Earlier this month Metro released their 2013 Service Guidelines Report [PDF]. This report, which replaced the Route Performance Report in 2011, was released in the spring in 2011 and 2012, but Metro has decided to release it in the fall from now on to better align with the King County Budget process. STB covered the last report back in April. Notably, this is the first report since the launch of RapidRide C & D and the large restructure that went with it.

Some highlights after the jump.

Investment Needs
While we’re currently discussing severe cutbacks to service, Metro reports an even larger number of hours would be required to address overcrowding and unreliability than last year. Plus, due to a change in the methodology used to set target service levels, the number of hours required to meet those targets has also increased. All told, it would require an increase of 510,700 hours to hit the quality and frequency targets. In contrast, the cutbacks would reduce service by more than 600,000 hours.

The change to the methodology for target service levels was twofold: first, three relative thresholds for the number of households and jobs served were replaced by five fixed thresholds, and second, college and university enrollment are now counted as jobs. As a result, the number of corridors targeted for “very frequent” service has increased by 16 to 53, while the number of Frequent, Local and Hourly corridors has declined by 17 (this reflects the elimination of corridor 113, which was old Route 23).

Peak Analysis
Mostly as part of the RapidRide C & D restructure, Metro reduced the number of peak-only routes from 92 to 83. Peak routes are measured by two criteria: whether they carry at least 90% as many riders per trip as alternative service, and whether they are at least 20% faster than alternative service. Of these 83 routes, only 30 meet both criteria. 21 meet neither.

Route Performance

As before, Metro uses two measures of performance: Rides/Platform Hour, a cost-effectiveness measure, and Passenger Miles/Platform Mile, a measure of average load. They are measured during three service periods: Peak, Off-Peak, and Night.

Overall average performance increased slightly, except for routes that serve the Seattle core in the Off-Peak period. Considering this represents the launch of RapidRide C & D as well as routes like the 40 with frequent off-peak service, a dip here is understandable.

RapidRide C scores in the top 25% on Passenger Miles/Platform Mile during all three periods, but never cracks the top 25% on Rides/Platform Hour. Meanwhile, I noticed something interesting about NW Seattle peak service: RapidRide D is in the top 25% in both metrics in all time periods except Passenger Miles/Platform Mile in the Peak period. Meanwhile, the 15EX, 17EX, and 18EX score in the top 25% in both metrics. There was a suggestion the other day that replacing Metro’s asinine peak-only zone system with a peak route surcharge could cost Metro money by requiring extra peak buses on RapidRide. Based on these numbers I’m not concerned – D coaches are currently more crowded off-peak than on-peak, and I believe for the right price commuters are willing to pay extra for the significantly faster trip.


Systemwide reliability increased slightly in 2013, increasing by two percentage points to a still-mediocre 78.6%.

Reading this report it is mind-boggling how far Metro has come in just a couple of years. It has gone from an agency committed to infrequent one-seat-rides to everywhere to one that is marching inexorably toward a network of frequent service corridors. It has become an organization where service planning is based less on politics and preserving the status quo and more on a logical, data-driven approach to putting buses where they will be most useful.

Seattle just recently became one of only five American cities where most people don’t drive alone to work. It’s quite sad that instead of building on this success we’re poised to take a giant step backward.

23 Replies to “Metro’s 2013 Service Guidelines Report”

  1. Fully agree with the author that Metro’s planners know what they’re doing and have put together an impressive analysis and proposal within the depressing constraints they’re given. But note that apples-to-oranges comparisons are still baked in, as a bottom quartile “Seattle core route” performs better than a top quartile “non-Seattle core” route. Route-level cost per boarding is absent, but IMHO that seems like the best metric when considering service reductions, as working from the bottom up would get you the most cost reductions the most quickly, which could then be corrected for geographic coverage needs.

    1. The division between “serves Seattle core” and “does not serve Seattle core” is galling, but probably necessary in order to preserve the political peace. If it were absent, large swaths of the Eastside network would be at risk of disappearing. The distinction allows Metro to present core suburban service as performing well, so that won’t happen. This is important because two Republican, suburban Eastside councilmembers (Jane Hague and Reagan Dunn) have been supportive of Metro well beyond what would be expected for the typical suburban Republican, and their support is critical to keep Metro’s agenda moving forward in the council.

