SDOT Photo
SDOT Photo

Late last week Metro released initial ridership data for the newly-split Rapid Ride C and D lines. Though ridership had already been seeing Link-like growth rates hovering around 10%, Metro says that in the month since ULink opened and the restructure split the lines and added 50,000 new service hours, C-Line ridership is up a whopping 26% and D-Line ridership is up 21%. 

Extrapolating from 2015 weekday ridership numbers, this means C-Line ridership has grown from 8,300 to 10,500, while D-Line ridership is up from 11,700 to 14,200.  The two lines now combine for 25,000 riders per day, roughly what Central Link generated back in 2012.

Much of this can be attributed to the additional destinations now able to be served. No longer tied to the Viaduct, D-Line riders have a much better connection to South Downtown, the stadiums, and Pioneer Square. And of course, the C-Line now serves South Lake Union on the successful Westlake bus lanes, giving riders in the corridor a frequency dividend, enabling them to take whatever comes first between the streetcar, Route 40, and the C-Line.

71 Replies to “Rapid Ride C & D Ridership Way Up”

  1. Not surprised at all. I wonder if the C is cutting into SLUT ridership, though.

    1. Probably depends which one comes first, djw. Market Street in San Francisco has historic streetcars share the street-and positive wires. Block or to between stops, but people can look back up the street and see which mode is closer.

      Also, MUNI Metro light rail cars run one level downstairs, and BART two levels. Offhand I’d say that people going south of Stewart Street will generally take Rapid Ride. DSTT passengers will take the streetcar for closest stop.


    2. Walk, C or streetcar, whichever comes first.
      So when I board the C, usually 12-20 people are catching the bus within South Lake Union, so there’s your latent demand that Metro just served. (Who might otherwise walk to 3/Pike).
      As for the ridership numbers, the C was booming anyway without the Westlake transit lanes — because hundreds of apartment units just were finished in West Seattle and a fair proportion work in IT and biotech downtown. The growth is noticeable every month, and another thousand or more units at the C stops should be finished within the next year.

  2. I’m curious to see ridership numbers next month following the #viadoom closure, particularly on the C line.

  3. The C seems to be functioning fantastically in SLU. I’m seeing either a C, 40 or a streetcar every time I look. Box blocking actually seems to be much less of a problem. There are enough buses to effectively ‘sweep’ the lane now and the people in cars appear to be learning. I wouldn’t be surprised if streetcar ridership would actually benefit. It used to get stuck by box blockers, poor parking, etc. all the time but it’s much more reliable now. At least Southbound from I just take the 40, C or streetcar, whichever comes first.

    1. yes, have to agree that it works fantastic going south bound, but north, not so much…this is my frustration with lots of lines that go the same places – they don’t stop at the same places downtown…

      I know that the RRs are together for marketing and probably cost…but all the lines going to SLU should stop at the same stop. I stand between stops and furiously work my OBA app.

      1. Yeah, I feel your pain – I come down Denny on the 8 and play OBA DJ to figure out which route to Fremont I can make. For SLU, basically the two stops the C and 40 share are 3rd and Virginia and Denny and Westlake, and that’s a bit sparse.

        I know that for some people coming from West Seattle, the fact that there’s a same stop transfer at Mercer and Westlake shifts the bus commute from WS to Fremont from “impossibly hard” to “doable”. For that I’m glad.

      2. Ugh! I have the exact same problem going from downtown to North Fremont – why can’t Metro put the E / 62 / 5 on the same signs downtown??? They only overlap at a single stop in Belltown. This has been a long running gripe of mine with Metro.

      3. And I had the same problem a week ago when I was going to LQA – the D and 1/2/13 were at different stops.

      4. “I stand between stops and furiously work my OBA app.”

        Many many years ago before the OBA app existed and most people used the OBA web interface, I would chain together several stops so that I would only need to load a single page.

        For example, I’d be at the corner of Broadway and John and wanting to get to the u-dist. I wanted to know if I should hop on a 49 (on Broadway) or a 43 (on John). So I would create a url like this (which still works):

        I wish a future version of the app would allow a user to combine two stops or have a one-click way to flip to the last viewed stop so I could quickly ping-pong between two stops.

