This study revolves around a computational algorithm called SDEL (Stochastic Difference Equation in Length) that generates approximate protein folding trajectories on the atomically detailed resolution scale. The protein studied is Barstar-a barnase inhibitor. Because of the protein's interesting structure (four alpha helices, three beta strands) and relatively small size (89 residues), Barstar is an optimal choice for running complete folding trajectories on a computer. 12 pathways were generated with SDEL, starting from a structurally wide selection of unfolded conformations, yet all ending with the native configuration. We tracked hydrogen bonds, dihedral angles, native and non-native contacts, and energetic along these folding pathways. The resulting trajectories show: 1) Barstar follows the Hydrophobic Collapse folding scenario, 2) native α-helices begin forming earlier in the trajectory than the β-sheets, 3) particular residues maintain a propensity for helical structure in their unfolded state, and 4) specific non-native contacts persist during the folding trajectory. Strong correlations were found between the SDEL pathways and data from NMR, CD, and other experimental studies.