Due to their persistent severe winds, derechos threaten human security and properties, and they are as hazardous and fatal as many tornadoes and hurricanes. However, automatic detection of derechos is challenging due to the complex criteria used to define the phenomenon and the lack of spatiotemporally continuous observations. This study proposes a compromised definition of derechos, which not only contains the key features of derechos described in the literature but also allows their automatic identification using either observations or model simulations. The automatic detection is composed of three algorithms: the Flexible Object Tracker algorithm to track mesoscale convective systems (MCSs), a novel machine learning algorithm to identify bow echoes, and a new algorithm to classify MCSs as derechos or non-derecho events based on our proposed definition of derechos. Using the automatic detection approach, we developed a novel high-resolution observational data product of derechos over the United States from 2004 to 2021. The dataset is analyzed to document the climatological statistics of derechos in the United States. Many more derechos (~20 events per year) are identified in the dataset than in previous estimations, but the spatial distribution is similar to earlier studies with a peak of occurrences in the central Great Plains. Most derechos occurred between April and July, contributing 12% of damaging wind gusts on average in the United States from 2004 to 2021.