Why You Perform Worse When It Matters Most

You have done this a hundred times. In the empty gym, in the practice room, in the version where nobody was watching, it came out clean and you did not have to think about how. Then comes the one that counts, and somewhere between the start and the middle your hands turn into equipment you are operating rather than limbs you have.

Afterwards, the standard diagnosis is that you wanted it too much, or that you are just not someone who performs under pressure. Both explanations are useless, because neither tells you what to do differently. There is a better one, it is specific, and it comes with a training method that worked in the lab.

Choking Has a Precise Definition

Roy Baumeister's 1984 paper is where the term gets its technical meaning: performance decrements under circumstances that increase the importance of good or improved performance.

Note what that excludes. Choking is not being bad at something. It is not nerves. It is the specific case where the importance of the occasion goes up and your performance goes down from your own baseline. If you have never done it well in practice either, you did not choke. You were just not ready, which is a different and more fixable problem.

The model Baumeister proposed is that pressure increases conscious attention to your own process of performing, and that this increased attention disrupts execution that had become automatic or overlearned.

Six experiments supported it. Three showed that increasing attention to your own performance process produced decrements on its own, with no pressure needed. Three others produced the same decrements through situational pressure instead: implicit competition, a cash incentive, and an audience.

Worth sitting with, given how much goal advice runs on raising the stakes. The incentive was doing its job on motivation and taking the skill apart at the same time.

Only Automatic Skills Can Choke

The mechanism got sharper seventeen years later. Sian Beilock and Thomas Carr ran four experiments that started somewhere unexpected: what expert golfers can remember about their own putts.

Experts had impoverished episodic recollection of specific putts they had just made. They could not tell you much about them. Which told the researchers something important, that skilled putting is encoded in a procedural form that runs without step-by-step attentional control. The expert cannot describe it because the expert is not doing it consciously.

Then they tested who chokes. Golf putting, a proceduralized skill, against alphabet arithmetic, a task analogous to mental arithmetic that had not become automatic.

That is the part that reframes the whole experience. Choking is not a symptom of being under-prepared. It is a symptom of being prepared enough that the skill left conscious control, combined with a situation that drags it back.

The Training That Worked, and the One That Did Not

Beilock and Carr then tested whether practice conditions could prevent it, and the two results together are the most actionable thing in either paper.

Training under a distracting second task did nothing. Choking in putting was unchanged.

Training while self-conscious, under observation, eliminated it.

The logic follows the mechanism exactly. If choking is caused by attention turning inward under observation, then practising while observed is practising the failure condition. Distraction training rehearses a different problem, so it protects against a different problem.

The results support explicit monitoring theories of choking and, as the authors put it, the popular but infrequently tested belief that attending to proceduralized skills hurts performance.

Two limits worth stating. The self-consciousness result is specific to putting in these experiments, not a demonstrated universal remedy. And there is a separate literature on choking through distraction in tasks that lean heavily on working memory, which these experiments did not test. This is the account for automatic motor skill, not a theory of every bad performance.

There is a second reason this pairing is worth knowing. Distraction training is the intuitive preparation, and it is what most people reach for: rehearse with the television on, practise in a noisy room, get used to chaos. That builds something, but it is not protection against pressure, because pressure does not work by taking your attention away. It works by pointing your attention at yourself. Those are opposite failures, and the training that solves one has no reason to solve the other.

The Detail Nobody Expects

Baumeister found that people low in dispositional self-consciousness were more susceptible to choking than people high in it.

Backwards, on first reading. But it sits comfortably with the mechanism. If you are habitually somewhat aware of yourself performing, the spotlight of an audience is not a new state, and it changes less. If you normally operate without any self-observation at all, the sudden arrival of it is a large change to the conditions your skill was built under.

The unbothered natural is not immune. They may simply have further to fall.

Outside the Golf Course

Almost none of us are putting for money, and the pattern shows up everywhere anyway, in the same shape.

The presentation you have given fluently a dozen times, delivered badly to the people who matter. The technical interview where you cannot do work you do daily. The recital, the driving test, the conversation you had rehearsed and then narrated to yourself while having it. All the same structure: a competent automatic process, plus a reason to watch it happening.

There is also a quieter implication for how goals get built. If raising the stakes can degrade skilled execution, then commitment devices and hard deadlines are not universally safe. They are excellent for behaviours that need to happen at all: showing up, starting, not skipping. They are riskier bolted onto a performance you already know how to do, where the added pressure buys motivation you did not need and adds monitoring you cannot afford.

What To Actually Do

Rehearse the conditions, not just the content. Run it once in front of one person. Record yourself. Perform the version where you can feel yourself being observed, because that is the variable the research says matters, and it is the one people leave out of every practice schedule. One observer is usually enough. What you are acclimatising to is not the size of the crowd, it is the sensation of being watched while executing, and a single colleague in a chair produces that at close to full strength for a fraction of the trouble.

Give attention somewhere else to go. Since the failure is attention landing on your own mechanics, the counter is a specific external focus during execution: the target, the next line, the other person's face. Not "relax", which is an instruction with nothing to do.

Do not troubleshoot mid-performance. The urge to fix the mechanics while executing is the mechanism itself, arriving disguised as diligence. Note it and repair it later.

Separate the two failure modes. Under-prepared and choking look similar afterwards and need opposite responses. If it was fine in practice and fell apart when it counted, more practice of the same kind is not the fix. If it was never fine in practice, this article does not apply to you and the work is the work. The diagnostic takes one honest minute: pull up what your unobserved version of this actually looked like, recently, and compare. People get this wrong in both directions, either grinding more reps against a problem that was never about reps, or blaming their nerve for a skill that had not been built yet.

Reframe the physical signal. The racing heart is not proof you are about to fail. We covered stress reappraisal separately, and it pairs well here: what arousal means to you is not fixed, and the meaning changes the response.

Keep a record of the ones that counted. Choking is situational, so the useful data is which situations produce it for you, and that is invisible without notes. It is one of the reasons Future You puts journaling next to goal tracking, free on iOS and Android.

The Reason This Is Better News Than It Sounds

The folk theory says some people have it and some do not, which leaves you nothing to work on. The research says something narrower and more useful: pressure makes you watch yourself, and watching yourself wrecks anything you had made automatic.

That is a mechanism with handles on it. You can practise being watched. You can pick something outside yourself to look at while you execute. You can notice that the biggest occasion is the worst possible moment to start improving your technique, and leave the mechanics alone until afterwards.

None of it makes the occasion smaller. It just stops the occasion from turning you into an observer of your own hands, which is the only thing that was ever actually going wrong.