Avgeek Joe/Flickr

4Q 2012

It’s old news given the December report ST already released, but the 4th quarter report is now out, and it has new information beyond the number of riders per mode.

ST Express increases are inflated by the change in counting downtown rides. Nevertheless, the 550, 554, 555/6, and 577/8 all increased over 20%. The 550 also did it from a huge base of 505,000 boardings, far and away highest in the system. The 555/6 is much smaller, but wasn’t affected by the downtown change and had the largest increase at 31%.

The loss of the ride free area had surprisingly little impact on reliability. ST Express on-time performance slipped from 88.7 to 85.6% from the same time the previous year; Central Link (which uses a tighter standard for reliability) went from 90.8% to 90.2% over the same period.

The other notable event was the drop in scheduled Sounder trips that operated. Largely thanks to a record-shattering mudslide season, ST operated 92.8% of its trips, down from 99.3% (including both lines).

ST Express and Central Link continued their back and forth battle to have the cheapest operating cost per boarding (setting aside Tacoma Link). As Central Link’s ridership peaks in the summer, ST Express was once again cheaper in the 4th quarter, but for the year as a whole Central Link wins 2012 ($6.07 to $6.51).

January 2013

ST also released its January Ridership Summary. Central Link’s Weekday/Saturday/Sunday average was 24,938/16,025/12,718, a 12.6% increase on weekdays over January 2012. Mudslide-infested North Sounder lost 28% of its total boardings, while South Sounder increased an astounding 35%, partially due to the opening of Lakewood Station. 11,142 riders boarded a Sounder train on an average weekday.

ST Express boardings for the month were up over 18%, partially a result of counting former ride free area trips.

51 Replies to “ST 4Q 2012 and January 2013 Ridership Report”

  1. What exactly changed with regards to counting downtown rides? Did they not count them at all before? And if so why?

  2. Seems like that 35% increase in South Sounder would have to include more than just Lakewood to account for it as that represents 53,000 boardings in a month.

    Also “South Line…represents 90% of ridership”. Interesting…

    1. South Sounder ridership being nine times as large as North Sounder ridership is nothing new.

      But if you want more Sounder trips, we need seats filled. That means get Metro to end the 158/159 duplicate-head for when Sounder moves to 20-minute peak-of-peak headway this fall (assuming that ends up happening), and get the Kent feeder routes timed to meet the trains.

      A useful dataset here would be empty seat ratio for each run of Sounder. Is it actually approaching capacity on any of its runs? Are all 592 bus trips being charged a bargain basement $3.50 undercutting Sounder?

      1. One thing about the express buses is that they start in the north part of downtown (Belltown) and pick up passengers all they way through until they get on the highway. Then after stopping at Kent Station they continue to multiple destinations up East Hill.

        Even if Sounder is faster, there is something about taking one bus from near your work, grabbing a seat and taking it as close as possible to your home. It’s as close to, or even better than, driving. However, taking a bus to a train to a bus with two wait times…not as desirable to many.

      2. How much local all-day service is Kent willing to forego to keep these expensive (albeit charged less than Sounder, for some reason) one-seat rides?

      3. Last time I rode into Seattle on the sounder (which was a few years ago) It was standing room only as we entered the city.
        Now there’s like 3000-4000 more people per day riding it, and I don’t think more than one or two more trips, so, probably still standing room only.

      4. Whoa Pardner! Before you gleefully ax the 158/159 let’s disclose where those riders are getting off in the Seattle CBD. (hint: the vast majority are north of the KSS walkshed)
        Now, factor in average wait times for both transfers (Kent & KSS) and give me the time penalty for killing the one seat ride.
        Next, look up the net subsidy of a Sounder ride and the same 158 ride. (no hints this time).
        I think you’ll discover that filling seats on Sounder may be admirable, but not if you live in Kent, and spare us the marginal cost rationale – we all know what that is.

      5. Really a big problem is the completely clumsy entrance that Sounder makes into Seattle.

        The lack of an easy way to get to the bus tunnel.

        Not being able to ride one or more stops further into the city (could they add more downtown Sounder stops…is it possible to access the BNSF tunnel or is it completely sealed after King?)

      6. But you can’t just ignore the marginal cost rationale. The marginal cost of filling more seats on a train that is already running is negligible compared to the cost of operating a bus trip. And even more so given that the peak period is the most expensive time to operate a bus trip.

