A little bit of statistical sleight of hand there, under the heading of cherry-picking. The US cities they chose for comparison were selected from among the most dangerous cities in the United States. Sure, they left out Newark, because that would have been too obvious, but they also left out NY (which is extremely safe, as big cities go). Then they compared those cherry-picked cities to Mexico's national murder rate. Wikipedia puts that rate at 23.7 per 100,000, not 14, and the equivalent number for the United States is 4.7, or about 20% of Mexico's. If I picked, say, Tijuana, Chihuaha, Culiacan, and Victoria de Durango, and compared them to tourist hotspots like St. Louis (never been), I think that relationship (i.e., 5:1 ratio of homicides) would probably hold up.
But if you are trying to decide between a trip to generic Mexico and a vacation in sunny Baltimore (to Eat Bertha's Mussels, perhaps?), then they have provided some useful insight, perhaps.