King County Metro buses (image: Zack Heistand)

Last year, a larger percentage of Seattle residents than ever reported using transit to commute to work. Seattle has made remarkable progress this decade in substituting away from drive-alone commuting. Booming employment growth in downtown Seattle and South Lake Union made it easier for workers to access jobs via transit. Large investments in bus and rail service added the capacity to get them there.

But there’s a less celebrated narrative in more recent data. Most of the progress this decade was before 2015. Light rail ridership was lower last quarter than a year ago. Bus ridership has been moving sideways since 2016. Despite large investments in off-peak service hours, non-work trips by transit aren’t growing.  Where transit ridership is growing, it’s not always keeping pace with population growth.

The decline in Link ridership last quarter wasn’t large, just -0.6% vs Q3 2018, but is still remarkable only three years after opening several new stations in a fast-growing city. Ridership for the year to date is a massive 12% below the Sound Transit budget plan, as an expected boost to rail ridership after removing buses from the tunnel failed to materialize.

Ridership on Sound Transit Express is down too. The worst declines can be attributed to a specific set of issues with Eastside routes over the I-90 bridge that were disrupted by East Link construction and the move of ST 550 to surface streets. But ridership generally hadn’t grown since 2015. It’s not for lack of investment. Sound Transit operates 9.5% more trips now than in 2015 and the cost of operations has risen even more with traffic delays lengthening trips and higher costs from partner agencies. Meanwhile, ridership is 4% lower than four years ago.

King County Metro fixed route service had healthy ridership growth through 2015, before falling with the U-Link opening in 2016 as rail supplanted several very productive routes. Since then, Metro bus ridership is up just 1%. Vanpool and Access ridership has declined. The challenges have been widespread, with 51% of routes seeing fewer riders. The declines are shared across all services; RapidRide, frequent and less frequent.

Recent Metro analysis highlights that the falls in ridership are off-peak and are particularly pronounced on Saturdays. This explains in part why the Census commute figures still look comparatively healthy. The loss of ridership is mostly non-commute trips.

Change in average rides by day and period, Fall 2016 to Fall 2018 (image: Metro)

The graph below shows annual changes in transit boardings by agency in King County. Before 2015, all agencies were showing solid growth, and Link ridership was boosted in 2016 and 2017 by new station openings. However, there is little net change in bus ridership after 2015. Total transit boardings are nearly flat by 2018 with the diminished growth on Link nearly offset by declines on ST Express.

Annual change in King County transit boardings by agency since 2010 (Image: Metro, click to enlarge)

Seattle and King County continue to grow fast, although both population and employment growth rates have slipped from their peaks. Employment growth peaked in early 2016 and population growth in 2017, though Seattle remains among the fastest growing major cities. Some slowing in transit expansion was to be expected with slower growth in downtown employment as the Amazon boom tapers off. What we’ve seen recently goes beyond that. Even with employment growth around 3%, Metro ridership growth is under 1% in each of the last three years. Countywide transit usage per capita is declining. Metro ridership has lagged population growth since 2013, and transit trips generally have started to decline relative to population since at least 2017.

Annual transit trips per capita in King County since 2010 (Image: Metro, click to enlarge)

There are a few interpretations one might draw from this.

Seattle is not immune to the headwinds depressing transit ridership in other cities since the recession. These include higher incomes in the city, the migration of lower income workers and jobs to suburbs where transit is less convenient, less immigration, higher car ownership boosted by inexpensive cars and a lower cost of driving. Together, they mean fewer people are transit-dependent. Most cities are seeing lower transit ridership at all times. Downtown Seattle is dense enough that driving is inevitably inconvenient and parking expensive. Most downtown workers with reasonable access will prefer transit for commute trips. But even transit commuters appear to use their cars more for non-work trips.

Our transit investments may have bumped up against diminishing marginal returns. 77% of STBD service hours are expended on weekend, evening, and night owl service. Perhaps that’s contributing to lower car ownership in the city, though it’s hard to see the benefits in off-peak ridership. Yet we don’t know the counterfactual of where ridership would have been absent the STBD spending. Maybe it helped us dodge the disastrous falls in ridership in some other cities.

On the other hand, at least some of the secular trends against transit ridership may be nearly played out. The I-90 bus issues will be resolved with East Link, if not earlier. Most importantly, a series of large rail investments will come to fruition early in the next decade adding another 200,000 daily rail trips.

158 Replies to “Seattle transit ridership pauses after years of rapid growth”

  1. I have a bunch of friends who use Lyft/Uber for a large percentage of their non-commute trips, whereas a decade ago they would have used the bus. They could have used taxis more in the past, but smartphone apps have vastly lowered the effort involved in getting a car to pick you up and drop you off.

    1. And taxis make a profit. Uber/Lyft never have. Would you friends still ride Uber if it became 30% more expensive?

    2. Not to mention the fact that busses in King county and the greater Seattle area are regularly late over crowded by the homeless and the bus lanes are filled with cars with no repercussions to drivers it’s sad. As someone who needs to be at work on time the Lyft and Uber app are way more reliable than the county metro these days.

    3. I’m one of those people. Partly because taxi service in Seattle 10 years ago was horrific. They’d demand an exact address, gave pickup estimates of “between 30 minutes to an hour” (if they even show up at all), always pretended their card machine was broken, frequently rip you off by taking weird detours… I don’t miss it at all.

      If a shared Lyft ride costs ~$8 and takes 10 mins while the bus takes over 45 mins and requires at least one transfer the choice is pretty obvious. When traffic is light or if you’re going anywhere other than downtown the bus gets less enticing.

      1. Or even going downtown from anywhere or during any time not served by an express bus. I understand the need for neighborhood service, but even with no transfers I can beat the bus on my bike 9 times out of 10. And that doesn’t even start on how confusing and impossible it is to pay fare if you aren’t on a monthly pass. There are no occasional riders on ST because the occasional rates are a nightmare to figure, paying involves putting money on a card far enough ahead to have it ready to use but not so far that it gets swallowed by the system, and none of the bus drivers know how to connect on their systems, much less between systems.

        I am a transit advocate. I DROVE for Trimet in Portland for two years and spent 3-4 years primarily using public transit for my commute in Boston. But in Seattle, I have tried it three or four times and something has gone wrong each time and so… so I gave up. The system is opaque to new residents, tourists, students, and anyone not on their main commute and not an expert level bus navigator. It serves for stretches and the leaves off, or is scheduled such that it’s faster to take a winding neighborhood bus during rush hour than to take an express. There are no route maps in the bus mall for someone without a smartphone app to route them (or with a dead battery).

        And since new people don’t start using the system occasionally, they don’t get to a comfort level where they opt in to using it consistently. And since Seattle is getting richer, people have other choices or they don’t live in Seattle.

        Northgate Link will help. It will make an obvious backbone route for the whole city. Making Orca work like the more simple and versatile fares in Portland, Boston or NYC, or even the moderately mixed but okay fares in San Francisco or Denver, would help too. We need to make simplicity a guiding principle, make everyone involved with transit in the Seattle area vastly more knowledge about connections, and put up more useful maps (more in both quantity and quality) on the bus mall and at transit centers. Then, curious people won’t get scared off but will grow to love transit.

  2. I think you are missing one large component, technology. More people are working from home. The percentage of remote positions will likely increase in the 2020s. Restaurants and stores are closing in Seattle. Food delivery and Amazon, makes it easier for people to stay home , if they wish, and avoid the crowds. I know I’ve done all before.

    1. Your comment runs counter to the available data, and my daily experience. Peak ridership continues to grow, as Dan excerpted above, so people are still commuting. The ground floor retail in most of the new buildings I see leases up quite rapidly. Restaurants close, yes, but the space turns over quickly, and usually the new restaurant is more upmarket from the previous. Sucks if you want cheap food, but it doesn’t indicate that nobody’s eating out any more.

      1. I think part of the problem is that Link is packed during commute hours. The city could help this in part by working with employers on staggering shifts or differing 4 day work weeks

    2. More people are working from home.

      True. Here are the percentages of commuters, from 2010 to 2018:

      Seattle: 6.5 to 7.2
      Other King County: 6.4 to 6.8
      Snohomish County: 4.4 to 6.5
      Pierce County: 3.5 to 5.7
      Kitsap County: 5.4 to 5.7

      So there is significant growth in workers working from home, especially from suburban counties. It is quite possible that what we have is a multifactorial situation (like so many now). Lots of people are now walking, instead of taking the bus. Lots of people are working from home. Transit to downtown sucks right now. There are more options for lots of people (fewer people depend solely on transit). None of these factors are enough to explain the fact that transit growth is not keeping pace with population growth, but together they could easily explain it. The population has grown 3%. Transit has grown 1%. The other 2% has gone to folks who work from home, walk, take a cab or just drive their own car to where they want to go (either because transit kinda sucks right now, or they can afford to drive and it has always been faster).

      1. I’ll take this a step further. There are plenty of stores closing in downtown Seattle. Macy’s will be gone. Forever 21 is gone. Barnes & Nobles will close in early 2020. The pacific place shopping center has a significant number of empty storefronts. Reasons to come to downtown, in terms of traditional sight seeing and shopping, are becoming fewer. If employers, imagine Amazon, decide to expand work from home positions, it will accelerate transit losses. There are many factors, but the world is changing, technology is playing a role. Amazon/Postmates/Caviar/InstaCart – make it easier for the top 10% to stay home. More and more people moving into the city, walk to work and make higher salaries. There is less need for city residents to slug it on a bus, when they can work from home/walk to the office/ order groceries on Amazon/ and have Dim Sum delivered on Postmates.

  3. Income inequality, and the spike in homelessness might also be an issue. Even if there isn’t an actual decrease in safety, the perception that there is may be deterring people from riding transit at night. Or just an avoidance of uncomfortable situations that Uber provides a way out of.

    If Seattle and other rich, liberal cities don’t lead on this issue, than no one will. Clearly not the feds or the state. And cities will become less livable and sustainable for both the rich and the poor.

      1. how would he make money that way?

        (yes I know, he doesn’t need more) but point is, no customers, no amazon.

  4. In addition to I-90, West Seattle buses have also gotten worse with the viaduct closure. What makes the West Seattle corridor especially bad is that the worst of the traffic is local to the transit route, meaning that, if you drive, you can bypass it.

    Uber and Lyft are also probably big factors. Off-peak trips are much easier to use Uber and Lyft as a transit substitute than peak trips. Not only do they happen less often (so less impact on the budget), but they are also more likely to happen in groups, allowing the cost of the Uber/Lyft car to be split.

    1. …and the restarted montlake construction has me driving again.

      The bus transfer to/from 520 – Link has become unusable. I don’t have 90 minutes to get to/from work.

      The upcoming 542 -> 544 debacle is yet another deterrent.

