The second year
12-24m · Updated 2026-08-23
One thing dominates the second year: the move from two naps to one. It is the biggest, sloppiest transition in all of childhood sleep, and knowing its true shape spares you months of second-guessing. The headline numbers: at twelve months, 17% of children are on one nap; at fifteen months, around half; at eighteen months, 77%; by twenty-one months, 88%. The median lands near a year and a half. So if your friend's fourteen-month-old dropped to one nap and yours hasn't — both children are dead on schedule.
What the transition actually looks like
Not a switch — a wobble. Weeks, sometimes a month or more, of "one and a half naps": some days two, some days one, with a rough patch parked in the late morning. The nap that survives is always the midday one — the morning nap, which was first to form back in infancy, is first to die, because a good early bedtime slowly absorbs the sleep it used to carry.
The mechanics that make it survivable, straight from the books this app is built on:
- Move bedtime earlier the moment naps wobble — twenty to thirty minutes, temporarily. Every dropped nap anywhere in childhood is paid for at bedtime, and this one especially.
- Protect the midday nap, not the morning one. If the morning nap is so good it's killing the midday, cap it (wake after an hour or so) or nudge it later in small steps until it becomes the midday nap.
- Let mixed days be legal. A two-nap Tuesday after three one-nap days isn't backsliding; it's the transition working as designed.
- Don't call it early. A couple of rough days around the first birthday is not readiness — before twelve months a "2→1 transition" is almost always something else, usually a bedtime that has drifted too late.
Once it settles, the single nap is a substantial thing — typically about two to two and a half hours, midday — and the Zurich percentiles have total sleep around thirteen hours with nights holding near eleven.
The rest of the year
Expect possible dust-ups around twelve and eighteen months (the regression windows are real but brief, one to two weeks, and skip half of all children entirely), disruption whenever walking is being mastered, and — late in the year — the first boundary-testing at naptime, a preview of toddler negotiation that has nothing to do with sleep need. The witching-hour arousal, if it visits, now lives around 4–5 p.m.
What the app does at this age
NaniNani treats the transition as weather, not breakage. Mid-wobble, the forecast honestly widens — a child who hasn't decided between one nap and two cannot be predicted more precisely than that, and pretending otherwise would be theater. In my own validation, the model's misses clustered exactly in these transition weeks, which I count as the model being right about a child who hadn't made up his mind (the scorecard is in the notebook). When the wobble ends, the water sharpens again on its own. Your job during the messy middle is the same as the app's: hold the midday anchor, spend the earlier bedtime freely, and wait it out.
Where this comes from
- Marc Weissbluth, Healthy Sleep Habits, Happy Child (4th ed.). Read in full; wake-window rules, turning points of sleep maturation, second-wind physiology.
- Alexis Dubief, Precious Little Sleep. Read in full; wake-time tables, the sleep-drive balloon, nap-transition mechanics.
- Ivo Iglowstein, Oskar Jenni, Luciano Molinari & Remo Largo, "Sleep duration from infancy to adolescence: reference values and generational trends", Pediatrics 111:302–307 (2003). The Zurich Longitudinal percentiles for total/night/day sleep and nap counts by age.