In the previous post I showed that where I live the earliest sunset is 12 days before the winter solstice, and the latest sunrise is 13 days after. But the exact difference depends on latitude. Here's the earliest sunset and latest sunrise just outside the Arctic Circle:
All three dates are much closer to alignment. Here's the equator:
At 0° latitude the earliest sunset and latest sunrise are almost two months before and after the winter solstice.
This makes sense given that the asynchrony is caused by the interplay of the rotation of the solar terminus and the day-length impact of Earth's tilt. Near the poles, tilt causes the length of day to vary so much that the movement of the solar terminus is all but invisible, while closer to the equator, where tilt has little to no effect on length of day, the movement of the solar terminus dominates.
As a result, the flattening of the curves that appeared on the mid-latitude graph disappears as the graph moves north, shifting until it resembles something more like a sine wave. Moving south transforms the curves into two sine wave-like cycles. It's the combined effect of these two shapes that causes the asymmetrical curves in temperate zones.
To show the transition, here are overlaid curves for several latitudes from 0° to 65°N:
And here's just sunset, stacking the curves rather than overlaying them: