When Common Sense is Wrong, and Intuitions Fail
Bike lanes are clearly the problem. (5th Avenue, SounderBruce, Flickr)
A recent blog post by well-respected local meteorologist Cliff Mass, "Fixing Seattle's Traffic Mess," offered an anti-urbanist grab bag of bad ideas. Bemoaning the current state of traffic, Mass distills Seattle's traffic woes to 9 problems:
- Road diets that "promote congestion and substantially reduce maximum throughput"
- Poor road conditions resulting from "Seattle Council members paying [less] attention to the traffic-producing bad roads than kayaking out to oil platforms destined for Alaska"
- Excessive draw bridge openings
- Sounder is too unreliable, and the trains are "less than half full"
- Distracted driving
- Link serves the Rainier Valley: "It takes forever to travel that segment and sometimes the trains get into accidents with cars."
- Undersupplied parking at Link stations: "Folks need a place to park if they are going to use the train"
- Bike lanes: "The only safe way to commute is to be totally separate from cars, not the side lanes of the "road diet" streets"
- "Continuing lack of bus service"
Many local outlets piled on the criticism, including The Stranger, but I see no reason to single out Mr. Mass. His assertions are widely held, intuitive, and derived from common sense. They are also completely unsupported by data.
To wit, road diets haven't greatly increased travel time or reduced throughput, drawbridges must open by Coast Guard mandate, Sounder is 95% reliable and carries 500 passengers per trip, Sounder mudslides have gone down markedly due to intensive work by WSDOT, a Duwamish Bypass for Link would cost $1B and only save 3-4 minutes, transit parking is a niche product that cannot scale, and our local and regional bus service levels are at historic highs.
Just as Mass' diagnostic skills are lacking in his post, so too are his 3 prescriptions: passenger ferries, flexible app-based carpooling, and a Big Data approach to signal timing, etc. Perfectly reasonable sentiments, but none of them remotely sustainable. To the extent that app-based carpooling diluted transit ridership, it would make things worse. While Big Data can optimize flow at the margins, the fundamental use-of-space problem is immutable. And when it comes to passenger ferries on Lake Washington, King County's official report showed that they would suffer from low ridership and would incur costs three times higher than Sounder.
Collisions Down, Traffic and Travel Times Flat. Road Diets Work.
From the Q4 2016 Ridership Report
But if you're stuck behind the wheel, it's reasonable that you'd think "two lanes would be better than one". Link's Rainier Valley deviation feels slow, even when it only costs a couple minutes for people like Mr. Mass who likely view it as an airport shuttle. If you would love to ditch your car but the park and ride is full, you'll wish there were more spots. In each of these cases, people's lived experience contradicts what the data clearly says. Accordingly, we should cut such folks some slack, and do a better job of showing our work.
If you drive everywhere, it may well look like madness, and Seattle's urbanist policies are a visible and convenient foil. But if bike lanes et al were to blame for traffic, you wouldn't expect traffic problems in car-focused places such as Kennydale Hill, the Fife Curve, or Joint Base Lewis McChord. But we know that each of those places are equally choked by traffic. So if you find yourself thinking, without irony, "Nobody goes there anymore, it's too crowded," it's time to take a step back and learn to distrust your intuitions. It's hard to get things right, especially outside one's area of expertise. Mistakes are acceptable, but we shouldn't assert without evidence.