March2015dailyboardings

It looks like Link’s ridership growth is pulling out of last winter’s slowdown. Throw out last February’s Superbowl parade and so far this year Link is growing at around 9%. Last year’s ridership gains were so high (high teens!), I don’t think it possible to match them this year, but a return to low double digits (what it was averaging before last year) would be nice. It would also be enough to comfortably get us back to pre-launch (and pre-recession) estimates.

March’s Link Weekday/Saturday/Sunday average boardings were 32,893 / 21,898 / 17,452, growth of 9.9%, -2.2%, and 3.2% respectively over March 2014. Sounder’s weekday boardings were up 10.4% with ridership increasing on both lines. Tacoma Link’s weekday ridership increased 6.9%. Weekday ST Express ridership was up 3.0%. System wide weekday boardings were up 5.9%, and all boardings were up 8.0%. The complete March Ridership Summary is here.

Sounder is doing great, 25% growth in Jan, 10% in both Feb and March. ST Express is marching along. How much more can it grow before overcrowding becomes an issue? Tacoma Link seems to have finally turned around. New employer down there?

My charts below the fold.

March15WeekdayRidershipMarch15WeekendRidership MAR15WeekdayChange MAR15WeekdayMovingAVGMAR15WeekdayMovingAvgChange

74 Replies to “March 2015 Sound Transit Ridership Report – Taking off again?”

  3. All the plots have a different scale……. But another good report.

    When U-Link opens it will blow all this out of the water.

    1. I agree. I think we’ll see another big jump with Northgate Link (another stop at the UW + Roosevelt + Northgate).

    2. Yep, ASU was huge for Phoenix’s numbers. The North section is going to be one busy stub. I’m curious what the University Ave is going to look like without the 70 series buses using this route? Will there be a lot of complaints?

      1. It’s not clear what Metro will end up doing. But yes, there will be some complaints.

      2. There will be complaints no matter what Metro does. So Metro may as well do whatever serves the most riders most efficiently.

    3. I hope this isn’t drifting off-topic, but will the tunnel bus routes that are scheduled to be eliminated, truncated, or moved upstairs next March have to wait until March to leave the tunnel, or will they leave the tunnel the day U-Link opens for revenue service?

      1. I believe the U-Link restructure will happen at the same time University Link opens, not on some other seasonal-change date.

    4. Thanks. As to the different scales, do you have any suggestions? I pretty much just made these all up as I went along. Taught myself excel along the way. I think we can all agree they’ve improved a bit from my first ridership post (and first STB guest post!):
      https://seattletransitblog.wpcomstaging.com/2012/10/26/september-october-2012-link-ridership/

      And hey, here is some weekend nostalgia, here is the first year of ridership:
      https://seattletransitblog.wpcomstaging.com/2011/01/31/yearly-ridership-trend/

      1. I was mainly referring to the horizontal axis on all bit the first plot. The y axis is ok with different scales. But the x should look the same.

        Sorry. I’ve been chastised
        More than once for the same thing. I’m a bit sensitive now to how performance data is presented.

      2. Matt,

        The Visual Display of Quantitative Information is critical in not only making sure your numbers are believable, and enhancing and preserving your reputation, but it is important in critical analysis.

        Surely someone has the book down there, and definitely DO NOT go to the Marketing folks for assistance ! ;-)

      3. Lazarus,
        No worries, wasn’t trying to chastise. But I can explain a little bit.

        So the first two are the same b/c they are pretty straight forward. Both cover every year since opening up to the current year.

        The next chart is year over year growth rate. By definition that chart can’t start until a year after opening. However, even then I chop off the first couple of months b/c the Airport opened 6 months after the main line and overall the first few months saw HUGE opening growth. So much that it would distort the chart. If I were to start in July 10 (first month of YOY data) the y axis would have to go from 0-24% to 0-80%! All of what I consider the good bits (the sustained growth after initial surge) would just be little squiggles at the bottom of the chart. And the trend line would be totally useless. See here:
        http://stb-wp.s3.amazonaws.com//srv/htdocs/wp-content/uploads/2015/05/example.jpg

        After that is the twelve month moving average. Basically the last twelve months are averaged. The idea is to level out the seasonal highs and lows. You can’t start that chart until you have twelve months of data to get your first data point. Therefor the chart starts twelve months after opening.

