It's true that we do stuff up the forecasts now and again. Weather is far from being a perfect science. Given the movement of the weather up to the point of issue, the forecast is predicting the highest probability of where the weather will head next. Unfortunately there are many variables that alter the course of the weather, so it's important to understand that a forecast is a probability, not a certainty. We humans endeavour to predict bank interest rates, stock exchange prices, property prices, other peoples behavior and of course, the weather. We collect as much existing information as possible, apply formulas and couple this with past experience to offer a probable result of what might happen. Forecasts can be wrong, and this may be due to faulty, insufficient or incomplete observations, rapid changes in conditions, errors (both human and computer) and many more reasons. Interestingly, it's only in the last 50 years of human existence that weather forecasting has reached the level of accuracy now experienced. Some interesting article:

Simple view of how the weather works History of weather discovery.