Bode

Most mood apps explain yesterday.Bode tells you about tomorrow.

Sleep, daylight, how loaded your calendar is and which way the pressure is moving are all knowable in advance. So a difficult day can be seen coming rather than explained afterwards.

How it works

You tap one number. Bode reads the rest.

Logging is a single tap on a scale, several times a day if you like. Everything on the other side of the equation is collected on its own.

5 Level
BadGood

Whole numbers only. Nobody can reliably tell their own 7.3 from 7.5, so a decimal would record noise and then invite correlating against it.

Daylight

Hours of light and how fast you are losing them, worked out on the device from the date and your latitude. No network, no account.

of light today in

Weather

Barometric pressure and its direction. The drop matters more than the level — a steady 1000 hPa reads fine; the same figure falling does not.

Calendar

How many events, and the longest unbroken run of them. Titles, attendees and locations are never read or stored.

A forecast

Named reasons, never a score.

A single number invites you to argue with the app. Three legible causes invite you to agree with it, or dismiss it.

  • Sleep debt2h 10m under your baseline
  • Meeting load5 events, 4h 15m back to back
  • Daylight13h 44m of light, −16m on the week

The part most apps skip

When the evidence is thin, Bode says nothing.

Every forecast is a claim, and a wrong one costs more than a missing one. So Bode declines rather than filling the slot, and shows what it checked instead.

Thursday runs low

Three things line up against you. Worth protecting the evening if you can.

  • Sleep debt2h 10m under your baseline
  • Meeting load5 events, 4h 15m back to back
  • Daylight13h 44m of light, −16m on the week

Moderate · 47 similar days

Nothing stands out.

Friday sits close to your baseline. Sleep is even, the calendar is light, the pressure holds. No call to make.

Quiet days are most days.

Why you can believe it

Findings have to survive data the engine has not seen.

A pattern is measured on your earlier weeks, then has to reappear in later ones it was never fitted to. Without that step, testing enough factors guarantees a finding whether or not one exists.

Ten weeks of one person’s record.

The pattern is measured on the earlier weeks only.

It shows its working

Every insight arrives with the evidence that let it through, and a button to tell Bode it is wrong.

It publishes its misses

Forecasts are scored against what actually happened. You see the hit rate, including the categories that were dropped for barely beating chance.

Trends

What everyone’s record shows, if you choose to add yours.

Pooled findings from people who opted in. Every figure is a change measured within the same person over time — never a ranking of one city, age or income against another, because people who download a mood app are not a random sample of anywhere.

−0.5

The clock change lands late

Not on the day. About 48 hours later, once the first full dark evening has been lived through.

−0.6

Sunday evening, not Monday

The average person bottoms out between 6 and 9pm on Sunday, and has recovered by Monday afternoon.

+0.8

Out of winter by late March

Between the December low and the June high. The climb is steeper the further from the equator you live.

Privacy

Your record stays on your phone.

Analysis happens on the device. Contributing to Trends is off until you turn it on, it sends daily summaries rather than anything you typed, and it can be turned off again at any point. Mood alongside location and the people you see is about as sensitive as data gets, and it is treated that way.

Early access

It is being built now.

Leave an address and you will hear once, when it opens.

check-ins logged today, everywhere