        Last is another growth rate chart. My friend Joshua called it the ‘3rd Derivative’ and since he’s a Boeing Engineer and my only C in college was Cal II, I’ll take his word for it. It’s basically the year over year growth rate of the twelve month moving average. Like the other growth rate I chop off the first bit so the rest of the chart is visible.

        Hope that makes sense. It’d probably be better if I actually knew the math behind what I’m doing. I really just stumbled into all this thinking it through in my head and trying to get excel to show it.

      4. Try Tableau Public for charts like this, but interactive, that you can embed in any blog post. Can also help with getting the axes consistent.

        https://public.tableau.com/s/

        (disclosure: I work at Tableau, but this is our free product)

      5. Jim, thanks for the recommendation, but it looks like it isn’t available on Kindle. :/

      6. Well, trundle on over to the Seattle Library. (it’s “available in some locations”)

        It’s well worth the read.

  4. Are there any reports on the individual stations for Link? I am curious as to where the increase is coming from.

    1. If they ever want people to take Sounder serious they need to better integrate LINK/Express/Local connections to it, offer more rts and faster service. Otherwise they should let it die off. Cost per Boarding $14.13 is way to high to justify.

      1. If I were a rider on one of those routes, I would not want to be re-routed onto North Sounder unless and until it becomes much more reliable. There are still too many mudslides to force transfers.

      3. Community Transit really should advertise some of its specific trips to Everett Station that arrive before a Sounder run leaves as “Sounder Connectors” with appropriate schedule branding.

        In the near future, once Lynnwood Link is completed and all of the commuter routes are truncated there, it wouldn’t be a bad idea to shift some of the far-flung routes (421, 422, 424, 425) to feed into Sounder at Everett Station and Link at Lynnwood TC.

      4. Absolute agreement, SounderBruce :-). The current Community Transit marketing strategy gets an F from me. Yours gets a B, maybe a B+.

      5. It is better to just stay on the freeway to Lynnwood Transit Center and feed into Link for several reasons:
        – Lynnwood TC is much easier to reach from I-5 than Everett Station. Lynnwood, you’re got a direct ramp between the HOV lane and the P&R. Everett, you’ve got to get off the freeway at a regular exit and wait at a bunch of stoplights.
        – Sounder operates on every 30-minutes, and to maintain a reliable connection for the morning commute, you would need to allow a considerable amount of pad time. Enough so when the freeway isn’t congested, the average wait for Sounder would be as much as 15-minutes, which, during rush hour, would be much more than even the worst-case wait time for Link.
        – Sounder itself is unreliable and gets canceled whenever a mudslide hits. To truncate a Marysville bus at Everett, you would need to provide an answer for how everyone is supposed to get downtown when the train is canceled. Most likely, this would be to take another bus to Lynnwood and ride Link, in which case you may as well just take the busload of people to Lynnwood directly.
        – Sounder has just one stop downtown, at King Street Station. Link stops all over downtown, so if you’re actual destination is the north part of downtown, Link is faster, even on a good day for Sounder. And, of course, if you’re trying to get to the U-district, Roosvelt, or Northgate, Link gets you where you need to go, while Sounder is completely worthless.

      6. ASDF2, to be brutally honest with you, I think if there’s a way to get rid of Sounder North – do it. Short of that, let’s try to make the damn contraption work!!!

      7. I’m not saying we should truncate the buses at Everett, but simply have it as an extra stop before Lynnwood (just as Lynnwood is for these routes on their way to Seattle). The only routes that should terminate at Everett Station are the all-day local routes that already do so (except for the 201/202, which really should be split between its Ash Way section and its Marysville section).

      8. Today’s routes 201 and 202 do stop at Everett on the way to Lynnwood. It’s appropriate for off-peak service when demand is too limited to justify two separate routes to Marysville, but serving Everett on the way to Lynnwood really is a significant time sink. There’s a reason why the downtown Seattle->Marysville buses today do not stop in Everett.