      1. The combination of 30-minute frequency and a forced transfer transfer at Yarrow Point is not good. I do it about 1-2 times week, but mostly because I’ve discovered various out-of-the-box tricks so that I almost never end up waiting anywhere near the full 30 minutes at the connection point. The new 255, which goes straight to the UW, every 15 minutes, will be a welcome change.

      2. Same issue here. The 542 to 255 transfer on 520 is timed awfully. It’s near impossible to get between UW/Kirkland during midday or past 7pm.

      3. The schedulers attempted to time it well, scheduling a 7-minute connection window. But they overlooked the fact that the westbound 255 is often late, coming all the way from Totem Lake/Brickyard, and, for eastbound, Metro assumes a much longer travel time than Sound Transit does over the exact same same 520 bridge, thereby causing the 255 to be consistently early, unless it hits traffic downtown.

        One useful trick that most don’t know about is that if you miss an eastbound 255 at Yarrow Point, you can jog down the 520 trail to Bellevue Way and catch the northbound 234/235. The way the schedule happens to work out, you’ll typically have about a 5 minute wait, and get home way earlier than you would waiting for the next 255. The catch is that you have to jog the mile or so between buses for the timing to work out – walking is too slow.

    2. Off Peak to and from West Seattle/Downtown in the daytime – Bus

      Off Peak to and from West Seattle in the evening – The car for me: My bus experience commuting off peak, without the convenience of living close enough to the Rapid Ride C line, is having an local bus alternative that shows up once every 25-30 minutes and waiting around in the evening when its either cold or there are crazies hanging out by the bus stop is unacceptable to me.

      Maybe more density in West Seattle will bring bring about shorter headways in the off peak hours, but not holding my breath for such improvement in the wake of I-976.

      Lyft is good for getting to the airport, but too pricey to constantly use around the city if you have a car, which I absolutely need since I don’t live in Capitol Hill anymore with its light rail:(.

  5. I’m a little puzzled by the surge in PM Peak ridership without any corresponding increases in other Weekday time periods.

    Occasionally I take Metro to go somewhere for the evening and then take another mode home but that doesn’t seem like a common enough scenario.

    1. Long ago I used to do a lot of walk (downhill) to work, then take the bus home. In general in Seattle I think work is more likely to be downhill than home.

      Don’t know why that would be increasing, however.

  6. A couple of thoughts:
    1. Uber/Lyft
    2. We are sort of partially towards a grid system with buses, but transferring is still a pain and a number of one seat routes became two seat routes.
    3. Buses are stuck in the same traffic so why deal with the crowding and anti-social behavior. Regardless of feeling about the street person situation, no one enjoys sitting next to someone who is obviously mental or smells very bad.
    4. Working from home

    1. The problem with Uber/Lyft is that it’s pretty much impossible to compete with business that can hemorrhage billions of dollars with no consequences AKA a functioning tax write off.

      It used to be commonplace that I would take the bus with friends around the City. Nowadays, once you have three people traveling together, Uber/Lyft is very competitive with the bus, especially considering time. That’s reflective of both the fact that Uber/Lyft are subsidizes (privately and publicly) and of our higher than normal transit fares.

      When (if) the Uber/Lyft honeymoon is over, we’ll see what impact it has on transit ridership. Like ALEX mentioned above, people these days are less gregarious in their work and social lives these days.

      1. Uber actually does profit on taxi rides by quite a bit, they take a 24% commission on every ride and operations of the core product are very cheap. The NYC TLC report found a 600% profit margin on their commission. Then they take that money they stole from drivers and spend it on insane stock based compensation for executives and managers as well as funnelling billions to study self driving and hover cars. It’s drivers that subsidize the ridiculous capitalist stuff, not vice versa.

      2. Can you share your data? I’ve read that they don’t make money on the rides, after you account for driver bonuses and whatnot.

      3. Chris: I don’t have actual data either, but consider:
        1) Uber/Lyft don’t offer nearly as lucrative sign up bonuses as previously as the driver base is so saturated they don’t have to any more. Those 1-3 minute wait times for rides only happen when there are a lot of idle drivers around!
        2) During surges, Uber/Lyft used to pay the entire surge amount (e.g., above the normal base fare) to the driver, now they keep the surge amount and drivers are lucky to get a flat rate “incentive” of $3-5 on those rides.
        It may well be that the drivers are subsidizing the corporate higher ups!

  7. Often I don’t feel safe on the bus. So many homeless and mentally ill people. People tell me all the time taking the bus is scary.

    1. There’s a difference between hearing about bad things vs. riding and seeing it yourself. I actually ride the bus fairly frequently (during my college quarter) and rarely if ever saw any of the problems you say. Most people mind their own business and it’s very rare to see a fight or argument happen on the bus. I’ve only ever see a fight break out maybe 2 or 3 times in riding metro over the years.

      1. My experience has been similar. If people begin to perceive it as unsafe however, fewer folks will ride it, and an empty bus is sometimes a dangerous bus. Downward spiral.

        If you want transit to continue to be successful, especially off-peak, you absolutely need to get a handle on poverty and homelessness. Buses are warm and dry. I’m not the only one who realizes that.

        Seattle is a rich city. It can put a roof over everyone’s head, if the political will were there. NY has a roof for everyone who wants one, and their problem is 10 times worse.

  8. I think more riders are having more bad experiences in the form of very delayed buses and bad traffic, that it’s causing some to avoid public transit, if possible.

      1. @Al S. +10
        I think you’re spot on all of your points. Frankly, I couldn’t have said it any better myself.

        I posted a comment a few months back saying how ST’s latest quarterly ridership report was quite troubling and now here we are. The agency is going to miss several of their annual targets for 2019, some by a significant amount, as Dan has alluded to in his piece above. I agree that the board needs to get ST to drill down on this and not just sit there nodding their heads and simply accepting the agency/Rogoff’s talking points about increased ridership not coming to fruition in the rail-only DSTT. How did they phrase it? Oh yeah:

        “Year-to-date ridership was 88.2% of the target primarily driven by the anticipated increases in ridership from the conversion of DSTT to rail-only were lower than planned.”

        Give me break. This sort of magical thinking has no place in the planning/budgeting process and the board needs to call out the agency for it.

        Lastly, in regard to how ST has increasingly become less transparent about providing basic ridership data and other performance information, I’ll just add that I find it completely unacceptable that it takes the agency 10+ months before they release their fare revenue report for 2018.

      2. Sorry. My post above didn’t nest correctly. This was a reply to Al S.’s comment below that began with “It’s a subtle mention, but ST’s unrealistic budgeting based on a significant ridership growth on Link is cause for concern. “

  9. It’s important to remember that this is boardings and not full trips. Part of the 2016 surge was because some riders went from making one boarding (bus) to two (bus + Link).

  10. Seattle has done a nice job with small, incremental improvements to transit. But ultimately the buses are still mixed with traffic and offer no advantage over driving or Uber/Lyft except to go downtown during commute hours.

    Honestly I doubt that will change. As a region we have decided to invest in new right of ways (light rail) rather than remove car lanes or parking for transit. Uber/Lyft is still cheap with virtually no restrictions. Maybe e-scooters will help with last mile connections.

    Finally we need to have a serious conversation about reducing the price of transit or making it free and what the pros/cons would be to do doing that.

  11. I think that people are increasingly less patient. They don’t like waiting outside for awhile for a transit vehicle. It’s one reason Uber/Lyft have gotten popular. I shudder to think how many more riders would walk away from transit without real-time arrival info on phones. I’m not sure how transit could counter our increasing impatience.

    1. I agree with this assessment that people are increasingly less patient. The region’s transit providers need to flip the script. A couple of items that would be helpful is the move towards a frequency based schedule for bus instead of timepoints and and make transit free. People get stressed out on the bus when it’s not moving. Eliminating excessive dwell for timepoints and fare payment would help this.

    2. I agree, as many people, including myself, have been far less patient about long waits, and that frequent service definitely matters a great deal, as does real-time arrival apps.

      As an example, a few weeks ago, I was with someone going to Fremont/34th to downtown. They suggested Uber, but I was able to convince them to ride the bus by showing them OneBusAway indicating that the bus was almost around the corner, and will likely to come *sooner* than the Lyft/Uber car.

      With Lyft/Uber, I’ve learned that the apps tend to consistently lowball their estimated wait times, in a bid to attract users, and the actual wait times are often about 5 minutes longer than what they say. At the same time, a shared Lyft/Uber ride making just 1-2 stops can often take just as long as a bus making 10-15 stops, because buses are surprisingly good at getting large numbers of people on and off the vehicle efficiently.

      That’s not to say I don’t use Lyft/Uber. I will definitely use them in cases where I just miss a bus with 30-minute frequency, or if I’m traveling very early in the morning or late at night, or if I’m going places that are connected by a straight line with a road, but only very convolutedly with bus routes. But, when the transit alternative is a frequent bus traveling in a straight line, the dollars spent/minute saved metric just isn’t all that great.

      1. I agree with you. I live in downtown and travel often at SeaTac. When I fly to the east coast, I fly out before 7AM. I live near a Link station, but can’t take Link to the airport. The first trains don’t arrive until after 5AM. Add the walk to the terminal, and lines at security – transit is not an option for morning downtown to airport trips.

      2. I often like to fly back to Seattle on Sunday evenings, so I get the full day to spend at wherever it is I’m visiting. But, that also means the worst level of bus service when I get back. My last trip, the plane arrived at the gate around 11 PM, and taking transit to my home in Kirkland would have required an overnight stay. (I could have gotten as far as either downtown/UW, via Link, or DT Bellevue, via the 560, but that’s it). So, I ordered a Lyft.

        (Pro tip: to avoid congestion at the SeaTac pick-up area, you can ride Link for one stop and have the Lyft pick you up at Tukwila International Blvd. Station. You can even order the ride while still on the train to reduce wait time. You also save about $5-10 by avoiding various surcharges and fees associated with being picked up at the airport.).

      3. I’m in the same boat re the airport. Sunday night service is bad as it ends an hour early (despite it being a busy arrival time as asdf2 and I experience). At least for me, going to downtown/Belltown has a transit option with the 124 – unfortunately while it extends past TIBS to the airport during owl hours when Link is not running, it does not change its hours on Sunday meaning you must walk out to International Blvd and catch an A, then transfer at TIBS. Metro should extend the bus to the airport earlier on Sundays.

        I do not use Lyft/Uber except in extremis for many reasons but probably an easier way to avoid the congestion is just to walk through the Link station and across the bridge to the bus stop, and pick up your ride there. No wait for a train. If you’re already at the south end of the terminal it’s a fairly short walk from arrivals to International Blvd as well.

