You can look into the reclassification fear of yours. Typically murders are used to compare. Meaning, 1 murder and 10 robberies. Next year 3 murders and 1 robbery. Some of pattern like this with the murder rate up or flat, but the crime down otherwise. Generally the point is that people will do a good job reporting murders (hard not to) and in the short term variance in the other crimes may have more to do with reporting characteristics.

One big divide is that people aren't talking about the same thing. Person A says they're less likely to die in location B. Great! Stats say violent crime is down! But there are a million pick pockets and I get robbed without a weapon every time I go downtown. ^alt SF version; every Tesla gets a window smashed.

Point being is two people can observe that and person 1 celebrate the lack of murders and person 2 flummoxed how come no one cares about the kids running out of Target with a T.V or the petty crime.