43

The United States is 3rd in murders throughout the world. If you remove Chicago, Detroit, Washington DC, St Lois, New Orleans the United States is then 189th out of 193 countries in the entire world.

I'd like to know if the claims about murders are accurate.

The "3rd in murders" is clearly referring to absolute deaths by murder, as it can be seen here, USA holds the 96th position for rate of intentional homicide.

So if the murders from these cities were removed from statistics, would the USA change its position from being the 3rd to being the 189th of 193 countries (as countries we consider the 193 members states of the UN)?

scohe001
  • 103
  • 2
user2638180
  • 2,396
  • 3
  • 17
  • 27
  • 63
    Does this meet the "notable claim" requirement if it's just some sourceless meme? – jeffronicus Jun 09 '21 at 20:23
  • 13
    @user2638180: Please show where you found this so we can see it is widely believed, and to get context. – Oddthinking Jun 10 '21 at 07:06
  • 4
    What does the "x" in "All 5x Cities" mean? – Barmar Jun 10 '21 at 13:42
  • @Barmar I think it's just poorly written - that should mean "5 times" or something to that effect, but it makes no sense here. – Nuclear Hoagie Jun 10 '21 at 13:56
  • 9
    Skeptics requires a "notable claim" e.g. in proper media. That is because if a question were opened for every graphic meme seen on Instagram, it would be inundated with spam. You know this. -1 – Harper - Reinstate Monica Jun 10 '21 at 18:46
  • 2
    I've submitted the 4th close vote. Without some indication that this is a wide spread meme and that people believe it, I worry that we are debunking noise, not claims. Also, in light of some interesting turnabouts the internet has produced in the past, strawmen become of particular concern now too. –  Jun 11 '21 at 17:30
  • @fredsbend About time. (at)OP: Notability would be demonstrated by this instance of [13.K positive Retweets](https://twitter.com/RyanAFournier/status/974458374845853696) this ['fact check' for a 2013 version in 2015 on Snopes](https://www.snopes.com/fact-check/united-states-third-murders/) which seems to omit [the originator shown on this meme](https://me.me/i/paul-joseph-watson-march-15-the-united-states-is-3rd-68363a322d31473eb88e89ed058afc45) as [discussed on reddit](https://me.me/i/paul-joseph-watson-march-15-the-united-states-is-3rd-68363a322d31473eb88e89ed058afc45) – LangLаngС Jun 11 '21 at 21:32

3 Answers3

150

This is false no matter what way you try to frame it.

Using the 2019 crime stats:

Murders

  • US: 15020 [1]
  • Chicago: 491 [2]
  • Detroit: 273 [3]
  • New Orleans: 120 [4]
  • St. Louis: 194 [5]
  • Washington DC: 166 [6]

These cities have a combined census population of 4.67m (out of 308.75m for the USA), which is ~1.5% of the total US population [7]

Removing those 5 cities brings the murder total from 15,020 to 13,776 - an 8% drop (the total murder rate goes from 4.9 per 100,000 to 4.5 per 100,000).

Based on the link in the original question [8], the ranking of the US wouldn't change in absolute numbers:

  • Nigeria 17,843
  • United States 16,214
  • Colombia Americas 12,586

But would probably drop several positions in rate:

