How to forecast fuel volume at a gas station: combining OPIS, RBOB futures, weather, and competitor data.

Why fuel volume is harder to forecast now

Three shifts in 2024 and 2025 made the old “last year same week” forecast obsolete:

• Total volume decline. US gasoline demand has been on a slow downtrend since 2019 due to fuel efficiency, work-from-home, and gradual EV adoption. NACS reported total c-store fuel sales of $476.3B in 2025, down from $501.9B in 2024 (5.4 percent).
• Price volatility. RBOB futures swung from $1.95 to $2.85 within 2025 alone. Retail followed with a 2 to 4 week lag. Stations that priced off last week sat at wrong margins for half the year.
• Local competition shifts.Buc-ee's expansion, Wawa expansion outside the Northeast, and Costco fuel adoption all create discontinuous local shifts. Your trailing baseline misses these until the quarter after they happen.

Source 1: your own 13-week trailing baseline

Start with your own data. The 13-week trailing average is the standard baseline because:

• It smooths weekly noise (a slow Tuesday cancels a fast Saturday)
• It catches early trends in your own volume
• It is comparable across stores at different stages of life

Pull gallons sold per grade per day for the last 13 weeks from your fuel monitor (Veeder-Root, Gilbarco, OPW) or your POS. Calculate the daily average per grade. That is your starting point.

Year-over-year comparison: pull the same 13 weeks from the prior year. If you are up 4 percent YoY but the market is down 5 percent, you are gaining share (good signal). If you are down 8 percent against a market that is down 5, you are losing share (investigate).

Source 2: RBOB futures

RBOB (Reformulated Blendstock for Oxygenate Blending) gasoline futures trade on the CME. The price is what wholesale gasoline costs delivered to the New York harbor, in dollars per gallon, for forward contract months.

Why it matters: RBOB leads the retail rack price by 1 to 3 days and the rack leads street price by 1 to 3 weeks. A spike in RBOB today usually shows up at your pumps in 14 to 28 days.

Practical use:

• Pull the current RBOB front-month price from CME or your fuel supplier
• Compare to the 30-day average
• If RBOB is up more than 10 percent vs the 30-day average, expect retail to follow in 2 to 4 weeks, and volume to dip 4 to 8 percent during the price spike
• If RBOB is down more than 10 percent, expect a margin window of 2 to 3 weeks before retail drops, plus a small volume bump from price-sensitive customers

Most operators do not track RBOB directly. Your fuel supplier sends rack price daily, which is downstream of RBOB by a day. For better lead time, watch RBOB directly.

Source 3: OPIS LocationPro and AnalyticsPro

OPIS is the institutional fuel data provider. The two relevant products for forecasting:

• OPIS LocationPro: AI-powered network planning that publishes station-level volume benchmarks, demographic data, and traffic patterns. Useful for benchmarking your stations against the local market.
• OPIS AnalyticsPro: daily street-level pricing across millions of US stations, with margin benchmarks by market.

Cost: OPIS data is enterprise-priced. Most independents do not subscribe directly but get OPIS data through their jobber or back-office software. If you are a 5+ store operator, ask your jobber whether they share their OPIS feed. Many will for free.

Source 4: weather and travel data

Weather moves fuel volume more than most operators realize:

• A predicted rain weekend in a tourist market can shift volume 8 to 15 percent
• A hurricane forecast 3 to 5 days out can spike volume 200 to 400 percent in the panic-buy window
• A snow day in a non-snowy market can drop volume 30 to 50 percent
• A heat wave (over 100F) typically reduces daytime traffic 5 to 12 percent

Free weather sources: NOAA, Weather.gov, the regional National Weather Service office. Pull a 10-day forecast every Monday and overlay against your scheduled fuel deliveries.

Travel data sources:

• School calendars (district websites)
• State DOT traffic counts at nearby highway exits (most states publish)
• AAA holiday travel forecasts (free, published 1 to 3 weeks before major holidays)
• Local event calendars (sports, conferences, festivals)

Source 5: competitor scans

Competitor pricing moves your volume day by day. A competitor dropping 4 cents takes 3 to 8 percent of your volume within 24 hours unless you match. A competitor raising 5 cents you do not match gives you a 2 to 5 percent bump.

Get the data:

• Drive-by scans: manager or relief cashier drives the local 3-mile radius every morning and afternoon and writes down competitor prices. Free and accurate.
• GasBuddy crowdsourced data: good for a once-over but lags 1 to 6 hours behind actual price changes
• OPIS street price (paid): daily street price at every nearby station, updated within hours
• Friendly-competitor agreement: some local operators trade weekly pricing intel

The drive-by is the most reliable signal because it catches intent (a competitor lowering price by 6 cents in the afternoon is signaling weakness or a fresh load).

Putting it together: the 90-day volume forecast

Combine the 5 sources into a single forecast:

1. Start with the 13-week trailing baseline per grade per store. This is the "normal" expected volume.
2. Adjust for the RBOB and OPIS-implied price direction. Up: trim volume 4 to 8 percent. Down: add 2 to 5 percent and watch the margin window.
3. Apply weather and event adjustments for the specific 14-day window (hurricane forecast, school break, AAA holiday driving forecast).
4. Layer in competitor scan adjustments for the day-by-day call (do not change the 90-day total based on competitors, but do shift volume between days).
5. Sanity check against the prior year same-week and same conditions.

Practical accuracy: a forecast built this way lands within 3 to 7 percent of actual for most independent stations. That is good enough to right-size fuel deliveries, plan working capital, and avoid stockout or carrying cost.

What to do with the forecast

1. Size fuel deliveries

Take the 14-day forecast volume per grade per store, subtract current tank inventory minus working bottom (the unusable inventory at the bottom of the tank, typically 500 to 1,500 gallons depending on tank size), and order delivery to fill 80 to 90 percent of capacity.

2. Plan working capital

Multiply 14-day forecast volume by current wholesale rack price. That is the fuel prepay you need to fund in the next 2 weeks. Plug into your 13-week cash forecast.

3. Right-size labor

Fuel volume drives traffic, which drives inside basket. Big-volume weeks (holidays, summer peak) need more cashiers. Small-volume weeks (post-holiday slumps) can cut hours. Use the forecast to set the schedule.

4. Time price changes

If RBOB is signaling a 5 cent retail increase in 2 weeks, hold price for now to keep volume, then raise with the market. If RBOB is signaling a drop, drop early to capture share before competitors react.