Even in our timeline, New Zealand did not join Australia, and Western Australia could have feasibly been separate too. Newfoundland was only a few polling points from going its own way despite only having a population of a few hundred thousand.
And New Zealand went very close to join Australia, and Southern Rhodesia to join South Africa. Plenty of butterflies also exist that create potential for less, not more, Balkanization than OTL.
I know you have a strong bias towards more expansive countries, but we need to accept that, even in our timeline, the USA is an example of a country much larger than most. In fact, only about one in eight is above 50 million.
Actually, the division of British North America into one free Dominion and one slaveholding Dominion does not make for a radically reduced amount of Balkanization in comparison to OTL, mostly it shifts it around.
I may also point out that OTL history of Anglosphere settler nations does not show any great tendency to break up into regional polities, typically they tend to expand till they fill all politically-available space in continental polities or almost so.
The main dividing line that showed up (apart from the monarchist-republican political feud that we have to assume is butterflied away as per scenario requirements) was the socio-economic split related to slavery, and the model I proposed fulfills it by splitting BNA in different "free" and "slaveholding" polities (there is of course plenty of potential for political, socio-economic, and colonization butterflies placing certain regional blocks of states in one Dominion or the other, such as the Upper South or the southern West Coast).
I am also extremely skeptical and suspicious of the assumption that OTL is a special case, the extreme feasible point in any kind of social development.
I would expect the large power centres wouldn't want to be diluted into a much larger power structure, and there are likely going to be difference beyond the one grudge issue that defined our timeline. I would thus imagine something like a Northeastern federation including the Great Lakes, a Virginia-led Southern Atlantic one, a Gulf of Mexico one, a Californian one and a Pacific Northwest one (both extending to the Rockies), and possibly a Great Plains one.
Frankly, this just seems Balkanization favoritism for the sheer heck of it.


No significant mainstream drive for such extreme regional fragmentation of Anglo North America ever surfaced. There was the free/slaveholding divide, and the Anglophone/Francophone one. About the former, I made my point.
About the latter, well, depending on various butterflies, Anglicization of Lower Canada may be rather more efficient, or just as lackluster as, OTL in a united BNA. OTOH, we may expect that English-speaking settlers would end up being the vast majority of the population in Western Canadian states (Upper Canada would quite possibly become French-speaking if Lower Canada remains so, however). So we may or may not see a third North American Dominion made up of French-speaking Middle Canada.
It's also worth considering that the individual state borders are likely to be changed/divided/united from over time too.
No contention with that.