Waymo has started releasing statistics for its autonomous vehicles. Here's the link.

There are a number of important considerations when comparing human-driven and autonomous vehicles. For instance, the two have different definitions of a crash. AV operators have to report any kind of physical contact (property damage, injury, or fatality). Human-driven cars, on the other hand, don't typically report accidents unless it was bad enough to necessitate a police report. So there are nuances to keep in mind.

That said, there is an argument to be made that AVs are already safer than human-driven ones. Through to June 2024, Waymo had already logged over 22 million rider-only miles. And here is what it is now reporting in terms of airbag deployments, injury-causing crashes, and police-reported crashes:

All of them are lower than their respective benchmark crash rates.

One of the important things that I remember them drilling into our heads in business school was about how to write a business memo. This might not seem like a big deal, but it is. Emails, decks, and recommendations are ubiquitous in business.

I remember three main points.

One, use clear and concise writing. If you can use fewer words, do that. Two, be decisive. In fact, they used to tell us that being decisively wrong was always better than being vaguely correct. And three, be as quantitative as possible.

If you can replace words with numbers, you should do that. For example, instead of saying that something recently increased significantly, it is far more effective to say that something increased by 27% over the last 18 days.

I was reminded of this earlier today when I came across this:

Supposedly, it is what Amazon used to tell its employees back in 2018. I don't know the source, but the tips sound right and make sense. Be concise. Use data. Eliminate weasel words. And make sure you're communicating a "what". In other words, be decisive.

The US Bureau of Transportation Statistics has just published some recent data looking at average trip distances across the country. What it allows you to do is drill down to the county level and see exactly how many trips people take that are less than 1 mile, between 1-3 miles, between 3-5 miles, and so on. This is interesting, in my view, for two reasons.

One, it showcases the fact that most of our trips tend to be short ones (a trip is defined as being away from your home for more than 10 minutes). If you look at the data you'll immediately see this, which is, of course, a pretty good argument for trying to encourage other forms of mobility besides driving.

And two, it is yet another example of how much data our mobile phones are constantly off-gassing. I mean, how do you determine where someone's home is so that you know when they're taking a 10 minute trip away from it? You figure out where their phone spends long periods of time (particularly at night) and you likely have that person's home.

What would be even more interesting to see is how this data correlates with built form. In other words, to what extent are higher densities inversely correlated with trip distances? This should certainly be the case, but it would be cool to see the data.