Failure Preview

The practice interview goes badly on Tuesday. By Friday, you are not preparing for the real interview so much as waiting for Tuesday to happen again. Failure preview turns one miss into a trailer: the important event will fail because the smaller one did.

The two events may share a label without sharing an outcome. One happened in a particular room, with a particular level of preparation and luck; the other has not happened at all. Yet the thought "one failure means I'll fail again" asks the first event to write the second. A bad rehearsal before the real event feels like evidence, even when it is only one draw.

Sometimes the earlier miss really does carry information. The practice talk fell apart because the opening was unfinished, and the real talk still has that opening. The first pitch was rejected because the numbers were wrong, and the big meeting uses the same numbers. Then the first event is a clue about a cause that will still be there. The useful move is to fix the cause.

The bug is skipping that question. "It failed, so the important one fails" lets the outcome do the forecasting. A different room, a different task, and a different day get folded into one story: this is how it goes. One draw is doing the work of a pattern.

• "The practice interview was a mess, so the real one is already lost."
• "The test print jammed. The client meeting is going to fall apart."
• "I stumbled asking for the small favor, so the funding conversation is doomed."
• "The first date was awkward, so meeting their family will be a disaster."
• "I blanked on the quiz, so the exam will go the same way."
• "The hallway demo crashed. The launch review will crash too."
• "I got the time wrong for a minor appointment, so the important trip is going to unravel."
• "One pitch came back no this week. The big pitch is a no before I give it."

The rehearsal: a practice interview, a dry run, or a friendly audience goes badly. The real room is later and matters more. You spend the days between them replaying the mess, and you walk in already defending a loss that has not happened.

The morning of: a permit, a printer, a parking spot, or a short errand goes wrong in the hours before something you care about. The snag is local. It still gets promoted into a sign about the event it happened to sit near.

The early no: one reader, one buyer, or one small audience says no. A larger pitch is still scheduled. The no is one answer. It gets filed as the answer the important room will give, before that room has seen the work.

The shared missing piece: the rehearsal really did expose an unfinished opening, a wrong chart, or a step you skip when you are rushed. Here the first event is useful. It stops being useful when the story upgrades from "fix the opening" to "the important one is already lost."

The time between the two events gets spent inside an ending you already wrote. Preparation turns into damage control, or it thins out, because the result feels filed. You may shrink the ask, avoid the event, or arrive apologizing for a failure the room has not seen.

A real clue can get buried under the trailer. An unfinished opening, a wrong number, or a night of lost sleep is something you can work on. "I have no luck with this" does not say what to change. And when nothing from the first event will be in the room for the second, you still pay the fear.

Decisions follow the forecast. You cancel, postpone, or play safe on the event that mattered, using evidence from the event that did not. The cost shows up as a smaller life around the important dates, built from samples that were never asked to carry that weight.

A sample of one is easy to treat as a picture of the process. The miss already happened, in detail. The future event is only imagined. Matching them feels like consistency. The mind expects the next draw from "this kind of thing" to look like the draw it just watched, and it underweights how much a single episode can be noise.

The filing matters. A coin that lands tails is easy to call a fresh toss, or even a toss that "owes" you heads. A failed rehearsal feels like a fact about you. Once the miss is filed as how you do, the recent result is expected to continue. The important event inherits it.

Stakes finish the job. The more the future event matters, the more an ending in hand is a relief, even a bad ending. Uncertainty is expensive. A trailer is a way to stop waiting.

Tversky and Kahneman's 1971 paper on the law of small numbers described the habit underneath the trailer. People treat a small sample as highly representative of the population it came from, and they expect a second sample from that population to resemble the first more closely than sampling theory allows. In their own summary, the believer has undue confidence in early trends, such as the first few subjects in a study, and in the stability of patterns already observed. Their cases are research samples, not a bad morning before a hearing. Failure preview is that habit in ordinary clothes: one unlucky episode is asked to represent the process, so the important episode still ahead is expected to look like it. They also treated it as a way of seeing samples, active when the conclusion is welcome and when it is not. The forecast can feel like realism. It does not need a motive to expect the worst.

Ayton and Fischer's 2004 experiments separated two forecasts that feel like one. People predicted a roulette color, and they also tracked whether their own predictions were hits or misses. After a run of one color, they tended to expect the wheel to reverse. After a run of their own hits or misses, they tended to expect that run to continue. The two sequences were the same kind of chance. The difference was the label. A result filed as how the wheel behaves was expected to correct itself. A result filed as how I am doing was expected to repeat. Their study did not compare a rehearsal with a final performance. It showed why a failure preview sticks once the miss has been filed as yours, and why "I am due for a win" is a different error, the one people make about the wheel.

• The event you fear matters more than the one that already went wrong.
• The link you can say out loud is "because that one failed," and you have not named a cause that will still be there.
• A smooth small version would not have settled the big one the way the rough one unsettled it.
• Replaying the miss feels like it makes the forecast more solid.
• You are changing the important event - shrinking it, avoiding it, or arriving already beaten - on the strength of the smaller one.

Separate the episode that happened from the episode that has not.

• Name the two events on paper: what already failed, and what is still ahead.
• Ask what travels. An unfinished opening, the same wrong numbers, the same unread brief, the same injury. If you can point at it, the job is that cause.
• If you cannot point at a cause that will still be in the room, call the earlier outcome a sample of one. It is a weak forecast of a different event.
• Write two other plausible endings for the important event before you rate how likely the repeat is.
• Do one concrete preparation step, then rate the event again. A forecast that cannot survive one real change was leaning on the trailer.

What if the earlier failure shares a real cause?

Then use the cause. A rehearsal that fell apart because the opening was missing is evidence about the opening, when the real event still has that opening. It is weak evidence that the room, the day, or your luck has been previewed. Fix what travels, and let the outcome step down from trailer to clue. A shared cause is a reason to work. It is a poor reason to close the event.

Should I expect the opposite, since I am "due"?

A bad rehearsal is not a coin that owes you heads. Expecting a reversal because the first one hurt is a different mistake, the one people make about chance devices. The question here is narrower. Will anything from the first event still be true at the second? If nothing will, the second event is a new draw. If something will, work on that something. The correction is the traveling cause, not a promise that the next one goes well.

Against failure preview, the earlier miss is a sample to inspect, not a trailer to obey.

The practice interview

The jammed morning

The early no

Practice this pattern in the Reframing App - capture the trigger, label it (like Failure Preview), check evidence, and write a more balanced thought.