If you run a Monte Carlo simulation of electrons on a solid (simulation link attached), you will see the paths of the electrons at a fixed point and the interactions spreading into the tear drop shape.
As H. Changela described, MonteCarlo simulations show this, but in fact you are right. Not everything which can be calculated must be true, especially since all these simulations make some assumptions which are not always correct or first order approximations. However, the shape of the interaction volume is longer known than MonteCarlo simulations exist so that you perhaps search for papers published in the 60ies or even before in order to find some experimental proof. Or you simply believe it, since it is written in books or papers which e.g. S. Vasudevan referred to.