Formal charges can be useful but you have to be careful with them because as their name suggests they are in reality a formalism. Another way of saying this is that they are a tool we have created to explain certain observations, i.e they are imaginary.
First of all, formal charges tend to ignore electronegativity. In your example, boron has an electronegativity of about 2, and nitrogen has an electronegativity of about 3. The electron density should be greatest around the nitogen, giving it a partial negative charge. Yet the properly drawn structure gives it a positive formal charge. Another example would be the BF4- anion. Boron has a negative formal charge in this structure, yet fluorine has twice the electronegativity, and the electron density should be greatest around fluorine, which has a formal charge of 0.
When you run into situations like this where things break down, I recommend considering oxidation states instead. In the BF4- case, boron has an oxidation number of +3, and, and the fluorines an oxidation number of -1. Makes a lot more sense.