Like most design equations, Balanis' are stated under a given set of assumptions: for example, that there are minimal fringing fields due to the conductors being very close to each other. At 25mm, yours aren't very close.
There could also be simulation issues, for example, how did you feed the antenna -- is it identical to what Balanis assumes? Is the simulation appropriately meshed?
When working with antennas, it's usually a bad idea to assume that "design equations" will produce an exact result -- but they can be a good starting point.