      9. When we have Link going to certain stations, both north and south, it makes sense to terminate buses at stations even if it means the buses go a little further or overlap with other buses. That’s what a HCT-based network looks like: everything goes to the HCT. The reason is to minimize three-seat rides. And if they transfer again in Seattle, it becomes a four-seat ride. The ideal trip is “(bus+)train(+bus)” or, if we had more subway lines, “(bus+)train+train(+bus)”. Not “bus+bus+train(+bus)”. That was the basis of Aleks’ plan for south King County. Bus segments were shuffled to give every major bus a one-seat ride to a Link station. No more transferring at Kent Station or Renton for the major corridors: the buses continue through the TC to the Link station.

    2. Nice post. I notice it doesn’t say anything about cost, or compare its ridership with other lines.

      1. Thanks Brent, well I decided to keep it simple. I do at some point want to talk about cost again…

    3. There are two ways to look at Sounder South. ST2 will make it almost hourly weekdays, with 1.5-2 hour gaps before and after noon. Is Sounder going to develop into an all-day hourly service with maybe some evening and weekend? What about half-hourly? That would address the HCT problem in southeast King County. If we’re heading in that direction, then it would make sense to fully integrate it into the all-day transit network. But not right now; wait until all the ST2 runs have started so that it’s truly a daytime service rather than just a peak service.

      The other way to look at Sounder is if it doesn’t expand. Then it doesn’t make sense to spend significant resources getting other transit to it. Kent, Auburn, Puyallup, Tacoma Dome, and King Street/intl Dist are already transit hubs the other routes go. Sumner has the Bonney Lake shuttle and voted down all local transit (by seceding from Pierce Transit). That really leaves only Tukwila. Tukwila Station now has a frequent route due to the RapidRide F detour, which is as idiotic as routing RapidRide C to the Vashon ferry. So 90% of the RapidRide runs make the detour even though there’s no train there. Better to put an infrequent bus timed with the train there, and let the frequent bus go where the all-day ridership is.

      Sounder North is not going to be expanded, period. It has four peak-direction runs. ST2 did not expand it, and no ST3 project does. The only thing ST is doing now is improving some stations and surface parking. Sounder North has low ridership, misses the population centers, has barely a walkshed, is frequently cancelled due to mudslides, and is on a crowded single track that won’t be expanded because it would be extraordinarily expensive to cut into the hillside and destablize the not-quite-million-dollar homes on top. Everett Station is a transit hub and may become more so depending on how ST3 plays out. Mukilteo’s downtown is tiny and anti-growth. Edmonds’ downtown has some potential but it’s isolated. We should really concentrate on connecting Edmonds and Mukilteo better to the existing trunk transit on 99 and I-5.

      1. Mike, check your info. “frequently cancelled?” I recall one, maybe two instances this season… Am I wrong?
        The BNSF line between Seattle and Everett is double tracked except for a small section in Edmonds that is graded and ready to lay down track anytime BNSF wants to. Signals/gates are an issue because Edmonds wants to have a whistle ban, but currently can’t pay for the full enclosure.
        Downtown Edmonds is a better walkshed than Downtown Lynnwood. Connective transit and ferries create a walkshed on the other side of the Sound in Kingston and Clinton as well. The Sounder connection to the Ferry system includes Bremerton connections at Colman Dock and could be more important for regional connectivity with through routing. Without Sounder, peak trips from the ferries to our Regional CBD will involve an additional bus transfer. As you can see from Joe’s ridership numbers, the mudslides are a marginal issue for ridership. The true problem that Sounder North has is the lack of origins and destinations. Pulling riders from 3 suburban stations to one CBD station 4 times AM and returning them 4 times PM is simply a very poor service design. There are other potential stations that have been discussed, including one on the west side of Everett’s downtown. That could be opened within a year with minimal improvements and was actually in the original Sounder North plan. The others would require study and funds for a simple platform. Through routing trains would open new travel markets that will not be served by Link, which would attract more trips to the service and it would make better use of existing equipment. Yes, there are issues around adding density and preserving views (obviously no one here has been in downtown Edmonds to see the crane lately. It’s not going to be Ballard but we’re building more units.) And yes, transit supporters live there, and yes we take the train, and the bus. I would take the train to Everett sometimes if it ran reverse trips, I would take it to Ballard if there was a stop, I would take it downtown (someday I might work there and take it everyday, until then, meetings and gamedays) If through routed I would take it to a variety of South KC locations, and my friends down there could use it to come up and visit me.