      5. Keeping track of multiple stops is difficult with the OBA interface. OBA has a API for writing your own apps using their data, so I ended up writing my own OBA web app which makes it easier to look at multiple stops: It uses your phone’s GPS location and offers the nearest buses; you check off which ones you want and it shows all of the upcoming arrivals at all of the nearby stops (filtered by direction).

      6. That’s one thing I like about PDXBus (one of the several apps that works in TriMet Land): the author wrote it so that it is easy to add multiple stops to a single bookmark and display them on a single page.

      7. Try Transit App ( – it shows times for all the nearby buses (and streetcars, LRVs, etc) sorted by stop distance. It’s also multi-system and multi-city; I started using it when I lived in Boston, and kept using it here because I liked it better than OBA.

    2. I’ve noticed that just having the right turn light signal at Northbound Westlake @ Dennys, that allows right turns before allowing pedestrians to cross, has been a huge improvement for traffic flow in the bus lane. Great job!

  4. Wow, those are some encouraging growth rates! Good news for the BRT boosters out there.

      1. All it takes is the Mayor and SDOT having the political spine to start carving out and physically barricading some lanes as bus-only to start…

      2. Agreed (both QA and Joe). I think this is more frequent service boosters. There is a lot to be said for just boosting service on the popular routes and seeing what happens. In this case, ridership that is very high.

      3. I agree with you, but it also doesn’t really matter.

        It’s improved bus service, and is getting the results. Those results will create the political climate for further improvement, both in the form of pressure from riders wanting more speed (especially once the notion starts to take root that buses speed doesn’t have to equal car traffic plus stops), and from reduced opposition from drivers as they start seeing lanes full of full busses, rather than just empty pavement while they’re stuck in traffic.

        “BRT” and “not BRT” are just two sides of a line drawn on a continuum of transit quality. It doesn’t matter so much if you’ve got “BRT” or “not BRT”, as much as if you are improving quality and moving in the right direction on that continuum (assuming, of course, that more service is appropriate for the market; obviously BRT level service shouldn’t exist Cle Elum)

      4. I agree, EHS. Half-ass BRT, bronze level BRT, whatever. The point is improved service on a popular line becomes more popular.

        I do think Madison BRT will be very interesting, though,as it is by far the farthest on that spectrum that will exist in the area (and that includes Swift). It will be fast, very frequent service along a popular corridor. Roosevelt BRT, if implemented the way they can implement it, might be even bigger, just because of the larger distances involved.

    1. The lesson is obvious. If it’s not fully grade-separated rail, it doesn’t matter.

      (Not that there isn’t a place for that). But we need to stop looking down our noses at incremental improvements to the entire transit network.

      1. I agree. There is a tendency to get excited about the big projects, whether it is BRT or light rail. Madison BRT gets me excited — center running buses with off board payment and level boarding running every six minutes all day — but sometimes the little things matter just as much. Dozens of changes all across the city can really make things a lot better.

  5. It’s great to see metro’s investment paying off. 14,000 riders on the D is approaching the E’s roughly 15,000!

      1. The RapidRide version of route 7 actually splits it into two routes, according to the Seattle Transit Master Plan: One route from Mt Baker to downtown, and one for the southern portion on Rainier, with a terminal turn over to RBS. And then the southern portion is proposed to be joined with the current version of route 48.

      2. Good point Brent. I do find that interesting and debatable. If you look at that previous link of corridors, Corridor 3 (Metro 7) and Corridor 7 (Roosevelt) appear to be linked. The Metro service plan ( has them linked as well, as the 1071 (select the button for the year 2025, then select the red line under Yesler Terrace). This would be similar to how the C and D were linked a while ago.

        This makes some sense. You avoid paying twice for service through downtown. Meanwhile, the city (and county) is trying really hard to make the area around Mount Baker station work like a transit center (a place where buses end, and transfers are plenty).

        Personally, I don’t like it, for a couple reasons:

        1) While the city may make the transfer situation much better, Mount Baker really isn’t a major destination, or a dead end. Mount Baker is just a spot along Rainier Valley. There is no drop off in density or destinations there. There are great transfer possibilities, to be sure, but I’m sure there are plenty of people who ride the 7 from one side of Mount Baker to the other.

