We went looking for the average ADHD day. It isn’t there.
Not because we lack data. Because averaging ADHD brains together cancels out the exact thing that makes each one legible. That turns out to be the most useful finding we have.
Every number on this page is measured from real check-ins, including the ones we looked for and couldn’t find.
3,519 check-ins · 62 members active in the last 30 days · updated hourly
Three shapes. One flat average.
Ask the whole community when energy peaks and the answer is a straight line. Morning, afternoon and evening all land within a fifth of a point of each other, which reads as “ADHD energy is steady across the day.” It is not.
Sort those same people by their own peak window first, and the movement comes back. Three groups, three directions, crossing in the middle. Average them and they erase each other.
- Peaks morning (n=16)
- Peaks afternoon (n=14)
- Peaks evening (n=11)
- All 41 averaged (5.91 / 5.89 / 5.42)
Everyone who has logged across all three windows, 41 people. Ask when ADHD brains peak and the honest answer is: pick one.
You vary from yourself more than you differ from us.
Line up the members with enough history and their averages sit close together. Take any one of them and lay out their actual days, and the spread nearly doubles.
The distance between you and the person next to you is smaller than the distance between you on Tuesday and you on Thursday. Researchers have a name for it: consistently inconsistent. It is the reason a population average cannot describe you.
The top band is how far apart people are from each other. The bottom one is how far a single person travels across their own days. Same scale, roughly double the width. From 16 people with at least 20 logged days, averaging 77 days each.
Mood and energy move together.
Two sections arguing that averages mislead would be a stunt if we didn’t also show where they don’t. This is the one that survives. Across every check-in that recorded both, mood and energy track each other closely enough that one predicts the other.
Every dot is a check-in, and the bigger it is the more people landed there. The cloud leans, which is the whole point: tell us your mood and we can usually guess your energy within a point or so. From 3,511 check-ins that recorded both.
The gap between them leans on tomorrow.
Mood and energy travel together, but not perfectly, and the gap between them carries information. When your mood is running ahead of your energy, tomorrow tends to come up. When it lags behind, tomorrow tends to come down.
And this isn’t just low days bouncing back on their own. We compared days that started in the same place, so the only thing separating them is the gap.
Only the two starting points with enough days behind them are shown. Higher and lower than that we have a handful of days, which is not enough to say anything, so we don’t. This is the signal the app calls Mood Shield, checked against the people using it.
Two things we looked for and couldn’t show.
A page that only publishes its hits isn’t a findings page, it’s an advertisement.
Both of these are drawn to a scale we fixed before we saw the numbers. Stretching a flat result until it looks like a shape is how you lie with a chart without ever writing a false sentence.
The week has no community-level rhythm. Your own week may well have a shape; the crowd’s does not. Last 90 days.
506 days down, 561 days up. Even the steepest drops and climbs match: about 3 points either way.
We expected crashes to fall faster than recoveries climb. They don’t.
The questions we can answer with more data.
- Top activities that lift energy
Which actions correlate with higher next-day energy across the community? Walking, eating, meditation, social time, all of it.
- Worst combinations for crashes
Activity sequences that consistently precede a drop. The patterns nobody warns you about.
- What recovery actually looks like
Distinguishing rested-recovery from sluggish-recovery patterns. Which one your data resembles, and what tends to shift it.
- Aggregates only, never individuals. Every line on this page averages at least five people, and every panel at least 10. Below that floor the panels stay blank rather than infer from too few people.
- No chart auto-fits its axis. A flat result renders flat and gets labelled a null result. This is why the weekday chart looks different than it used to.
- Every number carries its sample size, on the panel, not buried in a footnote.
- We publish the misses. Two on this page today.
- Opt-out coming. We’re building an explicit opt-out into Settings so you can keep your data fully private without losing the rest of NeuroSpicy. Until that ships, contact us if you’d like to be excluded from the aggregates.
- No third parties. We don’t sell or share these aggregates. The patterns go back to the community that built them.
Help build this.
Every check-in adds a data point. Every consistent week sharpens the patterns. Download the app and start adding yours.