    3. RapidRide C and D started with a headway-based schedule and it was hugely unpopular and the public eventually forced Metro to publish a full schedule. Headway management is OK for sub-10 minute service, but when you apply it to 15-minute middays and evenings, then all the buses after the first late one will wait 15 minutes each because it’s b-a-a-a-d to go sooner. That had multiple problems:

      – You couldn’t plan multi-seat trips with it because you never knew what time to take it to guarantee you’d make it to your other route. And that route was most likely 30-minute frequency back then. You’d have to leave a full 15-minute between the routes just in case. Whereas if the first route had a fixed schedule, you might find your you’ll only have a 2 or 5 minute wait.

      – 15-minute headways translated to “minimum 15 minutes”. If a bus was 10 minutes late, the second bus wouldn’t leave until 15 minutes after it, because it would be b-a-a-a-d to to go earlier. This means one round of people would wait 25 minutes and the next round 15 minutes and the third round 15 minutes; instead of the first round 25 minutes, the second round 5 minutes, and the 3rd round 15 minutes.

      – If the first bus is overcrowded because it’s late and some people can’t get on, they’d have to wait 15 minutes more a bus they could get on to, rather than 5 minutes in the traditional scheduling.

      A fixed schedule makes it easier to predict multi-seat itineraries, and know when to go to the bus stop. It’s annoying when it’s late, but that’s better than never knowing when the bus will come even in good times. Moscow has all headway-based service, but all metro lines and inner-city bus routes come every 5 minutes so you just plan that into all multi-seat trips. It’s easier to include a 5-minute buffer than a 15- or 30-minute buffer.

      1. Now I’m confused between RapidRide and Link. Link had headway-based scheduling. Maybe that was the one that had the uproar. In either case, the complaint was that headway-based scheduling is OK for peak-hour frequencies but inadequate for off-peak frequencies.

      2. My preference is for buses to follow a fixed schedule at the start of the route, but once they begin, just drive. It’s not perfect, but it’s better than artificially slowing people down at random times in random places.

        If buses become bunched, at least you’ll have OneBusAway to tell you that, and if buses just drive, rather than wait in random places because they’re too close to the bus in front of them, OneBusAway’s arrival times become more accurate.

        Link, to my knowledge, also follows a fixed schedule, although, with mostly exclusive right-of-way, it’s better at keeping it than buses. One datapoint I have to justify this was a few years ago when I was with a large group of people getting on the train at Tukwila Station. One person ran up the escalator, saw the train coming, and blocked the door for several minutes to hold up the train long enough for the rest of the group to make it up (including those that had to wait in line to buy tickets). After getting off at Westlake Station, just 5 minutes later, while I was waiting for a bus transfer, came the next train.

      3. Maybe this happened with both Link and RapidRide at different times. Because it feels RapidRide didn’t have a schedule and then it did, and also Link didn’t have a schedule and then it did. But there was always an internal schedule for the drivers. But the agencies didn’t want to publish it because they wanted to reserve the right to modify the times to achieve even headways without people saying, “But it’s supposed to come at this time.” And passengers said, “But there’s no advantage to us to wait longer than the headway when the bus/train is late, without any compensating shorter wait after it. We end up just losing that run.”

      4. Totally agree regarding frequency. I meant it for high frequency service such as 10 minute frequency or less as you suggested.
        I also like the suggestion from asdf2 below about starting with a schedule but then just go. The problem with mid route time points is they are so hard to predict if there are large variations in transit speed and reliability.

  12. Yet we don’t know the counterfactual of where ridership would have been absent the STBD spending. Maybe it helped us dodge the disastrous falls in ridership in some other cities.

    We had to make these investments. For far too long, Metro was focused solely on commute peak, then Seattle comes along and says “OK, we will add funding, but not just on the routes that have traditionally been funded.” I remember the debates over what should happen to the last of the unwieldy 82/83/84 Night Owls and which routes should get what. (I wrote a Page Two article back then over how to get something like 18 bus routes to half-hourly or hourly service 24 hours a day and got absolutely harpooned by our two resident cynics at the time, yet then Metro and Seattle come along and do basically the same as my proposal.)

    In my experience–and I wish Metro would publish per-route boarding data a lot more frequently–routes that don’t go through downtown or that have more dedicated lanes are more popular, especially on weekends, not less. I ride at all hours of the day and night but, yes, I can’t be everywhere so this is anecdotal (hence my desire for more frequent route boarding information). But routes that used to go mostly unnoticed off-peak, like the 75 or 347 or 73, are more popular. Heck, the 40 (which went from 30 or 60 minute service on Sundays to 15-minute most of the day; that’s amazing) was standing-room only on one of my trips this weekend.

    I think we’re panicking a bit too early. This isn’t a decline in ridership, it’s a decline in ridership growth. We all knew that our transit system would pick up a massive boost once Link served the densest ZIP code in the whole state and, sure enough, it did. That was an easy win. Now, we have to do the hard things to get the more difficult wins. We need to stripe more bus only lanes and enforce them. We need more all-door boarding, especially at transit centers.

    Those are the obvious ones that we all know. I would also like to make the transfer experience better. Whether that means having frequent enough service that nobody has to wait more than 10 minutes for their next transfer or devising a way for bus riders to tell a nearby bus “hey, I need that one, please wait 20 seconds” (without beating on the side of a departing bus), I don’t know, but I want a transfer experience that’s more reliable and predictable.

    We have a lot of room to keep improving; let’s start doing it.

  13. There are several factors. Uber and Lyft. Gas prices remain low. Congestion on the limited access highways is worse and transit speed slower. Transit is less reliable. therefore, transit is less attractive. The agencies have played a role. I-90 service is slower due to East Link construction; the D-2 was lost. All routes in downtown Seattle are slower due to the premature end of bus service in the DSTT for the county sale of Convention Place Station (CPS) to the Washington State Convention Center. Seattle did not help by taking 1st Avenue off the table for the CCC Streetcar dithering. the state was building the SR-99 deep bore. the transit agencies did not help themselves by restructuring routes to meet South Sounder or Link, so as many routes are trying to go through the more constrained downtown Seattle as before the CPS sale. The DSTT is less useful for circulation with longer Link only headway.

  14. One factor that I don’t see mentioned is the suburbanization of poverty. As many of the once working class neighborhoods of Seattle have gentrified, many lower to moderate income workers have been replaced by workers of much higher means. Those higher income workers are going to be more apt to carpool, Uber, Lyft, drive solo, or telecommute to work. The lower to moderate income workers have been displaced primarily to areas with poor transit service, like Kent, Federal Way, Auburn, or Snohomish County. Now, those same workers who would have taken a half hour bus ride to work are faced with a choice between driving to a transit center at an obscenely early hour to snag a parking spot followed by a 1 to 2 hour hour trip by bus or train, and a 45 to 60 minute drive. Despite the high cost of parking, when accounting for other external factors like child or elder care, limited daycare hours, or fixed pickup times or school bus schedules at schools, the flexibility of driving solo will become a necessity for some percentage of these displaced workers.
    That said, the solution to the suburbanization of poverty and the forced solo commute by displaced residents is an increase in housing that’s been long discussed.
    In parallel with working on our housing crisis, the next big boom in ridership could be accomplished by providing “2005 Seattle” service levels to the suburban areas of King County. That is, 30-minute headway, all-day service to all dense-ish neighborhoods outside of Seattle. The emphasis should be on getting people to key transfer points, such as the Sounder, Link, and I-5 freeway transit centers, so they can easily transfer to different routes – most likely the ones heading to Seattle, Tacoma, Bellevue, and Everett.

    1. One of the reasons I gave up blogging is that Twitter seems to have reduced the attention span of readers to about 500 characters plus a witty meme. If you read Dan’s third-from-last ‘graf, you’ll see he does mention the suburbanization of lower-income people.

    2. Dan did mention it, but in a round about way. It is important to stress that while some people are moving out of the city, more people are moving in. So we really aren’t seeing an increase in suburbanization — quite the contrary. The city itself is growing faster than the suburbs. Unfortunately, with rapid job growth in the city, housing prices keep going up, and people who can’t afford to live here move to the suburbs.

      That leaves you with lots of well to do folks in the city, and lots of poor folks in the suburbs. As Dan mentioned, the poor people in the suburbs have bad transit options, so they will spend money on a car, even though they can barely afford it. The wealthy people in the city, meanwhile, can certainly afford a car, and have no qualms about using it.

      The problem is that we still have a 1950s Seattle mindset. The bus is seen as something that poor people take, or something that everyone takes if they work downtown. That clearly explains the drop in off-peak ridership. We used to have a lot of poor people *in the city* taking those trips, but now, those poor people are in the suburbs.

      1. I don’t buy it being a wealth thing. Buses in Seattle are full of wealthy people during commute times. People want to do the right thing (take transit) and will even sacrifice a little convenience to do it. But they will not take transit trips that add 30-60+ minutes as almost any weekend trip will do.

        At off-peak times people tend to be going from neighborhood–>neighborhood rather than neighborhood->downtown. And the weekend bus network does nothing to support that, it funnels people downtown just like it does during commute hours. So you are forced to transfer downtown.

        It would be interesting to see if hourly express buses on weekend would be popular between different Seattle neighborhoods, i.e. West Seattle Junction -> Ballard through the tunnel.

      2. Buses in Seattle are full of wealthy people during commute times. People want to do the right thing (take transit) and will even sacrifice a little convenience to do it. But they will not take transit trips that add 30-60+ minutes as almost any weekend trip will do.

        Yes, but ridership actually went down. As terrible as it was (and it was worse than it is now) more people took “transit trips that add 30-60+ minutes”. That is because they had to. They couldn’t afford a second car, or maybe a first.

      3. The suburbanization of poverty is a complex phenomenon. It’s a whole other post to explore it properly. There is definitely a migration of low-income workers to the suburbs where transit access is relatively weak. And there’s some evidence that the transit growth in Seattle was disproportionately in higher-income groups.

        What complicates this though, is that there’s also a pronounced migration of lower income jobs out of the city. Firms that need a lot of lower-skilled employees are finding Seattle an increasingly expensive place to be. Housing costs put a high floor on labor costs (even more so than Seattle’s high minimum wage).

        So if a commute from South Seattle to downtown becomes a Covington to Kent commute, yeah, he/she is probably going to start driving. It’s a doubly-hard market for transit agencies to serve.

      4. There are four groups:

        1. Middle/upper-class drivers. They don’t take transit because they can afford not to. This also includes people who take taxis a lot.
        2. Choice riders, who take transit for the environment, the sense of community, to avoid driving, or to save money.
        3. Working-class drivers who can barely afford a car. They drive because there’s no reasonable bus alternative for their trips or they really dislike buses. They don’t take taxis much because they can’t afford them.
        4. Working-class/poor riders who can’t afford a car or taxis. They have no choice.

