Solar Performance Ratio (PR): formula, averages and what is good
The headline efficiency number for a PV plant, independent of weather.
Performance Ratio (PR) is the ratio of a PV plant’s actual energy yield to the yield it would have produced at its nameplate efficiency under the irradiation it actually received. It normalises out the weather, so it is the single number that says how well the plant is converting available sunlight. Average PR for modern plants is roughly 80 to 90 percent (Fraunhofer ISE); measured weather-adjusted fleet averages in the US run 91 to 94 percent, while unmanaged distributed fleets can sit near 79 percent.
Definition
PR divides measured AC energy by the theoretical energy from the plane-of-array irradiation at STC efficiency. Because it cancels the irradiance the plant received, a falling PR points to losses inside the plant — soiling, degradation, downtime, clipping, thermal losses — rather than to a cloudy month. It is the EN 61724 benchmark metric for plant health.
Formula
PR = actual energy yield ÷ (POA irradiation × nameplate DC ÷ STC irradiance)
Typical range
Well-run utility-scale plants: PR ≈ 0.80–0.85. New plants can exceed 0.85 in mild climates; hot desert sites sit lower because of temperature losses. A PR below ~0.75, or a downward trend, signals recoverable loss.
Why it matters
PR is the term in nearly every O&M contract and performance guarantee, and the metric lenders track. A 2-point PR slip on a large plant is six figures a year in lost generation. Because it strips out weather, a falling PR is the earliest honest signal that something inside the plant — not the sky — is costing yield.
How NuraVolt tracks it
NuraVolt computes PR continuously from POA irradiance and AC output, decomposes the gap to nameplate into named loss buckets (soiling, temperature, availability, clipping, degradation), and trends each — so a PR decline is attributed to a cause and a euro figure, not just flagged.
Average performance ratio: published benchmarks
There is no single "average PR" because studies measure different flavours of the metric: raw PR includes temperature losses, weather-adjusted PR corrects them out, which is why fleet studies report higher numbers than raw rules of thumb. The table below only contains published, sourced figures.
| Fleet / era | Performance ratio | Source |
|---|---|---|
| Modern plants, industry-wide rule of thumb | About 80 to 90 percent (raw) | Fraunhofer ISE, Photovoltaics Report |
| Plants built before 2000 | About 70 percent | Fraunhofer ISE, Photovoltaics Report |
| US fleet study, 250 systems, 157 MW | 93.5 percent average (adjusted) | Deline et al. 2020, via US DOE performance report 2022 |
| California fleet, 2,200 reporting systems | 91.7 percent weather-adjusted average | Walker et al. 2019, via US DOE performance report 2022 |
| Germany, 100 systems, 2010 | 70 to 90 percent, median 84 percent | Reich et al. 2012, Progress in Photovoltaics |
| US federal buildings, 75 mostly unmanaged systems | About 79 percent average when running | US DOE performance report 2022 |
| 1980s-era systems | Around 70 percent | IEA-PVPS Task 13 long-term performance report |
Two more published numbers frame what "good" means. NREL’s O&M best-practices report (3rd edition) estimates that comprehensive O&M could lift the average age-and-temperature-adjusted PR of US systems from 91.7 percent to at least 95 percent, which is the recoverable-loss argument in one sentence. And kWh Analytics’ Solar Risk Assessment 2025 found US PV sites underperforming their P50 production estimates by 8.6 percent on average across roughly 34,000 system-months, meaning the typical plant has more recoverable loss than its owner assumes. Module degradation compounds this slowly: NREL’s Jordan and Kurtz compendium puts median crystalline-silicon degradation at 0.5 to 0.6 percent per year, with the mean nearer 0.8 percent.
PR in contracts: guarantees for IPPs and utilities
PR guarantees used to be the standard performance clause in EPC and O&M contracts, and they still appear, most often where the EPC and the O&M provider are the same company. SolarPower Europe’s O&M Best Practice Guidelines now recommend availability and response-time guarantees instead, citing a minimum guaranteed contractual availability of 98 percent over a year as best practice, because an O&M contractor controls uptime but not irradiance sensors, soiling regimes, or degradation. For an IPP the practical consequence cuts both ways: if your contract carries a PR guarantee, the metering and sensor basis of the PR calculation decides disputes, and if it carries an availability guarantee instead, PR becomes your own internal health metric rather than the contractor’s liability, which makes independent PR tracking more important, not less.
Frequently asked questions
See also
PR with the weather’s heat penalty removed, so real losses stand out.
Energy produced per unit of installed capacity — the cross-site comparator.
The share of nameplate a plant actually delivered over time.
Actual generation measured against a physics model of what the plant should have made.
A leading recoverable drain on PR.
Hardware loss that drags PR down and never recovers with rain.
The platforms that track PR, compared honestly.
See this on your own plants
NuraVolt turns your SCADA and BMS data into early fault detection, degradation-aware BESS analytics, and audit-ready reporting. A fixed-scope audit shows you what we’d find on your portfolio.