Roof direction, pitch, and shade: what they cost in output
Updated July 2026
Statewide sun figures assume a decent roof. Here is what "decent" is worth: facing the array west instead of south costs about 15%, a flat-ish pitch costs single digits, and a hard shadow across the array at 2 p.m. can cost more than both combined. These are the three dials, with real percentages on each.
Azimuth: which way the roof points
In the northern hemisphere, true south is the production maximum. The falloff is gentler than folklore says: southeast or southwest gives up only 2 to 5%, due east or due west about 12 to 18% depending on latitude, and even a north-facing array still produces 60 to 70% of the south figure, which occasionally pencils where power is expensive and roof is plentiful. One wrinkle worth money: west-facing arrays produce later in the afternoon, exactly when time-of-use rates peak, so on TOU tariffs a west roof can out-earn what it under-produces.
Tilt: the pitch of the roof
The rule of thumb says tilt equals latitude; the truth says the curve is flat near the top. Anywhere within 15 degrees of ideal costs low single digits, so almost every conventional roof pitch (4/12 through 9/12, or 18 to 37 degrees) is fine. The genuinely bad cases are flat roofs without tilt racking (soiling stops washing off, and summer-noon geometry is the only thing saving output) and steep alpine pitches, which trade summer for winter production, sometimes on purpose in snow country because they also shed snow.
Shade: the dial that goes to eleven
Tilt and azimuth are percentage games; shade is not. A shadow across one panel in a traditional string can drag the whole string toward the weakest panel's output, and a chimney shadow that walks across the array for two hours daily can erase 20-plus percent of annual production by itself. Hardware mitigates: microinverters and DC optimizers contain the damage to the shaded panel, at real cost. But the honest sequence is trim the tree, route around the chimney line, and only then buy electronics. A site survey with a shade-analysis tool (or PVWatts plus an honest eye on the yard at 10 a.m., 1 p.m., and 4 p.m.) beats every rule on this page.
Stacking the dials
Losses multiply. A west roof (0.85) at a mediocre tilt (0.97) with a modest shade line (0.90) delivers 0.85 × 0.97 × 0.90 = 74% of the table value, which turns a 22-panel calculator answer into 30 panels. That is not a reason to distrust the tables; it is the reason installers walk roofs. Use the state figure to size expectations and catch nonsense, then let the survey set the final count.
Questions people ask
Do solar panels have to face south?
No, they just earn most there. Southeast and southwest give up under 5%, east or west 12 to 18%. West facing even wins financially on time-of-use rates that price late-afternoon power highest.
How much shade is too much for solar?
A hard shadow across the array during the 9 a.m. to 3 p.m. window is the deal-breaker zone; that band carries most of the day’s energy. Early-morning or late-evening shade costs little. String systems suffer whole-string penalties from partial shade; microinverters confine it to the shaded panel.