        The suburbanization of poverty means that choice riders in Seattle are becoming compulsory drivers in the burbs. Those who already drove in Seattle continue to drive in the burbs. The new people moving into Seattle are both choice drivers and choice riders. Some of them stop driving when they move to an in-city home. But they’re still a minority. Even in southwest Capitol Hill with buses going every few minutes in almost every direction and almost everything is within walking distance, a surprising number of people drive everywhere. It’s so much that I assume all my neighbors drive unless they tell me otherwise or walk to work.

      5. “And the weekend bus network does nothing to support [neighborhood-to-neighborhood], it funnels people downtown just like it does during commute hours.”

        It does more than it did in 2010 or 2015. The 16, 17, 22, 25, 42, 54, 55, 56, 56, 66, 71, 72, 73, 106, 107, and others have all been restructured to strengthen the crosstown/neighborhood-to-neighborhood grid network. Routes 8, 31/32, 44, 45, 48, 60, and 75 have increased service.

        The ease of traveling neighborhood-to-neighborhood varies widely according to location and trip pair. In some cases it’s better; in other cases it’s worse. If you think everything is worse, you must live along one of those downgraded trip pairs.

      6. re Mike Orr rider categories: at the margin, all riders are choice riders; even those without a car, have options and choose how many trips to take by transit. the more attractive transit is, the more trips they will take. this is true for most rider categories except those wealthy who never ride.

      7. @eddie – Yeah, of course to a certain extend all riders are choice riders. But if your car is in the shop (or you simply don’t have one) then a thirty minute bus ride sounds a lot better than a forty minute walk (especially in this weather). If everyone with a car is busy, and you can’t afford a cab, then there is no choice, really: you are taking the bus.

        But if you car is solid, and dependable (even if it is a bit beat up) then driving sounds mighty appealing. You can be there in ten minutes. That choice, if you will, is responsible for at least some of the rides that would otherwise be on the bus. Gentrification — increased wealth — within the areas that represent the bulk of transit ridership (Seattle) is likely to be somewhat significant.

      8. Given the choice between a 30-minute bus ride and a 40-minute walk, I would personally choose the 40-minute walk, barring a heavy rain. Especially if the 30-minute bus ride involves riding a bus 10 minutes, waiting 10 minutes, then riding another bus for 10 more minutes.

        At least with the 40-minute walk, I:
        1) know exactly how long it’s going to take, and are immune to traffic and/or service delays
        2) Have the ability to jog portions of the route, potentially cutting the 40 minutes down to 20-30 minutes.
        3) Keeping warm by moving, rather than getting cold by standing at a bus stop
        4) Feeling better about myself by being more in control and getting exercise.

  15. The next major policy action we need (besides invalidating or negating the effects of I-976) is downtown decongestion charging. Downtown’s a schedule graveyard for every bus that goes near it, and the only plausible solution is to significantly reduce the number of cars on the road during the weekday. Evolving the bus system away from an almost-exclusively radial system has been great, but most parts of the city don’t have Link, and the highest ridership bus routes almost all intersect greater downtown. There has been radio silence on this topic from the Durkan administration since May.

    1. I don’t expect that to happen. The polling on that is terrible. Durkan has trouble with various constituents — she is going to struggle getting reelected. The last thing she wants to do is upset her base (liberal moderates who drive). I think there is a much better chance that the city will tax Uber/Lyft for rides, as well as rides in general. I could easily see a sliding scale sort of thing, with the highest rate being for trips to and from downtown during rush hour. This will do little to decrease traffic, but fund some other projects.

      For downtown, things will slowly improve on their own. The construction on SR 99 will finish, and pretty soon buses like the 41 and 550 won’t even go downtown. We should also continue to push for expansion of the bus lanes, along with better enforcement. The latter costs money, but I could easily see some of the Uber/Lyft tax going into that.

      In general, I don’t see commute trips being the big weak point here — at least not long term. The bigger issue is all day service (and I’ll offer up my opinion on that on a separate thread).

    2. I should have added “more off board payment” to the list of relatively cheap improvements that could make a big difference on a lot of routes.

    3. The city should establish monthly car-free days. Yes, this would take a ton of detailed planning, communication, and alliance-building (aka leadership), but it could be awesome. Maybe start on a Saturday, maybe make transit free that day. Run more trains but you might not actually need more buses since they’d run so much faster. Metro would have to make an entirely new schedule or base run times on late night schedules.

      Once people see what that looks like, maybe you’d see more support for decongestion pricing.

    4. Well, there was a discussion during budgeting because the mayor put in funding for congestion charge outreach, which is for designing an equitable policy and presumably improving the dismal poll results. But yeah, it’s polling horribly and it has to pass on the ballot. Yet it’s the centerpiece of the mayor’s climate plan. I don’t understand what the mayor is thinking. Is she waiting for a re-election to try? Is she avoiding controversial action by punting to voters?

      1. She’s weasling on transit and carbon emissions. She’s a status quo person, or she favors the current consensus, or she thinks it’s the best way to get incremental improvements done and get reelected. If she has any innovations they’re in non-transit, non-climate areas. Otherwise she just follows what the consensus says or any ideas she reads about in the paper. Her transit policy is to continue the policy refined by the Schell/Nickels/McGinn/Murray administrations. Her strong support for the streetcar is one aspect of this.

  16. Link ridership is likely skewed by the number of people who do not tap. I see it all the time when I tap my card and someone behind me doesn’t and they continue down to the platform and board the train. It seems like some just are willing to take the risk of getting fare checked rather than paying.

    1. Link ridership is based on Automatic Passenger Counters located above the doors, which are validated/calibrated by occasional manual counts.

      Failing to tap may skew ridership by station and metrics around bus/rail transfers, but shouldn’t impact the gross Link ridership number.

    2. each Link station has several card readers; you cannot know for sure they did not tap outside your vision.

    3. Fare evasion is 2-3% according to ST’s estimates. So for 100 people on a train, 3 may have not tapped and don’t have a ticket. Some of those people who intended to tap but they forgot because there was no reader right in front of them or they misunderstood its beeps/messages. I assume you’re not including those people because you said they’re right behind you and saw you tap. Still, if you subtract them, you get 1-2%. A tiny fraction of the ridership. Not enough to make much of a difference in ST’s overall numbers.

    4. If anything, Link ridership should have gone up, since the tunnel doesn’t have buses. In the past, a lot of people wouldn’t tap (since they were waiting for a bus) so it wouldn’t be crazy for someone to just get on the train, and start looking for the place where they pay. Then you have “first available” riders, who waited until the last minute. They could easily forget. Those days are gone. Everyone who rides on a regular basis is in the routine of tapping, while those that aren’t see almost everyone doing that somewhere.

      1. I realized I worded that wrong. What I meant to say that the rate of people tapping within the tunnel probably went up. Link ridership is a different thing.

    5. A lot of former tunnel riders were making intra-downtown trips, or going down to SODO where Link still overlapped with the 101 and 150. That meant a train every 6-10 minutes or a bus 5-10 minutes, so one way or another something was usually coming in 5 minutes or less.

      Now it’s only Link, and off-peak frequency is 10 or 15 minutes. Whereas the 7/14/36 on the surface come every 3-5 minutes. And now that 3rd Avenue is bus-only all day, it doesn’t get bogged down as much and operates more like the tunnel. So a lot of those intra-downtown trips have moved to surface buses.

      1. Yep. I mentioned that as a real possibility before they kicked the buses out.

        It isn’t clear whether there was a significant loss in ridership though. ST doesn’t have stop to stop data. Sometimes they have directional data, but they didn’t in 2019. In general, the ridership reduction was across the board (if memory serves).

      2. It was clear to most of us who live/work downtown that this would happen. I’m often darting out to meet somewhere at the other end of downtown and whereas before I would always just pop down to the tunnel as whatever came would get me there faster, I don’t ever do that any more. The buses on 3rd work just fine in that regard, the only issue being knowing enough to not catch a bus that turns off before you get to where you’re going! :)

        There’s nothing inherently wrong with that – I love the frequency of buses on 3rd (despite it being the most horrible street in the entire city in other regards); you can easily get anywhere from Seattle Center to Pioneer Square with little to no wait most of the time. Perhaps with 4 minute train frequencies that will change but hopefully bus service will also continue to improve with more lanes and enforcement of said lanes.

      3. “the only issue being knowing enough to not catch a bus that turns off”

        That’s why I emphasized the 7/14/36, for anyone who doesn’t know that they go all the way to International District Station, and they continue beyond it to 12th & Jackson. There the 36 turns off, and a block later the 7 turns off.

      4. If frequency in the DSTT is a factor in suppressing ridership, it would be in ST’s best interest to address this and begin operating the second line in the DSTT tunnel ASAP once the Northgate Link stations open (before all of East Link opens).

        I realize that this depends on train car storage and possibly early use of At least a portion of East Link tracks and that this is about a two-year period (something discussed several times in other posts recently). Another issue this would address is the impending overcrowding of Link after Northgate Link (even with four cars in a train) until the second line opens.

      5. Indeed, although my comment was more holistic in that you might be headed the other direction and should know what buses go all the way north to, say, Denny (D, 1, 2, 3, 4, 13, 29 more or less). If you’re going from one end of downtown (ID) to the other (Belltown/Denny) even the 7/36 don’t help much – the 14 does if you know it turns into the 1.

        I’m still bemused (and not in a good way) that the D does not extend all the way to Jackson to make its loop. It’s a major route to several different neighborhoods and you shouldn’t have to walk to 3rd/Yesler to catch it if you just got off Sounder/Amtrak… especially since that is not a particularly pleasant walk.

      6. It occurs to me that if people are sensitive to frequency Downtown, the First Ave streetcar project (Center City Connector) is a waste of money. Not only will the streetcar compete with better frequency of buses on Third Ave but a Link system running at 4-5 minutes until after 10 pm most days.

        If the lack of higher ridership can be blamed on lost riders making DSTT trips, we have real-world proof that a slow and less frequent and generally less direct service is not a wise use of transit funds. We should be building more and better entrances to Link stations, and better vertical circulation in Link stations instead.

      7. “If frequency in the DSTT is a factor in suppressing ridership, it would be in ST’s best interest to address this and begin operating the second line in the DSTT tunnel ASAP once the Northgate Link stations open”

        ST’s interest is to get Northgate Link, Lynnwood Link, and Federal Way Link open as soon as possible. That would solve many of ST’s operational and customer-satisfaction problems. Maximizing intra-downtown ridership in the meantime is not its priority. They can use the surface buses if the don’t want to wait 10 minutes.

      8. It occurs to me that if people are sensitive to frequency Downtown, the First Ave streetcar project (Center City Connector) is a waste of money. Not only will the streetcar compete with better frequency of buses on Third Ave but a Link system running at 4-5 minutes until after 10 pm most days.

