Tale Of Two Cities: Visual Analysis of Baltimore Crime

Back in 2015, I shared some information about Baltimore, regarding its crime statistics specifically, and how crime statistics in general are often mishandled.

In tracking the homicide rate for 2017, I noticed an unusual dynamic. For all of January and half of February, we averaged about one murder a day. While Baltimore's homicide totals have been between 200s and 300s since the 1990s (the rates escalated after the crack market epidemic of the late 1980s/early 1990s), we seldom had over 40 consecutive days with at least one killing.

Out of curiosity, I examined the homicide locations from 2001 through 2016 to see if there were any visual clues as to what may have caused this sudden spike; this video shows the patterns:

What jumped out at me, was that by 2016, central Baltimore did not record a single murder and that there were two distinct walls of murder on either side. It also appeared that the areas where the murders occurred were much more compressed than in prior years. Since homicide is a function of proximity and opportunity (i.e., people kill where they live), it makes sense that compressing the footprint of traditionally challenged neighborhoods increases the level of conflict.

The obvious question is, "How did that middle corridor not have a single murder?" Was it because there are now worse marksmen, faster ambulances, or better trauma surgeons in central Baltimore? I think there were three reasons.

First, for the last year, there has been an intentional effort by the city, Johns Hopkins University, and community development groups to designate central Baltimore as a business development and family-friendly corridor. There is a strong push to draw young entrepreneurs and families to this area. Second, and accordingly, the housing policy has changed. The city has legacy policies such as 'Move to Opportunity." This involved the city purchasing homes in the county and offering them as subsidized housing for the poor. But what is new is that young suburbanites are moving back to the city and redeveloping the houses and blocks in that corridor. Third, I believe that this corridor benefits from an augmented policing policy. In addition to coverage from the city police, the Johns Hopkins Campus police, community block watchers and walkers, cameras on the light poles combined for a better surveillance system. In effect, I believe that the multiplicative effect of employment opportunities for skilled young professionals, a revitalized housing policy, and a broader policing policy resulted in the central Baltimore being murder-free in 2016.

As wonderful as this is, there seems to be an unintended consequence. Specifically, the revitalized core means that the traditionally challenged areas could not expand, thus increasing population density. Areas already overwhelmed with crime, food deserts, poor public health conditions, and chronic unemployment, now had to absorb those displaced by the revitalized core strip. Here is the demographic irony that exacerbates the situation -- the young professionals and middle class families have the wherewithal to live in the county or the city. In fact, at certain income levels, to live in the city is a no-brainer because the cost of living is lower. However, the working class and poor cannot afford to move to the county because it is more expensive. Hence they are stuck and cannot trade places with the county folks who moved to the city. The result is a revitalized central Baltimore, but an increased murder rate; a tale of two cities -- the best and worst of times.

But with me being a native Baltimorean, I may be too close to the situation to see what these visuals are really saying. Please tell me, how do you interpret the movements of Baltimore's murders? Is this a valid use of visual analytics? Please share.

Bryan Beverly, Statistician, Bureau of Labor Statistics

Bryan K. Beverly is from Baltimore. He has a BA in sociology from Morgan State University and an MAS degree in IT management from Johns Hopkins University. His continuing education consists of project management training through the ESI International/George Washington University programs. He began his career in 1984, the same year he was introduced to SAS software. Over the course of nearly 30 years, he has used SAS for data processing, analytics, report generation, and application development on mainframes, mini-computers, and PCs. Bryan has worked in the private sector, public sector, and academia in the Baltimore/Washington region. His work initially focused on programming, but over the years has expanded into project management and business development. Bryan has participated in in-house SAS user groups and SAS user group conferences, and has published in SAS newsletters, as well as company-based newsletters. Over time, his publications have expanded from providing SAS technical tips to examining the sociological, philosophical, financial, and political contexts in which IT is deployed. He believes that the key to a successful IT career is to maintain your skills and think like the person who signs your paycheck.

Simple Tool to ID Data Fraud: Benford's Law

This simple, old school fraud detection tool for data can help weed out problems in a number of different scenarios.

The Three Social Functions of Business Jargon

The lingo or jargon you use in analytics or IT or business management is part of what makes you part of a culture. Here's a deeper look at the reasons why we use jargon.

