The Case for Speaking Badly in Public
By Dara Halloran · Community & Practice
The most common way people prepare for a talk is to run it in their head. The second most common is to say it out loud alone. Both are useful and both have a ceiling, and the ceiling is lower than almost anyone expects.
What rehearsing alone cannot train
The physiology. Alone, your heart rate is at rest. In front of people it is not. Every skill you have rehearsed in a calm body has to be re-executed in an aroused one, and some of them will not survive the transfer. You cannot train that alone, because you cannot summon it alone.
Recovery. Rehearsing alone, when you lose your thread, you stop and restart. Nobody watching restarts. What you need is the skill of losing the thread and continuing anyway, and you can only rehearse continuing if there is a reason not to stop.
Reading a room. Attention has a texture, and you learn to feel it only by being in front of it. Alone, you are always speaking to an audience that is politely fascinated, because you are imagining them.
Timing. Laughs take time. Points need a beat. Alone you compress all of it, then run long on the night and cannot work out why.
Practising under pressure is its own skill
The first point on that list has been tested outside speaking, in sport, where performing under pressure is measured precisely.
Raôul Oudejans and Rob Pijpers ran two experiments with experienced athletes.1 In the first, expert basketball players practised free throws over five weeks, some under conditions designed to make them anxious and some without. In the second, expert darts players practised throwing either from high on a climbing wall, which made them anxious, or from low on it. Then everyone was tested under pressure.
The pattern was the same both times. Players who had trained only in calm conditions got worse when the pressure arrived. Players who had trained with anxiety held their performance, even though they still felt more anxious, had higher heart rates and reported more effort. The authors' interpretation was that practising under anxiety lets people get used to the specific processes that come with it, so those processes stop derailing the skill.
Speaking is not darts, and the study does not prove that the same thing holds on a stage. But the shape of the finding matches what speakers report: a talk rehearsed only in a quiet room can fall apart in a full one, and the fix is not more quiet rehearsal. It is rehearsal with some of the pressure left in.
Why the low-stakes room matters
The obvious response is "so speak in public more." The problem is that most public speaking opportunities are high-stakes: the client pitch, the conference slot, the wedding. You get few of them, they are spaced far apart, and the cost of experimenting is high enough that nobody experiments. So people can give the same talk, the same way, for years.
What produces improvement is a room with real people and low consequences. Real people, because that is the part you cannot simulate. Low consequences, because otherwise you will play it safe, and playing it safe is the thing that stops you improving.
That is the design rationale for a weekly session. Motivation and accountability are side effects. The point is a supply of the one input that private practice structurally cannot provide.
Failing on purpose can teach more
There is a second argument for a bad talk, and it comes from education research. Manu Kapur studied what he called productive failure.2 Secondary school students worked in groups on physics problems. Some groups got well-structured problems they could solve. Others got messy, ill-structured problems that they mostly failed to solve well. Afterwards, working individually, the students who had struggled and failed did better on new problems than the students who had succeeded in the first place.
The lesson is not that failure is good in itself. It is that an attempt which stretches past what you can currently do produces information an easy success does not. A talk that goes exactly as rehearsed teaches you that you can deliver that talk. A talk where you tried something new and it half worked shows you where the edge is.
Deliberately spending a session badly
The people who improve fastest often treat some sessions as disposable. They go in planning to try a thing that might not work:
- Open with no notes at all
- Tell the story you usually cut because it is too personal
- Deliberately pause twice as long as is comfortable
- Argue the side you do not hold
- Do the whole thing without a single filler word, and stop dead every time one starts
Some of these produce a bad talk. That is the point. The session is a place where a bad talk costs you very little and teaches you something, a trade that is hard to find anywhere else in your speaking life.
Pick one experiment per session, not five. If everything changes at once, you will not know which change did the work. Tell one person beforehand what you are trying, and ask them afterwards only about that.
The reframe
Stop treating practice sessions as small performances. They are a different activity with a different goal.
A performance is for the audience. Practice is for the speaker. If you leave a practice session having performed competently and learned nothing, you have spent the evening on the wrong one.
There is also less at stake in a stumble than it feels, which makes the experiment cheaper still. The people in the room notice far less of it than you do, as the research on the spotlight effect shows. Go in planning to be a little bad, and see what you learn.
References
- Oudejans, R. R. D., & Pijpers, J. R. (2009). Training with anxiety has a positive effect on expert perceptual–motor performance under pressure. Quarterly Journal of Experimental Psychology, 62(8), 1631–1647. https://doi.org/10.1080/17470210802557702
- Kapur, M. (2008). Productive Failure. Cognition and Instruction, 26(3), 379–424. https://doi.org/10.1080/07370000802212669