Notes

The crime statistics were released for 2018 last week by the FBI. They measure a broad range of different crimes, but here we’ll primarily focus on the homicide data. There are many narratives that could be crafted with these numbers. Often completely different narratives can be created with the exact same data, as I’ll show in each section below. As always, it’s important to understand the full context and have the ability to look at the data yourself, to keep from being mislead by media reports or memes.

All statistics come from the various tables within the FBI’s 2018 report, and when relevant the change from the prior year includes data from the 2017 report.  Feel free to peruse and get comfortable with the many data tables on your own as there is tons of data to sift through, but here’s the recap of highlights that I found were important and noteworthy in the data.

Total Homicide Victims

• 14,123 people were murdered (or victims of non-negligent manslaughter)
• This was a drop of 1,006 from the year prior, a decline of 6.6%

Narratives: 

• Less people were murdered in 2018, a positive for Trump, and the trend continues for the 2nd straight year.
• This continues to reverse the troubling increase at the very end of the Obama term in 2015 and 2016.
• The decline in murders continues the broad downward trajectory in the US over the past few decades in murder and violent crime, enjoyed by several administrations. Thus, it could be likely attributed to broader cultural changes and/or law enforcement.

Victims by Race/Ethnicity/Sex

• White: 6,088 (7.5% decline)
• Black: 7,407 (5.7% decline)
• Hispanic: 2,173 (7.7% decline)
• Male: 10,914 (8% decline)
• Female: 3,180 (1.3% decline)

Narratives: 

• Murders declined for all major racial/ethnic groups by roughly similar amounts.
• Black people continue to be disproportionate victims of murder and violent crime.
• Men get murdered more than 3 times as much as women, but male victims declined much more than female.

Offenders by Race/Ethnicity/Sex

• White: 4,884 (4.7% decline)
• Black: 6,318 (2% decline)
• Hispanic: 1,576 (4.7% increase)
• Male: 10,306 (3.4% decline)
• Female: 1,443 (unchanged, literally the exact same number in 2017!)

Narratives: 

• Both black and white murderers declined, while Hispanic murderers increased.
• Black offenders continue to commit a disproportionate amount of murders.
• Male murderers declined more than females, who remained unchanged. But males commit murder more than 7 times what women do.

Murder Type by Race

• White on White: 2,677 (6.4% decline)
• Black on Black: 2,600 (1% decline)
• Hispanic on Hispanic: 655 (12% increase)
• White on Black: 234 (11% decline)
• Black on White: 514 (11% decline)
• Black on Hispanic: 143 (13% increase)
• Hispanic on Black: 83 (20% decline)

Narratives: 

• Most murders occur between members of the same race.
• Whites are responsible for less than 10% of black murders, and killed appreciably less black people this year.
• Black people kill whites more than twice as much, but this number also came down appreciably, suggesting perhaps less conflict between the groups.
• Hispanic on Hispanic murders rose considerably, however they are still far below black on black rates. (Note: since Hispanics can be white, I didn’t compare Hispanic on white, or vice versa as it gets too confusing.)

Murder Type by Sex

• Male on Male: 4,073 (3.8% decline)
• Female on Female: 184 (7% increase)
• Male on Female: 1,731 (unchanged)
• Female on Male: 521 (1% increase)

Narratives: 

• Both sexes kill males far more than females.
• Males commit far more murders, but male murders are in slight decline while female murders are increasing.

Type of Weapon Used

• Total firearms: 10,265 (7.2% decline)
• Handguns: 6,603 (6.4% decline)
• Rifles: 297 (24% decline)
• Shotguns: 235 (11% decline)
• Knives: 1,515 (5.8% decline)
• Blunt Objects: 443 (6.1% decline)
• Hands, Fists, Feet, etc.: 672 (5.3% decline)

Narratives: 

• Murders from firearms declined across the board, without any new major gun control laws.
• The vast majority of firearm murders come from handguns, which cause more than 20 times the amount of deaths as rifles.
• Knives account for 6 times the murders that rifles do, and “personal weapons” (hands, feet, etc.) account for more than double.
• Despite the focus on “assault rifles,” there was a steep decline in murders from rifles in 2018, with no legislation. In fact, rifle murders fell more, percentage-wise, than any other firearm.
• 2016-17 saw a spike in rifle murders, so the decline could just be returning to normal. 2018 levels are still above 2015.