      We’ve had recent progress in this respect, when 40-40-20 was abandoned as part of the Service Guidelines process. Going further will happen but will take time.

      1. The outside/inside core, or something like it, is pretty much required to avoid having the metrics force you into a hub and spoke system. It’s a legacy of the fact that Seattle’s land use has failed miserably in providing affordable family appropriate housing close to jobs.

    2. Many suburban routes provide vital coverage beyond their performance, but it’s important to compare them by performance also.

      The real difference between “Seattle core” and “non-Seattle core” (whether Metro folk intend it this way or not) is that “Seattle core” routes serve at least one significant area that was built and grew up around mass transit and walking. The other routes don’t, and operate within land use and infrastructure that’s largely indifferent or hostile to transit and walking; they’re at an obvious handicap. Still, many suburban leaders (especially at the county level) have goals to change these areas as they inevitably grow so this growth can occur without massive expansion of freeways and vertical parking (and thus VMT and pollution). That change can’t happen without effective transit, so routes that perform well in the suburbs, even if not to Seattle standards, are necessary for the change we desire.

      1. There’s an interesting use of the passive voice in your comment. These areas weren’t built, people built them and the people who built them knew exactly what they were doing. The people who moved there actively moved there. They knew where they were moving. They were trying to get away from all the grit and all the other people and all the noise of the buses and be in the “leafy” suburbs. A lot of dollars went into the infrastructure to accommodate them. A lot of dollars are still going into the infrastructure to accommodate them (see the special session planned for Olympia). There’s a lot to be said for telling them they don’t get to have their cake and eat it too while urban riders who cost significantly less per route mile are standing cheek by jowl on an urban bus route.

      2. The two-level grouping is clunky but at least it addresses the fact that the suburbs need a service level appropriate to them, and competing directly with Seattle routes is not fair because it would lead to deleting almost everything except the 120, 150, 169, and 358. (Sorry 255 and 271.)

        The main thing I would change is to put the suburban-downtown peak expresses into the outside-core category rather than inside-core, or make a third category for them. I think they may be getting too much privilege being in the inside-core category, although I’m not sure about that. The priority needs to be: (1) Seattle all-day routes (including those terminating in Burien or Aurora Village), (2) suburban all-day routes (including those going to Seattle like the 150), (3) Seattle peak-hour routes, (4) suburban peak-hour routes.

      3. @Breadbaker: The suburbs aren’t going anywhere. There will be people living in them as long as there are people living in Seattle. If we have any hope of making a dent in the massive energy cost of transportation here suburbs, which house a majority of the people and will for the forseeable future, must contribute, and therefore the way they grow is of utmost importance to all of us.

        Of course, it’s also true that suburban political support is necessary to fund the system Seattle has — even with many low-performing routes, the suburbs still subsidize Seattle’s bus service in net terms. It’s probably a good thing for everyone (including themselves) that they do so, but suburban business owners that see sales tax surcharges on their receipts funding a transit system that draws people away from their businesses might not see it that way.

      4. Anyway, that’s why I think it’s absolutely worth caring about suburban transit and infrastructure. Worth insisting that there’s a robust transit network, worth insisting that infrastructure, land use, and maintenance are planned around transit access, worth flaming land use planners on STB about, etc.

      5. Another thing is that not everybody in the suburbs is there because of leaves and lack of grit. A lot of people were born there, have family there, have jobs there, can’t afford to move, or are under 18. If you gut suburban transit it leaves all those people out. And those people alone are probably the size of Seattle’s population, because Seattle is only 1/3 of King County and 1/6 of Pugetopolis.

      6. The two-level grouping is clunky but at least it addresses the fact that the suburbs need a service level appropriate to them, and competing directly with Seattle routes is not fair because it would lead to deleting almost everything except the 120, 150, 169, and 358. (Sorry 255 and 271.)

        I agree with your general point, but I would point out that of the routes you mentioned, all but one *are* grouped with the “Seattle core” routes. (The only exception is the 169.)