        Yep. The folks touting the streetcar seem to ignore that it is only two lines. Two overlapping lines, each running 10 minutes sounds great, until you realize that there are dozens of overlapping, similarly frequent buses running a couple blocks away. Oh, and one of the two overlapping lines is prone to major delays, making streetcar bunching (i. e. 10 minute frequency) likely. Of course there will be riders; but there won’t be that many, because many of the potential riders will simply walk a couple blocks (and in some cases, one block) instead of waiting (and waiting, and waiting …).

  17. Another factor might be due to ReachNow, LimePod and Car2Go (Sharenow). I often don’t want to deal with the slow routes 3 & 4 in the morning when I am running late so I take a free floating car share vehicle to work. With the 3 & 4 you get so many homeless and mentally ill too that it can be scary or they end up slowing the bus down even more.

    1. The 2, 3, 4, and 12 are absolutely ridiculous: it shouldn’t take half an hour to go less than a mile between two of the highest-ridership areas in the county. I’ve learned to take a Pine/Yesler/Jackson route if I don’t want to get caught in it. I used to live a block from Harborview and a 3/4 stop. Still, I rejoiced when the 27 came first because I could avoid the slog up James Street. If I’m downtown and going to somewhere around Madison/Seneca, I often take a Pine Street route and walk, both to avoid the bottleneck and because the overlapping routes are 2-3 times more frequent.

      1. the eastbound Route 2 was improved by the SDOT bus lane and signal at 6th Avenue. yes, the James Street situation is sad.

      2. I don’t ride route 2, but I have ridden the Trailhead Direct bus, which used the same bus lane and signal at 6th that the 2 uses. I was not impressed, as the signal seemed to give clear priority to the cars, while the bus waited and waited. The bus probably could made better time by not even using the bus lane, in order to piggyback on the car priority.

    2. Still, the problems with the 2, 3, 4, and 12 have been the same for decades. This ridership loss is only in the past few years.

  18. I wonder how much walking might be part of the equation. As more housing opens in the SLU/Denny area, there are more people walking to work. Presumably that’s the demographic that would be car free and transit friendly, but if you have a 15 minute walk to work, you probably don’t bother with the bus most days.

    From Dan’s prior article, Walking mode share increased by 1.5 ppts. If that’s for the whole city, that would be a pretty large number relative to marginal transit growth?

    1. It was just Seattle that saw a substantial increase in commuters that walk. Still, that could easily change the numbers, especially for midday trips. There are a lot more people living in the greater downtown area — an area that has increasingly become more “all day”, rather than “9 to 5”. I could easily see how way more people just walk to their destination if they don’t have a car, the way folks do in New York. You would have to see an increase in driving to be really concerned.

      1. part of the pedestrian increase may be due to increased employment in First Hill and SLU that are near concentrations of residences. the sidewalks of Uptown, SLU, Capitol Hill, and First Hill are busy.

    2. Yes, but many of them are living on floors that didn’t exist in 2010. One building was three stores with large 1950s apartments with large balconies and was so run down it was practically empty, replaced by a seven-story building with smaller units. The rise in SLU jobs and downtown walkers-to-work coincided with tripling or quadrupling the number of units per building. So you could say all the walkers are in the new units. That’s not technically true because some walkers are in older buildings or replacement units, but numerically they cancel each other out.

    3. This is a good point. I also think that would affect off-peak transit ridership as well. If people live in a walkable, dense neighborhood, they’ll likely drive and bus less for day-to-day errands or to go out to eat or drink. I certainly have noticed a lot more pedestrians in the central city over the last few years – especially since the denny triangle and south lake union have filled in with apartments and offices.

      Like others, I’d be more concerned if overall off-peak motorized trips have risen while transit ridership has fallen, and it will be interesting to see how these trends change as more Link extensions open up.

  19. It’s a subtle mention, but ST’s unrealistic budgeting based on a significant ridership growth on Link is cause for concern. Being off 12 percent because of an overly rosy ridership increase projection without any new stations or efforts/ restructuring to add riders is pretty irresponsible.

    The “feel good” Board members and senior management shouldn’t be shrugging this off. It comes with a clear double-speak about performance data, saying that more measures are needed like how clean bathrooms are while cutting back on public disclosure of the fundamentals of boardings and revenue (like no longer providing monthly ridership reports).

    1. Short-term ridership may not meet expectations but long-term ridership will probably exceed it, because of the factors ST isn’t counting. It can’t count upzones that weren’t committed before the ridership plan was calculated, and there will likely be a societal shift toward transit but nobody knows when or how much. There has already been a partial shift, with many tech workers and urbanists choosing transit who twenty years ago they would have driven and lived in car-dependent areas. And all kinds of other people commuting to work on transit who twenty years ago would have driven. This trend is likely to accelerate in the next 20-30 years — regardless of what ST or Metro does — as long as the transit doesn’t become significantly worse. For instance, if Seattle’s TBD supplementation disappears long-term and base Metro service is cut 20-25% and ST suffers a similar loss, then the increased ridership would reverse. Especially given the continued population growth and congestion increase.

      1. I can’t by the notion that what was missing is a lack of densification through upzoning. 2019 looks pretty similar to 2018. The housing market here, while still squeezed, has been able to consistently add around 2 percent to our population for several years and our construction boom is still happening.

        I agree that long-term densification will add riders, but there are many other countervailing forces at work. Work-at-home, work at hours not convenient for using transit, part-time jobs that don’t offer Orca, on-line shopping and food delivery and buses being slowed because we gave lanes to bicyclists instead of buses are all factors — in a city that has already had decades of excellent transit so that expecting a systems growth of double digits in a single year (unless a new line opens and takes riders from elsewhere) just won’t happen in reality. It’s very unrealistic budgeting.

      2. I was thinking of the ridership estimates for 130th Station, the Aurora alignment, Pacific Highway, etc. In every case ST said the option wouldn’t add riders. But if they had large urban villages — many seven-story buildings — then they would. Vancouver even has small highrise clusters around its Skytrain stations. The current generation of city leaders won’t approve it, but maybe a future generation will. Many people want to live within walking distance of a Link station with a variety of walkable businesses in the neighborhood, and they would ride Link. Only a few can because of stunted zoning. But if you build significantly more urban villages around Link stations, they would fill up with pent-up demand. Even without Link, those villages could thrive with high-quality RapidRide feeders. And more people could live in the kind of neighborhood they want (which is currently skewed one way), and they would be a larger percentage of the population. There’s no downsize to urbanizing Aurora and Pacific Highway, just the lack of political will.

      3. People who work at home still go to the grocery store, errands, social groups, bars, parks, etc. It would be nice if they could do all these easily on transit. Currently a significant minority can, but we should aim higher. Everybody should have the opportunity to live in that kind of situation if they want to. And even if they move there and drive, they’ll still find it so convenient that sometimes they’ll walk to a destination or take transit. At least they can, whereas now they can’t, unless they can score one of the top 10% of apartments.

      4. In general, ST has underestimated urban ridership, and overestimated suburban ridership. I will say though, that there are way too many moving pieces to come up with a good estimate.

        The biggest problem is the fact that there are two different agencies involved. Coordination with Metro is abysmal. You would think ST would first talk with Metro (by far the largest transit agency in the region) before picking station sites, but that isn’t their style. Metro would have preferred 155th over 145th (saving ST millions with their BRT plan) and certainly would have requested a station at 130th. It cuts both ways. ST doesn’t know whether Metro will truncate or not. They assumed (incorrectly) that Metro was going to truncate at Mount Baker. But that station was so poorly placed, that Metro didn’t. They figured (again incorrectly) that Metro wasn’t going to truncate many of the routes at the UW (buses like the 73 would keep going downtown). But Metro took an aggressive restructure, which probably lead to an extra 5,000 riders a day. Without knowing who is going to restructure, it is pretty easy to get the estimates off.

        To the north (and east) it will become a lot more clear. I don’t expect any buses on the north part of I-5 or I-90 once Link expands. To the south it is a lot trickier. ST will probably truncate their own buses, but buses that come from places like Renton will continue to just go downtown. Much of our system is dependent on feeder buses, but it isn’t clear whether they will exist or simply parallel the train line.

        Then there is growth, as Mike pointed out. Rainier Valley stations have grown because the area has grown, but that is a mixed bag (in part because of the weird issues involving the land around the stations). Trying to guess whether those (or other) stations will suddenly have a lot more people is tough.

        In general, I think it is very difficult to estimate future ridership for those (and other) reasons. I cut them some slack in that regard. But assuming that there would be a 12% increase in ridership (a very high number) just because the buses are no longer in the tunnel is not only overly optimistic, but bizarre. I really don’t understand their reasoning. Budgeting under that assumption is irresponsible.

      5. “ST doesn’t know whether Metro will truncate or not.”

        That was true when Lynnwood Link was planned, but it hasn’t been true since Metro released its long-term plan in late 2015. The future network may not be exactly like the plan, but major issues like which routes will be truncated will probably remain stable. It’s not clear which changes will land in part 1 (Northgate Link) and which in part 2 (Lynnwood Link), but that will be settled by 2024. Minus any reductions and postponments related to I-976.

      6. Oh, and the full plan depends on the countywide tax measure next year. Without it, probably some of the Rapid lines would be downgraded to Frequent, and some of the least-plausable new corridors may disappear (e.g., the Aloha Street routes).

      7. @RossB
        “But assuming that there would be a 12% increase in ridership (a very high number) just because the buses are no longer in the tunnel is not only overly optimistic, but bizarre. I really don’t understand their reasoning. Budgeting under that assumption is irresponsible.”

        Exactly. Not to beat a dead horse here, but the ST board shouldn’t be in such a hurry to bury said horse. As I stated above, the agency’s reasoning doesn’t make much sense imo and speaks to some sort of magical thinking being used in the planning/budgeting process with regard to ridership projections. In other words, I think the explanation given for the “failure of the increased ridership in the rail-only DSTT to materialize” is mostly bs and the board shouldn’t just take the agency’s selected narrative at face value.

        FYI, the agency just published its 2019 Q3 financials the other day and in this report ST continues this same narrative.
        What was more alarming to me, however, was the canary in the coal mine contained in the report’s YTD data on capital projects. I won’t go into all the details (since it’s rather off-topic for this thread), but the numbers highlight a considerable lack of progress in the capital program for 2019 thru the third quarter.

      8. That was true when Lynnwood Link was planned, but it hasn’t been true since Metro released its long-term plan in late 2015.

        So the U-Link Metro structure came after that? Seems like folks at ST were fairly surprised (as were most people) when they went full truncation.

        Anyway, the long range plan isn’t written in stone. For example, they have — in 2040 — a route on Delridge that goes all the way downtown (Rapid Ride no less). That means no truncation at Delridge. Will that really happen? If so, then West Seattle ridership will essentially be cut in half. Call me skeptical. I think they will send all of those buses to Alki, and ask people to transfer. I guess we’ll see.

