One simple trick to avoid being a victim of crime...
The typical homicide victim in America has 10 or more prior arrests, and probably more than half have themselves committed a serious violent crime.
Written by Inquisitive Bird.
In 2017, the average homicide victim in Baltimore had 11 previous arrests on his record. Nearly 50% of victims had previously been arrested for a violent crime, and about 70% for a drug crime. Similarly, a 2018 analysis found that 81% of homicide victims in Chicago had prior criminal records. In addition, a 2021 analysis that homicide victims and suspects in Washington DC were remarkably similar, both demographically and in terms of criminal history: “most victims and suspects with prior criminal offenses had been arrested about 11 times for about 13 different offenses by the time of the homicide.” This similarity is shown in the chart below.
These numbers illustrate a phenomenon that is well known among criminologists: there is substantial overlap between criminal victims and offenders. Several other studies have made similar findings (Jennings et al., 2012). If we assume (not unreasonably) that a non-trivial fraction of perpetrators manage to avoid arrest, these findings suggest that well over half of victims of homicide have themselves committed serious violent crimes. The vast majority have been extensively involved in crime.
The risk of violent victimization is substantially raised by criminal involvement and other problem behavior. Childhood problem behavior is associated with both the likelihood of criminal offending and the risk of violent victimization later in life (Koepp et al., 2023; Beckley et al., 2022). In their review of crime correlates, Ellis, Beaver, Wright (2009) conclude that the evidence “unequivocally supports the conclusion that criminal offenders are more likely to be crime victims than are persons in general.” Serious delinquency and crime predict later homicide victimization.
More broadly, victim-offender overlap is found for most types of crime (Jennings et al., 2012), with even a significant fraction of intimate partner violence being bidirectional (Langhinrichsen-Rohling et al., 2012). Overlap is also found in non-criminal contexts: for example in the significant overlap between bullies and victims of bullying among middle school students (Haynie et al., 2001).
Correlates of both offending and victimization
If engaging in crime is a major risk factor for victimization, then correlates of violent offending should also be correlates of violent victimization (Farrington & Loeber, 2011). This is almost invariably the case:
Sex. Most violent offenders are male, as are most victims.1
Age. Violent victimization and offending have broadly similar age curves. The risk is lowest before puberty and in old-age, and is highest between the ages 15 and 40.
Ethnicity/race. Crime is typically intra-racial, so disparities in offending rates are mirrored in victimization rates. In the United States, victimization/offending rates can be ordered from lowest to highest as follows: Asian, white, Hispanic, black.
The fact that victims and offenders are demographically similar was already confirmed in the previously mentioned analysis from Washington DC. To show the effects of the three major demographic factors at the national level, the chart below displays homicide victimization rate by age, sex, and race/ethnicity. As you can see, homicide victimization risk is strongly demographically differentiated. Similar patterns are observed for risk of being killed by police (Edwards et al., 2019) and for imprisonment rates (see, e.g., Table 10, Carson, 2020).
Aside from basic demographic characteristics, many other individual-level correlates of offending are also correlates of victimization (Beckley et al., 2017). Some examples are:
Mental/psychiatric disorders. Psychiatric disorders are associated with both violent victimization and perpetration. Within families, the sibling with a psychiatric disorder is 3-4 times more likely to perpetrate or to be subjected to violence than their sibling without a psychiatric disorder (Sariaslan et al., 2020). Consistent with the idea that offending itself increases victimization risk: among people with mental disorders, violent victimization risk is much higher for those who recently perpetrated violence towards others (Silver, 2006) or have criminal records themselves (Short et al., 2013; Dean et al., 2018).
Alcohol/drug use. Alcohol and drug use disorders are well-known correlates of both offending (e.g., Yukhnenko et al., 2023) and victimization (Dean et al., 2018). A substantial fraction of homicide victims test positive for alcohol (Kuhns et al., 2011; Naimi et al., 2016), as do many homicide offenders (Kuhns et al., 2014). A meta-analysis found that alcohol/drug use is associated with victimization and perpetration (Duke et al., 2018).
Intelligence. Low intelligence is a well-known correlate of violent offending (e.g., Frisell et al., 2012; Schwartz et al., 2015; Jacob et al., 2019). It is also a correlate of violent victimization (Beaver et al., 2016) and not because victimization itself caused the lower intelligence (Danese et al., 2017).
Conclusion
The image of an innocent person being attacked, like unsuspecting prey falling victim to a predator, is certainly a vivid one. But it’s not particularly realistic. In actual fact, violence is often mutual or retaliatory, and mostly committed by repeat offenders. The typical homicide victim in America has 10 or more prior arrests, and probably more than half have themselves committed a serious violent crime.
If a person wants to avoid violent victimization, their best bet is to stay away from crime. Of course, we shouldn’t ignore the fact that many victims of violent crime are entirely innocent – even if they’re in the minority. Most people would probably agree that such victims deserve greater concern and sympathy. It’s therefore not surprising that the image of the innocent victim takes up more space in our collective minds.
Inquisitive Bird specializes in data analysis, with a particular interest in clarifying social patterns. You can find his analyses and writing on his Substack.
Follow us on Twitter and consider supporting Aporia with a $6.99 monthly subscription:
Sex is sometimes called the “fundamental correlate of crime” (DeLisi & Vaughn, 2015). Approximately 90% of homicide perpetrators are men (Fridel & Fox, 2019). According to the CDC, 80.3% of homicide victims in the United States were male and 19.7% were female from 2018 to 2021. While violent crime broadly fits the typical victim-offender pattern, rape/sexual assault is one obvious example that doesn’t. Here, instead of victimization and offending being more frequent in the same group, women are more likely to be victimized but much less likely to be perpetrators.
Women are the primary readers of the true crime genre, of which I am one. Of the stories I've read a majority of female victims are prostitutes or drug addicts or frequent places where crime is more likely, i.e. bars, dangerous neighborhoods, and large blue cities. I'm ever mindful of my surroundings and I don't take careless chances, but my concern about being a victim of violent crime is negligible.
Status competition among males is the primary driver of individual and group violence. It should hardly shock us that the violent competitors for status mostly murder one another.