  • United States 4.96
  • Kenya 4.93
  • Angola 4.85
  • Micronesia 4.67
  • Lithuania 4.57
  • Niger 4.44
Cotton Headed Ninnymuggins
  • 1,025
  • 2
  • 8
  • 13
CJR
  • 5,347
  • 2
  • 19
  • 24
  • 45
    Interestingly, the drop from 15,020 to 13,776 still leaves the US at the same place in the list. – IronEagle Jun 10 '21 at 04:31
  • 8
    Can you include the population side of your calculation? Google lists the population at 328.2M for the US and 4.8M for the 5 cities, which results in slightly different numbers. – CodesInChaos Jun 10 '21 at 06:18
  • @CodesInChaos it probably depends on the source I guess. I found out that the same https://ucr.fbi.gov site has different numbers for the same year in a different location. For example, [this source](https://ucr.fbi.gov/crime-in-the-u.s/2019/crime-in-the-u.s.-2019/tables/table-16) says it's 14014 murders in 2019 (rate of 5.0), while [this source](https://ucr.fbi.gov/crime-in-the-u.s/2019/crime-in-the-u.s.-2019/tables/table-12) says it's 15020. – Dan Jun 10 '21 at 11:04
  • Did you correct for the reduced base-population too? – Deduplicator Jun 10 '21 at 11:11
  • 16
    Tangentially, the claim is further complicated by the fact that while Chicago has strict gun laws, northwest Indiana does not, and many gun-related crimes can trace those weapons back to Indiana (pg 9 https://www.chicago.gov/content/dam/city/depts/mayor/Press%20Room/Press%20Releases/2017/October/GTR2017.pdf). 60% of all recovered crime weapons orginate outside of Illinois (pg10). – Sam Jun 10 '21 at 13:50
  • To answer the question, this needs to address how the US's murder rate compares to other countries. – Vaelus Jun 10 '21 at 14:15
  • 1
    @CodesInChaos Edited the population source into the answer (the preliminary 2020 census results). That's perhaps not the most accurate estimate, but it's the most comparable between different cities, and I would bet that the geographical difference between the regions defined for crime and the regions defined for population would exceed any real difference between population metrics. I did remove the city populations from the 4.5 per 100000, but it would have rounded to the same value either way. – CJR Jun 10 '21 at 14:43
  • It would be nice to attempt a reconstruction: how did they come up with this? (eg: American (US) Virgin Island *did have* the 3rd highest homicide rate in 2017). I am quite sure that however twisted/wrong etc, by some calculation/error 'they' did arrive at such numbers. – LangLаngС Jun 10 '21 at 15:36
  • You're answer needs drastic improvement. The population of Chicago is far different than the population of the Chicago Metropolitan Area. The CMA population is more than double what you've provided as the total population for the 5 listed cities, the number of murders in those areas is also drastically different than your supplied answers. Extrapolate for all 5 cities, and the statistics are drastically different https://data.cityofchicago.org/Public-Safety/Crimes-Map/dfnk-7re6 Answer coming soon. (PS, not saying it's true, just that this answer is lacking) – TCooper Jun 10 '21 at 22:52
  • I don't think the link I posted maintains filters, but you can add a filter for "primary description" "is" "homicide" - also no answer coming, I think the second answer down does a fine job of effectively disproving the claim – TCooper Jun 10 '21 at 23:12
  • @LangLаngС Paul "prisonplanet" Joseph Watson is a far-right propagandist, it's a reliable heuristic that anything he says is extremely misleading or outright false. – pjc50 Jun 11 '21 at 10:03
  • McCarthy demonstrated that simply making up random but specific numbers is an effective tactic. There's no need for any factual basis, all you need to do is create the impression that there's a factual basis. And arbitrarily specific numbers do that better than approximations. – barbecue Jun 11 '21 at 14:02
79

This is not possible based on the information given. The link you provided has the US at 6th with 16,214 murders and 189th place is a 3 way tie with 9 murders each.

Using this data that would require that those 6 cities accounted for all but 9 murders in the country.

If you look at the overall stats for the country you will find this is impossible.

https://www.cbsnews.com/pictures/murder-map-deadliest-u-s-cities/2/

The murder rate in Baltimore is 58.27 per 100,000.

Joe W
  • 4,799
  • 2
  • 25
  • 37
  • Could it be they’re twisting it, and it would drop from 3 by total number, down to 189 by rate? – Tim Jun 10 '21 at 16:27
  • 2
    @tim even that would be impossible as the rate in those cities isn't a majority of the country. My answer lists the number two city which wasn't on the list. – Joe W Jun 10 '21 at 16:33
  • 16K murders in a country with a population of over 300 million still seems amazingly low to me. You have to consider population as well. – Neil Meyer Jun 10 '21 at 18:31
  • 10
    @NeilMeyer most people are not murdered, which is true for literally every country based on the data I can find. Going by rates the US is about the middle of the pack, definitely not "amazingly low". – Kamil Drakari Jun 10 '21 at 18:54
  • 3
    This is a far better than the accepted answer. The simple statement "that those [5] cities account for all but 9 murders in the country." is all you need with the given counter example. It goes much further than interpreting the answer in one strict way and slapping numbers that are easily refuted onto the page. Although, adding murder rates per capita info might be helpful/ I assume it will be similarly easy to prove from this perspective. +1 – TCooper Jun 10 '21 at 23:00
  • Indeed, since we failed to fix the Q earlier: please amplify that the info sold as 'extracted from WP' is *wrong* to begin with. (But add the exact terms ["6th for absolute numbers" or sth]) // But crucially, it is OP that gives a WP page, not the meme. You mix claim & OP data, which are no match from the start. Is the meme – which is, as you've shown, *incompatible* from the start with "3rd place", eg based on the same year data as WP (obviously no, but which then? When & where does this meme originate etc?) – LangLаngС Jun 11 '21 at 01:12
31

The meme is unclear

It is not clear whether the meme is talking about absolute numbers or per capita murder rates. I'll talk about both.