        It’s certainly true that most non-core routes would be in the bottom 25% in at least one category, if not for the grouping. The exceptions appear to be A, B, 140, 164, and 169. Only the A would be in the top 25%. Conversely, both the 255 and 271 would be in the top 25% in the non-core routes, but because they’re part of the core routes, they’re both middling performers.

  2. “the number of corridors targeted for “very frequent” service has increased by 16 to 53”

    This is an encouraging sign, and it augments Metro’s previous reports about routes that underserve their market (e.g., evening 5, 40, 41). In order to improve service, first Metro and the council need to recognize what the underservice is, then they need to find funding for it. Now that Metro is speaking more clearly and loudly about its underservice needs, and the Seattle and Bellevue transit master plans complement this, it puts the focus more squarely on doing something about it. Metro should point out that several of our peer cities in North America have a service level comparable to what Metro is asking for.

    1. I’m also happy to see that Metro’s definition of “very frequent” has changed from “15 minutes all day” to “15 minutes *or better* all day”.

  3. I’ll focus on this statement for my comment:

    “It has become an organization where service planning is based less on politics and preserving the status quo and more on a logical, data-driven approach to putting buses where they will be most useful.”

    I call BS on this! I had several e-mail exchanges with someone at metro years ago who said service improvements to my bus line (the 101) were coming soon. With each shake up I heard the same thing – it’s coming. And yet I watched as, how shall I put this, the more affluent areas within the system, where there are more bus options had MORE service added, while we out here in the less affluent south end have had to endure standing room only (SRO) buses all day long – yes, even mid day I have ridden a SRO bus both directions! All the while I’ve seen Metro create all the rapid ride ___line (fill in the blank) buses, again in the more affluent areas of the system. Oh and seeing people who take these complain about them just irks the crap out of me!!! They run every 7-10 minutes ALL DAY, something we’d give our eye teeth for out here in the south end where the ONLY time we see that level of service is in 2 short windows of the day.

    Don’t even get me started on the weekend service!!! NO BUS into Seattle until 8:13AM on Sunday (at my stop location) on such a busy route is RIDICULOUS! On the days I need a bus on Sunday there have at times been at least 15 to 20 people at my stop, and the bus is ALREADY at SRO! Then there’s the fact that while the 106 runs until the 1AM hour from DT, the 101 cuts off at 11PM! This is our ONLY option for a bus in this corridor, and since we are made up of a lot of folks who can barely afford bus fare, let alone a car, we likely are the ones feeding the fare box, instead of saying to heck with it, I’ll drive.

    So, HOW is this a “data driven approach”? If it is, then the data is VERY flawed!!!

    All these stats mean nothing to those of us out here that have just one bus option. We are on a major highway and the other so called options are NOT easy to get to (the nearest 106 stop is over a mile from my stop and the 150 via Tukwila P&R is at least that far and not a safe walk!!)! So Metro better fix the issues out here in the south end before someone is hurt riding one of these extremely over full busses and cost them, or I mean cost tax payers when they settle the lawsuits that will come from someone suing.

    Just my 10cents worth.

    1. They run every 7-10 minutes ALL DAY…

      No, they don’t. Not even close.

      You were lied to. TransitNow voters were lied to. RapidRide-restructured areas were lied to. There’s been a whole lot of lying in the recent past.

      The forces of sanity at Metro are just starting to play catch-up, but it’s not remotely surprising to me that the agency finds its support among the populace strained.

    2. Several of us have advocated strongly for making the 169 full-time frequent and extending it on the 101’s path to Rainier Beach Station. The former is one of the highest-ridership routes in south King County, and the latter would give Renton frequent access to downtown via Link. (This would replace the 101 off peak. A shorter route would allow it to run more frequently for the same cost.) As for the 106 and 107, I’m not sure they need changing in this scenario, but certainly they should run later.

      My own anecdotal observations is that the 150, 169, and 164 are busy all day (standing room only on part of the route), but the 101, 105, and 106 seem pretty lonely evenings and weekends (only three or four people besides me). So I’m seeing high all-day ridership in Kent but not in Renton, and that has perplexed me. This could be one reason for the neglect of Renton.

