You said twenty minutes, and it is an hour later, and you are not done.
It was the slides for Thursday. You have made slides like them a dozen times, and every one of the dozen took an hour, and you still said twenty minutes with a straight face. It is 6:20 now and dinner was at six. The lost hour is not the problem. The problem is you will say twenty minutes again next week, and mean it.
Everyone guesses short; you have more riding on the guess
The planning fallacy is the name for guessing short about your own tasks while knowing you always have.
Three researchers gave it the name in 1994 after asking students to predict when they would finish their own work. The students knew they had run late before. They guessed short anyway, and ran late again (Buehler, Griffin and Ross, 1994). This is not an ADHD finding. It is a finding about people, and the students were ordinary students.
What is different with ADHD is what the wrong guess costs. A guess is how you build an afternoon. Four tasks at twenty minutes each is a plan for a good afternoon. Four tasks at an hour each is one task done by six and a story about a wasted day. The story is the expensive part, and it is false. You were never lazy. You built the afternoon out of guesses, and the guesses were built wrong.
Time is also felt differently with ADHD, which makes the guess harder to check from inside. Time blindness has its own page. This one is about the guess, and the one thing found to improve it.
Why the guess is wrong: you build it from the plan
A guess made from the plan sees the steps and none of the interruptions.
When you ask yourself how long the slides will take, your mind plays the plan: open the deck, paste the numbers, tidy the fonts, done. Twenty minutes. The plan is a story about the future, and stories have no printer jam in them. No colleague at the desk. No number you have to go and find. No reply you wait for. The past has all of those in it, every time, which is why the past says an hour and the plan says twenty.
You have the past. You do not consult it. This is the whole of the planning fallacy, and it is the whole of the way out.
The one thing found to help: remember last time, first
Before you guess, recall how long it took last time.
In a series of lab experiments, people who first recalled how long a past task took predicted the next one better than people who guessed cold (Roy, Mitten and Christenfeld, 2008). Nothing about their will changed. What changed was the material the guess was built from: the memory of a real run, instead of a story about a clean one.
Two things follow. First, the order matters: recall, then guess, because the point is to build the guess out of the memory. Second, a written number beats a remembered one. "Last time felt long" is a feeling, and feelings about time are the thing in question. "Last time: 58 minutes" is a number, and a number does not care how the hour felt.
So the whole method is a record. Guess, time it, write both down. Next time, read the record before you guess. The guesses get closer, and the task gets smaller in your head, because a task you can time is a task with an end.
Then race the record, and nothing else
Once you have a real number, the number becomes a game you can win.
The catch is who you race. Racing a target someone else set, or a stranger's time, runs out fast. Racing your own past holds up: in a longitudinal study of a VR exercise game, people who raced ghosts of their own earlier runs pushed harder and enjoyed it more, and it lasted (Michael and Lutteroth, 2020). One study, exercise, not slides. But the shape is the useful one. The only opponent who cannot make you quit is the person you were last Tuesday, and your record is their time.
Starting is the win. A slow run is still a run you timed, and a run you timed is a better guess next week. The whole game rule is here.
How Kaapi does it
Timi asks you to guess, times the task, and scores the guess, not the speed.
Games are off until you turn on "Make it a game". Then Timi, the pebble at the bottom right, offers at most one a day, chosen from what is on your list, with no menu. One of them is the guess. "Guess how long this one takes?" You pick a number of minutes, do the task, and when you finish, the score is about the guess: "Close call. Guessed 10, took 12." Or "Longer than it felt." Or "Quicker than you thought." A slow task can be a good guess, so a slow day is not a bad one.
We show you how your guesses run only after three of them, because two guesses are an anecdote. After the third, you see how long your ten-minute tasks take in practice, from your own finished runs and not from a rule of thumb.
The second game is the race. Once you have finished a task you do often a few times, Timi knows your usual. "Your usual is 12:00. Race it?" Then: "2:00 faster than your usual." Or "1:30 over your usual. Still done." Your own past is the only ghost on the track. No points, no leaderboard, and nothing is counted back at you.
How to do this without Kaapi
You need a stopwatch, a pen, and one column you never had before.
- Before a task you do often, write your guess in minutes on the row.
- Start a stopwatch. When you finish, write the real number next to the guess.
- Next time, read the real number before you guess. Say it out loud, then guess.
- After five runs, take the middle number as your usual. Race only it, and write "started" on the days you stop early.
In a week you will have a record. In a month the record will be closer to the truth than your gut ever was, because the record does not remember the plan. It remembers the hour.
Frequently asked questions
Why am I so bad at guessing how long things take with ADHD?
Everyone guesses short about their own tasks; a 1994 study named it the planning fallacy. The guess is built from the plan, which has no interruptions in it, instead of from last time, which had all of them. With ADHD, time is also felt differently, so the guess is harder to check from inside.
How can I get better at estimating time?
Recall how long the task took last time, before you guess. In lab experiments, this one step improved the next prediction. Keep a written record of guess and real time for the tasks you do often, and read the record before each new guess. The guesses close in over weeks.
Should I double my time estimates?
A flat multiplier is another guess. Your own record is better: after five runs of a task, the middle number is your usual, and it beats any rule of thumb because it came from your hours and not from advice. Before you have three, recall last time and guess from it.
Does timing myself make tasks feel worse?
It depends on what you score. Timing to beat a target you did not set, or a stranger's time, wears out. Timing to see how close your guess was, and racing only your own usual, held up in a 2020 study. Score the guess, not the speed, and count nothing against yourself.
Sources
- Buehler, R., Griffin, D., and Ross, M. (1994). Exploring the "planning fallacy": Why people underestimate their task completion times. Journal of Personality and Social Psychology, 67(3), 366–381. doi.org/10.1037/0022-3514.67.3.366. The planning fallacy; students.
- Roy, M. M., Mitten, S. T., and Christenfeld, N. J. S. (2008). Correcting memory improves accuracy of predicted task duration. Journal of Experimental Psychology: Applied, 14(3), 266–275. doi.org/10.1037/1076-898x.14.3.266. Lab experiments.
- Michael, A., and Lutteroth, C. (2020). Race Yourselves: A longitudinal exploration of self-competition between past, present, and future performances in a VR exergame. CHI 2020. doi.org/10.1145/3313831.3376256. One longitudinal study, VR exercise.