What we found
Harris, Texas leads this eligible sample by total reported customer-hours. Otsego, Michigan leads after dividing by modeled customer count and two calendar years. Its volume rank is #87. Scale changes which places appear at the top.
Why the two lists answer different questions
A total asks where the largest aggregate customer-hour burden was recorded. Dividing by a fixed modeled customer count asks how large that recorded burden is relative to estimated account scale. Neither metric by itself establishes which county has the worst utility reliability or where a particular household will lose power.
Largest totals in the eligible sample
| County | 2024–2025 total | Annual hours / modeled customer |
|---|---|---|
| Harris, Texas | 185.82M | 50.8 |
| Los Angeles, California | 51.33M | 6.8 |
| Hillsborough, Florida | 40.49M | 20.3 |
| King, Washington | 32.37M | 16.8 |
| Dallas, Texas | 22.21M | 11.7 |
| Allegheny, Pennsylvania | 20.72M | 15.5 |
| Cuyahoga, Ohio | 19.65M | 16.3 |
| Fort Bend, Texas | 17.78M | 12.8 |
| Sarasota, Florida | 16.96M | 27.3 |
| Manatee, Florida | 16.73M | 31.4 |
Largest customer-adjusted burdens in the same sample
| County | Annual hours / modeled customer | Volume rank |
|---|---|---|
| Otsego, Michigan | 100.7 | 87 |
| Coffee, Georgia | 80.9 | 65 |
| Lowndes, Georgia | 79.3 | 16 |
| Ware, Georgia | 69.4 | 129 |
| Toombs, Georgia | 64.6 | 176 |
| Russell, Virginia | 64.0 | 202 |
| Richmond, Georgia | 62.9 | 11 |
| Greenwood, South Carolina | 62.6 | 50 |
| Wayne, Georgia | 61.2 | 204 |
| St. Croix, US Virgin Islands | 60.5 | 93 |
What the screen excludes
1,337 observed counties do not satisfy the full eligibility rule. The screen excludes small denominators, missing denominators, counties without positive records in both years and counties with any above-MCC value. This is a selected comparison, not a ranking of every U.S. county. Passing the screen does not prove a county’s source data is error-free.
Sources and reproducibility
WattSizer calculations from the local EAGLE-I annual CSVs. Read the full methods and quality review. Original sources: Figshare historical data, Brelsford et al., Scientific Data (2024; 2014–2022 descriptor), and ORNL’s 2025 release. The descriptor explains the collection system; it does not describe observations from 2023–2025. The source manifest contains local input SHA-256 hashes, file sizes and timestamp bounds.