Recent Penn Economics graduates Mike Chirico (BA'11, PhD'17) and Pau Pereira (PhD'17) were two members of a winning team (Team Kernel Glitches) in the National Institute of Justice's Real-Time Crime Forecasting Competition. The goal of the NIJ competition was to maximize two different crime hotspot scoring metrics for calls-for-service to the Portland Police Bureau (PPB) in Portland, Oregon during the period from March 1, 2017 to May 31, 2017. The team's solution to the challenge was a spatiotemporal forecasting model combining scalable randomized Reproducing Kernel Hilbert Space (RKHS) methods for approximating Gaussian processes with autoregressive smoothing kernels in a regularized supervised learning framework. This highly flexible modeling approach out-performed traditional crime forecasting methods and a range of other contemporary approaches, especially over short forecast periods and for sparse spatiotemporal events. The team tied for 1st place in the large organization competition with 9 wins and $135,000 in prize funds. For more details, see their recent paper on their winning solution--https://arxiv.org/abs/1801.02858.