For sure, the original intent of the 787 was to enable flights that skip hubs[1], but for whatever reason, in practice that hasn't really happened[2].

Also, it's totally possible that there's a better example of a rote enabled by the 787 than New Orleans/London. That's just one that came to mind.

[1] - https://www.forbes.com/sites/marcbabej/2014/12/11/airbus-a38...

[2] - https://centreforaviation.com/insights/analysis/787-network-... (though you only get a lame summary of the article without an account)

Direct flights to more destinations from any hub still means a superlinear increase in the number of slot pairs required. Are you suggesting that that wouldn't also be problematic at hubs in North America?

I'm suggesting, as I took sjm-lbm to be originally, that there are more direct flights that are not using the hubs, therefore less need for larger planes to support aggregated passengers. If flights from New Orleans to New York used to go through Atlanta and the Atlanta to New York flight was full of aggregated smaller flights to Atlanta, that would require a larger plane. If more flights are going direct from New Orleans to New York using smaller planes, that lessens the aggregation in Atlanta, and lessens the need for another 747 at that location.