At 11:40 you stand in the hallway with one shoe on the baby and one in your hand. The plan was simple. Pharmacy, post office, home by 12:30, nap at 12:45. You gave it a generous fifty minutes. It is now 11:40, you have not left the house, and somewhere in the back of your mind a small, cold voice says what you already know: you are going to do this again.
By 12:50 you are in the car. The baby is rubbing both eyes with their fists, and you are doing the arithmetic nobody warns you about. Every minute past the window costs more than the one before it.
It is easy to read this as a personal failing. You think you are bad at time, or disorganized, or that other parents have some secret. You aren't, and they don't. You are running into one of the most reliably documented quirks of human thinking, and it has a name.
The planning fallacy, in a diaper bag
In 1979, Daniel Kahneman and Amos Tversky described the planning fallacy: people predict how long a task will take by imagining the best-case path, and then they are wrong in the same direction over and over. Later work by Roger Buehler, Dale Griffin, and Michael Ross in the 1990s showed how stubborn it is. People underestimated how long their own projects would take, even when they knew that similar projects of theirs had run late before. They treated the past as a different kind of case, one with unusual circumstances that wouldn't repeat.
Kahneman called the habit of building a forecast from the specifics of the current plan the inside view. The outside view asks a duller question: how long do tasks like this usually take? Humans prefer the inside view because it feels like knowledge. You can see the whole trip in your head, and it runs smoothly.
In that mental movie, nobody has a blowout in the car seat. The baby does not decide to eat at the moment you reach the door. The parking lot is not full. The pharmacy line doesn't have one person who wants to talk about their prescription. Your imagination edits out every friction point, and with a baby the friction points are the main event.
Why a baby makes the fallacy worse
The fallacy is bad enough on its own. Add an infant and three things stack on top of it.
Hidden subtasks. Researchers who study time estimation have found that breaking a task into its parts tends to produce longer, more accurate estimates than judging it as one lump. "Go to the pharmacy" feels like twenty minutes. Unpacked, it is: feed or top off, change the diaper, find the other shoe, load the bag, buckle the car seat, drive, park, unbuckle, carry, wait, carry back, buckle again, drive home, unload. Almost none of that appears in the sentence "quick errand."
Unpredictable interruptions. A baby is not a fixed-duration task. A feed can run twelve minutes or thirty. A diaper change can be routine or can need a full outfit change. The variance matters as much as the average, because you are planning against a hard edge: the end of the wake window.
An asymmetric penalty. If you finish a work task an hour late, you apologize. If you finish an outing twenty minutes late with a baby, you can end up with a baby who has built up a lot of sleep pressure and is now fussing in a car seat. A baby who is too tired often struggles to settle, so the nap you were racing toward gets harder, shorter, or skipped. The cost of lateness is lopsided, so the right amount of padding is more than your instinct says.
None of this means errands are doomed. It means the fix is not trying harder to hurry. The fix is a different way of planning.
Swap the inside view for your own data
The most useful finding from this research is that the outside view works. When people forecast using the record of how similar tasks actually went, rather than a story about this one, their predictions improve. Kahneman and Tversky called this approach using a reference class, and later planners adopted it for large projects. You can use it on a hallway and a car seat.
You don't need an app or a spreadsheet to start. You need three or four real measurements.
Next time you leave the house, note the time you decide to go and the time the car door closes behind you. That gap is your real "getting out" time. Note it again for the return: car door opens to baby asleep. Do this for a few outings. Most parents are startled. The number that comes back is usually well above the one in their head, and it is remarkably steady from day to day. A steady number, even a large one, is something you can plan around.
The unpredictable part is smaller than it feels. The routine part is just bigger than you admit.
Plan backward from the window, not forward from the door
Most of us plan an outing like a story, starting at the beginning. But the nap is the fixed point, so plan from the end.
Start with the time the baby should be down. Subtract your measured return and settling time. Subtract the errand itself. Subtract your measured getting-out time. Subtract a buffer for the unknown. The result is your real departure time, and it is usually earlier than feels reasonable, which is exactly the point. The planning fallacy is the feeling that there is more time than there is.
Two refinements help.
First, set the deadline as "back in the car" rather than "back home." The last leg is the one you can least compress, and a mid-errand deadline gives you a decision point while there is still time to act.
Second, decide in advance what you will drop. A plan with a pre-chosen cut is easier to follow than one that requires a decision when you are already late and frazzled. Psychologist Peter Gollwitzer has studied this kind of "if-then" commitment, called an implementation intention. A rule you wrote while calm, such as "if we are not in the car by 12:00, the post office waits until tomorrow," holds up better than a judgment call made at 12:10 with a baby starting to wail.
Your next moves
- Time two real outings this week. Write down when you decide to leave, when the car door closes, when you return, and when the baby is asleep. Use those numbers, not your guess, as your baseline.
- Plan backward from nap time. Write the nap start at the bottom of a note, then subtract return time, errand time, your measured getting-out time, and a 15-minute buffer. That total is your departure time.
- Pick a "turn around" time and one errand to drop. Decide now which stop gets cut if you are behind, so you don't have to decide mid-trip.
- Pre-pack the night before. Keep a second diaper bag restocked and by the door, with a spare outfit in it. That deletes the most variable part of your getting-out time.
- Do the one outing that matters most at your baby's best time of day. If your baby is usually at their calmest in the first half of the wake window, put the most important stop there and leave the flexible ones for later.
One safety note: if the baby does fall asleep in the car seat on the way home, that is fine for the drive, but once you arrive, safe-sleep guidance from pediatric bodies is to move them to a firm, flat sleep surface rather than leave them sleeping in the seat.
The kindness in better planning
There is a quiet emotional payoff here, and it matters more than the logistics. Each time the day goes sideways, a parent tends to file it as evidence: I can't manage this. I'm bad at being organized now. The planning fallacy says otherwise. Your mistake wasn't a character flaw. It was a prediction made by a mind that is built to underestimate, now working on four hours of broken sleep.
Once you know the bias, lateness reads differently. It becomes data. You were not careless. You used the inside view, which everyone does. Next time you will use the outside one.
This is the part of Drowsy that earned its place in our own routine. It keeps a running picture of your baby's actual wake windows, so the next nap time on screen is built from your baby's own pattern rather than a chart average or an optimistic guess. When you plan backward from the window, the window is already there. If you would like to try it, you can find it at https://drowsy.lumenlabs.works. Either way, take the stopwatch idea with you. You will be surprised how much calmer a day gets once your plan matches the day you actually have.