      9. I may have the timelines wrong but Metro’s LRP was available during ST3’s solidification in December 2015-April 2016 because it was part of that debate. U-Link opened summer 2016 and its public planning started around either December 2014 or December 2015. So Metro’s LRP may or may not have been available then, but if it was it was just published and getting initial feedback.

        The LRP focuses on 2025 and 2040. I’m assuming “2025” means Lynnwood Link and RapidRide G. U-Link is a halfway holding pattern, and Metro implemented some of its long-term ideas but held off on others until Northgate Link.

        “Seems like folks at ST were fairly surprised (as were most people) when they went full truncation.”

        I was surprised. ST probably heard it from Metro earlier because they tell each other what plans they’re drafting.

        “the long range plan isn’t written in stone. For example, they have — in 2040 — a route on Delridge that goes all the way downtown (Rapid Ride no less). That means no truncation at Delridge. Will that really happen?”

        You’re assuming it won’t. On the one side we have an officially-blessed draft network. On the other side we have one amateur skeptic. I interpret RapidRide H as “the C is gone because Link has three stations between AJ and Delridge, and the Fauntleroy-SLU express for the ferry gives a secondary one-seat ride to almost downtown. RapidRide H remains a one-seat ride because Delride doesn’t have multiple Link stations along it and there’s no express route.” Both the H and the express route are an attempt to find a compromise between Link-transfer efficiency and the one-seat ride activists. Both of them are iffy: they may be modified after public and political feedback. Still, it’s the only comprehensive and officially-blessed plan we have at this point.

        I may have overstated when I said Metro knows what ST will do. ST knows what Metro’s overall goals are. That’s what I meant. It knows they may be modified after political input, just like ST’s own alignments are modified after political input. The point is that before the LRP, nobody know what Metro would do from decade-to-decade because it never said anything long-term and proposals were canceled when one constituent objected.

      10. And budget restrictions. The express route may not make it if the countywide measure fails and I-976 prevents Seattle from raising enough to save it. The H to downtown will probably be prioritized in Metro’s base budget, I’m assuming, so it would cut other things first.

    2. @Al S.
      I agree with your comments here. Please see my complete reply above as it didn’t nest correctly when I posted it for some reason.

    3. I don’t see why they felt that kicking the buses out was going to lead to a lot of new riders. The trains run a little better, but that was the only improvement. It is not like the trains are a lot more frequent. I suppose you do have the possibility that since the buses are more messed up than ever, people might gravitate towards the train. For example, instead of catching the 41 towards downtown in the evening, folks (like me) take the 373 to the UW and then the train. But that really doesn’t work well, and certainly shouldn’t lead to significant increase.

      Meanwhile, you have a lot of things that are worse. For example, transfers from buses like the 41 to places south are worse. Northgate to Beacon Hill used to be pretty smooth – now it isn’t. In general, transfers involving the train are worse (buses are more crowded downtown and spread farther away from Link stations). Using the tunnel as a way to get from one part of downtown to another is worse. Link only got a portion of those rides, but a portion of a lot is better than 100% of nothing. Those riders won’t bother waiting forever for the train — they just take surface transit.

      Overall, I would have expected transit use to be flat, if not decrease a little once they kicked out the buses. I’m not sure why ST thought otherwise.

      1. “I don’t see why they felt that kicking the buses out was going to lead to a lot of new riders. The trains run a little better, but that was the only improvement. It is not like the trains are a lot more frequent”

        I generally agree with your stance here but let’s not forget how much of a PITA it was to be standing on the platform looking at your train stuck up the tunnel while one bus finished a long loading cycle so that another bus could pull up to it’s bay further along the platform to start it’s long cycle and oh great, the wheelchair ramp is coming down, this guy is paying entirely in nickels stored in various pockets, and the platform is really starting to get crowded… Or even better: being stuck on that train in the tunnel with no information and no idea what is holding the show up.

  20. Is peak ridership growing up off-peak not?

    Link has capacity issues at peak, particularly during Connect 2020, which really won’t be resolved until East Link opens. When I was at ST, there was discussion around stepping back from investing dollars & time in advertising / customer outreach to add riders to an already overloaded system, instead focus on outreach around construction impacts, reroutes, etc., and go back to investing in growth once you have a bigger capacity to sell.

    Similarly, I don’t think Metro is trying to get new riders for its best routes, and those are already crowded, nor is WSDOT telling people to use already full P&Rs.

    So less advertising could be depressing growth, particularly relative to all the fanfare around Ulink and Angle Lake

  21. Most of the numbers here don’t concern me. These numbers tend to go up and down. We have endured (and still endure) a mess downtown because of all the changes. Kicking out the buses early was bound to hurt overall ridership (I have no idea why ST thought otherwise). You have buses substantially delayed, while service within the tunnel has very low midday frequency. West Seattle and Bellevue got hammered by the changes to SR 99 and I-90 respectively — no wonder ridership took a hit. Even the biggest improvement to the system (UW to downtown rail service) came with questionable changes (the restructure in Capitol Hill as well as the truncations). Overall the system is better, but quite a few people came out behind. Despite all those problems (most of which are temporary), commute share is holding steady.

    It is ridership in the middle of the day and Saturday that got hammered. OK, maybe that is too strong a word. If I’m not mistaken, that is system wide. For a weekday, there are about half a million riders — so a drop of 700 is bad, but perhaps a reflection of the entire system, summarized quite well in the headline (we have paused). Still, it isn’t good.

    This statement, from Metro, was interesting:

    During the off-peak periods, routes with frequent service (defined as routes scheduled every 15 minutes or less) had smaller declines in ridership since 2016 than those with less frequent service. Frequent routes on Sunday realized a 1.8 percent increase in ridership.

    So frequent routes on Saturday took a dip, but frequent routes on Sunday went up. Less frequent routes were down both days. In my opinion, the takeaway is that except for trips downtown, riders demand very good transit. If they don’t get it, they won’t use it. There was a time — not too long ago — when a substantial number of riders had to use it (they had no other choice). Now those days are gone.

    We need to keep chipping away at efforts to build a real transit network. We need to keep working towards a grid. The geography of the city make that tricky, but it is the only way to get high frequency routes that don’t involve a tremendous amount of backtracking. Someone in South Lake Union should not have to go downtown first to get to First Hill (they should go straight across, on Boren). Someone trying to get from Lake City to just about anywhere in northwest Seattle (Bitter Lake, Greenwood, Crown Hill) shouldn’t have to detour all the way down to Northgate, then back up again. In greater Ballard we should have a bus on 65th — Market and 85th shouldn’t be the only east-west bus.

    It is highly unlikely that any of these changes will happen before Link gets to Northgate (and many won’t happen after). Judkins Park Station, and Madison BRT will also be cause for a shakeup. Like transit itself, a grid scales. The more frequency you have, the less painful the transfer, allowing you to avoid costly redundancy (e. g. sending all the buses downtown). The city is borderline, in many ways. We don’t have quite the density to justify a lot of the transit that we want (i. e. a grid that allows frequent, fast service everywhere to everywhere). But the more we push towards that goal, the better these numbers will be.

    For the suburbs, it is trickier, but there are similarities. When 520 construction is done, I could see those buses all being truncated at the UW. It is unlikely you would have enough ridership to justify a good grid in the area, but you could have additional express service to the UW along with East Side Link locations in Bellevue and Redmond.

  22. I might not be an expert on ST express routes, but the declines are mostly in non-peak and weekend rides and ST Express routes. What’s the Venn on those two riderships?

    Anecdotally, I’m definitely in the commuter usage bucket. As a northender, most of the friends and/or shops I frequent are diagonal in Greenwood or Ballard. There’s not that much interesting south of me until the Udistrict. Or LCW, which requires at least 1 transfer, even though it’s less than 2 miles to the business core there. East-west routes are pretty weak in Shoreline. The result is that I bike for most weekend trips.

    1. East-west routes are pretty weak in Shoreline.

      They are pretty weak in most of Seattle as well. The few that are very good are very slow (8 and 44).

    2. I’m not a numbers person but the 550 was hit especially hard. Its northeast-most southbound stop is 5th & Union (a longer walk from Westlake, southwest Capitol Hill, or the Denny Triangle). It lost the Rainier freeway station and South Bellevue P&R. It runs on surface streets and on the I-90 regular lanes in SODO/North Rainier. It often comes late now. It used to just meet RapidRide B for transfers east but now they just miss each other.

      Lake City to the U-District worsened with the loss of the 72, but that’s a pretty unusual case. Lake City to Ballard and West Shoreline to East Shoreline have always been bad, so they don’t explain the loss since 2015. The 75 was split in 2012 if I recall (losing the one-seat ride between Lake City and Ballard), but that was before 2015 and during the time of biggest ridership growth.

      1. yes, Route 550 lost the South Bellevue P&R, the center roadway of I-90, the D-2 roadway, and the DSTT. it is about 10 minutes slower in each direction. that makes it less attractive to intending riders. for those oriented to the north end of downtown, Routes 271 or 556 and Link is a better option.

      2. The 271 doesn’t help when it comes at the same time as the 550 or five minutes later. That’s what happens on weekends when I’m coming back from Bellevue. I can take either route, and sometimes I might want to take the 271 and stop on Broadway on the way home. But if I have to wait as long or longer for the 271, it isn’t worth it.

      3. I’ve confirmed through experience that heading from Bellevue Transit Center to 5th/Pine on a Sunday morning, taking 271->Link vs. 550 all the way are nearly equivalent, in that you leave at the same and arrive at the same time. And, this is on a zero-traffic Sunday morning. When there is traffic, 271->Link will tend to perform better, assuming the Montlake exit ramp isn’t backed up.

        One of the culprits for the 550 is some ridiculous signal timing when the bus exits the freeway to 4th Ave. The bus seems to just sit there 2-3 minutes, waiting for that one light to change, while hardly any cars go by on 4th. Then, when the light finally changes, the next line turns red, and the bus has to wait another minute or so. There seems to be something wrong with the lights, where they always assume football traffic, whether there’s actually a football game or not.

    3. It is also worth mentioning that since about 2009, Sound Transit has made essentially zero investments in improving the frequency of their off-peak service. What limited service improvements they have made has been mostly at rush hour, and many of the new service improvements they’ve made have been sunk into more schedule padding for more rush hour traffic, so don’t even show up in the schedule.

