The Bowling Green Massacre: The Data Behind A Tragedy

Most of my posts on this blog are not political and usually cover personal finance/investing topics.  However, I want to talk about something that has been widely debated recently.  There have been a lot of stories going around the internet about what did and did not happen at the Bowling Green Massacre.  Today I am going to clear up any confusion by using the actual data we have from the event.  While anecdotes can be powerful for emotional appeal, in this instance, the underlying data will present a clearer picture of what actually happened.

Before I show you the data, I want to start by talking about the victims.  Stalin once said, “One death is a tragedy, one million is a statistic.”  This statement rings true when trying to summarize data about a tragedy.  I want to reiterate that I am not trying to downplay the tragedy by presenting aggregated statistics.  Please remember that the victims were real people and not just the numbers I present below.  If you want to support the victims and their families, please donate to the Bowling Green Massacre Victims Fund.  With that being said, here is the data of the number of victims killed in the Bowling Green Massacre.  Note that I did this over time to reflect the number of victims that have come to light as the story has unfolded:


While this is tragic as it is, once you add in those injured the numbers are almost threefold higher.  To sanity check this, take the number of victims killed and multiply by 3.


However, if we break these numbers out by religious status, you can see that the perpetrators of this tragedy clearly had an agenda:


Despite the numbers behind this tragedy, Kellyanne Conway was right:  No one in the media reported it.  This is a breakout of the total percentage of airtime dedicated to the Bowling Green Massacre on the day of the event:


I could do a more in-depth analysis using a machine learning model, but I think the numbers speak for themselves.


How Can We Prevent This Tragedy From Happening In The Future?

The best way to prevent a tragedy from happening is for it to not happen at all. In all seriousness, I hope everyone enjoyed this.  I will be back to my normal posting next Tuesday.

Thank you for reading!

This is post 07.  Any and all code I have related to this post can be found here with the same numbering:



Everything stated above is completely factual.  

6 thoughts on “The Bowling Green Massacre: The Data Behind A Tragedy

  1. This analysis is suspicious. First off, the timeseries graph reports the death rate as perfectly constant. We’d expect an event like this to show variance over time according to the Poisson distribution. It’s possible we’re looking at a naive inference from post-event aggregate statistics. Secondly, there’s no indication that the author tested the assumption that the death reports’ margin of errors obey a Gaussian distribution, and since media economics encourage detection & reporting of events with strong emotional entailments, I think this may result in an unhandled model skew to inflate the reported deaths.. If we assume that undetected false-positives follow a Weibull distribution with the same control parameters as similarly lethal events (e.g. those reported in live radio broadcast by Wells in 1938), I calculate the potential skew is between 0 and 0 (95% confidence.)


    • Dear Jacob,

      Thank you for reading. I see your point on the Poisson distribution. However, even after adjusting for any time varying effects, I get the SAME result. In particular, I used an exposure variable to adjust for the varying number of days in a particular year (i.e. leap years could have more events), and I record no difference in the outcome. Additionally, while you are right that I “assumed that the death reports’ margin of errors obey a Gaussian distribution”, your own skew measurement between 0 and 0 is not economically different from the figures I report. I agree that there may be 0 additional fatalities that my model is missing, but there is a 0% chance that this is NOT correct.

      I appreciate the rigor in which you have applied to my analysis, but, in the end, my result is robust and there is no economical difference when we use your assumptions to model the event.



  2. Nick:

    I wish you did not do this. Our political culture is toxic enough this kind of thing will change no one’s mind. It will only inflame those that disagree with you making you readership base smaller.

    I very much like your blog and think it has real promise. But if it going to descend into partisanship, I too might stop reading it.


    • Jim,

      Thank you for reading. I appreciate the comment. I agree that if this blog went down a partisanship road I would not read it either. This post was clearly a joke using data analysis. I am confident that most of my readers understood this. I was not trying to make anything “toxic” and I was not trying to change anyone’s mind about anything. How can I change someone’s mind about an event that never happened?

      Someone simply misspoke and I wanted to do something funny in response. Hope you understand.

      All the best,



  3. I have evidence that the Bowling Green Massacre was an inside job. Looking closely at the data (which the mainstream media will NEVER do) you start to see multiple curiosities – exactly the same number of men and women among the casualties for instance. Coincidence?

    But dig a little deeper… can we hear the 911 calls from Bowling Green that day? No – “none available” … mysteriously wiped, one assumes. Phone videos of the event and aftermath? Again – none available! (Can you imagine who needs to pull the strings at a network level to have individual smartphone data deleted from cloudservers and even from memory cards?)

    Most chillingly, how many relatives of senior CIA, FBI and White House staff were among the casualties? Indeed, how many registered Democrats? I don’t think you need me to tell you the answers. After all, you have the data.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s