ADHD Time Estimation: Why "20 Minutes" Keeps Turning Into an Hour

"This'll take twenty minutes." You've said it about the expense report, the email to your landlord, the quick fix before lunch. It was true in your head. Then the clock said seventy-five, the next thing started late, and by evening the day's promises had quietly turned into apologies.

Your estimate is a simulation of the best case

When you guess how long something will take, you don't look anything up. You run a quick mental movie of doing the task — and that movie is edited. It shows the core of the work, start to finish, uninterrupted. It leaves out finding the file, logging in, remembering where you left off, the message that arrives halfway through, the tab you open to check one thing, and the five minutes it takes to get your head back afterwards. The movie is accurate. It's just not the whole film.

Psychologists call this the planning fallacy, a term Daniel Kahneman and Amos Tversky coined in 1979: people predict from the "inside view" — the specific plan in front of them — instead of from how long similar tasks have actually taken before. Everyone does it. Students do it with essays, governments do it with bridges. It's not a character flaw; it's what estimating from imagination produces.

Why ADHD makes the gap wider

ADHD doesn't create the planning fallacy, but it tends to stretch every part of it.

• The overhead you don't imagine is bigger. Getting started, gathering what you need, and climbing back in after a drift are exactly the parts the mental movie skips — and for an ADHD brain they're often the most expensive parts of the task.
• The clock you'd check against is unreliable. Even while you're doing the work, elapsed time is hard to feel, so you don't notice the estimate failing until it's already failed. That's time blindness — perceiving time as it passes. Estimation is its sibling: predicting a duration before it starts.
• The evidence doesn't stick. Last week's "twenty-minute" email that took an hour should be data for this week's guess. But it rarely comes to mind as a number — just as a vague bad feeling — so every estimate starts from scratch, from the same optimistic movie.

There's one more quiet pressure: a small estimate makes a task feel startable. "Twenty minutes" is a number you can talk yourself into; "an hour and a quarter" is a number you avoid. The optimism isn't only an error — it's sometimes doing a job. Which is why the fix can't be "just be more pessimistic."

Why "just double it" doesn't stick

The standard advice is to pad every estimate — double it, add half, round up to the hour. The trouble is that a flat rule is somebody else's number. You don't know if it's right for you, so you don't trust it, so in the moment you shave it back down. And the rule depends on remembering to be pessimistic at the exact moment your brain is being optimistic, which is the one moment it won't.

Kahneman's actual remedy was different: take the outside view. Instead of asking "how long will this take?", ask "how long have things like this taken me?" That works beautifully if you have the record. The catch is that the ADHD brain is the one least likely to be keeping it. So keep it outside your head, and let it do the arithmetic.

Ballpark: guess first, then let the clock answer

Ballpark is a Mac app built around one loop. Type the task and your gut guess in minutes. Hit start, and a live timer runs while you do it. When you're done, stop it, and Ballpark records the truth next to the guess: estimate, actual, and the ratio — 2.0× means it took twice as long as you thought. Did the task already, or timed it some other way? You can type the actual minutes in instead.

After a handful of tasks, the big number at the top becomes your personal multiplier — the typical factor by which reality beats your guess. From then on, as you type a new estimate, Ballpark shows the adjusted version alongside it: guess 30, and if your multiplier is 1.8, it suggests budgeting about 54. You're no longer negotiating with a feeling. You're reading your own track record.

Two design choices matter more than they look. First, the multiplier is a median, not an average. Most tasks run a bit long, but now and then one explodes eightfold — a rabbit hole, an interruption spiral. An average would let that single blow-up inflate every future suggestion; the median ignores it and represents your typical day. Second, it reads your lean in plain words: under-estimator (pad your plans), over-estimator (things go faster than you fear), or fairly accurate. A small trend chart plots your recent ratios around the 1× line, so you can watch the guesses tighten. Cancel a timer and the task just sits in your history as pending — nothing lost, nothing scolded.

A one-week calibration that actually fits a real week

You don't need a spreadsheet habit. You need about ten data points.

1. Log ordinary tasks, not heroic ones. Emails, errands, a report section, the dishes. Your multiplier is most useful for the everyday stuff that silently eats the day.
2. Guess fast and honestly — don't pre-pad. The point is to measure your gut, not to outsmart it. A padded guess hides the very number you're trying to find.
3. Let the timer run while you work, or enter the actual afterwards. Either way, record it the same day while it's still true.
4. Read the multiplier, not any single task. One 4× disaster isn't a verdict on you; it's one point the median will happily ignore.
5. Spend the corrected number somewhere you can see it. Put the adjusted durations into Day Map, where each task is a block sized to its real length — if the honest day doesn't fit, you'll watch it spill off the edge at 9am instead of discovering it at 4pm.
6. Watch for the opposite surprise too. Plenty of people find their dreaded tasks — the phone call avoided for three weeks — take a fraction of the time they feared. That's worth knowing: a smaller, truer number makes the thing easier to start.

Keep logging now and then and the number keeps itself honest. The goal isn't perfect predictions. It's replacing a guess you can't trust with a correction you can.

Squirrel apps are wellness tools, not medical devices, and don't diagnose or treat any condition.