        Edmonds has had passenger rail service since 1893. It is not going away. Spend your effort on improving the world’s most beautiful commute, not hoping for its’ end.

    4. Forgot to say, if Sounder isn’t expanded, then we have to start with, who wants ride the infrequent runs from where and to where? Who wants to ride Sounder’s current runs who can’t do so now, and which specific other routes or directions do they want to transfer to? Because if you lay out the cases for Sounder South and Sounder North, there aren’t many. Rainier Valley to southeast King County is the most probable. But I don’t think that’s worth a $2 million BAR station and another Sounder station. What it is worth is a frequent bus from Rainier Beach Station to Renton and Kent.

      Where else do people in southeast King County want to go? If they’re going north of downtown, they should stay on Sounder to downtown, it’s much faster than transferring to Link at BAR. If they’re going to the airport, they should take an east-west bus to the airport. If they’re in Rainier Valley going the airport, they’ve got a one-seat ride on Link. Where else do people want to go that a transfer to Sounder would service? Especially those that can’t be acommodated by the existing Kent Station, Auburn Station, and RapidRide F?

      And on Sounder North, it only works if you live right near downtown Edmonds or Mukilteo or drive to the parking lots. If you live anywhere further east, it does not make sense to take a bus west to those stations. That’s no faster than taking the I-5 buses or future Lynnwood Link, and probably slower.

      2. One would hope that for all the support Edmonds and Mukilteo put behind the Sounder service, that they would be looking to enhance TOD in the station areas.

        Edmonds is constantly in the ‘building height’ (don’t screw up my million dollar view) restriction battle, particularly for the Edmonds bowl.

        Don’t know if Mukilteo has the same issues with their citizenry/council.

      3. Jim,
        I understand Mukilteo is redeveloping their waterfront. They “get it” Sounder North is a target for elimination – and many of us here at STB wouldn’t shed any tears…
        Joe

      5. Of course, how many from Edmonds or Mukilteo actually ride Sounder North, compared to how many ride buses?

      6. Of course, it’s the voters that will decide, not the commenters on the Seattle Transit Blog.

        Campaign for eliminating Sounder North and adding more buses in ST3, and we would all find out!

  7. I am guessing that LINK use will explode once Angle Lake opens.

    Also Sounder has the Sunday Mariners games runs that started last month.

    1. “I am guessing that LINK use will explode once Angle Lake opens.”

      Are you serious?

      Ridership will get an uptick, and every other efficiency metric will go down.

      1. 1. Look at the parking shortage at Tukwila.

        2. Look at the desire of Kent Valley riders to get into Seattle — all day long, nights and weekends

        3. Look at the potential for offboarding regional bus riders (150, 180, 168) to LINK from Kent Station and from Federal Way for that matter.

      2. John — ST is building a 1000 stall parking lot for $50M. The cost per rider of this garage is about the same as building the actual rail line.

        That said — 1000 riders/day (assuming these aren’t the same drivers using Tukwila now) or so plus whatever transit adds will be a short term, very minor, pop in ridership. Additions of TOD (both residential and jobs) will have a more significant impact, but will take years (or even decades) to develop.

        U-Link, on the other hand, will be a 100% increase in ridership for the entire system the day it opens. I suspect it will perform even better than expectations. ST needs to get serious about adding serious bike facilities.

      3. I actually agree with John on this one. We will see a pretty big jump. But nothing like when U-Link is added. The first is like when your band’s song is played at the local radio station. The second is like when your song is played on American Bandstand (yeah, I’m old).

      4. I agree that ST needs to get serious about bike facilities at Husky Stadium Station. Apparently after negotiations with UW (who owns the land), ST was only permitted to install the little bike racks. No bike lockers. No bike cages. And this is the only station on the Burke gilman route. I suspect there are going to be a lot of bikes boarding trains due to insufficient secure bicycle parking.

