I don't really love the hybridization model except as a tool to understand molecular geometry, but neither here nor there. I think the first thing to point out is that it's clear what you suggest doesn't happen, because the true geometry of water isn't consistent with the kind of bonding you propose. As to the why, well - there are a lot of ways you could answer that, some due to energy, some due to symmetry. I think most of those explanations would be, based on your questions, a little bit beyond your level. So let's just consider what a water molecule would look like without the hybridization: you'd have two hydrogens in perpendicular orientation, then you'd have a lone pair perpendicular in the third direction, and another lone pair (from the 2S orbital of oxygen) spread spherically around the oxygen nucleus, overlapping with the other lone pair and the electrons in the two O-H bonds. Do you see any problems with this, from an energy standpoint?