Doing It Again Tomorrow: How Good Days Become Repeatable

Something went well. Not luckily well, actually well: you were sharp, the timing held, the hard section went past without you noticing it, and you came out knowing you had been at the top of your range. Then somebody asks what you did differently and you find you cannot say. It felt like weather.

It was not weather. Days like that have inputs behind them, and the inputs are duller and more knowable than the day itself. Sleep, training, food, warmup, first move. Write those five down while the day is still fresh, keep doing it for a few weeks on ordinary days too, and a pattern shows up that you can then run on purpose. That is the whole method, and its real payoff is not one repeat performance. It is a floor that rises underneath every performance you give from then on.

How do I repeat a performance that went well?

Write down the five inputs behind it while you still remember them: sleep, training, food, warmup and first move. A good day is repeatable to the extent that its inputs are knowable, and they are far more knowable than they feel from the inside.

Be concrete. Not "slept well" but seven and a half hours, in bed by eleven. Not "trained" but lifted Tuesday, rest day Wednesday. Not "ate fine" but breakfast at eight, real lunch at one. The warmup is whatever you actually did in the last hour, and the first move is the exact thing you opened with, because the first ninety seconds set the tone for everything behind them.

Do not log how it felt. Feelings are the output you are trying to explain, so putting them in the same column as the causes is how the whole exercise turns into a diary. Five lines, ninety seconds, on the day itself.

What should I actually log, and for how long?

The same five lines on ordinary days as well as big ones, for about six weeks. You need some average days and some poor ones in the record, because a log that only contains your best performances cannot tell you what made them best.

Six weeks is a practical span rather than a scientific one. It is long enough that ordinary days and bad days both turn up in the record, which is what makes any comparison possible, and short enough that you will actually finish it. Keep it somewhere with no friction: the notes app, the back of a diary, one line in a spreadsheet.

Monitoring is doing real work here, not just record keeping. A meta-analysis of 138 randomized experiments found that people prompted to monitor their progress toward a goal were more likely to attain it, and that the effect was larger when the information was physically recorded or reported to somebody. So the log is not admin you do around the practice. It is part of the practice.

How do I tell which inputs actually mattered?

Put your best three days beside your worst three and look for the input that moves in the same direction every time. Usually two of the five do, and the rest is noise you can drop without a second thought.

For most people one of the two is sleep. When college basketball players extended their nightly sleep over several weeks, their sprint times and their shooting accuracy both improved, and adults generally need seven or more hours a night for the systems that hold attention and reaction time together. If your best days sit behind your longest nights, you have found your first lever and it is not a mysterious one.

The second is usually the one you would not have guessed: the size of lunch, whether you trained the day before, whether you spent ten minutes alone before you started. Look for it in the gap rather than the average, meaning the input that differs most sharply between your best three and your worst three, not the one you have the strongest opinion about. When you find it, write it down as a rule in one sentence. Two rules you actually follow beat a checklist of nine you read once and abandoned.

What do I do when a good day does not repeat?

Check the inputs before you rewrite the method, because two short nights and a skipped lunch explain more of the variation than most people expect. A rule that works four times out of five is a working rule, not a broken one.

Performance is not deterministic and the log is not a machine. What the five lines buy you is a shorter list of suspects. When a day comes in flat and you can see you slept five hours and skipped the warmup, you have an explanation and a fix instead of a mystery and a mood. When every input was in place and it still went flat, that is useful too: it means the variation is somewhere you were not looking, usually the room, the other people in it, or something you had no vote on at all.

Do not add a sixth input in response to one disappointing day. The log stops working the moment it becomes a chore, and every extra line makes it likelier you quietly stop keeping it. Give the two rules a month before you renegotiate them.

How do I make the two that mattered stick?

Attach each one to something that already happens at a fixed time, rather than to your intention to do it. Repetition in a stable context is what turns a decision into a default, and the honest timescale for that is weeks to months.

The number people quote is twenty one days, and it is folklore: it traces back to a surgeon's observation about patients adjusting to their new appearance, not to a study of habits. When researchers actually tracked daily behaviors, the sense of automatic action climbed and then leveled off after an average of about sixty six days, with wide variation between people and between behaviors. The same work found that missing a single day did not meaningfully set anybody back, which is the part worth carrying.

So do not resolve to sleep more. Set the phone on the shelf when you brush your teeth. Do not resolve to warm up. Put the ten quiet minutes immediately after you park, every single time, until you would notice their absence. The trigger does the remembering, which frees you from having to.

Expect the first fortnight to feel like effort and expect that to fade. That fading is the entire point of a habit: it is capacity you get back, permanently, for a fixed cost paid once.

A rising floor is what lets you keep aiming higher

Peaks are exciting and they are also mostly out of your hands. The floor is not. Raising the floor means your ordinary day gets closer to the day you thought was exceptional, and when the ceiling moves after that, it moves from a higher place. That is how a hard pace turns into a long career instead of one good year.

There is a fast way to find out whether your five inputs are real, and it is to hand them to somebody who has a date on the calendar this month and watch what happens. Teaching the routine is how you learn it twice: the parts you cannot explain out loud are the parts you were still guessing at, and a person asking you questions finds them faster than another month of solo practice will. You get a tested routine, they get a corner, and the next time you are the one walking in you will already know it works under pressure.

Sources

• PubMed, Does monitoring goal progress promote goal attainment? A meta-analysis of the experimental evidence
• PubMed, The effects of sleep extension on the athletic performance of collegiate basketball players
• PubMed, Making health habitual: the psychology of 'habit-formation' and general practice
• National Heart, Lung, and Blood Institute, How Much Sleep Is Enough?