The problem with statistics is that they often do not tell the whole story. For example the murder rate in 1933 was quite high (9.7 per 100,000 versus 5.0 per 100,000 in 2009 according to the FBI) due in no small part to Mob wars and motorized bandits like John Dillinger, Pretty Boy Floyd, and Baby Face Nelson. However, historical perspective must be taken into account. While the murder rate was quite high due to organized crime and bank robbers, these entities very seldom victimized ordinary people. By and large, the victims of these sorts of crimes were police, celebrities/very wealthy people (targets for kidnapping and ransom), or other criminals. The average person had little to fear from the violent criminals of the 1930's because they had nothing which the criminal was interested in taking from them.
Today, however, the world is quite different. The types of criminals that contributed to the violent crime rates of the 1930's are, for the most part, a thing of the past due to years of work from the FBI and other agencies. However, random crime against ordinary individuals was tolerated much less in the 1930's than it is today. This is due in no small part to the increasing urbanization of the population. In the 1930's, a much larger percentage of the population lived in either a rural area or a small town. While people were largely unconcerned about the victimization of some gangster or bank, a crime against an ordinary person was much more likely to be a crime against a family member, friend, or neighbor and that was of concern. Today, however, the majority of the population lives in larger cities. Larger communities means that the average person is well-acquainted with a smaller percentage of his or her community. The murder, rape, or robbery of an average person was much more likely to draw public outcry, and therefore be punished, in the 1930's than it is today.
My point in all this is that just because the violent crime rate as a whole has declined, that does not necessarily mean that the risk of victimization for the average person has decreased or even remained linear. Basing one's personal risk assessment on the raw numbers of crime rates is unwise because raw numbers are too easily misinterpreted or misrepresented.