Flying from one mid-sized American city to another usually means changing planes somewhere larger. That detour is the direct result of how many markets an airline can serve with a fixed number of aircraft.
The arithmetic of connecting everything
Serving every city pair directly requires an enormous number of routes, and the count grows far faster than the number of cities involved. Most of those flights would carry very few passengers.
Routing through a central airport collapses the problem. Each city needs only a link to the hub, and passengers combine two of those links to reach any other city in the system.
The result is that a modest fleet can offer service between thousands of city pairs. The trade is that most journeys involve two flights instead of one.
Banks of flights create the connections
Hubs do not operate at a steady pace through the day. Arrivals cluster into waves, sit on the ground together, and depart as a group so that passengers can transfer between any two flights.
These waves are called banks, and the gap between arrival and departure sets the minimum connection time. Shortening the gap raises the risk that late arrivals strand connecting passengers.
Bank structure also explains why hub airports feel alternately frantic and empty. Terminal staffing, concessions and security lines all pulse with a rhythm set by the schedule rather than by demand.
Delay propagates through the structure
Because aircraft fly several segments a day, a morning delay travels with the airplane. One weather event at a major hub can disrupt flights in cities that had clear skies all day.
Connecting passengers compound this. A single late arrival can cause missed connections across many onward destinations, and rebooking those passengers consumes seats that were already sold.
Airlines build slack into schedules to absorb the effect, which is part of why block times have lengthened. The padding is insurance against a structure that transmits disruption efficiently.
Geography determines where hubs sit
A hub works best when it sits near the center of the demand it serves, so that connections do not require long backtracking. Several large American hubs occupy that middle ground.
Coastal hubs exist for a different reason, serving as gateways where domestic traffic gathers before crossing an ocean. Their value is in the international segments they feed.
Airport capacity matters as much as position. A hub needs enough gates and runways to handle a bank, which is why the map of hubs changes slowly.
Point to point service coexists with it
Some carriers built networks around direct flights between secondary cities, avoiding hubs entirely. That model lowers connection risk and turns aircraft faster.
It also requires enough local demand on each route, so it works for leisure corridors more than for thin business markets. Most large networks now run both patterns simultaneously.