Last week I did some math on the risk/reward profile of exercising indoors (risking covid exposure) vs. outdoors (risking exposure to smoke from the CA fires), and found the numbers for the day I did the math (low-for-fire-season pollution outside, sparsely populated indoor gym) overwhelmingly favored exercise of any kind over not exercising, and any other factor was overwhelmed by how likely it was to create friction to exercising.
Over on Facebook, a friend pointed out that I’d left out the biggest cost of exercise: the time in which it took place. I then realized a full accounting would also include the time to get to the gym and the risk of getting hit by a car en route. And was I sure the micromort estimate for exercise incorporated the risk of injury? (no, because the data is hidden in an appendix BJM paywalled and sci-hub doesn’t have. If you have BJM access and would like to help me out by emailing me (firstname.lastname@example.org) the appendix for this article it would be much appreciated. EDIT: received and responded to. Thanks Steve!). But exercise has benefits beyond dying later, and I wasn’t fully accounting for any of those either. And time spent at or traveling to work out isn’t exactly lost: zipping around on my scooter is fun, and leaving my house regularly on some sort of schedule has been good for me. This gets unwieldy really quickly.
Nonetheless, time spent traveling and the accompanying risk of car accidents seemed really significant, so I updated the spreadsheet to incorporate it. Ignoring any positive effects beyond the microlives, this was enough to make going to my gym for cardio net costly (note: because the spreadsheet measures in micromorts a positive number is bad), although going to the gym for weights and my nearby friend’s backyard for cardio still come out ahead.
I still think gym cardio is net beneficial for me because I think my exercise is much more impactful than average. But I don’t think it’s so much more beneficial than my friend’s backyard treadmill, so I’m going to emphasize the latter except on very bad smoke days.