What Is Capacity Factor? The Number Behind Every Power Plant

If you want to turn a power plant’s headline rating into real energy, capacity factor is the number you need. A “100 MW” solar farm and a “100 MW” nuclear reactor share the same nameplate, but over a year the reactor typically produces seven to eight times more electricity. Capacity factor is the difference. Once you understand it, you can read press releases correctly, compare technologies on fair terms and sanity-check any energy calculation, including the ones on this site.

The definition and the formula

Capacity factor (CF) is the energy a generator actually delivered over a period, divided by the energy it would have delivered running at full rated power for every hour of that period.

You will use the rearranged form most often, because it converts installed capacity into annual energy. One gigawatt at a 25% capacity factor gives about 2.19 TWh a year; at 90% it gives about 7.9 TWh.

Keep power (kW, MW: a rate) and energy (kWh, MWh: an amount) apart. Capacity factor links the two. If the units are hazy, read kW vs kWh first, and use the energy unit converter for MW↔TWh conversions.

Equivalent full-load hours

Many engineers, especially in Germany and across European industry, express the same idea as full-load hours (Volllaststunden): annual energy divided by rated power. In solar, the same quantity is called specific yield and is given in kWh/kWp.

A German rooftop PV system yielding 950 kWh/kWp therefore has a capacity factor of 950 / 8,760 ≈ 10.8%.

Typical capacity factors by technology

The ranges below are indicative for modern plants. Fleet averages move with weather, market conditions and policy, and any single year can fall outside these bands.

The ranges come from the U.S. EIA Electric Power Monthly (fleet averages by fuel), the NREL Annual Technology Baseline (new-build assumptions) and IRENA’s Renewable Power Generation Costs (global weighted averages for renewables). The EIA, for example, reports US nuclear fleet capacity factors above 90% in most years since the early 2000s. IRENA has tracked the global weighted-average capacity factor of new onshore wind rising from roughly 27% in 2010 to the mid-to-high 30s in recent years as rotors grew.

Our country profiles and rankings show national generation by source, from which you can infer capacity factors.

What drives capacity factor

There are two very different kinds of driver, and confusing them causes most of the misunderstandings.

1. Resource-limited technologies

For wind, solar, run-of-river hydro and tidal, nature sets most of the capacity factor. A solar panel can’t produce at night. A wind turbine produces nothing below its cut-in speed (around 3 m/s) and is capped at rated power above roughly 11–13 m/s. Engineering choices still make a big difference:

• Wind: put a larger rotor on the same generator (lower specific power, in W/m² of swept area) and it catches more energy at moderate wind speeds, raising CF, though it gives up some output in very high winds. Taller towers reach faster wind; see the wind speed at height calculator.
• Solar: tracking, tilt and orientation, temperature, and the DC/AC ratio. Many utility plants install 1.2–1.4 kW of panels per kW of inverter. The inverter clips at noon, but it runs close to full output for more hours, which raises the AC capacity factor.
• Location: the biggest lever of all. The Energy Explorer and the world energy map show how much sun and wind vary from place to place.

2. Dispatch-limited technologies

Gas, coal, oil and reservoir hydro plants could run most of the time, but they run only when the market or grid operator calls on them. Their capacity factor reflects economics, not engineering. A gas peaker at 5% isn’t broken. It is doing its job, which is to be available for the few hundred hours a year when demand is highest. In many European countries, coal plant capacity factors dropped sharply in the late 2010s and 2020s as carbon prices rose and renewables took the low-cost hours; Ember’s electricity data tracks this year by year.

Nuclear sits in between. Physically it can run almost continuously, and since its fuel is cheap relative to its capital cost, running it flat out also makes economic sense.

Capacity factor vs availability vs efficiency

People often mix up these three terms:

A wind turbine can be available 98% of the time and still have a 35% capacity factor, because in most hours the wind is below rated speed. A solar panel’s 22% module efficiency has no bearing on its 12% capacity factor. A more efficient panel needs less area for the same kW, but per kW installed it produces about the same energy.

Worked examples

Example 1: Home solar in Bucharest

The solar panel calculator estimates about 7,300 kWh a year for a 6 kWp south-facing system in Bucharest.

Example 2: How much capacity replaces a reactor?

A 1,000 MW reactor at 90% CF produces about 7.9 TWh a year. How much onshore wind at 35% CF would produce the same annual energy?

Equal annual energy does not mean equal firm capacity. The wind and solar fleets would produce at different times from the reactor, so a real system also needs storage, interconnection or flexible demand. For reliability, grid planners use separate metrics such as capacity credit or effective load-carrying capability.

Example 3: Checking a headline

A press release claims a new 300 MW onshore wind farm will “power 250,000 homes”. At a 33% CF it produces about 300 × 8,760 × 0.33 ≈ 867 GWh a year. Spread over 250,000 homes, that is about 3,470 kWh each: plausible for a European household, but too low for an average US home (about 10,500 kWh, according to the EIA). With capacity factor you can reverse-engineer the assumption in seconds.

Why capacity factor matters for cost

Capacity factor sits in the denominator of the levelised cost of electricity. Spread the same capital cost over twice as many MWh and you halve the capital component of the cost per MWh. That is why a solar farm in Chile can sell power at a fraction of the cost of an identical one in Scotland, and why wind developers chase a few extra tenths of a metre per second in mean wind speed. Try different capacity factors in the LCOE calculator to see the effect, and read LCOE explained for the full method.

There is a less obvious consequence too. As a grid adds large amounts of one variable technology, its dispatchable plants run less. Gas plants that once ran 60% of the year may run 30%, and their cost per MWh rises even though nothing about the plant itself has changed.

Common mistakes

• Using nameplate as output. “We installed 10 GW of solar” says little about energy until you multiply it by roughly 0.10–0.25.
• Mixing DC and AC ratings. US solar plants are usually reported in AC MW, European ones often in DC MWp. The same plant can show 20% (DC) or 26% (AC).
• Comparing one year with a long-term average. Wind output swings by roughly ±10% from year to year, and hydro can swing much more in drought years.
• Treating low CF as low value. Peak-shaving and seasonal output can be worth far more per MWh than baseload energy.

Key takeaways

• CF = actual energy ÷ (rated power × hours). Annual energy = MW × 8,760 × CF.
• For resource-limited sources (wind, solar), weather and design set the CF. For dispatchable plants, markets set it.
• CF is neither efficiency nor availability.
• Before judging a technology, look at CF together with cost (LCOE) and timing (when the energy arrives).

Figures in this guide are typical ranges drawn from the sources above; your own numbers depend on location, prices and equipment. Use the linked calculators for site-specific estimates.