Justin Fox

What’s Behind the Great American Murder Decline? Homicides are plummeting in many places. The explanation may be the sheer volume of different efforts to reduce violence.

Great thoughts from @byjustinfox.bsky.social on what's behind the Great American Murder Decline www.bloomberg.com/opinion/arti...

Good times

NYC Is One of the Safest Places in America New York City should be doing more to take advantage of being one of the least dangerous places in the US, explains Bloomberg Opinion columnist Justin Fox. (Source: Bloomberg)

www.bloomberg.com/news/videos/...

Yeah but if you think of it in terms of risk per minute, time on the subway seems to come with higher homicide risk than life in general in NYC. (Not totally confident in my calculations here; going to redo them a few more times to be sure.)

Death by AI One man's struggle with his mortality.

"To the best of my knowledge, I did not pass away last November 20. That is not just my opinion. In recent months I have been examined by two different licensed physicians, and if I had been dead, I'm pretty sure at least one of them would have mentioned it." davebarry.substack.com/p/death-by-ai

You can try to make them comparable, and I've done that. It's not the only way to do it, and I would agree that with crime my methodology has some problems (because people spend much more time outside the subway system than in), but that means I need to come up with better measures, not give up.

Enlighten me! If it's that one spends a relatively small amount of time on the subway, which makes comparisons with overall homicide rates somewhat suspect, I think that's probably right. But the comparison with traffic deaths seems fair.

Almost every other large city in America, plus most of the South

… but it’s the original that’s going around.

It’s a chart of annual deaths per 100K population, so yes it’s on the same per capita basis. What do you find confusing about it? (I genuinely want to know, just to help me with future chart design.) If it’s everything going to three decimals, I posted a new version that doesn’t do that …

Why not?

Its traffic fatality rate is actually kind of low for rural America (these are 2018-2021 numbers)

Yeah, that's also a valid way to think about it, especially when comparing the homicide rates. I don't the choice of 500 rides is the problem, it's more an issue of measuring per-minute risk rather than annual risk.

Sawyer County's traffic fatality rate is actually on the low side for rural America

Made a new version that's less confusing

I am so thrilled my daughter will by going to college in NYC, in large part because she will not be doing the single most dangerous thing a young person can do (drive), and will instead be getting around on one of the safest means of transportation ever devised (the subway).

Chart showing risk of being murdered in the subway if you rode 500 times last year (0.4 in 100,000) versus other risks such as that of dying in a traffic accident in the US last year (12.1 in 100,000).

The subway risk chart you've been waiting even longer for (revised so it no longer goes to three decimals). Sawyer County, Wisconsin, is where Sean Duffy is from, and the rates for there are annual averages over seven years. Gift link: www.bloomberg.com/opinion/arti...

The CDC's National Vital Statistics System (which gets its data from death certificates) has 19 motor vehicle traffic fatalities among Sawyer County residents from 2018-2024 and 23 occurring there. I used the former number for calculating the mortality rates.

I went to three digits because otherwise lightning risk would be 0.0 or 0.00, but I agree that it causes problems. Am working on a new version.

The problem is I went to three decimals because otherwise lightning-strike risk would be 0.00, and in our do-it-yourself charting software I can't then not go to three decimals for everything. Will see if I can get somebody in graphics to fix it

it's a decimal point, not a comma

Those are decimal points not commas

That’s why I used a seven-year average!

Rural counties in general do have by far the highest traffic fatality rates, though (this is for 2018-2021 because for some reason CDC Wonder doesn't have populations for later dates)

Yeah the localized data just aren't available to do an injury chart like that

Yeah, I went to three decimals because otherwise the lightning risk would come out as zero, but it clearly led some people (not just you) to read the decimal point as a comma

Maybe this is confusion about how we use decimal points (versus commas) in the US? The chart is saying that the motor vehicle traffic mortality rate in Sawyer County is ~15 per 100,000

These are mortality rates per 100,000 population (which is the standard way to report mortality rates), not percentages

Yeah I went out to three decimals so the lightning risk wouldn't be zero, but I can see how that would be confusing