Justifiable Homicide:

• Law Enforcement: 410 (7.5% decline)
• Private Citizens: 353 (4.3% decline)

Narratives:

• Police are killing less people, even “justifiably”. This could be due to better training and/or awareness, or just because there’s less violent crime overall.
• Private citizens killed less people in self-defense. This runs counter to the notion that more guns available and more conceal carry ability would increase this.

Murders in Major Metropolitan Areas

• Atlanta: 341 (14% decline, murder rate fell to 5.7 per 100,000)
• Baltimore: 373 (9.7% decline, rate fell to 13.3)
• Boston: 93 (26% decline, rate fell to 1.9)
• Chicago: 762 (15% decline, rate fell to 8.0)
• Dallas: 328 (14% decline, rate fell to 4.4)
• Denver: currently unavailable
• Detroit: 345 (1.9% increase, rate rose to 8.0)
• Houston: 440 (unchanged, rate fell to 6.3)
• Las Vegas: 169 (29% decline, rate fell to 5.5)
• Los Angeles/OC: 619 (3.6% decline, rate fell to 4.6)
• Memphis: 231 (5% increase, rate rose to 17.2)
• Miami: 385 (2.6% increase, rate rose to 6.2)
• Minneapolis: 70 (26% decline, rate fell to 1.9)
• New Orleans: 198 (9.2% decline, rate fell to 15.5)
• New York: currently unavailable
• Orlando: 139 (11% increase, rate rose to 5.4)
• Philadelphia: 540 (8.9% increase, rate rose to 8.9)
• Phoenix: 222 (18% decline, rate fell to 4.6)
• San Antonio: 135 (15% decline, rate fell to 5.4)
• San Francisco/Oakland: 196 (1.5% decline, rate fell to 4.1)
• Seattle: 131 (14% increase, rate rose to 3.3)
• St. Louis: 361 (2.0% increase, rate rose to 12.9)
• Tampa: 119 (7.2% increase, rate rose to 3.8)
• Washington D.C.: 314 (12% increase, rate rose to 5.0)

Narratives: 

• The decline in nationwide murders was lead by major cities (which tend to be Democrat controlled).
• Notorious problem areas like Chicago and Baltimore fell substantially. However, there were significant exceptions to this (Philly and DC).
• Some major cities, like Minneapolis and Boston, have murder rates surprisingly low and on par with areas in Europe. Others, like New Orleans and Memphis, have appallingly high rates on par with the third world.
• The vast majority of murders continue to come from major cities, likely due to gang and drug violence.
• The safest city (over 200,000) was Green Bay, WI with a murder rate of just .3/100,0000. The most dangerous was Memphis, TN with a rate of 17.2

Best and Worst States

• The safest state was South Dakota, with a murder rate of just 1.4/100,000. The most dangerous was Louisiana, with 11.4.

Notes

• It’s always important to note the population change when comparing FBI crime data to different years. From 2017-18, the population analyzed covered 2 million more people. A slight increase, but less than 1%, so its effects should be minimal in this analysis.
• There are many murders where some aspects are unknown. Thus, numbers in different areas might not add up to the total amount of murders (example: adding men + women murders won’t equal the total murders).
• Of the firearm murders, a significant number (almost 1/3) are unknown or unstated. In my opinion, these are likely handguns, as many gang related crimes like drive by shootings go unsolved and the suspect and weapon never found. So the FBI numbers likely understate the percentage of handgun murders, but technically they can’t be classified without evidence.
• Statistics for “murders” is technically “murders and non-negligent manslaughter” according to the FBI. I shorten that in this analysis for brevity.
• “Hispanic” is technically “Hispanic or Latino”. I shorten it in this analysis for brevity.
• Hispanic and Latino is classified as an ethnicity, not a race. There are white and non-white Hispanics.