      If you just look at the evening/weekend schedule now vs. 10 years ago, you can see very little improvement, and some routes have actually gotten worse. From my memory, here’s a route by route summary:

      512 – consolidated 510/511 into a more frequent corridor, but didn’t add any new service hours, and travel time to Everett got 10 minutes longer. 15-minute service is during Saturday daytime hours only.
      542 – previously did not exist, now every 30 minutes
      545 – was every 30 minutes, unchanged
      550 – gained 15-minute service during Saturday daytime hours. Still every 30 minutes evenings+all day Sunday.
      554 – was every 30 minutes, reduced to hourly on weekends before 10 AM and after 7 PM
      577/578 – was hourly Saturday/Sunday. Now, half-hourly Saturday, hourly Sunday.
      594 – was every 30 minutes, unchanged

      King County Metro invested in more service and got more riders to show for it. Sound Transit, on the other hand, didn’t and didn’t.

      1. ST is dedicating any extra money to Link construction to get it open faster. Metro doesn’t have anything comparable.

      2. The 554 was every 20 minutes, and was supposed to be every 15 minutes (something ST promised when they moved the route from serving Eastgate Station to the freeway flyer stop).

      3. asdf2, amen. ST could run Link more often at off-peak times; it would make transfers and network restructures easier. why wait? Yes, off-peak headway on routes 512, 535, 545, 550, 574, and 594 should be 15-minutes. in fall 2019, the evening span of Route 522 was reduced. the service budget is a relatively small portion of the total for ST.

  23. Glad to read that the Courthouse entrance to Pioneer Square Station has been secured and re-opened. The message being broadcast by its abandonment was worth whatever was spent to turn off. And keep it that way.

    However, I think that one change of focus in commentary on disorder and violence will go a long way to straighten out the politics necessary for a cure. Term “Homeless”now accurately applies to thousands of people whose most recent change of circumstance, read rent, has turned them into polite and presentable managers whose shaving mirror is now above their dashboard.

    For about a year, every single discussion beginning with the “H-word” has got to start with the hundred percent truth that The State of Washington no longer has a mental health care system. Only approaching piece of public medical jollity was recent announcement of a hundred simultaneous virus cases at a single local school. And we just had a hospital close all its emergency rooms too, didn’t we? Over a long-known infection. Suggestion: let’s start calling every germ a terrorist and at least the politics will start to come right for treatment.

    For today’s main theme of falling ridership, would be interesting to compare our own figures with other systems at this stage of construction and conversion over the course of History. The regional electric railroad we’re building is still mainly a bus line. Let’s hold off talking decline ’til rail hits at least Lynnwood and Bellevue.

    Mark Dublin

    1. It burns me up when people talk about how we just need to increase substance-abuse services and mental-health services and instill a sense of self-reliance/responsibility/morals will solve most of the homeless problem. Most of the homeless are not panhandlers sleeping on the sidewalk and spending all their money on drugs; they’re just people who can’t afford rent. Some of them have jobs but the low wages are not enough to cover housing. Or they can afford a monthly rent but they can’t afford the move-in fees on top of it. Or they have bad credit or a conviction or an eviction. An increasing number are whole families. Here’s a new slogan: “It’s the housing, stupid!” Give them housing they can afford and most of the problems would go away, including loitering in buses and parks. Then focus on those substance-abuse and mental-health services which are a small part of the problem.

      And note that tiny-house villages aren’t scalable. A village of 5,000 units would take a lot of space. They need to be in multistory buildings, at least some of them. We’ve already got a model for similar-sized housing: apodments. Build them, don’t ban them.

      Still, the total number of homeless are around 5.000. Maybe 10,000. But not 100,000. So it’s still a small fraction of the population.

      1. I don’t think we disagree that much on what needs to be done, Mike. We’ve all got our flash-points, that’s all. I don’t think it’s all that political for a transit system to protect both its passengers and its workers from personal physical violence on transit premises.

        All my working years, I have had trouble keeping my own temper in the presence of other workers who habitually impugn the character of people whose luck has been a fraction worse than theirs.

        Right now, though, my worst political fury is for professional politicians on my side who think public housing with actual rooms is too extreme to advocate. Honest, nobody ever told me we owe Yesler Terrace to the Khmer Rouge. Or how many brave Americans died fighting for the right to medical care that’s mostly co-payment notices.

        Since Franklin Roosevelt died a few months before I was born, most definitely an age thing. Which gives me an idea about what to call the Downtown LINK station containing the symphony hall.

        I’m pretty sure the DSTT entrance with the war memorial park has room for two bronze statues, and also some of the treatment described here:
        Careful, disability questions raised some problems for the designers.

        So there’s no locational confusion about the Roosevelt name, rename the facility, and the Link station, Franklin and Eleanor Park Memorial Park. “New Deal?” Well, monuments do turn Green, don’t they?

        OR: in a last-ditch attempt to heal our region’s partisan politics, let the Democratic Party donate same plaza to the Republican Party’s absolute best:

        An immigrant whose Right to Keep and Bear Arms I would have deeply supported.

        Mark Dublin

      2. Right on, Mike. 12 thousand and change was the last number I saw. Which is 10 times more than where I am, for a similar sized city. That thinks it has a severe homeless problem.


      3. Yep, “apodments” are very similar to a modern equivalent of low-end “residential hotels”. The city should embrace them.

      4. “Most of the homeless are not panhandlers sleeping on the sidewalk and spending all their money on drugs; they’re just people who can’t afford rent.”

        Assertions made without evidence may be dismissed without evidence. Conversely, the city’s own survey of homeless persons in 2016 ( obtained the following responses: over two-thirds of respondents (69%) said they were not from Seattle, a majority (51.1%) said Seattle was not the place they most recently became homeless, and a majority (54.7%) reported using alcohol or other drugs. In 2019, King County sued Purdue Pharma, Endo Pharmaceuticals and Janssen Pharmaceuticals, submitting documents claiming the majority of King County’s homeless population “…is addicted to or uses opioids.” (

        Our attempts to solve an addiction issue amongst recent, already-homeless arrivals as if it was actually a housing-affordability issue amongst long-time residents has — unsurprisingly — failed. We need a new approach, one which recognizes addiction for the disease it is, and treats victims of it without resorting to empty moral platitudes. Until then, get used to unpleasant rides on public transport, as even addicts want warm dry places to sit. (Who can blame them?)

      5. If it’s really just economics, being unable to afford the rents in the city, people will move further out and commute before sleeping under a freeway bridge.

        If you don’t have any money, and/or have mental problems preventing you from holding down a job, regulating the rent to $1000/mo. instead of $1,500/mo. isn’t going to make any difference.

  24. It’s surprising that most if the ridership increase ended in 2015, while U-Link was in 2016, Seattle’s TBD started delivering in 2017, and the booming economy has yielded additional runs.

    The most likely answer is the impacts to reliability due to many short-term factors. The I-90 impacts, 520 bridge replacement, Montlake area construction, expulsion of tunnel buses, building construction closing lanes, viaduct removal, 99 tunnel tolling, and on and on. It’s a general principle that ridership goes up and down in lockstep with the quality of the transit service. While Metro has added many off-peak runs, the unreliability of many routes has had the opposite effect.

    The restructures have been mixed. The first recent restructures were in 2012 to 2014 for RapidRide C, D, and E. The fact that that coincides with the biggest ridership increase suggests they were very successful. Those restructures did not just create RapidRide but also created routes like the 40, giving Fremont-Ballard its first frequent service and first service west of 8th Ave NW.

    After 2015, the U-Link restructure was good in many ways but it also made things harder for people on the Stevens Way routes, people near 12th-15th & Pine, people in Lake City going to the U-District, etc. Maybe it wasn’t overall as successful as the C, D, & E restructures?

    Other notable factors are the rise of Uber-like taxis, dockless bikeshares, walking to work, and the perception that bus stops and buses are increasingly unsafe.

    1. We can hope that you’re right, Mike, in predicting that when the “short-term factors” end, ridership will rebound. But it may be true that gentrification has changed the future of potential Seattle transit ridership, dooming the “all-day transit” Valhalla that everyone dreams of.

  25. Didn’t I just see a study that showed the more people are walking to work than ever before in Seattle. That I suspect is because many of those Apt complexes on Capital Hill have been completed and the tech workers have moved in. As a commute to S. Lake Union or downtown walking beats the bus. It’s good exercise for desk workers and at the distance involved it’s competitive on time.

  26. A bit speculative, but it may also be that people who live in Town at the lower income range where you mostly ride the bus because Uber/Lyft are too expensive might have less disposable income and therefore are taking fewer leisure trips? Housing costs of course went way up since 2015, which eats away at disposable income. Also some economists say we’re coming up on the next recession, which would be expected to reduce the total number of non-work trips taken via all modes (with driving somewhat offset by cheaper gas, but not indefinitely).

  27. The prices to ride a bus have become rediculous. We now pay the price of a gallon of gas each person to ride the bus one way! I used to take the bus on a daily basis to get to and from work. I can no longer afford to pay the rediculous prices! It’s nothing short of robbery. Why would i take a bus and give a bunch of theiving pirates the money i could be using to feed and cloth my children..

    1. I personally don’t pay attention to it, since I have an employer-funded pass for unlimited rides. But, it can become an issue when I have visiting family. For short trips, by the time you two adult fares, you may as well just ride Uber/Lyft because it’s only $1-2 more.

      One idea I’ve proposed to mitigate this is to tax Uber/Lyft to fund transit, and set the amount of the tax to match an adult bus fare, at least in the inner part of the city during reasonable hours. That way, at least when a single person makes the decision, the bus fare becomes immaterial since you’re paying it anyway. Based on some recent Uber/Lyft ridership stats published by the Seattle times, I ran some crude numbers to guess how much revenue such a tax could bring in. The conclusion was that it could, conservatively, Eyman-proof the city’s revenue stream by completely replacing the city’s $80 car tabs the city currently collects. Or, it could be in addition to the car tabs, allowing a lot of those cuts in the Move Seattle “reset” to be restored.

      1. We really have bifurcated STB readership/commenters into two camps – those who have employer-paid passes and for whom transit use is free, and those who have to pay to ride. $5.50 for two people, $11 round trip. Often, why not drive, especially if parking is free…and of course private cars can course through even congested areas for free. Why tax Uber and Lyft users, who at least aren’t parking a car, and let SOVs roam free? Uber and Lyft can be a complement to transit usage. What we should tax is employer paid parking, and if there is an Uber/Lyft fee it needs to be paid by SOVs too.

      2. Arguably, employer provided transit passes are no more “free” than employer provided health insurance contributions are. It’s part of the total compensation package. Like a tax, it’s something your employer pays for your employment but is money you never actually see as cash. Assuming there is little or no “out of pocket” expense on top of that tax, you’d be more inclined to use the service more often.

        Perhaps there’s a lesson in how mass transit is paid for compared with how health care is paid for.

      3. You don’t think that Terrible Tim will come up with an Initiative to ban TNC taxes? Oh ye of too much faith!

      4. I guess you could just ignore those of us who pay for our own passes and don’t drive unless it’s absolutely necessary. Or can’t even afford a car.