        2) For this new route, I don’t see much advantage of connecting the two routes. This is in contrast to the old C and D. Splitting those lines made sense from a reliability standpoint, but if the line had been more reliable, then keeping them together allows for some very nice connections (West Seattle to lower Queen Anne). For a trip like that, the C/D combo was as fast as you could get. But with the Roosevelt/Rainier Valley run, it isn’t as useful for longer distance travel because it curves around too much. In many cases (e. g. Yesler Terrace to Denny) it would probably be faster to take the bus traveling on Boren (the 1074).

        If we need to split this, I would split this downtown. Unfortunately, I don’t see a great way to do this. The C and D split was quite elegant, providing new service from both ends (especially the C). I don’t think this split would be quite as nice, but I think it still add some value:

        Have the Roosevelt BRT live loop on 1st. There isn’t a lot of value added here (a couple more blocks of coverage) but this is the heart of downtown (Lenora and 1st) so every block adds new riders. Have the Rainier Valley line stay on 3rd, go through Belltown and turn around in lower Queen Anne somewhere. This seems like a logical turnaround spot, as demand is high through that corridor, but drops off quickly as you head north.

        This would, of course, cost more money. I have no idea where to get the money in terms of service, and it might not be worth it (I would rather force a transfer than water down everything). If money could be found to do this along with everything else, it would mean that Rainier Avenue would have two BRT lines from Rainier Beach to 23rd (with one headed up 23rd and the other continuing downtown). This would allow an investment in the corridor (e. g. bus lanes) to be used by more buses, which is a really good thing. That might be overkill, but it might just be what is appropriate for the area (which is fairly dense and growing).

  6. That is great news in light of Metro LRP. I’ve ridden both the D and E for long stretches in the last couple of weeks, they each seem equally busy.

  7. If this holds true,

    The E line, D line, and C line all will surpass the issaquah line in 40 years, today separately on all lines.

    You can’t simply deny this. The e line filled up 7 miles out of town today and the bus before us didn’t even stop as it was too full.

    We need an investment on these lines. The riders have done what you asked , show up. Now support them.

    1. Today the Viaduct is closed. Is it also passing people up when the Viaduct is open? WSDOT provided mitigation funds for extra E runs during tunnel construction, but they weren’t extended when construction stalled so they ran out a year or two ago. I think Prop 1 too on funding those runs, but I’m not sure if WSDOT ever did replenish the mitigation fund for runs on top of that.

      1. The E is one of the lines with a consistent pass-up issue, even after two rounds of frequency improvement. How much of that is bunching and how much is a total capacity issue, I’m not sure.

        Prop 1 never funded any additional E trips because the route doesn’t meet the 80% rule, but it’s one of the “suburban” routes Metro invested in on its own for the current shakeup, even though the ridership crunch is entirely within Seattle and mostly south of 85th Street.

      2. Thanks for the info, David. Wow, if the E is one of those suburban routes, then it is my kind of suburban route. Ridership exceeding billion dollar light rail lines, with frequency far in excess of it, all day long. Too bad it suffers from what appears now to be an arbitrary border. If only there was an agency tasked with improving the transit system across agency boundaries. Oh, wait, maybe ST3 includes money for … HA! just kidding.

      3. Compared to most other routes in the area (the 5, 16/26, 28, and 40, at least), the E has surprisingly strong ridership north of 85th. I regularly use the N 100th St stop, and my experience is that on weekdays, I’m fairly lucky to find a seat.

      4. Yes, the only places I’ve seen the far north end of the E drop off is at the true north end, where there is the interface with Swift.

        I would say the biggest issue it has is the amount of time it takes to get across highway 99. Those lights create a long term project for pedestrian traffic. A few bridges would be helpful, since there is probably no way to get that light wait time reduced.

        Since 200th is the border between CT and KCM, it would be nice if both Swift and RapidRide could feed an express stop at 200th. Instead the express stop is at 145th: too far south for Swift and too far north for RapidRide to go there and not break the corridor.

      5. @Glenn: The border between King and Snohomish Counties is right around King County’s 205th St, or Snohomish County’s 244th St. The closest ST 512 stop to the border is Mountlake Terrace P&R, in the middle of I-5 near 236th (just on the Snohomish side). That’s actually pretty close! The obstacles to getting there from 99 are getting pretty far afield from the main post, however.

