Part of the Loopa iOS app guide.
The Pattern identified card is the small "PATTERN IDENTIFIED" panel that surfaces on Today's Page once your log holds enough signal. It names what moves what in your own data (sleep and next-day mood, protein and energy, and the like), drawn straight from what you've logged. See Using the iOS app for where Today's Page sits.
General wellness tracking, not medical advice.
What the card shows
The card is headed PATTERN IDENTIFIED on the left and FROM YOUR JOURNAL on the right, and lists up to three findings. Each finding pairs a small strength dot (filled for a strong pattern, half-filled for an emerging one) with a plain sentence like "Your mood runs about 0.6 pts higher the day after 7+ hours of sleep." A footnote reads: "Based on your own entries. The more you log, the clearer it gets."
The card only exists when there's something to say. Until your log has enough history, it stays hidden entirely. You won't see an empty or placeholder version of it.
What it correlates
The patterns come from a fixed set of "driver → outcome" relationships checked against your own days. Among them:
- Sleep → next-day mood and sleep → next-day energy: how you feel the day after 7+ hours.
- Protein → energy and protein → mood: your higher-protein days vs your lower ones.
- Hydration → energy and hydration → mood: well-hydrated days vs the rest.
- Movement → mood and movement → glucose: days you move vs days you don't.
- Sleep → resting heart rate and sleep → fasting glucose: both read lower the morning after a good night.
- Better-mood days → more sleep: the reverse direction, too.
A finding is phrased in your own units: points on the 1–5 mood and energy scales, hours of sleep, ounces of water, bpm, and so on.
From your own data, and deterministic
Nothing here is guesswork or AI writing. The pattern engine is deterministic and computed on the server from your daily log (the app fetches it as part of loading Today's Page). It works by bucketing your days by a driver (for example, nights of 7+ hours vs shorter ones) and comparing the average of the outcome between the two buckets. If the gap is big enough and backed by enough days, it becomes a card line; otherwise it's left out. Because it reads your real numbers, two people can see completely different patterns.
Why it needs enough history
The engine looks back about 90 days and holds every finding to a bar before it will state it:
- Your log needs at least a dozen days of data before any pattern is considered.
- Each side of a comparison needs a minimum number of days (roughly six), so a single unusual day can't invent a trend.
- The difference between the two sides has to clear a threshold: a real gap on the 1–5 scales, or a meaningful percentage on other metrics.
Findings that clear a larger effect size are marked strong; the rest are emerging. Only the strongest few ever show, so the card stays a highlight rather than a wall of text.
How to read it honestly
A pattern here is an association in your own log, not proof that one thing causes another. The same caution applies to Weekly insights. It's a prompt worth noticing, and the more consistently you keep up your Daily check-ins and other logging, the sharper and more trustworthy it reads. To see the underlying trends behind a pattern, open Trends on iOS.
