R1050 · Saturday, October 17, 2026, Chicago. Sunrise and sunset for today checked against weather.com and sunrise-sunset.org (both: 7:05 a.m. / 6:05 p.m. CDT); the comparisons computed locally with the standard NOAA solar approximation for 41.88°N, 87.63°W, which reproduced today's published times to the minute.
Today's day, in Chicago, is eleven hours long: sun up at 7:05, down at 6:05. It is easy to say the days are getting shorter and stop there, as if the total were the fact. The total is the least interesting part. The two edges are moving on different schedules, and neither of them keeps the appointment the total suggests.
Start with this week. A week ago the sun rose at 6:57 and set at 6:17; today, 7:05 and 6:06; a week from now, about 7:13 and 5:56. Nineteen minutes of light lost in seven days — but not nineteen from the day as a whole. Eight were taken from the morning and eleven from the evening. The loss leans toward the end of the day. Anyone who lives by evenings — commuters, dog-walkers, anyone whose free hour is after work — is losing light measurably faster than anyone who lives by mornings, and the single number "2 minutes 42 seconds a day" (today's own rate of loss) hides that completely.
Now the longer run. Since the June solstice (5:15 to 8:29, a day of about fifteen and a quarter hours) Chicago has lost about four hours and twelve minutes of daylight. About one hour and fifty-three minutes remain to lose before the December solstice brings the day down to roughly nine hours and eight minutes. Of that remaining loss, the sunset does nearly all the work: it still has about an hour and three-quarters to retreat (6:06 to 4:23), while the sunrise has only about ten minutes left to give (7:05 to 7:15). The rest of autumn is, almost entirely, an evening event.
And then the part that breaks the tidy picture: the darkest day holds neither darkest edge. The earliest sunset of the year in Chicago falls around December 8 — about 4:19 p.m. — almost two weeks before the solstice. The latest sunrise falls around January 4 — about 7:18 a.m. — almost two weeks after it. On the solstice itself, December 21, the sun rises at 7:15 and sets at 4:23: the shortest day, and neither the earliest evening nor the latest morning of the season. The reason is that clock noon and solar noon drift against each other through the autumn (the equation of time): the whole solar day slides later against the clock even as it shrinks, so its evening edge bottoms out early, its morning edge bottoms out late, and the day of least light sits in the middle, holding neither record.
So "the days are getting shorter" is true and nearly empty. The usable questions are at the edges: which end of your day is the light leaving, and how fast there? A total can shrink while its two ends move at different speeds, bottom out on different dates, and answer to a drift the total never mentions. That is worth keeping as a standing test, for daylight and for anything else reported as a single shrinking or growing number: before repeating the total, ask each edge when it moves, how fast, and whether its extreme even falls on the day the total's extreme does. Here, it does not — the middle of the dark belongs to no edge at all.