Also, mathematically speaking, you will have less control over initiating your turns if you mount them at core center. This might sound confusing, but I'll try to make it understandable. Every ski has a sidecut that is technically a segment of a circle. If you were to completely extend the curvature of a ski's sidecut, it would circle around completely until it returns to the starting point. There is an "inflection point" on the sidecut curve of a ski where the edge stops curving inward and begins curving outward. In a more understandable sense, skis are fat at the tip, they get skinnier as you go down the ski, and then expand outward again. The inflection point of the curve is located at the point on the ski where the width is the thinnest. On park skis, this "thinnest" section is at the dead center of the ski, the point exactly halfway between tip and tail. Thus, the tip and tail have the same widths because the curve of the sidecut is identical on the fore and aft parts of the ski. However, on all mountain skis, this inflection point is moved backwards, meaning that the front half of the ski's curve extends for a longer amount of time, allowing the tip to be wider, while the curve toward the tail is shorter, causing the tails to be skinnier. The binding SHOULD be mounted above the waist (thinnest part of the ski) because the force for a turn is initiated at the apex of the sidecut circle, making it the most efficient. By mounting a binding above or behind the waist, you have to put a little more effort into making the ski turn. So, for the best, most efficient turning ability, you should mount where it is recommended, at the waist.
In reality, though, it doesn't make much of a difference. I have my powder skis mounted 3cm forward from the waist/recommended mounting position, and there really isn't a physical change, nor do I feel like I have less control. Really, it's just another interesting thing to think about the math of a ski and how they work.