Interactive · 5 min to play · Aug 25, 2026
Spot the pattern: a two-week food log, and you're the detective
Every parent trying to find a food trigger eventually tries the same thing: pay closer attention. Watch what he eats, remember how he felt after, look for the pattern. It works for about four days.
Here's what those four days usually turn into: a phone full of half-typed notes, a symptom you're pretty sure happened the day after pizza but can't swear to, and a memory that gets fuzzier every day you don't write something down. So we built a small experiment instead of just telling you about it. Below is two weeks of logs for a fictional kid, written the way a tired parent actually writes them: some days detailed, some days rushed, a couple of days skipped entirely. Read through it and see if you can spot what's setting off the stomach aches. Then we'll show you what the pattern actually looks like once it's laid out instead of scattered across two weeks of memory.
The log
Below is a scan of two weeks from an actual (fictional) parent's notebook, coffee ring and all. Two ingredients show up again and again, and only one of them is actually doing anything. When you've got a hunch, pick a suspect and see what the data actually shows. Nothing here is timed, take your time the way you would if this were your own kid.
your hunch?
what's setting off the stomach aches?
Why that's harder than it should be
If that took some real back-and-forth, that's on purpose. Two ingredients travel together often enough in the same meals that either one looks like a reasonable suspect for most of the two weeks. The only days that actually separate them are the ones where just one ingredient shows up on its own, and noticing that takes seeing all fourteen days side by side, not scrolling through a notebook one page at a time.
There's usually a second twist buried in there too. Sometimes the amount matters as much as the ingredient, a little is fine, more of it (or two days running) isn't. Sometimes the log is honest but pointing at the wrong day entirely, because what shows up in a symptom or a stool this morning can be about yesterday's food, not today's. Both are real, boring, easy-to-miss reasons a two-week log can look messier than it actually is.
That's worth sitting with. Even in a made-up example built to have one clean answer, this fooled a careful reader for a while. A real log is messier still, with more foods, more possible triggers, and symptoms that don't always show up on schedule.
Even a pattern this clear can point at the wrong ingredient. That's exactly why it's a starting point for a doctor, not a verdict you reach on your own.
What BellyMystery does with the same two weeks
This is the exact problem BellyMystery's analysis engine is built to surface, not settle on its own. Every meal, symptom, and stool you log gets compared against everything else you've logged, automatically, checking dose, frequency, and timing lags the way this game just made you do by hand, so you're never the one holding fourteen days in your head. Untangling which ingredient, or whether it's more than one, or neither, is exactly the kind of question worth bringing to a doctor or dietitian with your logged data in hand instead of a guess.
Worth saying plainly: Theo and Carolyn aren't real. Even a pattern this clear is a lead, not a diagnosis, and it's meant to start a conversation with a doctor, not end one.