      Also, the “data-driven approach” is only 1 1/2 years old. That’s when the council repealed the 20/40/40 rule and told Metro to make route decisions based on performance. So what you heard about the 101 may have been before that, and also for the past five years Metro has been a funding crisis and can’t afford to add service even where it knows it’s needed.

    3. Not lied to. Metro’s revenue has been cut out from under it repeatedly since Transit Now passed, both due to initiative 601 and the recession and fuel costs and now the legislature’s paralysis, and Transit Now money has gone to filling holes in basic service. If you’re promised two more runs but two runs are cut, then you’ve still gotten the two more runs but the net result is zero. That’s pretty much what has happened.

    4. All the while I’ve seen Metro create all the rapid ride ___line (fill in the blank) buses, again in the more affluent areas of the system. Oh and seeing people who take these complain about them just irks the crap out of me!!! They run every 7-10 minutes ALL DAY

      If only. Outside of peak RapidRide operates at 15-minute frequency. Check out the schedules for yourself.

      Appendix L shows that the 101 is one of the routes slated for “very frequent” service – an upgrade to 15-minute off-peak service and 30-minute night service. But the simple fact is that there isn’t a glut of underperforming service elsewhere in the county that could be cut to pay for it. And while we squabble over where to allocate the service we have, we’re confronting the possibility of shrinking the whole pie by 17%.

    5. The only route that’s every 7-10-15-15 minutes (peak-day-eve-night) is the 71/72/73. The only thing more frequent is Link at 8-10-10-15 minutes. Then RapidRide, 7, 44, 49, 26/28 and maybe a very few others at 10-15-15-30. Most “frequent” routes are 10-15-30-30 (and 30 all day Sunday).

      As a beneficiary of the 71/72/73, I appreciate it, but it’s not a piece of cake, as buses are regularly standing room only, leave people behind several times a month, and are subject to bunching (where one bus is so late it comes at the same time as the following bus). That’s what people complain about. It also argues that since this route trio has such high ridership, it needs all the frequency it can get.

      1. The 3/4 between downtown and Harborview run about every 7 minutes all day; I think that’s the single most frequent corridor outside of downtown.

        The 7 and 36 both have 10-minute all-day frequency.

        There are a few other routes that come more often than every 10 minutes during peak (off the top of my head, there’s routes RRC/D, 41, and most of the popular cross-lake routes). There are plenty of routes that come every 10 minutes during peak.

        Of course, all of these buses desperately need the frequency, and your broader point is absolutely correct — it would be a wonderful world if most of our buses came every 7 minutes all day.

      2. OK, I didn’t know the 3/4 had gone above 15 minutes (midday), and the last time I looked the 7 had the same frequency as the 49. And I wasn’t being precise about peak frequency because midday is what we have to watch for — peak takes care of itself.

        We need to get to 5-minute frequency on the core routes, but first we have to get to consistent 15 minutes all day and evening, then we can work on 10 minutes, then 5.

      3. We need to get to 5-minute frequency on the core routes, but first we have to get to consistent 15 minutes all day and evening, then we can work on 10 minutes, then 5.

        Totally agreed. I think RapidRide’s biggest success is not speed but legibility; each of the RapidRide routes runs with a single service pattern (no branching or turnbacks), and meets Metro’s frequent service definition for all time periods (15 minutes or better, 7 days a week, from the start of service until 9:30 PM). I look forward to the day when the entire frequent network meets that standard.

  4. Yes, but did anyone read the comments on the article about Seattle? Very depressing. Many of them were disbelieving in one way or another, and those comments got the most positive votes.

    Is there a city in the US to move to where transit and bikes isn’t viewed negatively by most people, as something that “liberals” are trying to foist on people? I’m serious – – it seems like it’s getting worse and worse here. I never realized how bad people in the Puget Sound think transit and transit riders were until I started reading these articles and especially the comments.

    1. You’re reading comments again. “Don’t read the comments.”

      The Times comments are a right-wing echo chamber. If they were representative of public opinion in the area, Rand Paul would be mayor and First Hill would have been bulldozed to make way for a new, 16-lane I-5.

Comments are closed.