The bike thing is about hedons. It feels directly better in the moment to go faster with less effort such that when you have a better bike you spend more time biking. This also applies to running and is why it is helpful to learn good running form.
Contra Taleb, I claim it is often perfectly reasonable to hold the following two positions concerning some set of journeys you have to make:
It is better to do them by bike than by car, mostly because you get more exercise that way.
Given that you’re doing them by bike, it is better to do them with a faster more efficient bike, mostly because it gets you there faster.
Suppose (it’s probably a tolerable approximation) that if your nicer bike gets you there in x% less time then it also means x% less exercise. Suppose that’s an improvement. What follows? Well, if everything is linear then it follows that a hypothetical superbike that gets you there in zero time with zero exercise would also be an improvement. But note that driving doesn’t get you there in zero time (but does mean zero exercise) so even the naïve linear model doesn’t imply that if cycling faster is a win then driving must be better.
And what’s the effect of the nonlinearities? Unless these cycle rides are rather long, taking utility proportional to time saved is probably not a bad approximation. But if the cycling is your main exercise, then (since the difference between “hardly any exercise” and “some exercise” is much bigger than that between “some exercise” and “twice as much exercise”) the gain from cycling at all is outsized compared to the gain from spending more time doing it.
I strongly agree with Romeo’s point too, of course, and while we’re bringing in other considerations Taleb doesn’t mention I’ll also remark that if you’re advocating cycling over driving because it saves fuel costs and reduces pollution, both of which are pretty common reasons given for cycling, neither of those benefits is less for faster than slower bicycles; in fact, since humans consume fuel and produce carbon dioxide they are actually slightly bigger for more efficient bikes.)
Agree and there’s a substantial effect due to marginal considerations. Your most marginal bike rides are also your longest bike rides. So if i have an effort threshold where I’ll easily choose bike if it’s within 20 minutes by bike, then the easier effort bike expands that physical range slightly and makes more trips happen by bike on the margin.
The bike thing is about hedons. It feels directly better in the moment to go faster with less effort such that when you have a better bike you spend more time biking. This also applies to running and is why it is helpful to learn good running form.
I don’t think it’s only that.
Contra Taleb, I claim it is often perfectly reasonable to hold the following two positions concerning some set of journeys you have to make:
It is better to do them by bike than by car, mostly because you get more exercise that way.
Given that you’re doing them by bike, it is better to do them with a faster more efficient bike, mostly because it gets you there faster.
Suppose (it’s probably a tolerable approximation) that if your nicer bike gets you there in x% less time then it also means x% less exercise. Suppose that’s an improvement. What follows? Well, if everything is linear then it follows that a hypothetical superbike that gets you there in zero time with zero exercise would also be an improvement. But note that driving doesn’t get you there in zero time (but does mean zero exercise) so even the naïve linear model doesn’t imply that if cycling faster is a win then driving must be better.
And what’s the effect of the nonlinearities? Unless these cycle rides are rather long, taking utility proportional to time saved is probably not a bad approximation. But if the cycling is your main exercise, then (since the difference between “hardly any exercise” and “some exercise” is much bigger than that between “some exercise” and “twice as much exercise”) the gain from cycling at all is outsized compared to the gain from spending more time doing it.
I strongly agree with Romeo’s point too, of course, and while we’re bringing in other considerations Taleb doesn’t mention I’ll also remark that if you’re advocating cycling over driving because it saves fuel costs and reduces pollution, both of which are pretty common reasons given for cycling, neither of those benefits is less for faster than slower bicycles; in fact, since humans consume fuel and produce carbon dioxide they are actually slightly bigger for more efficient bikes.)
Agree and there’s a substantial effect due to marginal considerations. Your most marginal bike rides are also your longest bike rides. So if i have an effort threshold where I’ll easily choose bike if it’s within 20 minutes by bike, then the easier effort bike expands that physical range slightly and makes more trips happen by bike on the margin.