Alpha particle interactions with crystalline silicon at approximately 400 mK, which produce phonons, are observed with Titanium Transition Edge Sensors (TiTES) on the crystal surface. A calculation of the expected mean free path suggests that the phonons should arrive diffusively, but a significant component of "ballistic" phonons, which experience few or no scattering, are observed. There are two sources of evidence for this. First, timing differences between sensors on opposite faces of the crystal are consistent with ballistic propagation. Second, the phonon pattern on one face of the crystal shows lobing indicative of phonon focusing, anisotropic propagation of phonons due to direction-dependent elasticity in the crystal. This pattern would be unobservable if the propagation was truly diffusive. This paper presents a comparison between the Monte Carlo simulations of phonon propagation and the experimental results, which have been previously described in detail.1 The sensors consist of a 400 angstrom thick line of titanium patterned into a meander. The width of the lines is 2-mu-m, and the pitch is 5-mu-m. The line is held just below the superconducting transition temperature and biased with a current. Phonons from a particle interaction in the crystal drive sections of the line normal, resulting in a resistance. Potential methods of increasing the sensitivity of the sensors include reducing the line width and using a superconductor with a lower T(c).