What the number actually reflects
Your MPG result is the ratio of two measured quantities, miles driven and fuel consumed, and both have error. Add in the conditions under which the driving happened (cold or hot, city or highway, alone or loaded), and the result is a single number that has to be interpreted against all of those things at once.
The factors below are in roughly the order they move the result, starting with the ones that move it the most. None of the factors are independent; the cold-start fuel enrichment in winter happens at the same time as the winter oxygenated fuel blend, and the cold tires, and the heavier cold engine oil. Treat the list as a way to think about the result, not a way to control for each one separately.
1. Measurement methodology
Largest impact. Most "my MPG is wrong" complaints trace here, not to the car.
Three habits move the result more than anything else:
- Always fill to the first click. Topping off adds 0.5 to 1.5 gallons of variability between fill-ups. The first click is the most repeatable fill level.
- Use the same pump at the same station. Pump-to-pump calibration can vary by 0.5 to 1.0 percent. A 50 MPG car measured at a well-calibrated pump and the same car measured at a poorly-calibrated one can return 49.5 and 50.5 MPG; that is real measurement noise, not a real efficiency change.
- Photograph the odometer at every fill-up. This catches more errors than any other single habit. A 0.3-mile reading error in a 200-mile fill-up is 0.15 percent; the same error in a 30-mile fill-up is 1 percent. Photographing also catches the trip-meter-reset mistake.
The full procedure, including what to do when you can't start with a full tank and how to handle partial fills, is on the step-by-step guide.
2. Temperature
Cold engines run rich (more fuel per air) until the coolant and oil reach operating temperature, and cold air is denser (more fuel per stroke). Cold tires have higher rolling resistance. Cold winter fuel blends have different energy density. All four effects stack in the same direction in winter, and that direction is "lower MPG than the sticker says."
Roughly, a winter-to-summer swing of 12 to 25 percent is normal for most cars, and 30 to 40 percent swings are not unusual in very cold climates or for short-trip drivers who rarely let the engine fully warm up. The drop is not a defect; it is the engine working as designed in cold conditions.
Fuel density also changes with temperature, which is a separate effect on volume-based measurement. Gasoline expands about 0.00094 per degree Fahrenheit. A 15-gallon fill at 80°F is about 14.72 gallons at the standard 60°F. For most personal-tracking use, this is in the noise; for fleet reporting, the accuracy guide covers temperature compensation in detail.
3. Driving conditions: speed, terrain, traffic
The aerodynamic drag on a car scales with the square of speed, and the power required to overcome it scales with the cube. Doubling from 55 to 110 mph takes about eight times the power just to push air out of the way; a 70 mph cruise burns meaningfully more than a 55 mph cruise even on the same car on the same road. Most cars hit their most efficient speed somewhere in the 35 to 55 mph range, which is why the EPA combined rating weights city driving heavily.
Terrain matters: a 1,000-foot climb costs more fuel than a 1,000-foot descent returns (regenerative braking in hybrids partly closes the gap, but not fully). Stop-and-go traffic wastes fuel on idling and on the cold-start penalty at the start of every short trip. None of this is avoidable in the moment, but it is worth knowing that your highway MPG and your commute MPG can differ by 20 to 30 percent for reasons that have nothing to do with your car.
4. Vehicle condition and maintenance
A well-maintained car returns close to its rated MPG; a poorly maintained one does not. The maintenance items that move the number the most:
- Tire pressure. 10 PSI underinflated costs 2 to 3 percent. Check cold, before driving, monthly.
- Air filter. A clogged filter can cost 5 to 10 percent. Replace at the manufacturer interval, sooner in dusty conditions.
- Spark plugs. Worn plugs can cost 2 to 8 percent. Replace at the manufacturer interval.
- Oxygen sensor. A failing O2 sensor is one of the more common causes of a sudden MPG drop. Diagnostic codes usually catch it; the OBD-II guide covers what to look for.
- Wheel alignment. Misalignment increases rolling resistance and tire wear. Get an alignment if you see uneven tire wear or after any suspension work.
These are also the items most likely to cause a real, persistent efficiency change that the SPC guide will flag as a "special cause" rather than noise.
5. Fuel quality and type
For a car designed for regular 87 octane, switching to premium 93 octane gains 0 to 1 percent MPG in best case and nothing in typical case; the cost premium usually outweighs the gain. For a car designed for premium (and most modern direct-injection engines are), running regular can cost 1 to 2 percent and may trigger knock-retard timing that further reduces efficiency. The owner's manual is the right source for your specific engine.
Ethanol content matters more than most drivers realize. E10 (the standard US blend) has about 3 percent less energy per gallon than pure gasoline. E85 (85 percent ethanol) has about 28 percent less. Flex-fuel vehicles are calibrated to handle this; non-flex-fuel vehicles run E10 with no issue, but a non-flex-fuel vehicle mistakenly fueled with E85 will run poorly and get dramatically worse MPG until the tank is emptied. The EPA's ethanol effects page has the full numbers.
6. Load and weight
Every 100 pounds of additional weight costs about 1 to 2 percent MPG. A roof box at highway speeds costs more than 100 pounds of weight in the trunk, because the box also adds aerodynamic drag. A roof rack with no box on it still costs some MPG; the rack itself disturbs the airflow even when empty.
Passengers and cargo in the cabin or trunk cost fuel proportional to their weight. The cost of a temporary load (a vacation's worth of luggage) is small; the cost of a permanent load (tools, a stroller, a permanent roof box) is larger because it is paid on every fill-up.
7. Altitude and weather
Higher altitude means thinner air, which means less power from the same engine and slightly less aerodynamic drag. The two effects roughly cancel for naturally aspirated engines; turbocharged engines actually do better at altitude. The net effect on MPG is small (a few percent) compared to temperature or driving-style effects.
Headwinds cost fuel proportional to the wind speed; a 20 mph headwind at 60 mph cruise has the same drag cost as driving at 70 mph on a calm day. Crosswinds matter less. Tailwinds help, but you cannot count on them.
8. Measurement period length
Shorter fill-up cycles amplify measurement noise. A 1 percent error in the fuel pump reading is a 0.5 percent error in a 200-mile fill-up; it is a 5 percent error in a 20-mile fill-up. Aim for 200 to 300 miles between fill-ups; below 100 miles, you are measuring the pump's calibration as much as your car's efficiency.
9. Driving behavior
Aggressive acceleration, late braking, and high-speed cruising can each cost 10 to 20 percent MPG compared to moderate driving in the same conditions. The "hypermiling" techniques (drafting trucks, pulse-and-glide, engine-off coasting) can recover some of that, but the safety and wear tradeoffs are not worth it for most drivers. The most useful habit is anticipating traffic: looking two or three cars ahead, lifting off the throttle early, and avoiding unnecessary stops.
10. Equipment
Your odometer, your fuel pump, and (if you use one) your GPS all have measurement error. Factory odometers are typically within 2 to 4 percent; the main cause of inaccuracy is non-stock tire size, which is corrected by the accuracy guide tire-size formula. The fuel pump has the error sources discussed in section 1. A GPS gives an independent distance measurement, useful for cross-checking the odometer on a long drive.
How to use this list
The factors above are ranked by typical impact, but your situation is specific. The right way to use the list is to identify which one or two factors explain the gap between your measured MPG and the EPA sticker, then control for those, then move on.
For a typical car on a typical commute, the practical interpretation looks like this: the EPA combined rating is the lab result, and your real-world MPG will be 10 to 15 percent below it for reasons that are mostly temperature, traffic, and individual driving. If your MPG is 20+ percent below the sticker, something specific is wrong (low tire pressure, dragging brake, aging O2 sensor), and the accuracy guide walks through how to find it.