The charts are ordered from highest to lowest homicide rate during the last available date. For intentional homicides, kidnappings, and extortions the counts refer to number of victims, for car robberies the counts refer to police reports (averiguaciones previas which may contain more than one victim).Car robberies include the crime of robo de coche de 4 ruedas only. Homicides include feminicides.
INEGI homicide data tends to be more precise. Forensic coroners and the INEGI simply don't have the same incentives to lower the number of homicides that happen in Mexico as the police. In addition to adding deaths from Legal intervention, operations of war, military operations, and terrorism to the INEGI homicide count I used a statistical model to predict the intent of deaths of unknown intent similar to what I used in this post. The disadvantage of using the INEGI data is it is usally more than a year out of date.
Intuitively, you can think of the statistical model as if someone were to ask you to guess the injury intent in the death of a 70 year old Mérida woman whose cause of death was transport related. You would probably answer it was an accident. On the other hand, if you had to guess the type of death of a young adult male whose cause of death was a firearm, and the injury took place on a public street in Ciudad Juárez, you would probably guess it was a homicide.