Snow Day Predictor Methodology

📅 February 15, 2026 ✍️ Snow Day Predictor Team ⏱️ 6 min read

This page documents how the current prediction score is produced. The model is a rule-based weather score that converts forecast conditions into a probability estimate. It is intended as an early planning signal and not as an official closure decision.

1. Data Inputs

Predictions are built from forecast data returned by the Open-Meteo API, including hourly and daily fields for:

2. Location Resolution

When a postal code or ZIP code is entered, the site resolves the location using:

3. Scoring Framework

The app computes a weighted score by combining weather risk factors. Factors that generally raise the score include heavier snowfall totals, fast snowfall rates, strong morning accumulation, low visibility, severe cold, and blowing snow conditions.

Important: Probability is capped below 100% because forecast and operational uncertainty always remain.

4. Time Context Logic

The calculation uses a day-selection rule to decide whether to score today or tomorrow morning based on current time. Morning commute windows receive additional weight because that period is often most relevant to transportation safety decisions.

5. What This Model Does Not Include

6. How To Use The Estimate Responsibly

Use this score for planning (wake-up alarms, childcare backup, commute prep), then confirm the final status through official school board and transportation channels. A high score is not a guarantee. A low score does not eliminate late changes.

7. Change Log

Related pages: Official Sources and Editorial Policy.