Portland, Oregon is well-known in the U.S. for cycling leadership, though its cycling rates are much lower than those in many European cities. But what many fail to understand is that it is Providence, not Portland, that has many of the key land-use features that signal an ability to lead in cycling.
While Portland Oregon has a density under 5,000 per square mile, Providence's density is almost 10,000 per square mile.
More importantly, Providence's total land area (25 square miles) is much smaller than Portland's: PDX proper is more than five times larger by area (133 square miles) than Providence city-proper.
To test the idea of land use, I cobbled together figures for surrounding communities near Providence, assuming that commute statistics in the Capital City would have to rely partly on the city's relationship to its suburbs and satellite cities. If Central Falls, Pawtucket, Cranston, East Providence, and Providence formed a giant metro blob of potential commuting habits, that blob would still be less than 60% of the land area size of Portland, Oregon proper. The East Cranstofallucketence mass would also be denser than Portland proper--above 5,000 per square mile, while Portland falls a bit under 5,000 per square mile.
Density matters to biking, though not in the ways one might expect. Transit density is nodal, looking like a string of beads. The ideal density for transit is right around stops, with as little settlement as possible in between. A bus line or train line could operate with ideal ridership conditions even (perhaps, especially) if that string of beads were relatively long, so long as riders and the places they wanted to go were near stops.
Bike densities rely on less stringent demands. Because cyclists don't coordinate themselves as a group as riders of a bus do, a general blob of dense development can work. The overall length of many journeys through the blob may matter more than the specific density of development around nodes (although nodal development certainly is a plus, regardless). A village with three residents as well as a thriving metropolis of millions could equally support biking, so long as the distances of travel were reasonably short--density is a second-hand proxy of this, for the purposes of biking.
The Plum Pudding Model of atomic theory turned out to be wrong, but gives an idea of what this means. At the turn of the 20th Century, English scientist J. J. Thompson proposed a mish-mash of protons, electrons, and neutrons without any kind of nucleus as how atoms might be formed. Such a model would work horrible for getting good commutes going by bus--buses work need nuclei--but bikers would find this very acceptable.
The City of Groningen, in the Netherlands, holds the world record for journeys by bike. While that city is larger and less dense than Providence, the overall length of most possible trips is fairly short. Alongside policies of filtered permeability and separated bike infrastructure, it's estimated that over 50% of trips are made in that city by bike*, with appreciable trips also going into modes like transit.
As Portland reclaims its top-dog position for U.S. cities from challenger Minneapolis--a city larger and less dense than Providence, which spends five months of the year under blizzards, and the other months in hot, tornado-prone summers--the question that should be on everyone's minds is why Providence has yet to even step into the radar of people paying attention to this race.
*English-born Dutch biking blogger David Hembrow, who writes the blog The View from the Cyclepath, takes issue with the 50-60% statistic, saying that it's about self-promotion rather than reality. One breakdown he gives on why he disagrees with the statistic points out that many walking trips are undercounted--and perhaps make up half of trips, before biking, transit, and driving are considered--and to my mind this only reinforces that Groningen has things together, rather than detracting from its success.