Plug-In Solar in 30 US Cities: Balcony Output and Savings (2026 Data)

Most online estimates for balcony solar, including some of our own, start from a rule of thumb: take what a well-tilted panel would make and knock off about 30% for vertical mounting. We wanted real numbers for the way people actually install these kits, so we ran them. This page shows what we found and how we did it, so you can check or reuse the figures.

How we ran the numbers

• Tool: PVGIS, the European Commission’s free photovoltaic calculator (version 5.2 PVcalc, queried through its public API on September 23, 2026).
• Solar data: PVGIS-NSRDB satellite data, the long-term US solar record originally produced by NREL (now the National Laboratory of the Rockies), averaged over 2005-2015.
• System: 800 W of crystalline silicon panels, 14% system losses (the PVGIS default), free-standing mounting, the terrain horizon PVGIS computes for each location, and no shade from buildings or trees.
• Five positions: vertical facing south, east and west (a panel on a balcony railing), tilted 30° facing south (a stand, yard or deck) and flat.
• Locations: downtown coordinates for 30 cities in 25 states plus Washington, DC, chosen to cover the states with plug-in solar laws and the largest markets.
• Savings: yearly kWh × 75% used at home × the state’s average residential electricity price for June 2026 from the EIA. Exports earn nothing. The EIA table we used has no figure for DC, so DC has output but no savings figure.

Yearly output of an 800 W kit (kWh)

Rounded to the nearest 10 kWh. Modeled with PVGIS 5.2 and NSRDB data; no shade from buildings or trees. PVGIS puts the year-to-year variation at about 1-7% depending on the city.

Finding 1: a railing costs more in the South

A vertical panel on a south-facing railing delivers a median of 67% of what the same panel would make tilted at 30°. But the spread is wide:

The reason is the height of the sun. In the North the sun stays low for much of the year and strikes a vertical panel at a useful angle. In the South, and above all in Hawaii, the summer sun is almost overhead and grazes a vertical panel. Our old rule of thumb of about 72% holds in the northern states but is too optimistic for the Sun Belt, and we have corrected the articles that used it.

Finding 2: vertical panels peak in winter in the Southwest

A tilted panel follows the calendar you would expect, with its best months in spring and summer. A vertical south-facing panel does the opposite in sunny, southern places:

Monthly kWh for an 800 W kit, vertical on a south-facing railing. PVGIS 5.2, NSRDB data.

In Phoenix the railing kit makes about 104 kWh in December and 27 kWh in June. In Denver it is 91 against 42, and even in New York December (63 kWh) beats June (41 kWh). Only the cloudy Pacific Northwest keeps a summer peak: Seattle’s gray winters outweigh the better angle. For households with winter lighting and heating loads this is good news: the output arrives when the house uses more power. See a year of balcony solar, month by month.

Finding 3: flat beats vertical everywhere

A flat panel on a deck, flat roof or in a yard produced 83-93% of the tilted figure in every mainland city, and 97% in Honolulu. That is well above any vertical position. In Honolulu a flat panel makes about twice as much as the same panel on a south railing. If you have a flat, unshaded spot, it will almost always beat the railing. Keep panels off a balcony floor, though, where the railing itself shades them.

Finding 4: east or west depends on your weather

Across all 30 cities, a vertical east- or west-facing panel made 42-53% of the tilted figure, so the two sides are close on average. Local weather tips the balance:

• Coastal California favors west: morning marine clouds cut east-facing output. West beats east by about 16% in Los Angeles, 15% in San Francisco and 26% in San Diego.
• Inland mountain and desert cities lean east: afternoon clouds and storms give east a small edge in Denver (+9%), Salt Lake City (+4%) and Phoenix (+5%).
• The Northeast leans slightly east, by roughly 0-12% (New York has the largest gap).
• The Pacific Northwest favors west: by about 12% in Seattle and 7% in Portland, Oregon, where mornings are often gray.

Which side suits you also depends on when you use power; read balcony solar facing east or west.

Finding 5: price beats sunshine

Output only matters once it is multiplied by what you pay for electricity. Ranked by yearly savings:

Savings = modeled kWh × 75% self-consumption × the state’s June 2026 average residential price (EIA). Your utility’s rate, time-of-use plan and how much of the output you use can move these numbers a lot. Washington, DC is left out because the EIA table we used has no DC figure.

Honolulu tops the list because Hawaii’s electricity costs over 50 cents per kWh, but only if the panel is tilted or flat: on a south railing it saves less than half as much. California’s four cities come next. Boston, Portland (Maine) and New York beat Phoenix and Las Vegas, where cheap power undoes the desert sun. Seattle, Portland (Oregon) and Miami sit at the bottom, each for a different reason: cloud, cheap hydropower and the grazing tropical sun on vertical panels.

Finding 6: the data source matters

We first ran the same 150 cases with PVGIS 5.3, which for US locations offers the ERA5 weather reanalysis instead of satellite data. For tilted panels its yearly totals were within 5% of the satellite figures in about two thirds of the cities, but 10-18% higher in Denver, Chicago, Buffalo, Salt Lake City and San Diego. If you check your own address in PVGIS, choose the NSRDB satellite database where it is offered. PVWatts uses NSRDB data by default.

What this means if you are buying a kit

• Use a realistic factor for your region. On a south railing, plan for about 70% of the tilted figure in the northern states, 60% in the South and under 50% in Hawaii.
• Look for any tilt you can get. In most cities, tilting the same kit to 30° adds 40-85% over vertical, and more than doubles it in Honolulu.
• Do not dismiss a railing in the North. In Boston or Minneapolis a railing kit gives up only about 30% and produces steadily through the winter.
• Run your own address. The savings calculator gives a quick estimate, and what can my kit power shows how much of it you would actually use.

Using these figures

You are welcome to quote or chart these results with a link to this page. Every figure is modeled, not measured: real kits lose output to nearby buildings, trees, dirt, heat and inverter limits. If you run a plug-in kit in one of these cities and log its production, we would like to compare notes.

Frequently asked questions

Sources