Full swing · Follow-through

Does random practice actually work?

Somewhere in the last decade, golf absorbed a finding from motor learning laboratories and turned it into a commandment. Do not hit the same club twice. Randomise everything. Blocked practice is a comfortable lie.

The finding is real. The commandment is not, and the gap between them is one of the more interesting stories in sports science — partly because it is still being argued about in the journals right now, and almost none of that argument has reached the golf internet.

The short version, before the detail: randomising is one lever among several for making practice usefully harder, it belongs near the end of a progression rather than at the start of one, and how much of it you should be doing depends on which skill you are asking about.

1What the original experiment actually did

In 1979 Shea and Morgan had people learn three variations of a timing task, either in neat blocks or in a jumbled order. The jumbled group performed worse during practice and better on a retention test days later. This is the contextual interference effect, and it is one of the most replicated results in the field. Note what the task was, though: pressing keys in a sequence. Not hitting a golf ball.

2Why it should work, in principle

Two explanations compete and both are plausible. You may be forced to reconstruct the movement plan from scratch each time rather than reusing the one still warm from the last attempt, which is effortful and therefore memorable. Or the jumbling may make you compare the variations against each other and encode each more distinctly. Either way the mechanism is the same shape: practice that feels harder leaves more behind.

3Then researchers tested it on actual sports, and it wobbled

Brady reviewed twenty-five years of it in 2008 and concluded the effect is robust in basic research and considerably weaker in applied settings. A 2024 meta-analysis of 54 studies found a medium benefit overall but described the applied-setting benefit as almost negligible. Most pointedly, a 2023 review of 36 sport studies examined 183 pooled outcomes and found only about 20 per cent behaved the way the theory predicts. Its authors titled the paper around the word myth.

4The field has not settled this

It would be equally dishonest to tell you the effect has been debunked. That 2023 review drew published commentaries and rebuttals arguing about its methods and how it pooled outcomes, and the exchange is ongoing. What you can say with confidence is that anyone presenting this as established fact for golf has not read past the 1979 paper.

5The framework that makes sense of the mess

Guadagnoli and Lee proposed in 2004 that what matters is not difficulty in the abstract but functional difficulty — how hard a task is relative to your current skill. There is an optimal challenge point, and learning falls off on both sides of it. Too easy and there is nothing to learn from. Too hard and you cannot extract the useful signal from the noise. Interleaving is a difficulty dial, and turning it up only helps if you were below your challenge point to begin with.

6Which tells you who it is actually for

If you are grooving a genuinely new movement — a rebuilt grip, a changed takeaway — you are already at the edge of what you can process, and shuffling clubs on top of that pushes you past it. Block that work. If you are a competent player whose range form is comfortable and whose course scores do not reflect it, your practice is almost certainly too easy, and interleaving is a reasonable lever. The advice is not wrong. It is just conditional, and the condition is you.

7It is a destination, not a starting point

This is the framing the popular version loses. Randomising is the far end of a progression - groove it, vary it, interleave it, simulate holes, play - and arriving there matters because golf on the course is genuinely random. But you climb to it as a skill becomes competent, rather than beginning there and wondering why nothing sticks. Nobody learns a new grip in a jumbled sequence.

8And it is one lever, not the whole idea

Interleaving sits inside a larger principle - what the Bjorks called desirable difficulties - alongside spacing sessions further apart, varying the target rather than the club, reducing how often you check feedback, and adding consequence to a shot. Several of those are easier to calibrate than club-shuffling and less likely to overshoot. If your practice needs to be harder, interleaving is not automatically the first lever to reach for.

9What to do on Tuesday

Block the new thing, interleave the established thing, and put the interleaved work at the end of the session rather than the start. Since you are almost certainly at different stages with different clubs, expect to be doing both in the same hour. Then be sceptical of your own verdict for at least three sessions, because the entire finding rests on the observation that what feels productive in the moment is a poor guide to what remains.

Common questions

So should I stop randomising my practice?

No, but stop treating it as the whole of good practice. Keep it for skills that are already competent, because the course is random and you have to arrive there eventually. Drop it for anything you are still building, and stop expecting the transformation the popular version promises.

Why is the laboratory evidence so much stronger?

Laboratory tasks are simple, brief and precisely measured. Golf is complex, slow to learn and measured noisily. A complex task may already sit near the top of what a learner can process, so adding interference pushes past the optimal challenge point rather than toward it.

Does skill level really change the answer?

It is the best-supported moderator we have. Difficulty is relative to the performer, so the same practice schedule can be usefully demanding for a scratch player and overwhelming for someone six months into the game. Beginners generally need more blocked repetition than the popular advice allows.

Is this true of short game practice too?

The same logic applies, and chipping is a decent place to interleave because the movement pattern is stable and the variables that change - lie, distance, landing spot - are exactly the ones that vary on the course. Varying the target is usually more useful than varying the club.

Sources

Every claim about evidence on this page traces to one of these. Follow them and check.

Shea, J. B., & Morgan, R. L. (1979). Contextual interference effects on the acquisition, retention, and transfer of a motor skill. Journal of Experimental Psychology: Human Learning and Memory, 5, 179-187.
The original laboratory demonstration: random practice order depressed acquisition and improved retention and transfer.
Brady, F. (2008). The contextual interference effect and sport skills. Perceptual and Motor Skills, 106(2), 461-472.
Reviewing twenty-five years of work: the effect is robust in basic research and considerably weaker in applied sport settings.
Ammar, A., Trabelsi, K., Boujelbane, M. A., Boukhris, O., Glenn, J. M., Chtourou, H., et al. (2023). The myth of contextual interference learning benefit in sports practice: A systematic review and meta-analysis. Educational Research Review, 39, 100537.
Across 36 sport studies and 183 pooled outcomes, only about 20 per cent agreed with the predicted contextual interference pattern.
Czyz, S. H., Wojcik, A. M., Solarska, P., & Kiper, P. (2024). High contextual interference improves retention in motor learning: systematic review and meta-analysis. Scientific Reports.
A medium retention benefit across 54 studies overall, but an almost negligible one once the setting is applied rather than laboratory.
Guadagnoli, M. A., & Lee, T. D. (2004). Challenge point: A framework for conceptualizing the effects of various practice conditions in motor learning. Journal of Motor Behavior, 36(2), 212-224.
Functional task difficulty relative to skill level determines the optimal challenge point, explaining why the same manipulation helps some learners and hinders others.
Bjork, R. A., & Bjork, E. L. (2020). Desirable difficulties in theory and practice. Journal of Applied Research in Memory and Cognition, 9(4), 475-479.
Places interleaving inside the wider family of desirable difficulties, alongside spacing, variability and reduced feedback.

Last updated August 18, 2026