More Thinking on Golf Carts

(1,800 words, ~8 minutes)

The humble golf cart could be the future of American personal transport.

Cities

The most basic nature of the human settlement is formed through its relationships between time, energy, and space. With energy we can control travel time, and thus access to space. We should envision cities as a pool of destinations accessible per unit time. The spacing between these destinations is determined by travel speed.

The number of accessible destinations reachable in a given time in an area can be roughly modeled as…

[network penetration] × [area density] × π × [transit speed]²

Due to the fragile nature of the human body, each increase in speed requires an increase in space. Apparently, the average driver hits the brakes after around 2.3 seconds. At 20 km/h, the vehicle will have moved around 12-13 meters before that happens. At 60 km/h, 38 meters. At 120 km/h, around 77 meters. As a vehicle becomes faster, it requires greater padding both in itself (to protect passengers) and in the environment (to protect others). This can be in the form of rules or procedures, but it can also be in the form of physical buffers, grade separation, signaling equipment, and so on. We can also constrain the freedom of motion of the vehicle, such as putting it on tracks, reducing the amount of information that others have to process to safely avoid it.

The padding takes the form of space, but it also takes the form of time, because it’s largely about acceleration. (This suggests a theoretical maximum density of human development.)

But let’s set the full discussion of my theory of city network power aside for now.

Density Considerations

Suppose a car travels at 60 km/h (~37 mph, the average US car speed) and has point-to-point access to an American suburb with a density of 700 persons per square kilometer.

In 15 minutes, the car could travel 15 kilometers. During this time it can go anywhere within that radius, for a total accessible area of about 706 km². Let us imagine that destinations are directly proportional to population, and that the population is uniform. The car can reach about 495,000 destination… units, or 33,000 per minute. (In practice, it won’t hit the average of 37 mph unless it reaches the highway, which it won’t do in one minute.)

The typical walking speed is about 5 km/h (~3.1 mph). A pedestrian’s 15 minute range is around 4.9 km². As it happens, there are American university campuses of this size, with on-site restaurants, gyms, and other amenities, with parking decks around the edges. If destinations are proportional to population, we’ll need to pack around 101,000 people into the area to equal the suburb.

Note that this doesn’t include public transit. Even a New York City bus trundling along at 7 mph (~11 km/h) is twice as fast as a pedestrian, for about 21,000 residents / km² - not far off the density of some parts of New York.

A bicyclist without much training can (apparently) achieve an average of 20 km/h (~12 mph). This is 4 times faster than a pedestrian. If we take that as around 78 km², we get roughly 6,350 residents / km².

Golf cart maximum speed is 32 km/h (~20 mph), giving us a range of 201 km². That makes for a density of around 2,470 residents / km², which we’ll round up to 2,500.

How tall is that?

Assume each resident requires 80 m² of space, and 35% of land area is dedicated to residential construction (350,000 m² for each km²). The heights are then:

• 0.16 stories for cars.
• 23 stories for pedestrian-only.
• 4.8 stories for New York buses.
• 1.45 stories for bikes.
• 0.56 stories for carts.

Any number below 1.0 means we can (potentially) build a yard on the lot (if it isn’t taken up by parking). Multiply by 2 to divide the lot in half by building taller. (If you’ve noticed European cities with lots of 3-story buildings, well, that makes sense.)

Golf Carts

Both The Villages in Florida and Peachtree City in Georgia are golf-cart-oriented communities. Peachtree City has “multi-use paths” mandated throughout. These paths wind through scenic forest and reach from subdivisions to stores, restaurants, schools, and other locations throughout the town. Look in on the parking lots from Google StreetView and Maps, however - they seem to be dominated by full-size cars, with only a handful of carts in any given lot. The Villages, a retirement community, does better in terms of the number of carts, but still has plenty of sprawling parking lot despite its wide-reaching cart path network. Why?

Simply put, these are largely conventional developments that have been equipped with a thorough, but auxiliary, golf cart network. Peachtree City has a median household income of $93,000, 66% above the statewide median, and The Villages is an extension of the retirement village golf course concept elevated to a new scale.

The parking lots aren’t filled with golf carts because the developments aren’t dense enough. In order to have the same number of destinations reachable per unit travel time by golf cart, they would have be about four times the density - around that 2,500 number we proposed earlier.

If we do this the carts may start to become a real traffic issue for each other. With the smaller size and slower speed of the carts, however, one lane’s width of conventional road supports roughly twice as many golf carts, so road land use can be allocated to 1/3^rd “Micro-mobility Path” (to be shared with bikes, e-scooters, etc) and 2/3^rds conventional road.