        If people value the one-seat ride to the north end of downtown enough that they are willing to pay high enough fares to cover the entire operating cost of the bus, I would say fine, go ahead and keep it, since it doesn’t cost metro anything. But whatever marginal mobility you get by eliminating the transfer penalty (which is often canceled out by traffic delays on I-5) is a lot less than other places those service hours may be going.

        One option that could increase the marginal usefulness of 158/159 service without making people in Kent feel like they are getting shafted might be to adjust the schedules so that they complement Sounder, rather than duplicate it. For example, the last Sounder trip currently leaves Kent at 8:00 each morning, so an express trip to Seattle that left Kent at 8:30 or 9:00 would serve a market that Sounder does not already serve. Similarly, the first Sounder trip leaves downtown at 3:15, so you could have a Kent express trip leaving at 2:30. Or, since the last Sounder trip leaves downtown at 6:15, you could have a Kent express trip leaving downtown at, say 6:45.

      7. asdf:
        It doesn’t change your point, but if your schedule shows the last Sounder trip from Kent at 8:00, you should pick up a new schedule.

      8. An express bus that serves Kent and complements the Sounder schedule should probably serve Tukwila and Auburn stations as well. Maybe Sumner and Puyallup, but those could be done with a separate bus.

        Perhaps ST should plan such Sounder shadow service for the additional trips they plan to run for the ST2 Sounder South service expansion until such times as the trains are running. It could build the ridership base prior to the actual service expansion.

      9. Marginal cost is a bogus calculation. As marginal riders are added, the trains start needing additional runs to keep up. At what point do you call a person marginal, or the one that tipped the scale, then everyone after ’caused’ operating cost to soar. It doesn’t pencil out that way.
        You take ALL the costs, and divide by ALL the riders, to figure out your cost per rider. None of them are on a free ride.
        That’s just the way it works.
        That said, Sounder was billed to Kent and Auburn as a cheap fast way to get rail service going, which was going to lower the cost per rider over the buses. It’s not working out that way, so I caution anyone to shoehorn bus riders onto rails, and charge them more for the pleasure of having two transfers and a trip of equal duration.

      10. Even if the 158/159 are full, their capacity is negligible compared to the capacity of a Sounder train, so moving those people onto the train does not mean you need to run additional trains.

      11. If you really want to make the argument that buses are cheaper than trains, you need to compare Sounder with the alternative where there is no Sounder, with the money being spent on Sounder re-directly to express buses. You would need to include all the costs associated with single-direction travel, for instance, that nearly every trip would require a 20-30 mile deadhead, plus the purchase of a new bus and the hiring of a new driver. You would also need to somehow find layover space for all these extra buses downtown during afternoon rush hour, along with the street space for all these extra buses to traverse downtown to load passengers. I do not believe that if you improved all these costs, the per-passenger cost would be anything significantly cheaper than Sounder, and it would certainly result in a slower ride for most passengers. During the peak, buses are traversing downtown at least every minute, if not more. The time penalty for transferring is negligible. And if you have an Orca card, the fare penalty for transferring is zero is well.

      12. “An express bus that serves Kent and complements the Sounder schedule should probably serve Tukwila and Auburn stations as well.”

        We’ve been throwing around adding a Federal Way stop on the 594, and rerouting the 578 from Auburn-Federal Way-Seattle to Auburn-Kent-Seattle. It seems like it would add all-day Kent service without costing much more than now. And an all-stop 594 would be comparable to what they’re doing up north, where they’re going to combine all 510s and 511s into 512 next fall.

      13. The transfer is negligible? Who are kidding?
        If that were the case, then try this one. When N. Link opens, lets suppose all the buses going to Northgate end there, and LINK only goes as far as Westlake Stn, forcing two transfers, to one of the multitude of surface buses continuing southbound. And for argument sake, suppose most everyone is going to Pioneer or IDS areas.
        Now, how do you like them apples?
        This is one reason I hate the thought of a Ballard-Seattle line stopping short of downtown as a dead end project.