      5. *sigh* I knew this day would come. Next STB meetup we’ll have to have a cage match fight over the name. :p

        Maybe you’ve got a middle name you can go by? Hopefully it’s not James (my parents obviously weren’t that imaginative).

        True story, there was another Matthew James Johnson on my school bus back in the day!

      6. How about Matthew we give you a callsign… say Matthew “StatsRider” Johnson?

      7. @Keith/Kyle “1000 riders/day ” — You don’t understand my argument.

        I argue that Angle Lake is a completely different class of station. The kind of Regional Transit Rail station (like Kent Station) that people wanted from the get go.

        One that is attuned to the real needs of a regional system.

        It is distantly located in the cheaper land areas that make car culture possible.

        At the same time people there occasionally want to get to the centralized resources.

        I pointed to the full parking lots of Tukwila.

        But Tukwila is traffic bound, unlike Angle Lake.

        So it’s not just 1000 rush hour riders a day, it could be more per day, and per night and per weekend day and night.

        On the other hand, I think you are overestimating the effect of U-District. The problem there is its the opposite of Angle Lake. You have to get there to use it and getting there is extremely hard. Navigating traffic to the U-District versus jumping in your car (at night say) and parking downtown is going to be six of one, half dozen of the other.

        However, as a destination…from places like Angle Lake…so that students can have apartments, and attend classes in the expensive U-District. Well, again, that is the effect of the Angle Lake’s
        .

      8. The problem is that the UW – a public institution from which Sound Transit cannot take land via eminent domain – only cares about U-link with respect to its own students from the south getting to class. It does not care at all about people from the north taking the train to downtown.

      9. Ever since going to the metro NE Seattle Link connection meetings and finding no good bus to link connections proposed at husky stadium station I’ve been picturing hundreds of people riding down the trail on opening day and bikes chained to railings, signs, trees the husky dog statue… “cyclepocalypse”

      10. @John
        Care to make a bet? UW – Capitol Hill – Downtown will result in a much bigger junp because the people are already there. No need to drive. Plus you have major bus feeder action there.

        As I said, Angle Lake will be big (for the reasons you mentioned) just not that big.

      11. John – I understand your argument, I just don’t agree with it. Even if people are that much more likely to drive to Angle lake than existing stations, there are 1000 parking spots there (you tell me on the street parking situation.)

        It can also be a decent spot for feeder transit – but this is somewhat limited in the area (correct me if I’m wrong.)

        Without people who either live or work near the station there is a hard cap on how many people will use the station.

        Your U-Link comment is pretty out there. An enourmous number of people live, work, or go to school in walking distance to Cap Hill and UW station. Add to that the enourmous number of people who will transfer from transit and bike there along the Burke and you have a blockbuster peice if infrastructure.

        It is my opinion U-Link will blow its ridership estimates out of the water if ST gets serious about bike facilities and Metro goes ahead with its re-routes.

      12. Sorry to correct some of the math, but 1000 filled parking spaces means at the very least 2000 boarding riders a day (a driver has to get back their car). Assuming some vehicles with two passengers and some parking turnover, at it should be argued that we’re probably talking about 2300 – 2600 riders per day from 1000 parking spaces at a station.

        The other math that needs to be discussed is that U-Link is costing $1.9 billion. If 1000 spaces is $50M, that means that U-Link cost is the same as 38 of this parking garage. Assuming that they are all inducing just 2000 riders a day, that would be 76,000 more riders — kind of in the ballpark of the 71,000 riders forecasted by U-Link. Sure there are reasons to not overbuild park-and-ride, but declaring the garage as a wasted investment is not one of them when you look at the math.

      13. Al — True. It would max out at about 2200 riders per day and you helped make my point. U-Link will cost about the same per rider…. and you having even counted the cost of building the rail to angle lake yet — just the cost of building the garage. U-link will see ridership growth as the city grows. That 2200 riders/day will never be higher than the day it opens. And unlike the apartment building that could have been built for $50M it will be nothing but an expense for ST for its entire lifespan.

