Lawrence Guth teaches the sixth lecture of MIT's 18.156 Projection Theory, arguing that the projection of a rough function onto a line at a typical angle tends to come out smoother than the original function. The lecture builds on earlier sessions in the course, developing the analytic tools behind this smoothing phenomenon and connecting it to the broader study of projections used in harmonic analysis and geometric measure theory. Guth works through the reasoning at the board, laying out why almost every angle produces a smoothing effect even when the underlying function is badly behaved, and what exceptions can occur. At 78 minutes, the session is dense with technical detail aimed at graduate-level students already familiar with measure theory and Fourier analysis, part of MIT OpenCourseWare's full recording of the spring 2025 course.