Methodology
Every number in this tool is traceable. Here's exactly how schedules are built — and where the model stops being reliable.
1. The data: published ranges, not opinions
Wake-window and sleep-total ranges come from published pediatric sleep guidance used across sleep councils and certified consultants, with total sleep anchored to the ranges endorsed by the American Academy of Pediatrics and the US National Sleep Foundation (newborns 14–17 h, 4–12 months 12–16 h, 1–2 years 11–14 h, 2–3 years 11–13 h per 24 hours). We store ranges, never single magic numbers.
2. Continuous interpolation
Charts split ages into bands ("4–6 months"). Biology doesn't. The engine linearly interpolates between age anchors, so a 4-month-old and a 5.5-month-old get genuinely different windows. Anchors sit at 16 ages from 2 weeks to 36 months.
3. Window shaping
Within a day, the first wake window is scaled to ≈85% of the age's base window and the final window before bedtime to ≈115% — matching the widely observed pattern that mornings start short and nights are preceded by the longest stretch. Daytime sleep is distributed across naps with earlier naps weighted longer, floored at 30 minutes and capped at 2 hours.
4. Bedtime lock
In lock mode the engine works backwards from your target bedtime. If naps plus windows can't fit between wake-up and bedtime, it drops naps one at a time (to a floor of 1) and tells you it did so — it never silently compresses windows below their age floor.
5. Limits — read this part
This is an educational model of typical patterns, not a prescription. It doesn't know your baby: temperament, feeds, reflux, daycare, siblings, and sleep training approach all move real schedules. Prematurity: use corrected age until 2 years. If your baby's sleep is far outside the ranges here, or breathing/snoring concerns exist, talk to your pediatrician. We cite ranges so you can check us — we don't ask you to trust a course.