Redlining and Its Legacy
  • 5/12/2017 8:49:27 PM

Some have asked how people become trapped in terms of home buying or renting choices. The ground work was laid many years ago by Redlining. This article from 2015 (the year of the Freddie Gray riots) gives some insights. Parentheses are added so that the URL will post without an error, so please delete them in your browser. (www).nytimes.com/2015/05/10/opinion/sunday/how-racism-doomed-baltimore.html?ref=opinion&_r=0

Re: Pareto Efficiency/Optimization - As Things Get Better For Some, They Get Worse For Others
  • 4/24/2017 4:16:29 PM

@PC, To your point, it is wrong (hence the retraction by the commander).  However, to the commander's  unvarnished statement, it is very difficult to prevent the murders and to dissipate the culture of death unless you literally unwind and renovated those troubled communities. The best that the police can do is solve the murders but not prevent them. They do target suspects with long rap sheets and high recidivism rates, but even the police get tired of re-arresting the same perpetrators – hence like Freddie Gray, they are punished by getting a 'rough ride' in the wagon – bounced around to teach them a lesson. Those officers had arrested him before, and as quietly as it is kept, you have officers who get tired of making an arrest just to see the person released or given a light sentence for  non-violent offenses. To complicate matters, the high number of drug arrests means that the jails are over-crowded, so anything that the prison system can do to release non-violent offenders they will do to reduce the population. Hence, the commander was referencing the fact since they cannot prevent the murders, and they cannot deliver street punishment to repeat offenders (especially not now with body cameras), and since they get tired of the revolving doors on the prisons, causing them to re-arrest the same folks, they would just as soon let the market dynamics of the street drug industry handle the problem. By no means am I condoning the commander's views, just channeling the frustration of making an arrest, writing a report, going to trial, enduring the appeals, seeing the imprisonment and then seeing  the same person back on the street in a couple of years for good behavior – rinse and repeat. To that extent, some would prefer it if the problem took care of itself. So yes it is wrong to feel that way, but you have some underpaid, overworked and highly exposed to being killed officers out there, who are tired of being darned if they do and darned if they don't. It is very frustrating for officers and citizens alike, so if the problem could take care of itself, he would not complain.  Yes – I agree with you,  but want you to know the broader context of the statement.

Re: Pareto Efficiency/Optimization - As Things Get Better For Some, They Get Worse For Others
  • 4/24/2017 3:14:02 PM

@Bryan - thanks for providing all of us a better understanding of the problem. 

Out of all of our conversation on this topic, the line that sticks with me most is one of yours from an earlier comment -

A few years ago, when asked about a soft effort to police an area with a high concentration of murders, a commander misspoke and said that the goal was to let attrition handle the problem - like a contained wild fire, you just let it burn itself out.

'Soft enforcement' seems to me to be a huge mistake. Letting a fire "burn itself out" when people are the fuel, seems like letting people just defend themselves, and fend for themselves. And that seems wrong.

Re: Pareto Efficiency/Optimization - As Things Get Better For Some, They Get Worse For Others
  • 4/18/2017 3:16:04 PM


@Lyndon_H - That 'better day' is quite possible, but two things are needed: (1) political will and (2) revenue stream.  Ironically, this Pareto tradeoff between utopian growth and dystopian density is related to these two requisites, such that the Pareto problem is also the solution. Let me explain. Around 1964, Baltimore began hemorrhaging residents. The creation of the expressway, beltway and suburban housing drove the white city residents to the county, and the MLK assassination and riots in 1968 sealed the deal.  From the 1970s to the present, blacks increased political power in the city, but unfortunately this steady rise pushed the tax base into the county. Hence you increased the number of black faces in public spaces, but unfortunately, black was not the new green – the predominantly black city lacked the revenue needed to make the city better. Many former city homeowners who moved to the county retained their city properties and thus the number of rental properties skyrocketed.  Rental properties are needed for the very young or the very old, but in general, homeownership and real estate taxes pay the bills. So what is happening now is somewhat of a tradeoff.  Central Baltimore redevelopment is being encouraged to get the tax base back after a 50 absence.  What was unintended was that the people who are stuck into more crowded areas are mimicking a Malthusian catastrophe scenario. In effect, this Pareto problem is also the solution. The problem of one group prospering at the expense of another is bringing political will and revenue stream in the process. This process is morally problematic, but pragmatically essential. However, I think that the 'better day' scenario that you articulated is a better outcome. But how to get there – that's the rub. To quote Thomas Paine "These are the times that try men's souls" – aptly from The Crisis (December  23,1776).

Re: Pareto Efficiency/Optimization - As Things Get Better For Some, They Get Worse For Others
  • 4/17/2017 8:51:11 PM


Via Wikipedia, Bryan explains Pareto efficiency:

Pareto efficiency or Pareto optimality is a state of allocation of resources from which it is impossible to reallocate so as to make any one individual or preference criterion better off without making at least one individual or preference criterion worse off.