The meme does not distinguish murder vs homicide

The first thing to note is that murder is a legal term, not equivalent to homicide. Murder does not include accidents or various manslaughter that led to homicide without malicious intent. In short, murder ⊂ homicide. The distinction isn't made in the meme.

For simplicity, I'll focus on murders only.

The intentions of the meme creator is politically-driven and fallacious

People who made the meme and who are spreading it are attempting to impress the concept that liberal-run cities with strict gun laws actually result in higher than average murder rates without providing additional substantive context, and thus committing a correlation fallacy, non causa pro causa.

Their point is that gun laws don't stop violent crime, as a means to maintain popular support for ownership of firearms, but they don't bother to analyze what percentage of the crime is committed with a gun, nor do they simulate what the violent crime rates would be if those laws were not in place.

The data do not support the meme

Putting this aside, the statistics simply don't support the assertion.

City/Country Murders Population Murder Rate
1. Chicago 491 [2019] 2,694,000 ~ 18.23 per 100,000 [2019]
2. Detroit 272 674,841 ~ 40.31 per 100,000
3. D.C. 166 [2019] 692,683 ~ 23.96 per 100,000 [2019]
4. St. Louis ≤ 194 [2019] 308,174 ~ 62.95 per 100,000
5. New Orleans 121 390,845 ~ 30.95 per 100,000
City/Country Murders Population Murder Rate
U.S. (all) 16,425 [2019, FBI] 328,240,000 [2019, World Bank] ~ 5.005 per 100,000
U.S. ( – 5 cities) 15,181 ~ 323,479,500 ~ 4.693 per 100,000

So removing those 5 cities does impact the murder rate nationwide, but not as significantly as the meme framers are professing.

In the fairyland tale where the U.S. doesn't have the five cities, it might be better than Kenya or Angola, but other than that, there is no noticeable improvement in terms of world rankings or whatever:

enter image description here

FluffyFlareon
  • 503
  • 3
  • 5
  • 3
    It might be useful to note that not all murders (or homicides) are performed with guns. Also, from a quick look, intentional homicide rates seem to correlate more with population density than anything else. States without big cities, where gun possession is common, are among the lowest here: https://en.wikipedia.org/wiki/List_of_U.S._states_and_territories_by_intentional_homicide_rate – jamesqf Jun 10 '21 at 16:41
  • This is an aside, but when looking at that list, I noticed that US Virgin Islands is #2 most homicidal entity on that list, just behind El Salvador and way ahead of places like Mexico and South Africa. I always thought of that place as a Carribian Tropical paradise, WTF is going on there? – Eugene Jun 10 '21 at 16:45
  • @Eugene US Virgin Islands has only about 100,000 people, and the extremes of any "per capita" measure tend to be small countries because random singular events have such an outsized influence. All of the 0-rate countries are micronations, and notice that Anguilla is #14 with only 4 murders. Montserrat is #21 with *one* murder! – JounceCracklePop Jun 10 '21 at 17:17
  • @JounceCracklePop I wonder if (homicides^2)/(population^2) would give better rankings by reducing the impact of singular events in small nations. – Tim Sparkles Jun 11 '21 at 00:20
  • Editing in a source for your last table (by eye, looks like WP?) would be nice. – nitsua60 Jun 11 '21 at 02:00
  • 2
    @TimSparkles: Wouldn't that give exactly the same rankings? – psmears Jun 11 '21 at 11:54
  • @JounceCracklePop 52 out of sample size of 106K is more than enough to even out some outliers. Even if the rate was, reduced by 40% due to outliers, the US Virgin Islands' murder rate would still be higher than Mexico, with a bloody, raging ongoing drug-war. – Eugene Jun 11 '21 at 18:58
  • @TimSparkles The only wway to reduce the impact of singular events would rather be to base the numbers e.g. on decades instead of years – Hagen von Eitzen Jun 13 '21 at 22:06