      14. to include all the costs associated with single-direction travel, for instance, that nearly every trip would require a 20-30 mile deadhead, plus the purchase of a new bus and the hiring of a new driver. You would also need to somehow find layover space for all these extra buses

        Sounder avoids the horendously high cost of a deadhead run by leaving the trains idle. A bus would either deadhead or layover and since it isn’t tied to the BNSF tracks there are lots of choices outside of DT. For example an ST bus coming from Auburn could change it’s reader board and take over peak trips on the 550 for the rest of the commute. During the day there are lots of layover spaces at existing bus bases. You also have to account for the wider area that buses can serve meaning a bus can pick up from several small leased P&R lots and head directly DT. That’s not much more expensive than if the bus has to instead drop off passengers at Sounder. That said, last time I looked Sounder was cost competitive with ST Express from Auburn with respect to operating cost but that’s only because it pulls in riders from Tacoma, Kent and now Lakewood which are much cheaper to serve by bus. Sounder North of course is nothing but a money pit and always will be.

      15. “We” means a few of the commentators. David L was the first one I saw to propose this tradeoff.

        Earlier I had suggested just rerouting the 578, but that would cut Federal Way’s service by half. Some of us think Federal Way is getting disproportionately more expresses than the rest of South King and needs to share, but politically it’s hard to reduce service because Federal Way is the symbolic halfway point on I-5 between Seattle and Tacoma and the interface between Metro and PT, and the 577/578 are a replacement for the previous 194. But South King is too wide for one north-south line to fully serve it. and Kent has more all-day ridership than Federal Way and anywhere else I’ve seen in South King, as well as being centrally located.

      16. I suggested the 578 reroute (without the 594) to an ST rep at one of the September service-change open houses, and he was afraid it would make what’s already a very long route even longer, and cost more. I couldn’t figure out why he thought so unless he thought the route would have to go up to 405, whereas I would use Kent-Des Moines Road. So that may be an indication of what the rest of ST thinks of the idea. Or it may mean that one person is making a guess in the absence of a study.

      17. Just because one ST rep pans an idea in a nonsensical off-the-cuff way the first one or two times you suggest it doesn’t mean it won’t happen. Somewhere along the way, ST had an “Aha!” moment with the 512.

      18. Mic: Marginal cost is the only meaningful calculation. Get a clue, please.

        Now, marginal cost is pretty bumpy. The marginal cost of each additional rider on Sounder is zero… until suddently you need another run, at which point the marginal cost of the next rider is pretty large (due to BNSF charges for making an additional run). Then it goes back to zero for a while, and then you get to the point where you need another trainset, and then the marginal cost of the next rider is very, very large. Then it goes back to zero for a while…

        So what you do is to fill the trains until they’re full. Then add some buses until it would be cheaper to have an additional train run than to run all those buses. Then replace the buses with a train run. Fill that train up until it’s full. Et cetera.

        Of course, you get completely wacky results if the buses charge less than the train, given that the marginal cost of additional riders is lower on the train than on the bus (…until you hit that “train full” point). They should charge the same amount generally, to avoid price-based distortions in behavior.

      19. “That said, last time I looked Sounder was cost competitive with ST Express from Auburn with respect to operating cost ”

        And Sounder South will get more competitive on operating cost with ST Express as time goes on.

      20. Total Babble. Roll the clock backwards, and please identify which riders were included in you cost calcs, so we can all know who the marginal ‘freebie’ riders were.
        Cost divided by Riders = Cost/Rider. It’s really that simple for transit.
        Now, if Sounder was selling seats, and at some point quit selling like the airlines do, or a packed theater, then sure, you can sell tickets at deep discounts when you KNOW you’ll have empties and the marginal cost of selling those empty seats is nearly zero.

      21. Any time someone expresses a desire to sit for a long period through a known traffic bottleneck rather than transfer, you can rest assured it is because the transfer is awful. You never know when you’ll happen onto the bus that will miss every light, that will attract a zillion cash-payers, that will get the wheelchair loading one stop and exiting the next, that will suffer a driver change in the middle of downtown.

        The solution is not to double down on one-seat rides, but to fix the transfer once and for all! That means pursuing a cashless zone in the center city, and then kicking buses out of the tunnel as soon as possible to provide trustworthy and legible zip-across-downtown option. (Shame not to be able to do anything about the prohibitive tunnel access times, though wouldn’t it be nice if we could at least learn from that mistake?)

        No one in the major transit cities in the world expects every bus and train they could possibly use to deliver them to within two blocks of their precise downtown destination — especially if they’ve used a 20-mile-plus commuter line to arrive at Grand Central, North Station, or the Ogilvie Transportation. The benefit of a comfortable ride that bypasses all traffic more than makes up for having to transfer or walk slightly further once downtown.