The shorter length makes a difference much more for parking than it does on the road. A Toyota Corolla covers a minimum of ~8.27 m², while a Yamaha Drive 2 golf cart covers around 2.87 m², a factor of almost 2.9x smaller.

Is it walkable?

Golf carts move about six times faster than pedestrians. Cars move around 12 (average) to 24 (highway) times faster. A medium speed for a horse is apparently around 10-17 mph (~3-5.5 times faster). Golf cart and bicycle-oriented development is therefore more analogous to the speed of a horse than it is to the speed of a pedestrian or a car, even though everyone becomes a pedestrian at either the beginning or end of their journey.

…but this isn’t necessarily the wrong approach. A town in which every resident has their own personal horse, cowboy hat, and six-shooter is very American in scale. It’s also a suitable scale for bicycles.

You might have noticed that in some U.S. cities with high housing demand, there’s some new construction going on in a more “New Urbanist” style, with townhomes being built with walking access to small stores or a grocery store. Their streets and parking lots are still scaled to automobile traffic, however, and actually reaching that grocery store from the center of the development might require walking one kilometer and crossing four lanes of loud, full-scale traffic. The small amount of groceries that can be carried on foot require more frequent trips. Most Americans would probably choose to take the car, as cargo bikes aren’t really a thing in America (though they should be) and don’t have much room for kids. Of course, if the grocery store is on the same side of the road, this is all easier.

…but if the New Urbanists are seeking to build old forms from the America that existed before the automobile, they may unintentionally be implicitly building for the power, mass, speed, mobility, and scale… of the horse. (And also the locomotive.)

Since there is a vehicle that roughly matches that description (the golf cart), and the modifications required would be relatively small (widening of sidewalks into multi-use paths, addition of golf cart lanes to roads, or similar measures), they would make an excellent choice to retrofit. And if we’re building for golf carts, we’re also building for bicycles, electric bikes, pedal quadricycles, electric scooters, and similar very light and comparatively slow vehicles, as long as we’re willing to throw in some small bike and electric infrastructure.

Cars

If you’ve been following so far, then you probably understand why Americans love cars. They have extremely high network penetration. They have extremely high speed. This allows either tremendous utilization of space or enormous compression of destinations per unit time. They’re air conditioned, armored, weaponizable, and just highly mobile in general.

They also get stuck in traffic.

Anyhow, with a car of your own you can go almost anywhere you want in a city, without asking other people, having to wait for public transit with public transit schedule and reliability, or needing to share your bubble of space with loud people. It’s a lot of freedom, despite the costs that come with it.

Though slower and with a shorter range, golf carts offer similar radical personal powered point-to-point mobility. A cart-based community should also offer pockets of sufficient density for public transit and cart-based park & ride - residential development will be on the interior of blocks, while businesses will be on the exterior, which makes not having a cart or bike at the destination not as big a deal for commuting.

There is one other matter to deal with.

I like this has lockable doors, my biggest issue with bikes is they get stolen all the time.

- YouTube comment on a video for Veemo, an electric-assist cab bike

It might be more common than you think.

“We should make it so people don’t need bikes as much to get around so it won’t matter as much when they’re stolen,” or “if people were more secure in their income, they wouldn’t feel a need to steal bikes,” or “it’s your fault for not properly locking up your bicycle” are not going to cut it. If you want the environmental benefits of the biking, the scooting, the golf carting, the accompanying density and decrease in infrastructure volume, and the reduction in fuel usage (and therefore CO2 emissions), you must arrest and prosecute bike thieves, and you must arrest and prosecute cart thieves.

Government programs such as drug clinics to keep addicts off the streets and job assistance for the poor are compatible with this transit mode, but people must be confident that their bicycles, quadricycles, e-scooters, and golf carts will not be stolen. If they are not confident of this, they will mouth all the correct words to you, then either move to some place where the median household income is $93k and the population density is 550/km², or just use a car.

The ones that won’t are the ones that can’t, and are those who are the least able to afford losing their vehicle - the poor.

We can make security mechanisms a part of our transit strategy, such as locking bikestands, or custom keys for golf carts, and registration at city hall for all these vehicles, but all of these micromobility options are inherently lower security than cars, and crime within the town must be handled accordingly.

TL;DR

Math regarding density and speed of transit suggests that a town with the density of 2,500 residents / km², multiple times the density of U.S. suburbs (with numbers like 500 / km² or 700 / km²) could be economically viable, if it had a second transport network based on golf carts, bicycles, e-scooters, and so on. This level of density would allow each resident to have a yard and maintain the car-like point-to-point mobility, despite lower use of resources and space.