Bryan then writes:

I have referenced this concept in my responses because, unless I am wrong, these visual plottings of the murder locations suggest that increasing the breadth of better neighborhoods is squeezing the people at the bottom into more confined spaces.


It seems to me this Pareto optimality dilemma springs fom some kind of scarcity of available resources. But why is there scarcity of decent housing, adequate public services, essential infrastructure, and other social necessities? America has a massive industrial base, a huge workforce, and many other resources. But all these resources seem to be mismanaged, squandered, and misallocated, leading to these serious inequalities and the Pareto dilemma.

Maybe one of these days enough people will perceive the bigger picture and pushback will be the result.


Re: Pareto Efficiency/Optimization - As Things Get Better For Some, They Get Worse For Others
  • 4/17/2017 11:49:38 AM

@PC - Lot of challenges make it prohibiting for these folks to move:

  1. During the Great Migration to the North, those were mainly people who did manual labor in the south, going to factory jobs in the north. Baltimore cranked out steel for years at Sparrows Point/Bethlehem Steel. However, those manufactoring jobs (that the president promised to get back) are gone. Many moved below the border (moving those jobs below the border not only reduced labor costs but was intened for immigrants to stay there and not come here - believe that NAFTA is tied into this). Factories now also use computers and robots (works well for Amazon). Therefore shifting to a service economy based on knowledge, marketable employees need a high school diploma, if not college degree. Today, high school drop outs in Baltimore (of which there are many) have few employment options. They cannot move and cannot commute. Interestingly, privately run prisons are doing quite well. Decent profit margins and labor costs zero. 
  2. Another factor is that lack of income increases the probability that one has committed a felony (robbery, drug sales, etc.). On most job applications nationwide, there is a question "Have you ever been convicted of a felony?". This question plus the ability to access court records on line further reduces employment options. If you tried to move to a good place, your background check would be a deal breaker.
  3. Transportation is a problem. Auto insurance is about 3 times higher in the Baltimore City than Baltimore County because insurance rates are set where accidents are more likely to happen than where the drivers live. Hence drivers can come into the city and have a fender bender but enjoy the lower county rates. But the city residents have to pay higer rates because most of the collisions happen in the city - residents and commuters. Hence the lack of ability to drive to work is a challenge.
  4. Some folks were moved to the county by the city (Move to Opportunity > The city purchased county properties and quietly moved in some city residents) However all social connections, baby sitters, beauticians (not every beautician in the county knows how to care for the hair of city residents) and jobs were back in the city.  Taking a bus requires at least two transfers (of course this is not applicable for the middle class blacks who intentionally moved to the county). So if metaphorically you sleep in the county, but the bulk of your life is in the city, then living in the county is a challenge.
  5. Many of these folks are un-banked or non-banked.  Hard to move ahead without a checking account.
  6. Above all, a lot depends on whether your family history represents multigenerational progress. It takes a special person a a series of divine interventions to rise above these challenges and escape. There is a Grand Canyon gulf between middle class and poverty culture; in Baltimore, its not quite a caste system, but it takes a lot of thrust to get off the ground and into orbit.  For east and west Baltimore, if you have ever seen The Wire, those scripts were written from the events posted in the news papers and police anectdotes. Once multigenerational hopelessness sets in, it is hard to break that cycle of dysfunction. In effect, you wake up each day knowing that you will likely die where you live, and if redevelopment causes your rent to go up and forces you to move to a cheaper place, then it does not make your mental health and coping options any better. "Baltimore born and Baltimore bred and when you die, you are Baltimore dead".

Therein lies much of the social unrest. As humans, we can live within 5 miles of each other, but live in two different worlds. An 8 ounce glass that contains 4 ounces of water can be seen as half full and half empty. So it is sociologically in Baltimore. Some can see a crime-free space and say that they want more - and they would be right. Some can see their neighborhoods being developed for houses they cannot afford to buy and are increasingly angry - and they would be right and turn anger inward. They turned it outward during the Freddy Gray riots, were labeled as thugs and condemned for not doing more to help themselves.

Hence based on these patterns, I am hypothesizing that things are getting better for one group as they get worse for another group. I think Baltimore has reached a state of Pareto Efficiency/Optimization.

Re: Pareto Efficiency/Optimization - As Things Get Better For Some, They Get Worse For Others
  • 4/17/2017 9:57:33 AM

@Bryan - Thanks to your detailed explanation, i can see the problem as you do.

