Whenever we talk about removing bus network inefficiencies, deviations are almost always a big part of the discussion. It’s a mathematic no-brainer that routes operating in a straight-line are safer, faster, and more reliable than their zig-zaggy counterparts. However, decades of bad land use planning and car culture have resulted in lots of destinations that are simply out of the way. Park-and-rides, like Eastgate and Federal Way, are some of the more egregious offenders that have likely cost Metro millions in operating costs over the years.
Most agencies have some sort of a deviation standard that weighs riders served by a deviation against through-riders. There’s a naturally flawed assumption built into this model– that riders accessing the destination are assumed lost if the deviation is eliminated. The broader assumption behind this is that the pedestrian environment has no impact on a rider’s choice to take transit. These types of deviation formulas would treat auto-oriented and urban contexts equally, assuming ridership generation remains constant.
For example, Metro’s deviation standard (according to the new service guidelines) is as follows:
More below the jump.
It’s well known, however, that good transit encourages farther walking distances and that better pedestrian environments are conducive to transit use. In areas with really good walkability, deviations become less practical. It’s partly why routes like the 5 and 358 can still “serve” Seattle Center even when stopping three blocks away*. It also justifies why I think Metro could get away with not sending every downtown Bellevue bus to the transit center. Many of the riders you would assume lost in a classic deviation formula actually turn out to be willing to walk the extra distance (within reason, of course).
It perplexes me, however, when people use walking as an excuse to justify a really bad route when planning rail. This was the case when Bellevue councilmember Kevin Wallace unveiled his Vision Line proposal for East Link, citing the Eastgate Freeway Station to justify the station’s long walk distance to downtown Bellevue. And similar arguments are also bound to crop up as Link extension planning to Lynnwood progresses. This will continue to be the case as long as you have the choice between routing a line through a major activity center or routing alongside it.
So why does it seem more acceptable to have rail deviate to major activity centers, but not buses? The answer actually has more to do with the guideway than the mode. Buses operating in mixed traffic are subject to turns, lights, and roadway configurations– all the things that make deviations slow, unsafe, and unreliable. Grade-separated fixed-transit modes, on the other hand, have the luxury of avoiding most of those things, all of which add time and cost for all passengers, but with disproportionately greater impact on through-riders**.
Take the 554, for example, which currently serves Eastgate via the I-90 direct-access ramps and the freeway stop. A deviation into the park-and-ride loop and back out would add eight turns and one 180-degree turnaround, all of which make for precious minutes Sound Transit can ill afford to waste. A grade-separated transitway, on the other hand, could greatly minimize the travel time penalty by gently veering off the freeway and back on, thereby eliminating every turn. There are obviously significant construction challenges and costs to doing so, so we’re willing to tolerate a longer walk out to the freeway station in the meantime***.
What’s more, forms of transit that make this kind of deviation more palatable (i.e., fixed-guideway, grade-separated) also tend to be high-capacity services which make fewer stops, but higher-ridership, higher-value ones. Plugged into the deviation formula, the proportionately greater number of riders generated by each stop helps justify a deviating routing when it means serving major activity centers. It’s an important mathematical principle to remember as we continue to expand rail regionally, even as the temptation to build freeway-running routes remains high.
*This particular stop is temporarily closed until SDOT is finished with Mercer construction.
**This is the case from an operational standpoint. There are admittedly additional barriers to constructing the infrastructure not mentioned in this post.
***Eastgate happens to be one of those cases where its poor walkability and road connectivity are both attributable to the surrounding topography. Nonetheless, it still stands as one of the least transit-friendly facilities in our region.