7.63 Million Miles Per Crash

Tesla launched a brand new replace to its voluntarily printed quarterly Car Security Report, revealing information for the primary quarter of 2024.

Because it seems, the variety of miles pushed per accident registered when utilizing Autopilot expertise improved noticeably to a brand new document degree. Nonetheless, the variety of miles when not utilizing Autopilot expertise worsened. In the meantime, the U.S. common stays the identical as a result of the latest information from the NHTSA and FHWA is for the 12 months 2022.

The information provides us glimpses of how Autopilot improves automobile security, but it surely has some limitations because of the methodology (extra on that on the backside of this put up). For instance, Tesla counts all accidents, whatever the trigger, although greater than 35% of all Autopilot crashes happen when the Tesla automobile is rear-ended by one other automotive, the corporate says.

Based on Tesla, the variety of miles pushed per accident registered in Q1 2024 utilizing Autopilot expertise amounted to 7.63 million miles.

That is a 47% year-over-year enchancment and a brand new document. The common can also be some 11 occasions greater than the U.S. common for all vehicles.

The earlier highest outcome was famous in Q1 2022—6.57 million miles per accident registered. In 2023, it was between 5.18 and 6.18 million miles.

With out utilizing Autopilot expertise, Tesla autos statistically notice extra accidents. The common amounted to 0.955 million miles pushed per accident registered.

The result’s 13% worse than in Q1 2023 and the worst ever recorded. We do not know why this statistic worsens as a result of a number of elements have an effect on it. Extra miles are pushed with Autopilot, probably leaving more difficult driving situations for drivers.

There’s a important distinction between the 2 modes of driving (with or with out Autopilot), and the distinction continues to extend. In Q1, it elevated by nearly 70% year-over-year to eight.

Nonetheless, we should notice that the 2 information units shouldn’t be in contrast aside from offering us with some concept of the numbers.

It is essential to notice that the outcomes are comparable solely inside a specific class (Autopilot or with out Autopilot), not between the classes because the enter information may be broadly completely different (like easy freeway driving or advanced metropolis driving). In different phrases, we are able to solely see whether or not the lively security programs are enhancing over time (and it is also solely a tough comparability), however we will not examine Autopilot to non-Autopilot driving.

We assume that the correct use of Autopilot improves security, however Tesla’s report doesn’t enable us to guage the distinction.

We now have hooked up beneath a full chart with all of the numbers offered by the producer, plus NHTSA and FHWA information in regards to the common distance between vehicle crashes in the USA.

The distinction is fairly large, however as soon as once more, we should not examine the numbers immediately. The U.S. common features a a lot older automobile fleet (with fewer security programs), which alone is an enormous issue. The U.S. common is 0.670 million miles (as of 2022, since there is no such thing as a data for 2023-2024 but).

Tesla has its personal model of the chart:

Necessary elements:

• information for every setting may be collected at completely different driving situations (like easy freeway driving or advanced metropolis driving), which makes the outcomes incomparable between the classes
• outcomes may be affected by numerous elements, together with seasonality (decreased daylight, climate circumstances), much less driving throughout lockdown
• NHTSA and FHWA common for the U.S. (up to date hardly ever) contains all vehicles, additionally it is outdated
• Tesla’s data in regards to the methodology of registering accidents:
  “We accumulate the quantity of miles traveled by every automobile with Autopilot lively or in guide driving, primarily based on obtainable information we obtain from the fleet, and accomplish that with out figuring out particular autos to guard privateness. We additionally obtain a crash alert anytime a crash is reported to us from the fleet, which can embrace information about whether or not Autopilot was lively on the time of affect. To make sure our statistics are conservative, we rely any crash during which Autopilot was deactivated inside 5 seconds earlier than affect, and we rely all crashes during which the incident alert indicated an airbag or different lively restraint deployed. (Our crash statistics should not primarily based on pattern information units or estimates.) In observe, this correlates to almost any crash at about 12 mph (20 kph) or above, relying on the crash forces generated. We don’t differentiate primarily based on the kind of crash or fault (For instance, greater than 35% of all Autopilot crashes happen when the Tesla automobile is rear-ended by one other automobile). On this manner, we’re assured that the statistics we share unquestionably present the advantages of Autopilot.”
• Assuming the methodology has not modified, we are able to see how every class improves over time.