        Uber and Lyft are not realistic ways to complement transit. In theory they could be, but 95% of the time that’s not how they’re used. And they’re WAY out of reach for low income people on any sort of regular basis (I can’t even use them myself, I tried once and they locked me out until I can provide a second credit card, which I don’t have.) This is why I’m really prickly about them being suggested as ‘last mile’ solutions or of any real importance to improving transit in the region.

        Parking absolutely should be taxed though.

  28. I’ve started driving to work more from South Seattle mainly for two reasons:

    Time savings – It takes almost 3 times longer to take transit, longer if buses are late/MIA.

    Comfort level – Metro buses are incredibly uncomfortable (any bus on 4th Avenue South is horrid) and packed to the gills at peak hours. If Metro buses were even remotely as comfortable as Sound Transit’s, I’d be more inclined to ride them. My female roommate regularly gets harassed and considers it a lucky day if she can ride the bus in peace.

    Light rail is also packed at peak hours. It is frustrating to wait for a train leaving downtown only to not be able to board because it’s completely full. It doesn’t help that it goes to the airport either. People stow their luggage in the aisles, seats and handicapped areas, anywhere but under the seats where it should be.

    1. South-central Seattle has some of the worst Metro service if you’re not on the 124. I take the 131/132 to Costco and they are notoriously late: always 5-20 minutes late except after 8pm and weekends before 9:30am, and thrown off-schedule whenever there’s a ballgame. I’m surprised there hasn’t been a larger uproar that forced Metro to improve it. And the 3rd & Union stop is probably the worst one midtown. The two north of it (the C stop and the 36 stop) are not as bad.

      Brent takes the 60 to Beacon Hill Station, if that works for you at all.

      1. yes, the route 26-28-131-132 is awkward. yes, routes 36, 60, and 107 serve the Beacon Hill Link station.

  29. In our infinite wisdom, we are about to decimate off-peak ridership on another Metro route, Route 255. It’s may be 15-18 minutes from S. Kirkland P&R to/from 5th & Pine off-peak. We’re going to turn that into a 35 minute trip when we truncate it at the UW. We are going to add the unpredictability of Montlake Bridge openings and Husky stadium events. Are we going to be surprised when people vote with their feet and abandon the route 255?

    We took what was a good transit facility at Montlake and closed it without replacement while letting the state DOT widen the freeway footprint. We make so many decisions in favor of cars and without valuing transit function. The entire design of Montlake and UW/Husky does not make for a viable and efficient transfer. And then crazy, we keep the peak-only express routes going downtown and want to send the weekend and night riders to UW. Will anyone be willing to reconsider when the ridership evaporates? Next move will reduce frequency to match ridership, the death spiral.

    1. Not everybody is going to downtown. Some people are going to North Seattle or to places on Link, or taking Link through downtown and transferring to a bus south of it. All those people are ignored in the current routing.

      1. ST 540 started out as a 7 day/week all-day route. There simply was not the ridership to support it outside of peak (and even the peak ridership is fairly weak) so it was progressively cut, first to 5 day/week and then to peak only. Some 255 riders will remain, for whom the transfer works well, but for a good number of riders it is way less convenient, and I expect them to either switch to driving or find other service. The 542 vs 545 also provides some empirical market data. The majority of riders are continuing to choose 545 vs. 542/Link. Weekends, too, when service levels are the same.

    2. the March 2019 Route 255 has already led to it attracting fewer rides. it is slower and less reliable, especially at off-peak times. The orientation with Link may help. It will not have to sit on the I-5 general purpose lanes or in downtown traffic.

      1. Carl,
        yes, Route 540 was restructured in February 2008 with the new Redmond TC. ridership had been quite low. Route 545 frequency was improved and its ridership took off. since then, Route 545 ridership has responded well to all service increases. new Route 248 mitigated the arterial tail of Route 540. the network was much better. the next logical step is to truncate Route 545 at UW Link and make it more frequent still.

      2. The 255 routing through downtown is horrid and has certainly hurt ridership. Not only is it needlessly slow, the stops are less convenient to many destinations. The 255 should serve the same stops downtown as the route 545. Cynically, I wonder if this was done in order to lower ridership so that the drop-off when the 255 is truncated doesn’t look as bad.

      3. The 545’s routing through downtown is not particularly fast either. If anything, I think the 255’s routing is slightly faster, at least on weekends (albeit, at the cost of worse transfers to/from Link and the 3rd Ave. bus corridor). The 4th Ave. bus lanes revert to parking on weekends. And the section from 4th/Pine to 8th/Olive is particularly slow.

        As to references to the 540, I don’t think it’s a valid comparison. The 255 has consistently run much more frequently than the 540, and 540 is notoriously unreliable, so choosing the 255 – even for those going to the UW – tends to result in much less waiting at the bus stop. The 540’s schedule was also set before Link to UW even existed, and it was assumed that it wouldn’t absorb any of the Kirkland->downtown market. Even with Link to UW, the 540’s poor frequency still makes it relatively unattractive for many compared with the 255 where you can just show up at the stop and go, without needing to pay attention to a schedule or worry too much about the punctuality of any one particular bus.

        The 540 also ends at Kirkland Transit Center and doesn’t serve Juanita and Totem Lake like the 255 does.

    3. “We” did not close the “good transit facility at Montlake”. “We” is the Washington State Department of Transportation, not Seattle Transit Blog readership. You should have used the egregious, but in this case absolutely correct “They”.

      1. True. But, the long-term transit network doesn’t really need the Montlake Freeway Station to begin with because the long-term future is to send buses to Husky Stadium, not to fight traffic on I-5 and the streets of downtown. They’re starting with the 255. By the time East Link opens and all the freeway construction is finished, I’m fully expecting the same to happen with the 545 and most of the other routes (maybe a few peak expresses to SLU/First Hill remaining, but not a lot).

        The freeway station also had some big problems, in particular, the eastbound stop not being wheelchair accessible.

        I can understand the argument that the Montlake Lid is supposed to be designed for the needs of the long-term, not the short term, so the freeway station shouldn’t constrain the design. Even so, you can see that there is clearly no room for the freeway station during construction.

    4. I’ve experimented with going downtown both ways – direct bus and bus to Link. In practice, during off-peak hours (no backup on either I-5 or Montlake exit ramp), the difference in travel time works out to be the wait for the Link train. That is, if the train comes right away, it’s a wash. If you have to wait 5 minute for the train, switching over takes an extra 5 minutes. Barring very unusual traffic situations, like a Husky football game, the extra time won’t be anywhere near 20 minutes.

      And, of course, sometimes the truncation will make you come out ahead. Particularly, when I-5 is backed up or the downtown streets are backed up. Just last weekend, I boarded a 255 that spent 15 minutes crawling in traffic down 6th Ave. So, even with I-5 wide open, riding Link to Husky Stadium and catching the 255 bus there still would have been faster.

      The truncation also funds a big frequency boost, so the extra 5 minutes in the switch, you get back by not having to wait as long at the bus stop. Not only do we get 15 minute service instead of 30 minute service, but on Saturday/Sunday evenings, the pre-upgrade baseline is hourly service. The better frequency should boost ridership, in and of itself.

      As to the peak expresses like the 252 and 257, I do expect to see those truncated too, sometime in the next few years. Metro is probably trying to not do everything all at once, just in case things don’t go smoothly and they have to adjust. In practice, I think it will be fine. And, when the construction on 520 finishes and Link’s off-peak frequency between Husky Stadium and downtown improves from 10 minutes to 5, it’s going to get even better.

      1. And, when the construction on 520 finishes and Link’s off-peak frequency between Husky Stadium and downtown improves from 10 minutes to 5, it’s going to get even better.

        That will make all the difference in the world. When the 520 work is done, there will be an HOV lane from 520 right to the edge of the Montlake Bridge. During rush hour, the bus will skirt by the traffic on the ramps. Outside of rush hour, even if the bridge is up, a bus won’t get caught in a big lineup after the bridge goes down. Bridges really don’t spend that much time going up and down — it is the traffic built up behind them that is miserable. Buses will avoid all that.

        There is no way that transfer can get timed, so going from ten minute frequency to five is also a huge change. Then there is the fact that the transfer opens up new trips. There aren’t as many people going between Kirkland/Redmond and Northgate/Roosevelt, but there are some. I could even see some of the very costly one way routes that deadhead becoming bidirectional.

      2. I’m thinking about things like attending a show at the Paramount. Go to 8th & Olive, hop on the 255 and you are at S. Kirkland in under 20 minutes. It’s implausible for it to be faster to walk to Westlake, down three flights, wait for a train, ride 8 minutes, rise three long escalators, wait for a traffic light to cross Montlake, wait for a bus…

        And I’m still wondering what the service pattern is going to be during Husky games, opening day of boating season, etc, Marathons. A logical thing would be to make a transfer at either Westlake station or Cap Hill, but based on how Metro handles these things today, it’s entirely plausible they will reroute to 45th St which can’t be an improvement. With the exception of stadium attendance, there isn’t much demand for the U area on weekends.

      3. Evenings, the 255 is quite fast from 8th and Olive if you’re lucky and don’t have to wait for it. But, what if you’re not? Evenings, the route only runs every 30 minutes, and if you’re show is on a Saturday night, it will be running just once per hour by the time it’s over. Then, there’s the fact that the bus is coming from the other end of downtown, so, even if you’re not riding that section, any traffic delays there will still delay you. If you assume a wait time of 15-20 minutes, you’re mostly looking at a wash compared to walking to Westlake and riding Link to UW Station.

  30. Link ridership during commute hours can’t grow because it is at capacity. I have less experience with bus commuting, but that seems to be the case for them as well. Trains are literally packed tight to the point that people are left on the platform (platforms which are generally too small for real lines to form, meaning that the less aggressive folks will continue to miss trains over and over despite waiting longer than others).

    This has been a complaint of mine for a long time with ST. The existing station areas aren’t fully developed yet, and we are also further expanding the lines to the north and south, and yet we are already at capacity. This system was not built for the future. It isn’t even adequate for now. As far as I can tell it hasn’t been designed to accommodate express trains, which could help alleviate this issue. There is no way to expand service during commute hours other than adding one more car to each train (this capacity will fill up very, very quickly).

    By the way, with enough switches and computer controlled trains ST could have express trains that switch to the opposite track to bypass other trains at stations, but, especially with the at-grade crossings in Rainier Valley, I don’t see this happening.

    1. The current capacity crunch is because ST underordered trains for ST1 and U-Link; it decided to wait until Lynnwood Link and East Link. More trains have now arrived, and Northgate Link will have 4-car trains. Lynnwood Link will have two lines alternating between Northgate and Intl Dist, or 3-5 minute frequency. There’s a debate whether Link between U-District and Westlake will reach capacity during Lynnwood Link’s planning horizon (2040).

Comments are closed.