.5% | 400.8 MW | 95.8–96.1% |
The cutoff is relative to each unit’s planned month: for an October start, it is the end of November. The 2025 ranges reflect conflicting entries. The follow-up inventory gave two Texas units and one California unit operating dates in or before the baseline September, when the units were still listed as planned. The low end excludes those units from the operating total; the high end credits them as on time. Both retain them in the original cohort. Removing each state cohort’s largest plant also leaves the ordering intact.
By the end of the third month after the scheduled month, the Texas shares rose to 64.5% in the 2024 cohort and 37.8–60.3% in the 2025 cohort. These are preliminary records from two autumn cohorts. The differences describe how rated megawatts were distributed across reported dates; they do not establish why the dates differed or give the odds of delay for an individual unit.
The date that governs the commitment
ERCOT’s historical planning guide treats energisation, synchronisation and commercial operation as separate stages. California ISO’s market monitor also distinguishes entering service from the date a battery’s resource adequacy contract takes effect. The EIA operating month therefore cannot stand in for a particular contract milestone or the amount of capacity a buyer can count.
When comparing Texas and California portfolios, test the cost of carrying a dated commitment beyond one extra month separately for each. The Texas cohorts give a reason to examine a longer scenario; the California shares show why a pooled benchmark can mislead. For each commitment, the question remains which milestone governs it and what evidence supports that milestone’s date.
Sources
US EIA, preliminary monthly generator inventory, September 2024–2026 workbooks; ERCOT Planning Guide, §5.5; California ISO Department of Market Monitoring, 2023 Special Report on Battery Storage, §2.10.1.