We don't think the urban renewal of central Baltimore is a bad thing, by itself, but the living conditions in the surrouding areas do seem to be getting worse.

Why is it that people feel trapped here? Why don't they just move out?

There are several news articles just this month about recent census data showing that Americans are moving less than we did in the past. Both renters and owners. Both short distance moves and longer distance moves are happening at lower rates. 

Here's an interesting quote from a NY Post article on the topic: 

Especially after WWII, millions of Americans with limited resources — southern blacks — moved hundreds of miles from home to take up industrial jobs in the north. At the peak, more than 30 percent of southern-born blacks moved north, from 1920 through the 1960s.

To break the Pareto Efficiecy, people in the worst areas should be making decisions to leave for better opportunity elsewhere. What makes this more difficult today than it was for southern blacks in the 30s?

Re: Baltimore's murder (rate) mystery
  • 4/17/2017 8:00:37 AM

While the police are generally thought to have the mandate to protect and serve, perhaps society could also use that same mandate in it's rules and governance. Putting some emphasis on serving, not only the briefcase crowd but the economically disadvantaged might result in less crime in those area we find in isolated neighborhoods of cities large and small.

Pareto Efficiency/Optimization - As Things Get Better For Some, They Get Worse For Others
  • 4/16/2017 1:17:21 PM

According to Wikipedia, Pareto Efficiency/Optimization is defined as follows:

Pareto efficiency or Pareto optimality is a state of allocation of resources from which it is impossible to reallocate so as to make any one individual or preference criterion better off without making at least one individual or preference criterion worse off.

I have referenced this concept in my responses because, unless I am wrong, these visual plottings of the murder locations suggest that increasing the breadth of better neighborhoods is squeezing the people at the bottom into more confined spaces.

Interestingly some see these visuals as a proven way to remove urban blight. Using blocks of real estate acquisition to reinvigorate the city and raise property values is good news to home owners.  But if the price of that means that circular, vertical or horizontal firing squads (metaphorically speaking - based on the fact that people kill where they live) increase the number of shootings and murders (as of 4/8/17, there were about 232 shootings which included 83 fataliities - much more than one shooting a day), then the ethical/moral questions are raised:

  1. If the data visualizations suggest that a city has reached the point where improving life for some means the quality of life will diminish for others, then is that OK?
  2. If people are going to shoot each other any way, based on economic and social dysfunctions, then should progress stop for the community investors?
  3. Better yet, what should be the role of the data analyst or data scientist in all this? Just present the numbers and then ask for more research money, or suggest that the research money be directed into reduce problems without hindering growth? In short, is there a point where research money should be redeployed into fixing problems rather than continued studies of the problems? Are data scientists complicit in exacerbating the problem by advocating for more research instead of policy changes?

The Pareto Efficiency/Optimization problem for challenged cities like Baltimore - Your thoughts please.

Re: Baltimore's murder (rate) mystery
  • 4/14/2017 2:23:59 PM

@PC - In principle, totally agree with you. However in terms of public safety policy, there is a difference.  The officers of the court - police officers - seldom have fatal encounters with the folks with the briefcases.  Much of the unrest in Baltimore in 2015 was based on the belief that the officers were tasked to protect and serve the folks with briefcases, even though they too fell under the rubric of "we have all sinned and come short of the glory of God".  In fact, the states attorney caught a lot of flack for indicting the officers involved in the Freddy Gray case; the implicit understanding wsas that they should never be questioned in their roles as puiblic servants. So it is in that light that the good people of East and West Baltimore become disheartened and frustrated when it seems that they do not get equal protection under the law. 

Oh I fully agree that each person must guard their hearts and search their souls because all of us have flaws, weaknesses and moral failure points.  So to your point that the real criminals may carrry firearms or briefcases, the sociological flashpoint is that the folks with the briefcases are seldom considered to be criminals requiring immediate police officer intervention . So compound that nuance with an expanding redevelopment path that forces dispalcement, 'land locked' citizens who want to get out but cannot, chronic unemployment with multigenerational economic stability challenges, etc. and the result is a high murder rate, and folks like me who are sympathetic (especially since I was stopped by the police with drawn weapons in 1981 because I matched the description of someone who robbed a store. I was carrying a bag, but it had a new Bible and not cash and jewely), but would like to see this growth increase, and who enoys the protection of the City Police, Community District Patrol, Johns Hopkins patrol and neighborhood walkers.

Back to the place where I have no answers. Perhaps the true criminal is in my mirror.

Page 1 / 3   >   >>