      6. Agree , but ridership is not decreasing. And the line is standing room 8 miles sb/nb constantly. The spacing is so short it becomes a banana line.

        Today was only of things to come.

      7. “Ridership exceeding billion dollar light rail lines, with frequency far in excess of it, all day long.”

        The flip side is the 45-minute travel time end to end. Link will get to 185th in 20-25 minutes. I think it’s feasable to bring the E’s travel time down to 30 minutes is doable with full BAT/transit lanes and a stop diet. Shoreline has earned its close stop spacing by zoning TOD around every station but Seattle hasn’t.

        “it would be nice if both Swift and RapidRide could feed an express stop at 200th.”

        Swift is planning to go south to 185th and east to 185th Station when Lynnwood Link opens. That will allow for a same-stop transfer northbound or southbound without detouring to Aurora Village and back. I think the stop would either be at 185th (largest destination in the area) or 192nd (Shoreline P&R and sometimes-proposed transit center).

  8. Does anyone have stop-by-stop boarding numbers for the D and 40? I’m interested in seeing how they compare in Ballard, as we prepare to build an expensive train up 15th that only serves two stops.

    1. Anecdotally, very few people get on or off on 15th. Frequently a mostly full bus will skip stops because nobody wants to get on or off. Light rail station spacing on 15th is about development potential, not about what’s actually there today.

      1. I assume you mean “on or off 15th before Ballard”. Fair enough. Still, like William, I would like to know more. I can kind of guess the pattern on the ‘D’, especially with your input (lots of people getting on and off in lower Queen Anne, then a big gap, then lots of people getting off at Market, then a few at every stop after that) but the 40 has to be more interesting. I often grab at a beer at The Outlander (on Fremont) and watch the 40 go buy with people on it both ways all afternoon and evening long.

      2. Yep, my bad – “15th in interbay” would be more accurate. Obviously 15th in Ballard is huge ridership. I don’t ride the 40 as much, but I agree, it is more interesting. Especially when the Fremont bridge goes up.

      3. I was quite surprised to see 6 others when I waited for the inbound D at Dravus on Friday. The D showed up first, half of us got on, presumably the others were waiting for the 32, which doesn’t make much sense because the two are stop-for-stop equivalents until the 32’s terminus near Mercer.

    2. From my experience, a good percentage of the bus gets off the D line at the first few Ballard stops (Leary and 15th; Market and 15th). The 40 sees a huge dumping of passengers at Market and Ballard Ave, but that whole corridor is pretty busy from Fremont all the way up into Crown Hill.

      1. (Not asked, but…) The C similarly unloads about 10% of WS-bound passengers on Avalon and in Triangle, and 60% at the Junction and… well that 60 includes me so I don’t know where the other 30 go. California has a few more blocks of 3-4 story apartments and condos south of Alaska, though I bet a lot of those folks just walk from Alaska, given the bus’s detour on 44th.

  9. I wonder how the Link opening affects demand. The lines are both longer, and the riders just north of Downtown could be gravitating to using RapidRide to get to Link rather than take a slower bus.

    Another factor may be induced demand since high-frequency, faster service gives a rider a broader choice of destinations within 15 minutes. SLU workers may be making more midday transit trips into Downtown Seattle rather than just walk around the SLU neighborhood, for example.

    1. I would guess that Link is pretty much irrelevant for these routes. It may play a part in increased ridership from West Seattle to (one part of) Capitol Hill, but that is about it. If you are headed to the UW, then the 71,72, 73 worked just as well. So, yeah, maybe a couple hundred new users — tiny in the grand scheme of things.

      Your other point is spot on. Better service leads to higher usage, for sure.

      1. I’m a rare West Seattle to UW commuter. I’ve been disappointed to find that it’s about the same time by Link as it was by 71/2/3 or 255. There’s a huge improvement in my blood pressure though! It’s a lot more pleasant to zoom reliably through a subway than stop and go traffic. That southbound commute from UW to downtown in the afternoon (after losing the power of the express lanes) was pure hell by bus. I still get flashbacks to waiting through multiple light cycles just to cross Olive Way into the bus tunnel ramp because those f***ing SOVs would block it over and over again…

        But timewise, it’s pretty reliably 25 minutes from Uni St. entrance to any actual campus building, like my department on the quad or where I teach on Campus Parkway. 8 minutes of that is usually travel time, the rest is getting in and out of tunnels and finding a way, bus or bike, across the astonishingly long distance between the “UW” stadium and UW itself.

      2. Yeah, southbound evening travel is really where Link shines. For the 41 (from Northgate) it is the same story. It takes forever. Link will make a huge difference as it gets farther north. HOV lanes can be slow, but they are sure faster than nothing, which is what folks get if they go against the grain.

        As for the UW, I agree, it is a shame this round doesn’t include the U-District. It is a long ways for a lot of people. Those heading to the hospital or lower campus are in luck, but everyone else has to wait a few years.

      3. I was coming home (to east Wallingford, barely west of I-5) from the airport a few days ago and the trip planner told me it was faster to transfer to the 26 downtown than to stay on the train and take the 44. Now, this was on a Sunday evening, when the 26 flies and the 44 still hits every stoplight on 15th and 45th, and there probably was some luck with the timing of the transfer to the half-hourly 26. In other circumstances the 44 might win. But by the time you’re over by Aurora you’re just about always going to be better off riding the E Line than taking the 44 or 48 to Link. Maybe Roosevelt Station changes that for people in 48 Land; I’m not sure U District changes it for those in 44 Land.

        Heaven help us all (by “us all” I mean “people that take buses downtown from the north, west of Lake Union”) if Belltown ever turns into SLU. Every proposal to build more parking there should be met with shotguns.

  10. Great news! This is exactly how transit is supposed to work: frequent lines support the land use and develop the ridership to support BRT; BRT supports the land use and develops the ridership to support rail.

    Such high ridership surely makes further ROW acquisition easier politically, which improves the route.

    If only it didn’t take 22 years to build rail…

    1. I agree, except (as said above) I wouldn’t call this BRT. Maybe BRT-lite, or maybe just frequent service. I’ve heard lots and lots of people talk about Link in northeast Seattle, and they often mention frequency instead of speed. To be fair, a lot of those buses (that traveled on the HOV express lanes) were pretty fast to begin with. But for those whose trips are not especially convenient with Link, the fact that the train comes quite often is a huge selling point. At this point, serving only a smidgen of the area, with a very inconvenient transfer, you can’t expect Link to do a stellar job (which is why many predicted that Metro wouldn’t make a major change for this go round). But frequency is huge (I know it is for me) and a lot of people like it. Service matters. A lot.

      I remember Frank wondering a while back what could be done with just paint. A bit extreme, in my book (I want to do some more tunneling, that’s for sure) but an interesting idea. What if you just take lanes (especially parking) and see what happens. I wonder what would happen if we just make things more frequent. Spend the money on more service, and see what happens. My guess is that both, combined, would not be ideal, but is sure would be nice. Buses traveling faster (although not super fast) and much more frequently. Sounds good to me.

      1. Oh, I wholeheartedly agree. BRT lite, it is. I suspect it will be close to full BRT by the time Link gets up and running on the corridor, though.

        Also agree: If we wanted to be cost effective, we’d buy paint and frequency, and we could do a hell of a lot. But the political cost of that paint is often greater than the $ cost (though encouragingly, not so on Westlake where it really, really makes sense). It’s political realities and future-proofing for capacity that leads to tunneling for rail, far more than any technological limitations of buses. But political realities are real.

      2. But the political cost of that paint is often greater than the $ cost

        I’ve heard the same thing over and over (and thought the same thing) but I’m beginning to wonder. I think people conflate the argument against zoning changes with the arguments against “taking lanes”. I really can’t think of any Seattle project, anywhere, that has been slowed down because of reluctance to take those lanes. Madison is the closest thing, but the folks clearly stated that they didn’t take all the lanes because it wasn’t necessary, not because it would hurt local businesses, or make traffic worse. When you consider what they did (carve out center running lanes in the heart of the city) I think this shows that SDOT is wiling to make some pretty bold moves to get buses moving.

        The state is different, of course. It would be really easy to change HOV 2 to HOV 3, but so far they seem unwilling. HOT has been weakened, but the politics behind it are more complicated (class warfare has something to do with it). For the most part, the state is pretty much sticking with what they have, even if it means HOV 3 on 520, even though the original reason (narrow streets) is no longer valid (they just redid the entire roadway).

      3. @Ross: What about the “d.p. Memorial Stop Light”? That’s not a pure “lanes” issue, but SDOT explicitly stated the reason they refused to shorten wait times for buses there was to keep through-traffic flowing on Elliott/15th. What about the part-time nature of the D and E Lines’ bus lanes? If you mean “slowed down” in terms of transit vehicle speed, I’d count these. If we end up deleting bus lanes from the waterfront that would count here, too, but I don’t think we’re going to do that.

        A couple years ago there was an SDOT proposal to paint bus lanes near the One Bus Stop In Lower Fremont; it hasn’t gone anywhere, possibly because of practical problems with buses turning to and from those lanes, but it was also opposed by local merchants (who want today’s parking and loading zones preserved). If you mean “slowed down” in terms of project implementation time this is the only recent one I can think of.

        But SDOT has been fairly cautious about what they even propose; it’s mostly carved out spaces for bus lanes that don’t make a big difference to drivers, and therefore aren’t all that contentious. There are lots of bottlenecks along popular on-street routes where speeding up transit would have a huge impact on general traffic flow. Move Seattle will be a big test. Implementing the transit part of the plan will involve real tradeoffs between speed of transit vehicles and general traffic, in neighborhoods that will, for a variety of reasons, resist more than First Hill did to bus lanes on Madison.

      4. I hope that’s the case, Ross. I do think people are fed up enough with traffic, and supportive enough of transit, that taking lanes is likely easier to do than we realize, because we (and our politicians) calibrated our scale several years ago when driving wasn’t so obviously an unviable solution. If we can get bus lanes on any road that has more than, say 10 buses an hour, we could absolutely transform transportation.

        But I fear that it only looks easy because we’ve so far taken easy to take lanes (parking, the third lane on a road rather than the second or the only lane). I think it’s way easier to take a lane on Aurora, which leaves you drivers with two travel lanes and a turning lane, than to take 50% of the car capacity on a road that doesn’t have a nearby alternative (eg the gridded arterials in residential neighborhoods). Especially because if you reduce a four lane road to a two lane road, the waits for people trying to turn could be horrendous.

        But, yes, I think opposition to zoning changes probably runs deeper than opposition to lane taking, and certainly when that lane is parking.

      5. I should mention that Seattle has also been pretty aggressive with their road diets. I can’t think of an instance where they went back (and added back lanes). But they aren’t implementing road diets everywhere. Don’t expect Denny to be one lane each direction any time soon. I think signal priority as well as taking general purpose lanes gets tricky. Do it wrong and you can screw up the buses (that don’t run in bus lanes), along with general purpose traffic (including freight). I think that is why a lot of projects haven’t been done. it’s not that the city is afraid of push back, but because they haven’t studied it yet (or if they’ve studied it, the results have been very negative). Your Fremont case Al, is a good counter example (if it was rejected because of negative feedback).

        I think Roosevelt BRT should be very interesting. If done at the levels they want (so called “Full BRT”) it could be amazing. But if it is done in a half-ass way, it may be hard to figure out why. Was it because of push back by local merchants, or because the city didn’t want to spend too much money on that project?

        One of the interesting things about both projects is that they can be improved over time. Madison is certainly poised for that. Six minute headways all day, center running (dual doors) and all that, mean that it will be popular from the beginning. The city also clearly stated that the lack of BAT lanes to the east wouldn’t be a problem. If it is, then folks will demand some more paint (and probably get it fairly quickly). This is in contrast to the streetcar, which, so far I know, will have no improvements made to it.

        I am sympathetic to folks that have to move their goods around town during the day, and want a place to park. I think money can often solve this problem, though. Sometimes it means a carve out for parking, which is common with road diets. Sidewalk bulbs replace a general purpose lane (that also allows parking) with a lane that is only for parking.

        I would support efforts by the city to pay the merchants or look for similar parking improvements if it comes to that. I would much rather see us spend a bit of money on that, then throw up our hands and say “we can’t do it — the locals didn’t want it”.

      6. I would take bus lanes or a busway any day over “BRT”. Bus lanes are actual continuous infrastructure whereas BRT is a bus painted red with custom stations. Very rarely does BRT ever even include bus lanes or a busway.

      7. @poncho — I think you are confused by the term BRT. Saying you will take bus lanes over BRT is like saying you will take an IPA over a beer. So far as I know, there is only one agency that judges BRT systems, assigning grades based on certain criteria ( Their scorecard gives out more points for avoiding congestion than anything else:

        * Dedicated Right-of-Way (8 points)
        * Busway Alignment (8 points)
        * Intersection Treatments (7 points)

        On the other hand:

        * Platform-level Boarding (7 points)
        * Off-board Fare Collection (8 points)

        A bus painted red?

        * Branding (3 points)

        So, for example, Madison BRT will score very high on this standard because it will operate in its own lane for much of the way (and have off board payment, etc.). If they forget to paint the buses, it won’t matter (it will still be considered BRT by ITDP). RapidRide, on the other hand, fails on most of these measures, which is why it would get the “Not BRT” rating (despite doing quite well in the branding category).

        Side note — Kirkland proposed a BRT system for their area with ST3. They specifically called out ITDP and said they wanted to build a “Gold Level BRT System”. I would imagine this was specifically done to answer critics of RapidRide (or BRT in general) who say the term is meaningless. They wanted to make sure that the BRT system would have all the advantages of a light rail line (along with the advantages of a bus system). Kirkland was rejected in their efforts, and pretty much got nothing in the area.

  11. Does anybody know what percent frequency increase those 50,000 service hours plus faster travel bought? If it’s about the same as the ridership increase, that would suggest that we’re still capacity limited in these services. If thats’ the case, we should keep boosting frequency until ridership doesn’t keep up.

    (a really interesting question is what the relationship is between the gains you get from increased frequency and the fullness of the buses. Obviously if buses are 100% full both before and after a change in frequency, then you’ll see a 1:1 relationship between the change in ridership and the change in frequency. If you had no increase in ridership, you’d have completely tapped your market, as even with shorter wait times, the bus isn’t useful to any more people (we can dream, right?). I bet you there’s a pretty simple and reliable mathematical relationship between the how full a typical bus is and the relationship between frequency and ridership on a route.

    If so, it would be a handy way to forecast ridership due to frequency changes, and to argue for utilizing service hours on high ridership routes. “Look, this route is typically 2/3 full; that means that if we increase frequency by x%, we can expect 0.7x% increase in ridership. This route is only 1/4 full – if we increase frequency on it by x%, we expect only 0.2x% increase in ridership.

    Obviously there are much more sophisticated models based on population density and job density, pedestrian friendliness, etc. But in my field, at least, people often do better with a simple model that targets a clever, representative variable, than they do with a complex model that tries to incorporate all the causes. You can go ahead and come up with complicated measures of density weighted for distance from transit, of walkability based on street width, sidewalk width, street connectivity, number of trees, size of storefronts, average gradient etc, only to find that your model missed the influence of brick facades and awnings, which really define where people walk. Or you can just look at transit ridership, which already incorporates the effects of all of the stuff you might try to measure, and use that. Then you can define walkability as “the stuff that effects ridership that isn’t density, frequency, or what the route connects to how quickly”. (obviously what the route connects to is huge, and not easy to measure – but I’m advocating this simple model purely as a way to evaluate changes in frequency, in which density and alignment don’t change. It’s not gonna be useful at all when you want to look at changing routes, but I bet it would be great for changes in frequency.)

    OK, that was a rambling stream-of-consciousness post. Hope if you made it this far, you found it at least somewhat enjoyable.

    1. Some of this is going to depend on many / badly people got pissed off by poor service in the past.

      In the pre-RapidRide world, I took the 15 from eastern Magnolia up to Carkeek Park a couple of times. Coming back south was a horrific pain because the schedule was oriented entirely around downtown Seattle commuter traffic. Afternoon southbound frequencies were 40 foot half hourly or some such, and buses northbound were very frequent and big articulatedes. I wound up having to go all the way to Seattle Center and then ride back north as the southbound bus was too crowded to even try to get off as it went through Interbay.

      I’m sure that a number of people experienced that level of crowding and decided “never again” but now RapidRide is changing that bus trip to be frequent in both directions. People are probably discovering the trip doesn’t suck anywhere near as bad as it used to.

      1. Yeah, for sure – averaging ridership and frequency for the whole route isn’t ideal, because there are time of day and direction of travel effects. But it wouldn’t be hard at all to split it up into an hourly analysis. I’d be really curious to see if there are any conditions where an increase in frequency leads to a larger increase in ridership than the increase in frequency – if it did so much to make the service more useful that buses got fuller, even though there were more of them. If you found a particular set of circumstances where more buses meant MORE people per bus, that would be amazing.

    2. Hope if you made it this far, you found it at least somewhat enjoyable.

      Sure did.

      I would imagine the concept has been studied, and someone has the numbers you are looking for. but browsing through Human Transit, I couldn’t find it.

      1. Haha, thanks Ross! I’ve read Human Transit (the website, not the book) pretty throughly and haven’t seen it, and that’s where I’d look first, too. But I’m sure somebody’s tried doing it.

      2. I took a course from the FTA long ago on ridership forecasting, with elasticity formulas for fare, frequency, etc, but going through their catalog, I can no longer find it. Same for APTA. You might try a more thorough search to find what’s offered now.
        The full bus factor sounds very interesting – kind of like a ‘Market Clearing Price’ is for economists.

      3. The book references a number of studies, and even has a chapter dedicated to frequency, but I couldn’t find a study like the one we want. I would imagine one issue is bus bunching. The more a bus moves along the BRT spectrum, the less this is an issue. But with signal priority, you can still get buses to platoon. But in general, I think if a bus is crowded, ridership will simply increase as you keep adding service. The 99 B-Line in Vancouver started with 10-20 minute headways, but ridership was so high they kept increasing it. Now it runs ever 2-4.5 minutes, and carries 54,000 a day. I think a lot of the buses are very crowded (it should be converted to rail).

      4. I’ve ridden TransLink’s 99. It desperately needs something more than articulated buses.

        I was only able to get on it by using the card reader at the rear door, and had to get off several times to let people off.

        It’s not really even BRT.

    3. I bet you there’s a pretty simple and reliable mathematical relationship between the how full a typical bus is and the relationship between frequency and ridership on a route.

      It may be mathematically modelable but I doubt it’s simple. The Frequency/ridership relationship is obviously going to be different in different neighborhoods and corridors, based on a) differences in the needs of the local ridership and, especially, b) the number of riders who are choice v. necessity. Perhaps we could figure those things out from available socioeconomic data or something else easy to plug into a formula, but it wouldn’t surprise me if it’s more difficult than you suggest here.

      1. Yeah, but what he is saying is that it is quite possible that those factors balance each other out. In other words, let’s say you had a few dozen graphs of various routes. They certainly wouldn’t all follow the same pattern, but it is quite likely that the bulk of them look fairly similar. That similar pattern can be the basis for your numbers. If applied in this instance (or any instance) you might find yourself as one of the outliers (for the reasons you mention) but at least you have a pretty good starting point. Furthermore, trying to control for all those other factors (i. e. figuring out if you are going to be an outlier or not) may not be worth the effort.

  12. For what it’s worth, I have to say this blog post made me look at using RapidRide again. But RapidRide C doesn’t serve well the stops the Seattle Streetcar does along the Westlake Avenue that I use – namely the Whole Foods and the Westlake Link. Occasionally Lake Union…

    That said, some pictures of the road diet along Westlake Avenue would be nice. Thanks.

    1. As an occasional visitor to Ballard and West Seattle, I find the C and D more convenient than their predecessors. It’s nice knowing that there’s always a bus within 15 minutes and the display shows how many minutes. The transit lanes aren’t what they should be but it’s still a better experience than riding a regular bus, and makes me more willing to go to those areas for dinner or a stroll.

  13. Mildly frustrating to see the NB buses on Westlake at Mercer (C and 40) not using the bus lane. It seems the priority signal for buses does not always trigger, so some drivers would rather block up the main lane to get across Mercer… Not sure if the trigger could be better timed in the sequence… But it is amazing how traffic has actually seemed to get better crossing Westlake since the bus lane changes…

Comments are closed.