[llamaorama]

I don't think total system ridership tells you what kind of physical infrastructure you need to provide a certain type of service.

City A might actually have pretty high ridership but the geography favors a network typology of hundreds of local bus routes each with a few thousand riders per day. City B might have low transit ridership, but a large proportion of it is in one or two corridors and rail is needed for capacity. An extreme example might be a random city in Brazil that doesn't have a metro yet versus somewhere like Lausanne, Switzerland.

I think with Dallas, the city's sprawl would favor a grid of local buses. But that would also be too slow for cross town travel. And if you are going to invest a lot of money into dedicated cross town infrastructure, you might as well funnel all your riders on to it, and if you do that it might as well be rail.

The dilemma for Dallas is that probably needs both DART rail and a local bus grid at the same time, but it couldn't afford both. Houston "cheats" by having some freeway transit lanes that were paid for using highway money. But I wonder how much those HOV lanes really cost per mile and when you consider how few people use them I wonder if a DART rail set up would actually be better.

Ultimately what might work best for Dallas IMO is to keep up with what they have, and then make up for what they lack in terms of local bus routes with better last-mile pedestrian and bike infrastructure. How close do local bus routes actually have to be to one another? If sidewalks and walkability is poor and there are physical barriers like a highway or a drainage ditch then you need more than one route in that general area. But if the last mile stuff was better people would be willing to walk further. For disabled or indigent patrons, the city would subsidize more partially privatized ridesharing type stuff with handicap van taxis. Treat local bus routes almost like BRT lite with stuff like signal priority and queue jumpers too, and then a single bus with one driver could do more trips per day and increase frequency with less overall capital investment.