        (This is why I support East and Lynnwood Links, by the way, despite the paucity of all-day demand and lackluster land use. I do believe that the option to bypass the worst and most entrenched bottlenecks — I-5, the lake — should be available. It’s the further spindles, whose demand could be easily accommodated by buses in HOV lanes, that are indefensible.)

        Mic, I’m actually relieved that they’re going to be studying a Ballard subway option without being forced to make it a through-route under downtown. At the very least, this will reduce the apples-to-oranges cost comparison that will inevitably be made against a streetcar to a factor of 6, rather than a factor of 11.

        I have no problem transferring, as long as the transfer isn’t to a fraudulent bus “rapid” transit, a fraudulent “rapid” streetcar, or to something else that is needlessly slow, infrequent, or out of the way. (The Ballard Spur concept works better as a through-route, for example, because the speed and lack of transfer penalty compensate for the additional miles traveled, whereas a 10-minute transfer penalty in the U-District would make the trip as slow as RapidRide.)

      22. I can’t claim credit for the 578-to-Kent/594-to-Federal-Way idea, although I’m a vocal proponent and think the merits of the idea are obvious and far outweigh the drawbacks. I saw it mentioned in a comment here at some point, but I just can’t remember who made the comment.

      23. I totally agree with d.p. on transfers.
        There has to be something in it for both riders and agency to make sense to do.
        RIDERS: Faster or cheaper or more comfortable or more predictable service,… and all 4 would be Nirvana.
        AGENCY: Lower capital and/or operating costs over time, and higher ridership and higher revenue. Better OE/OR ratios, etc.
        Do for any other reasons, and it’s doomed to failure.

      24. “Sounder was billed to Kent and Auburn as a cheap fast way to get rail service going, which was going to lower the cost per rider over the buses.”

        Yes on the first part, I don’t think so on the second part. The public debate was almost entirely about mobility and capital costs, not operational costs. People wanted something faster and more frequent than the pre-ST network, and they wanted some pretty trains. ST promised to get ST Express and Sounder running quickly and inexpensively, and to build expensive Link. Most of the public does not even know the operational cost of these modes, much less which one is cheapest. They have no concept of what expenses are involved. The only thing they know is what their car costs them — and there they often forget infrequent expenses.

        Among those who care about operational costs, it’s widely believed that Link will be less expensive long-term per passenger (in operational costs). We can debate all day whether that’s true, or whether it applies to the suburban tails as well as the core, but we’ve done enough of that already. I don’t think anyone has ever claimed Sounder would be cheaper than buses or Link. Maybe some commuter rail somewhere is, but not Sounder. Sounder’s operational cost was just ignored. The anti-tax people said capital costs are all that matter; Sounder riders said they want it no matter what it costs, and we’ll vote against ST if it doesn’t include Sounder; and that was that.

      25. Also, the true comparison of Link’s and Sounder’s capital costs is not more buses or buses’ operational costs, it’s widening the freeways. The whole purpose of Link and Sounder is to get freeway-like mobility without adding lanes and special bus connector roads. More buses that get stuck in traffic will not help us. Just adding capital costs to operational costs is misleading, because the infrastructure is a public good as well as a convenience to today’s passengers. So at most, only some of the capital cost should be charged to the per-passenger account.

  3. One piece of data I hope ST, Metro, and CT are tracking is zone-juggling volume (number of requests to reset zones on ORCA readers to be charged for only one zone). Assumedly, zone-juggling has gone up since the end of the RFA, but I don’t know.

    ST, unlike Metro, has had to deal with zone-juggling as an all-the-time problem (if it is a problem) for three and a half years. Metro has had the luxury of it being a problem only during peak, but that is likely to change in October after Metro’s fare restructure. The signals from Metro are clear that they want to end the “off-peak discount” as a large source of new fare revenue.

    So, data would be of much use, both before and after the October transition. Metro will be having its own before-and-after RFA elimination report due in June. I hope both ST and Metro can track more than just ridership numbers for their respective reports. The comparison of multiple types of data will be quite valuable to transit planners far beyond the northwest.

    1. If you’ve got a multi-county pass on your ORCA, there really isn’t any impetus to trouble the driver with changing the reader.

      1. Right. The question is how many riders are not multi-county, how many are asking for fare-zone resets, and if this has any significant impact.

        Don’t be deluded by the OTP data. Travel time through downtown during PM peak has grown. But the planners have guessed fairly precisely how much they would grow by.

        Note that before October, fare-zone resets mostly just affected the bus on which a reset was requested. Now, it affects every bus lined up behind it in the Central Business District.

