I've got two suggestions.
First, there's more randomnity in this than you think. We just don't have enough examples to draw really powerful statistical conclusions. And you're ignoring one: Papua has had agriculture for quite a while now, but has never developed civilisation, even when it was in contact with other civilisations.
Second, if there are specific factors, I'd suggest that crop yields may be a big one. In Ye Olden Days, China was a wheat-and-millet culture. Wheat and millet are not particularly well-suited to the Chinese climate, which meant that a) overall yields were a fraction of what they were in, say, Greece, and b) nowhere were there yields as high and as concentrated as they were in the Nile Delta, or in Mesopotamia. Less food, and less concentrated food = fewer cities.
Iirc, the big change (adoption of rice, which allowed population densities up to ten times higher than millet) came during the Han Dynasty, so it doesn't entirely explain why they had come good by then. but perhaps there's no mystery there: eventually their low-density crop did yield large populations and urbanisation, it just took a lot longer than higher-density crops did elsewhere.
A bigger objection might be: how sure are you of your pattern that you think China breaks? North America took 5000 years to develop the rudiments of civilisation, and arguably never developed cities. Papua New Guinea has had agriculture for 9000 years - towns developed eventually, but never cities. Central America - more than 5000 years from agriculture to the first known civilisation. Peru took less than 4000 years it now seems - but that first civilisation appears to have been a dead end, with continuous civilisation needing another 2000 years. Egyptian agriculture never produced civilisation - it died out and was later replaced by mesopotamian agriculture. Ethiopian civilisation developed only under contact with other civilisations.
Remember also: records so long ago are scant and rely on luck in finding them. Maybe we haven't found the earlier Chinese towns. Or maybe chinese towns just didn't leave such obvious marks - in many places it's only the big projects, the canals and the earthworks and so on, that are the surviving signs of early towns. Maybe chinese towns just didn't have convenient ring-ditches around them and burnt the wood of their houses when the houses fell down.
A final suggestion: russia famously had no towns for a long time. Or it did, obviously, but they were extremely rare: it had a few big trading cities, and everyone else lived in a network of small villages - far more densely packed than european villages, but without the bigger towns that europe had. It may simply be that certain terrains and climates tend to favour dense populations and other terrains and climates tend to favour spread-out populations.
But the river tripped on her by and by, lapping
as though her heart was brook: Why, why, why! Weh, O weh
I'se so silly to be flowing but I no canna stay!