As others have said you can't tell anything from a single tank and the way you measured it you are most likely to amplify any error in filling it. The reason is you can easily have a ±0.5 gallon difference on filling the tank up depending on lots of factors including but not limited to: the angle the car is at parked at the pump, the temperature of the fuel remaining in the tank, the speed at which the fuel is pumped (creating foam or not) and for that matter the fuel itself and how foamy it is.

As careful as you think you are you you can NEVER fill a fuel tank to precisely the same level every time. If the car is sloped ever-so-slightly away from the pump it will take more fuel than if it is sloped towards it.

So, by measuring after using only a few gallons your error is amplified because a 0.5 gallon error out of 5 gallons would be 10% or about ±3 MPG while that same 0.5 gallon error on a full 20 gallon tank would only be 2.5% or about ±0.75 MPG. The most accurate readings therefore would be taken after burning as much fuel as possible from the tank between fillups and then averaging several of them.

If you did it this way you'd probably discover that the first trip was an anamoly and you'd really be getting around 27-28 MPG on 100% highway driving...and that's about typical for these models.

__________________
Marty D.

2013 C300 4Matic
1984 BMW 733i
2013 Lincoln MKz