How is ClearSkys forecast accuracy measured?

Every day ClearSkys records its own score for nights 7, 3, and 1 days ahead across 165+ tracked locations. Once each night has passed, the actual hourly weather is fetched from the ERA5 reanalysis archive and the score is recomputed with the identical scoring engine. The difference between predicted and actual score is stored and aggregated on this page.

What does 'within 10 points' mean?

The ClearSkys score runs 0–100, and 10 points is roughly the width of one quality band (for example the difference between a good and a great night). A forecast within 10 points of the verified actual would have led you to the same go or no-go decision.

Why does accuracy improve closer to the night?

Weather models sharpen as the forecast horizon shrinks. A 7-day-ahead score tells you which nights look promising; the 1-day-ahead score is the one to trust for setting up a telescope. The verified numbers on this page quantify exactly how much each lead time can be trusted.

Verified accuracy

We check our own homework.

Every forecast ClearSkys makes is recorded, then checked against what the sky actually did. No hand-picked examples, no marketing claims — the numbers below are computed from every verified night and update automatically as new nights are checked.

63.1%

of next-night forecasts within 10 points of the verified actual

±10.4

average score error, one day ahead (0–100 scale)

1,033

city-nights verified since 2026-06-21

Accuracy by lead time

Weather models sharpen as the night approaches, and the score inherits that. Use the 7-day view to spot promising nights, and trust the 1-day score for the go/no-go call.

Forecast made	Nights verified	Avg score error	Bias	Within ±10 pts	Within ±15 pts	Avg cloud error
1 day ahead	620	±10.4 pts	+7.6	63.1%	75.2%	±18.8%
3 days ahead	413	±10.6 pts	+6.5	60.3%	74.3%	±22.8%

VERIFIED NIGHTS 2026-06-21 → 2026-06-26 · UPDATED DAILY AS ERA5 DATA BECOMES AVAILABLE

Bias is the average signed difference (verified minus predicted) — negative means the forecast ran optimistic, positive means nights turned out better than promised.

The go/no-go question

Point error is abstract; what matters is whether the score would have made the right call. So we also check the decisions the day-before score implied:

98.1%

of nights scored 70+ the day before were verified 60+ (432 go calls)

62.0%

of nights scored below 40 the day before were verified below 50 (50 no-go calls)

THRESHOLDS ARE FIXED AND STATED, NOT TUNED: GO = PREDICTED ≥70 VERIFIED ≥60 · NO-GO = PREDICTED <40 VERIFIED <50

How verification works

01

Record the prediction. Every day, the score for each tracked location is computed and stored for the nights 7, 3, and 1 days ahead — exactly as the app would have shown it, using the same scoring engine.

02

Let the night happen. Nothing is touched until the night has passed and the ERA5 reanalysis archive — a globally complete, satellite-and-observation-derived weather record — has caught up, about six days later.

03

Rescore on reality. The actual hourly conditions for that night are fed through the identical scoring engine. Same code, same weights — the only difference is forecast inputs versus what actually happened.

04

Publish the difference. The gap between predicted and actual score is stored per night, per location, per lead time, and aggregated into the table above. Bad nights count the same as good ones.

Honest caveats

ERA5 is reanalysis, not a telescope. Ground truth here is the best available global weather record, not a human standing under the sky. It is spatially complete and independent of the forecast models we score with, but it is still a model of what happened.

Moonlight is held constant. The moon is identical on both sides of each comparison (same night, same place), so it cancels out. These numbers measure the weather component of the score — which is the part a forecast can get wrong.

Rain is the weakest comparison. Forecasts express rain as a probability; the archive records what fell. Cloud cover — by far the largest factor in the score — compares directly.

Questions

What does “within 10 points” actually mean?

The score runs 0–100 and 10 points is roughly one quality band — the difference between a good night and a great one. A forecast within 10 points of the verified actual would have led you to the same decision about setting up.

Why publish this at all?

Because every forecast site claims to be accurate and almost none show their working. ClearSkys exists to answer “is tonight worth it?” — and that answer is only useful if you know how much to trust it at each lead time.

How is the score itself calculated?

Cloud cover across three altitude layers, moonlight and its overlap with darkness, transparency, wind, and more — the full breakdown is in How the Stargazing Score Works.

See tonight's score for your location →