Real-World Solar Production in Los Angeles: Measured on 89 Systems
A typical rooftop solar system in our Los Angeles-area monitoring fleet produced 1,456 kWh per kW of panels a year — the median of 385 complete system-years measured from 2019 to 2025. The middle half of results ran from 1,311 to 1,557, about 4.0 kWh per kW a day on average. Below: how results spread across systems, how they changed year to year, and how to compare your own system.
How much 1 kW of solar produces in LA
Each system’s median year, all 89 systems.
| Measure | p10 | p25 | Median | p75 | p90 |
|---|---|---|---|---|---|
| All system-years, kWh per kW | 1,220 | 1,311 | 1,456 | 1,557 | 1,666 |
| Per system (each system’s median year) | 1,242 | 1,310 | 1,449 | 1,531 | 1,644 |
| Average per day, kWh per kW | 3.3 | 3.6 | 4.0 | 4.3 | 4.6 |
| System-years: n = 385. Systems: n = 89. p10 means 1 in 10 results were lower. | |||||
Read it as a benchmark: if your system made less than 1,220 kWh per kW in a full calendar year, it did worse than 9 in 10 of the system-years we measured. That is worth a look — shade that has grown, a string or optimizer offline, or an inverter fault. Above 1,557 is better than three-quarters of them, usually a south- or west-facing roof with little shade.
To get your own number, divide a full year of production from your monitoring app by your system’s DC size in kW (the size on your contract or permit). For a single month, use the monthly benchmark.
In kilowatt-hours: a 6 kW system at the median rate makes about 8,740 kWh a year (7,870–9,340 for the middle half). Other sizes, single panels and sizing from your usage are in output by system size.
Year by year: output moves with the weather
Median and middle half per year.
| Year | Systems | Median kWh/kW | Middle half | Change vs prior year, same systems |
|---|---|---|---|---|
| 2019 | 11 | 1,478 | 1,364–1,550 | — |
| 2020 | 34 | 1,498 | 1,366–1,555 | -0.7% (n=10) |
| 2021 | 53 | 1,554 | 1,409–1,629 | +2.3% (n=34) |
| 2022 | 69 | 1,550 | 1,451–1,658 | +1.1% (n=47) |
| 2023 | 75 | 1,416 | 1,305–1,508 | -10.7% (n=63) |
| 2024 | 75 | 1,396 | 1,263–1,480 | -1.8% (n=70) |
| 2025 | 68 | 1,410 | 1,276–1,498 | +0.6% (n=64) |
Comparing each system with itself removes the effect of the fleet growing. On that basis 2023 stands out: the median system produced 10.7% less than in 2022, and even the best-performing tenth fell by 2.1% or more. Almost all of the shortfall came from January through June; July through December 2023 stayed close to other years. When nearly every system moves together, the cause is shared — weather, smoke or dust — rather than any one installation. Panel aging cannot explain a change this size: typical home systems lose 0.5–1.3% a year (degradation data). The practical lesson: judge a system against the same months of other systems, not against last year alone.
Real rooftops versus an idealized roof model
PVWatts, NREL’s free solar model, describes one ideal roof: facing due south, tilted 20°, never shaded. Real roofs are not like that.
This comparison sets every roof in the fleet against that one idealized roof. It is not a comparison with any project’s own estimate, which should reflect that roof’s direction, tilt and shade.
| Period | Measured median | PVWatts ideal roof* | Measured as % of ideal |
|---|---|---|---|
| Full year | 1,456 | 1,685 | 86% |
| April–July | 611 | 646 | 95% |
| November–January | 233 | 326 | 71% |
*PVWatts v8, downtown Los Angeles, 20° south-facing, 14% system losses — Cali Energy runs published in How much sun LA gets (Northridge models 1,710 kWh per kW a year). Measured: monthly medians summed over the period.
In late spring and summer, real roofs come close to the ideal (95%). The gap opens in winter (71%), when a low sun stretches shadows from trees, neighboring houses and chimneys, and east- or west-facing panels lose more of the short day. If an estimate you were given assumes an unshaded south-facing roof, ask how shade was modeled. The winter study has the month-by-month picture.
Methodology
How the numbers were produced.
- Data source
- Monthly AC energy recorded by SolarEdge inverters in Cali Energy’s monitoring fleet. These are readings from the monitoring system, not from utility revenue meters.
- Systems
- 96 monitored systems; 4 excluded (1 customer system we service but did not install, 1 expanded mid-period, 2 whose output does not match the capacity recorded in the monitoring account). 89 systems have at least one complete system-year.
- Complete system-year
- One system over one calendar year, 2019–2025, with production recorded in all 12 months. Years below 75% of the same system’s own median were treated as downtime and kept out of the benchmark (3 of 388). Result: 385 complete system-years. 2026 is incomplete and not used.
- Unit
- kWh per kW: yearly or monthly energy divided by the system’s DC nameplate rating (kW) as recorded in the monitoring account. It removes system size, so a 4 kW and a 12 kW system can be compared.
- Statistics
- Medians and percentiles (p10, p25, p75, p90), not averages, so a few unusual systems do not move the result. “Middle half” means p25–p75.
- Geography
- Assigned by ZIP code: 49 systems in the San Fernando Valley and Burbank, 21 elsewhere in Los Angeles County, 4 in the Santa Clarita Valley, 4 in Ventura County and 11 elsewhere in Southern California.
- Limits
- Not a random sample of Los Angeles roofs: these are systems we monitor, weighted toward the San Fernando Valley. Roof direction, tilt and shade vary from system to system and are not adjusted for. Customer names and addresses are never published; only aggregates are.
- Updates
- Recalculated when a calendar year closes. The URL stays the same; this page shows the date of the last update.
Download the data
https://cali-energy.com/research/los-angeles-solar-production. Aggregated data are free to use under CC BY 4.0 with a link to this page.Frequently asked questions
How much electricity does 1 kW of solar produce in Los Angeles?
In our measured fleet, a median of 1,456 kWh a year per kW of panels, with the middle half between 1,311 and 1,557. That averages about 4.0 kWh per kW a day.
How do I compare my solar system with this benchmark?
Take a full calendar year of production from your monitoring app and divide it by your system’s DC size in kW, from your contract or permit. Compare the result with the median of 1,456 and the middle half of 1,311–1,557 kWh per kW.
What counts as low solar production in LA?
A full year below 1,220 kWh per kW was lower than 9 in 10 of the complete system-years we measured. It is a screening signal: check for new shade, offline panels or inverter alerts.
Did solar production drop in 2023?
Yes. On the same systems, the median 2023 result was 10.7% below 2022, and the shortfall came almost entirely from January through June. Nine in ten systems fell together, a fleet-wide pattern.
Does panel aging explain year-to-year changes in solar output?
Only a small part. Typical home systems lose 0.5–1.3% a year as they age, so a drop of several percent in one year points to shared causes such as weather.
More research
Next steps
Below the benchmark, or planning a system?
Prepared by Cali Energy, October 7, 2026. Figures describe systems in our monitoring fleet and are not a guarantee of what any specific roof will produce; orientation, tilt, shade, equipment and weather change results. Not engineering, financial or utility advice. See our Content Disclaimer.