    2. The Angle Lake P&R may get well used until KDM Station opens, if TIB P&R is full. As I’ve said before, there are people who drive from Auburn and Federal Way to TIB Station to go to Seattle, at least off-peak when Sounder and the peak expresses aren’t running. And that’s all of Seattle, not just downtown. For instance, I attend a Vegetarians of Washington dinner at the Mt Baker Club some Wednesday evenings, and met an Auburn couple who does that.

      However, Metro has not made any noise to acknowledge Bailo’s dream of a Kent route to Angle Lake Station. So you’ll have to drive to Angle Lake, or take the 180 to SeaTac. One way that would not be worth it is the slow 166 to Des Moines and RapidRide A to Angle Lake: that would probably take twice as long as the 180.

    3. More on Angle Lake. Why is taking a Kent bus to Angle Lake better than taking the 180 to SeaTac? The 180 is already as fast as an express could be, since northwest Kent and Orillia Road have few on/offs and are wide and uncongested. Going to SeaTac or KDM is a pretty straight shot, but Angle Lake is hard for a bus to get to. The people driving to Angle Lake or transferring to a bus there are completely separate transit markets, so there’s no reason that both uses have to be at the same station. There’s also the large number of people going to the airport: it seems better to combine airport travelers and Link transferers at SeaTac station rather than park-n-riders and Link transferers at Angle Lake Station. It’s a very short distance: 24 blocks or just over a mile, so it’s not closer to Kent in any practical sense, especially since buses would have to make a longer C shape to get to it. A bridge across the freeway would help, but you’d have to get WSDOT do do that and it has other priorities.

      It seems like making the 180 15-minute frequent full time would address all your concerns. Or, extend the 168 to SeaTac halfway between the 180s. That would still be slower than the 150 though!!! The 150 is 45 minutes to Kent off-off peak, 60 minutes to Kent daytime. Link is 37 minutes to SeaTac, the 180 is 20 minutes to Kent, plus the transfer. That means the optimal transfer is equal to the slowest 150 run. Angle Lake Station would not be faster to Kent; it would be slower. So it seems like everybody who’s willing to take the long way around (Link+180) is already doing it, and a 15-minute 180 would bring in the rest, but it still wouldn’t be as convenient as taking the 150. So maybe the 150 should become 15-minute full time (or at least until late evening).

      1. The only way Angle Lake Station beats the bus for someone coming from Kent is if you are driving to Angle Lake (meaning no walk to the bus stop or waiting for a connecting bus).

      2. “The people driving to Angle Lake or transferring to a bus there are completely separate transit markets, so there’s no reason that both uses have to be at the same station”

        Illustration: If your car is at the station, you don’t need a bus. If you left your car at home, then it doesn’t matter whether you transfer at the P&R station or another station, it’s one half dozen to the other.

  8. Ridership is quite seasonal, isn’t it?

    * These nice steady growth charts for Link will all be blown out of the water in early 2016 when U-Link opens.

    * Sounder North and Sounder South really need to be displayed separately.

  9. Pingback: North by Northwest Big Interview 01: Mukilteo Mayor Jennifer Gregerson

  10. There have been several comments about the lack of sufficient bike facilities at Husky Stadium for Link. In my mind, that station may resemble the ones I would see in Holland. I personally, will switch from going all the way downtown on my bike to just riding down to the station and locking up for the day. My Dutch friends did this a normal part of life.

    Are there estimates as to how many bikes husky Station will hold?

    1. If memory serves from passing by there recently, they have installed about 15-20 bike racks underneath the new pedestrian bridge that connects the station with the Rainier Vista triangle. Each of the bike racks probably has capacity for up to 10 or so bikes. Since UW is so hostile to bikes on the surface, maybe Sound Transit could carve out some space within the station.

      – The other Matt Johnson (cage fight not necessary, my middle initial is D !)

    2. I’m likely to ride to Husky Stadium as well rather than suffer the 11 (so long as I can count on space; nothing like going to work and having to backtrack to catch a bus you were trying to avoid in the first place!). If so, riding the two miles and catching the train is likely to be faster than taking the 11 and walking a few blocks. We’ll have to see how that works in practice however (and riding in the Arboretum sucks…now there’s a place for some bike infrastructure from Madison to Montlake!)

Comments are closed.