The last three years have been a bonanza for gravitational wave astronomy. The first gravitational wave signal ever detected---GW150914, the merger of a 30- and a 40-solar mass black hole---has been followed by five more announced black hole mergers (the latest, GW170814, observed for the first time by both the LIGO detectors and the Virgo detector). Slightly more than a year ago, the first observed merger of two neutron stars (GW170817) was also detected in the electromagnetic spectrum, in a collective effort involving almost 4000 astronomers around the world. Each of these events carries a wealth of information about the merging objects, their progenitors, their environment, and the universe; however, to extract this information we must often resort to statistical methods that extract features collectively from the entire population of events. I will run through a few highlights from individual detections (including the measurement of the Hubble constant from GW170817), and then discuss recent results in modeling the population of merging compact objects including: using the distribution of spins in the population to distinguish among competing formation mechanisms; measuring the distribution of merger events in redshift to determine that we live---and gravity propagates---in three spatial dimensions (who knew?) and (eventually) measure the star formation rate to sub-percent precision; and measuring the maximum mass of stellar-mass black holes and using this feature in the mass spectrum for cosmography.