This text explores how new data streams, such as GPS and mobile event data, can be used in conjunction with spatial analysis methodologies to better understand and address safety concerns on road networks like those in New York City. By analyzing collision data from the NYPD's open dataset, researchers have identified patterns and trends, including spikes in collisions during early morning hours and midnight, which may be attributed to drunk or tired driving. They have also used spatial autocorrelation to detect clusters of high or unusual activity, which can help identify areas for investigation and potential mitigation strategies. Additionally, they have applied decision tree methods to identify common features among crash-prone streets, such as the presence of parking lanes and travel lanes, and found a connection between traffic jams and collisions, with pedestrians being more frequently injured during traffic jams than outside of them. The study highlights the importance of leveraging new data sources and methodologies to gain a deeper understanding of urban mobility and improve road safety.