        I don’t know if reset requests are rare or frequent. That’s why I hope data is being collected.

    2. Does that mean the off-peak fare would rise to the one-zone fare, or would two-zone off-peak return?

      1. I don’t know the answer, but the tea leaves I’m reading suggest that the peak fares would become the all-the-time fares, including on 2-zone routes.

        I’ll say more about this, and how pricepoints could affect whether to ditch the zone system, tomorrow.

  4. I have to question ST’s definition of on-time performance for Link. If a PM counter-peak trip takes nine minutes longer than average, but that longer travel time is planned for, should that really count as on-time?

    Perhaps OTP is not the best measure for the effect of eliminating the RFA. Average travel time during peak, each direction, seems like a better measure. But this OTP data should give us umbrage that the planners know what they are doing.

  5. Does anyone know if Metro or ST keep data on the percentage of trips that are paid via Orca vs cash, either system-wide, or on a per-route basis?

    If they do keep those statistics, do they make them publicly available?

  6. I hope these numbers help ST (and Metro, whose cooperation will be needed) realize that the current situation on the 550 is unsustainable for the decade between now and the opening of East Link. The 550 currently appears to be the single route in the Metro and ST systems that has the worst mismatch between demand and capacity at the peak of the peak. It is an abnormal occurrence when a passenger can get on at IDS or even PSS between 4:45 and 5:45 p.m., and people get left at Mercer Island on a regular basis in the morning as well (hardly the recipe for convincing them to accept tolls because the transit alternative is so good).

  7. Yo, anyone else notice that in Q4 the 522 went gangbusters and was ST’s 4th best route, beating the 511 (which has a much better off-peak schedule and weekend service).

    I’ve actually only used the 522 a few times. Does anyone have opinions or data? My feeling is that the 522 should get better network connections in Seattle and become an all-day trunk route (land use and the layout of Highway 522 aren’t perfect but, hey, it’s probably better than I-5), with the rest of the network adjusted for this (probably no reason to run the 372 much north of Lake City after that, but more importantly, combining the odd patterns of routes between the U District, Lake City, and Northgate into a few corridors really worth investing in). But this feeling is just a feeling, no data.

    1. The 522 is my morning ride to work every day and my evening ride home from work about 60% of the time (later in the evening it’s sometimes the 41, which runs more often). I also ride it at other times from time to time. These are my thoughts…

      The route is in a bit of a holding pattern because it will surely change when North Link opens. I think ST is likely reluctant to invest in it or change it too much for that reason, even though it will need to serve as is for at least the next eight years.

      The route’s ridership pattern is somewhere between that of a true all-day core route (such as the 41) and a strict commuter route. It runs SRO, but doesn’t typically pass people up, during the peak in the peak direction. It has relatively poor performance in the counter-peak direction. It is underserved outbound in the evening and at night, when it is frequently SRO between downtown and Lake City. Service levels are appropriate on weekends.

      An extra stop around 80th or 15th would considerably enhance the route’s usefulness for connections, but would likely push peak passengers from the 77, which has spare capacity, onto the 522, which has little extra capacity at peak times.

      In the long run (after Link opens), I think it would be better to try to create a true 15-minute north-south corridor on the 372, while relying on various east-west connections and a streamlined 72 to get downtown passengers to Link outside of peak hours. Peak-hour service would be provided on the current Metro 312 and 306, with extra trips. The corridor has high demand, but is relatively expensive to run.

      1. Ah, so forward-peak it’s sort of like in the I-90 corridor where some routes need to bypass Eastgate so the outbound Issaquah passengers have a chance to get on the bus (or why the north I-5 corridor is so messed up). A challenging situation. It’s always a shame when peak capacity concerns get in the way of network concerns. These sorts of situations represent real missed opportunities but it’s hard to fulfill them without more money sometimes…

        The network concern is that it’s so ridiculously slow to get to Lake City from many places west of I-5 on transit. If you could Spider-Man-Transfer from the 44 or 48 this problem would be less. The basic infrastructure exists for a 44 transfer but bi-directional all-day HOV lanes would be good to have on I-5 first (I used to use the 511 reverse-peak and sat in lots of evening inbound traffic jams while the outbound jams were even worse).

        Oh, well. When we blow up I-5 we’ll have to rethink the whole thing anyway.

  8. I have been told 3 out of the 9 trains are mostly always standing coming into Seattle now. 6 more months until the 10th roundtrip starts!

Comments are closed.