Get more out of the forecast.

Two free public sources. Wholesale: daily closes of front-month RBOB gasoline futures (the U.S. wholesale benchmark), from Yahoo Finance's public quote data. Retail: the U.S. Energy Information Administration's weekly average price for regular gasoline, via the St. Louis Fed's public FRED service. Both are linked from the app so you can inspect them yourself.

Gas stations price off what it costs to replace the fuel in their tanks. When wholesale gasoline rises, pump prices typically follow within one to two weeks; when it falls, pump prices drift down (usually more slowly). Watching wholesale therefore gives a few days of warning.

Two numbers, both in cents per gallon. Momentum: the change in wholesale futures over the last five trading days. Pressure: the app measures the typical markup between wholesale and the pump over the last eight weeks, then checks whether today's pump price is above or below the level today's wholesale price implies. The two are blended into an expected pump-price move for the next several days; more than about three cents in either direction becomes a rising or falling call.

We replayed the exact shipping model over 21 years of public price history — every trading day from April 2005 through August 2026, 5,381 daily forecasts, scored against EIA's weekly U.S. average pump price. On days the model said rising, the pump price was higher a week later 80% of the time (1,643 of 2,053 days), versus 48% of all days, with a median rise of about 4¢. On days it said falling, the price was lower a week later 85% of the time (1,543 of 1,818). On steady days, the average moved less than a nickel 83% of the time. The bigger the wholesale move, the stronger the follow-through: after wholesale jumped 10¢ or more in five trading days, pump prices were up within two weeks 84% of the time (691 of 824 days), by a median of about 9¢.

Honest caveats: these are backtested statistics, not a live track record; daily forecasts overlap, so those days are not independent events; and the score is against the national average, not any one station. The model reads the ordinary week-to-week drift of prices — it cannot see a refinery fire or hurricane coming, and small expected moves are within noise. That is what the steady call and the confidence level are for.

In the same 21-year replay, when the model called rising or falling with high confidence, the direction was right 84% of the time (2,974 of 3,554 days). With medium confidence, 67% (212 of 317 days). Treat low confidence as no strong signal — it usually means the two signals point in opposite directions or the source data is stale.

The forecast is for the U.S. average. Your station's price also reflects state and local taxes, local competition, and how recently its tank was refilled. Local prices usually move in the same direction as the national average, but the size and timing vary.

About once a day, which matches how often the underlying signal meaningfully changes. The app refreshes when you open it (if the current forecast is more than a few hours old) and attempts one background check per day. Pull down on the main screen to refresh manually.

Alerts are sent only when the forecast is rising — never for steady or falling. Also, iOS decides exactly when apps may run in the background, based on your charging and usage patterns, so an occasional day can pass without a background check. Opening the app always fetches the latest forecast.

No. It never asks for location. Its only network activity is fetching two public price series — the same data anyone can open in a browser. There is no account, no analytics, and no server.

No. Fuel Forecast is an informational aid about retail gasoline prices. It does not provide investment, trading, or financial advice, and it should not be used to trade fuel or energy markets.

The app shows the most recent forecast it fetched, marked with when it was generated. If the forecast is more than a day old, a notice reminds you it may be out of date.

If the weekly retail series is unavailable, the app falls back to wholesale momentum alone and says so in the explanation. If the wholesale source is unavailable, the app keeps showing the last good forecast with its age, and explains that it could not update. Email support if a source seems to be down for more than a day or two.

There is nothing stored anywhere except a small cache on your phone (the latest prices, forecast, and your alert preference). Deleting the app deletes it, subject to any device backup retained through your Apple settings.