The sport after this one

"What is the best sport?" has no answer, and "what is the best sport at each age?" sounds like the fix. A child should play this, a teenager that, a fifty-year-old something else, a grandparent something gentler. The evidence does answer at each stage, but it never answers with a sport. Early on it says do not pick one. In midlife it ranks a few sports, by association only. Late in life it names a capacity, balance, that some activities train and others do not. So the useful choice is made by property rather than by name, and the property that runs through every stage is whether the activity leaves you with something to play at the next one.

The clearest advice about children is a refusal. In 2015 the American Orthopaedic Society for Sports Medicine convened a think tank on early specialization, the practice of narrowing a young child to one sport.¹ Its primary conclusion was that "there is no evidence that young children will benefit from early sport specialization in the majority of sports."¹ It summarised the available evidence as suggesting that specializing before the age of 12 is associated with more burnout and dropout and with worse athletic development over time.¹ One study it cites found that 88% of German Olympians had played more than one sport as children.¹

The statement came out of a roundtable, not a trial.¹ The studies it draws on are mostly observational, so "associated with" is the right verb. But notice what kind of answer it is. It does not say which sport a seven-year-old should play. It says the property worth having at that age is variety, and that the measured cost of narrowing early is stopping. Dropout is the hinge between stages. A child who burns out of one sport at eleven does not carry a sport into adolescence at all.

For adults there is at least one study that compares sports with each other. Pekka Oja and colleagues followed 80,306 adults in England and Scotland, with an average age of 52, and adjusted for potential confounders.² All-cause mortality was lower among those who played racquet sports (a hazard ratio of 0.53), swam (0.72), did aerobics (0.73) or cycled (0.85).² For football and running they found no significant association.²

This is the nearest the evidence comes to naming sports, and it should be held loosely. It is observational. People who play racquet sports may differ from people who do not in income, health and time in ways adjustment does not fully capture. How many people played each sport, and which confounders were adjusted for, would decide how far the ranking can be trusted, and the paper's summary does not give them. "No significant association" for running is not evidence that running is useless; it may be a matter of small numbers or of who runs.

At the other end the evidence is strong and points away from sport as such. A Cochrane review of 108 randomised trials, with 23,407 people living at home and an average age of 76, found that exercise reduces the rate of falls by 23%, with high certainty.³ What mattered was the kind. Balance and functional exercises cut the rate of falls by 24% (high certainty); programmes combining several types, usually balance plus resistance work, probably by 34%.³ Tai Chi may cut the rate of falls by 19%, a low-certainty estimate, and reduced the number of people who fell by 20%, with high certainty.³ For walking and dance programmes, the reviewers were uncertain of any effect on falls.³

Walking, whatever else it does for an older person, has not been shown to protect against falls, and neither has dance. The property that matters late is balance, trained on purpose. Of the activities people take up for pleasure rather than as a programme, Tai Chi is the one with evidence behind it.

Read across the stages, the answers line up as properties: variety in childhood, something a person will keep doing through adulthood, balance in old age. An activity that carries more than one of them has a claim to span stages. Racquet sports are a tempting example: the largest association in the adult study, and a game that requires constant changes of direction and is played with other people. Nothing here shows that they train balance in a way that prevents falls, and it would be a mistake to read the hazard ratio as a prescription.

What the stages do share is a failure mode. The child who specializes and burns out, the adult whose team sport ends when the league does, the older person whose only exercise is one that does not protect against falling: each has chosen something good for the stage they were in and found nothing waiting at the next.