People chose between mindfulness and Tai Chi, and choosing did not make it work better

There's one question I get more than any other, and it almost always arrives in the first week. Which one should I do. Yoga or Pilates. Tai chi or the sheet of exercises the physiotherapist printed out. The Feldenkrais Method® or the meditation app her daughter installed on her phone in July and has now asked about twice.

I have never had a good answer. A trial published this week in the Journal of Clinical Oncology has one, and it isn't the answer anybody was hoping for.

The study is called MATCH, run out of the University of Calgary with the Princess Margaret Cancer Centre in Toronto and the Osher Center at Harvard. It enrolled 587 adults living after cancer treatment, three quarters of them women, average age just under 61, and instead of assigning each person a program the way almost every trial does, the researchers did something that still strikes me as slightly radical: they asked people what they wanted.

If you had a preference, you got it. If you didn't, you were assigned by chance. Then everyone, chooser and non chooser alike, was sorted into either an immediate group or a waiting group, so each program had its own comparison. About two thirds walked in with an opinion. Of those, 57 percent wanted Tai Chi and qigong and 43 percent wanted the mindfulness course, a structured program called Mindfulness-Based Cancer Recovery. The remaining third shrugged and took the coin flip.

The first set of results appeared in the summer of 2025. Both programs lifted mood against their waiting groups by small to moderate amounts, and choosing your program did not make it work better than having it handed to you (Journal of Clinical Oncology, 2025).

The paper published this week reports what happened to sleep, fatigue and pain (Journal of Clinical Oncology, 2026). Both groups improved on fatigue by around three and a half points on the questionnaire, and both lowered how much pain interfered with daily life by a little over a point. One place they came apart. Against the people still waiting, only the Tai Chi and qigong group showed a clear gain in sleep quality, just under two points, and it held after the authors accounted for age, sex, cancer stage, and whether the person had chosen. Preference, again, changed nothing.

Read the next line slowly, because it's the one that matters. Compared directly against each other, the two programs did not differ. So the careful version is this: the moving practice cleared the bar against doing nothing on sleep, the still practice did not clear that particular bar, and the distance between the two was never wide enough to call a winner. Both statements are true. The second one keeps the first one honest.

This is roughly what happens every time somebody lines movement methods up against each other and looks properly. Yoga, Pilates, Tai Chi, qigong, the Alexander Technique, Feldenkrais: each runs on its own mechanism, each has things it does unusually well, and when they meet head to head in a group large enough to see straight, the ranking tends to dissolve. What survives is the non result. The choosing did not help.

That one deserves more attention than it will get. Everything we assume about motivation says the woman who badly wanted Tai Chi should have done better than the woman who was handed it. She believed in it. She had a reason. She had, in the language of the field, expectancy on her side. In 587 people it did not show up.

What I see again and again in my classes is people spending an extraordinary amount of time on the decision. Months, sometimes. They read comparisons, they ask which method is best for a stiff lower back, they watch videos of three different approaches and then do none of them. And the deliberating is not really about mechanisms; it's almost always about something quieter, which is that they have tried things before, the things did not hold, and they are now waiting, politely and indefinitely, to be convinced that anything will. You haven't lost movement. You've lost trust that movement will help. Comparison shopping is what that loss of trust looks like from the outside. It's the same shape of question I get about Feldenkrais for anxiety: people want to know whether the gentle thing is the right gentle thing, when what they're actually asking is whether anything will hold.

I do want to sit with the sleep finding, carefully, because it's the only place the two came apart at all. An hour spent slowly changing how you lie down, turn over and roll leaves something different in the body than an hour spent watching the breath. Both are worth an hour. Only one of them hands your nervous system new information about being horizontal, which is the position it then has to hold for the next seven hours. That is my guess. The trial did not test it, and I would not want it repeated as though it had.

A few cautions about carrying any of this too far. These were people after cancer treatment, and their fatigue and pain are not the same animal as the fatigue that follows a long stretch of not moving much. The improvements were real but modest, a point or two on questionnaires rather than a different life. And gentle movement of any kind sits alongside the clinical care a person is already receiving, never in place of it.

What does travel is the knot. Sleep, fatigue and pain moved together in this data, which the authors read as an argument for treating them as one cluster rather than three separate problems, and anyone who has watched a bad night turn the next day's back into something else entirely already knew that in their bones.

The people I keep thinking about are the third who had no preference. They walked in with no opinion, got handed whatever the draw gave them, and did exactly as well as the ones who had spent the drive over hoping for Tai Chi. Nobody designs a trial to find that out. It's usually the variable you adjust away and never mention again.