Last week, we reported on the under-performing Ride2 services in West Seattle and Eastgate which have experienced low ridership and outsized costs per rider. Yesterday, Metro announced they were ending both pilots effective December 20. (The news was first reported by West Seattle Blog).
The Ride2 services were created as one year pilots, and the end comes as the West Seattle version reaches that milestone. Eastgate, which experienced a change of provider from Ford subsidiary Chariot to Hopelink in February, ran for 14 months.
There were some interesting new details on how the services performed from Metro. Over seven thousand users had downloaded the phone app, though fewer than 15% had used the service within the last month as customer interest failed to develop.
Average ridership in West Seattle was just 29 per day, despite serving an area with 53 thousand residents and 12 thousand jobs. That low ridership drove an extraordinarily high cost of $84 per trip. Eastgate did a little better with 82 daily riders and a $35 cost per rider on a similar sized footprint. The West Seattle service was funded through the Seattle TBD and the Eastgate service by Metro. Metro characterized the Eastgate results as within expectations, but it needed other funding to continue.
Metro will continue the Via service which has been far more popular and cost-effective, serving nearly 1,000 daily riders. The Rainier Beach service has been particularly busy. Metro believes part of Via’s success is that it connects to very frequent transit service like light rail and very frequent buses making transfers easier. Eastgate bus service is more sparse off-peak and parking remains available until late in the AM peak. The frequency of the water taxi in West Seattle is yet more limited with 30 minute service in summer and less in winter.
Via has operational advantages over Ride2 because it serves several connected areas including five rail stations in Seattle and Tukwila. That model scales more efficiently than the Ride2 shuttles serving isolated islands because Via vehicles can cross between service areas. It’ll be interesting to see if they also succeed as a feeder to other rail stations or frequent bus routes in future.
I can’t speak to these 2 services, but I’m curious how the Shoreline/Lake Forest Park community services are doing. I personally think they’re pretty useless but maybe others find them useful.
1 service is nights only and must start and end in those 2 cities, but at least has an app.
The other service is only for groups and requires emailing someone 2 days in advance.
And Metro doesn’t seem to have advertised either of them.
“Over seven thousand users had downloaded the phone app”
That gives us a minimum for how many people knew about the service. To put it in a familiar context, that’s 140 bus-fulls of people. So the theory that few people knew about it and there wasn’t enough outreach doesn’t seem to be borne out.
Two things that differentiate Via and Ride2. Both Ride 2 West Seattle and Bellevue don’t serve light rail like Via. And, I believe, West Seattle’s and Bellevue’s Ride 2 areas have a higher income than Via’s areas.
But if it were up to me, micro transit would also be cancelled in the Rainier Valley and Tukwila, two areas with some of the best bus service in King County. If you want to ride in a car or van to a transit station, get an Uber and pay for it yourself.
“But Sam, we built such a pos light rail system that it requires a fleet of minivans to function!” Your words, not mine.
Alternative hot take – we built a light rail line through a lot of distinctly so-so land use. And there’s lots more where that came from.
So after spending a fortune building light rail there, we should double down and supply extremely expensive microtransit as well? Sorry, that makes no sense.
Before this whole things started, the four stations in Rainier Valley carried about 10,000 riders. So this increased it by about 10%. 1,000 riders (sorry, nearly 1,000 riders) is not very impressive. The 7 carries about 10,000 riders. The 50 carries over 2,000, while the 36 carries over 9,000. For the most part, this is just Cadillac service that replaces having to walk to the 50 or one of the other buses. The 50 is not that frequent. At noon, it only runs every half hour. My guess is that you would see the same, or better increase in transit ridership by simply improving frequency of the 50.
Or, as Sam implies — you could focus the efforts on another area. This is an area that for the most part *has* all day coverage. So if coverage is the goal, there are plenty of places more deserving. If the goal is to increase ridership, then there are much better values. This is a nice experiment — there is likely some good data here, perhaps arguing for new cross town routes — but this is not a good value, nor is it likely to be a good value in the future. Ridership per hour of service is terrible, well below even the weakest bus in the area (the 50). We have the data, the next step is to use it to build a better bus system.
Many parts of SE Seattle are not close to light rail, and can be as far as 1.5 miles. There are hills like other parts of Seattle. Transfer environments are horrible and ST did no station connectivity beyond rebuilding existing crosswalks. Crosstown routes are not so productive or good to reach local commercial districts. Route 7 is great but except for Mt Baker’s terrible transfer it doesn’t connect to Link in SE Seattle. All of these structural problems in an area where transit dependency is higher than many other parts of Seattle.
I’ve pointed out before that the distances in SE Seattle are shrunk when compared to NE Seattle by Seattle Subway diagrams. Mt Baker to Rainier Beach is about the same distance as UW to Northgate, but the latter distance is shown to be 40-50 percent further. These kinds of diagrams send a message that the area is smaller than it really is. Some even complain that Link is too slow in SE Seattle because there isn’t an awareness about how large the area is.
There is a systemic theme that Link + Routes 7 and 106 equals great coverage. That’s just not the case.
With all the future Link light rail lines, you can say about all of them, many areas will not be close to the light rail station. You can also say about many of them, there will be hills. Bridal Trails won’t be near East Link’s 130th Station. Beaux Arts won’t be near South Bellevue Station. South Mercer Island won’t be near M.I. Station. The list is endless. And there will be hills involved in all of them.
And what are we going to tell the Bridal Trails resident, for example, who wants to get to East Link’s 130th street station a few years from now? We’re going to say, you can walk, ride a bike, take a cab, have someone drive you, etc. That’s what we’re going to say to them. My question is, why don’t we tell the Rainier Valley and Tukwila resident that, as well?
Yeah, that’s the problem. It sets a horrible precedent. The thinking is something like this:
OK, we built this very expensive train, but we don’t have enough riders from low density surrounding areas. We must do something.
I know, we’ll have connecting bus service. That way folks can not only get to the train, but other places as well. Oh wait, we can’t afford that.
Oh, I know, we’ll just run cabs to them. Call it a breakthrough in technology, even though the basic idea goes back over 100 years. Hold on, let me do some math.
Oh no, that is too expensive. It turns out that fixed route buses — even very poor ones — perform much better than jitneys.
Oh wait, we’ll just pay the drivers less than a living wage. Brilliant. It is still more expensive than a typical bus and contributes nothing to the overall network, but as long as it is as cost effective as the worst fixed route bus, we’ll declare it a success. Don’t bother seeing if these riders are simply avoiding other buses or compare it to increases in regular bus service — call it a success and keep it.
I feel sorry for folks who used to ride the 17 in the middle of the day. Where is there free taxi?
Southeast Seattle does not have the best bus service in King County. The places that do are downtown, Pioneer Square, Little Saigon, Belltown, Uptown, part of Capitol Hill, and the U-District.
Link in Rainier Valley functions fine without Via and gets a lot of passengers. Via is just an attempt to give the furthest and most obscure parts of the valley access to Link. You can make an equal argument that the Central District should have Via too. The only reason Rainier does and the CD doesn’t is policy decisions.
Via is just an attempt to give the furthest and most obscure parts of the valley access to Link by sidestepping the labor agreement, hiring underpaid drivers, and giving a relative handful of riders expensive, special service.
There is a hybrid option called time-of-day service. Maybe micro-transit should run only between 11 am and 3 pm, and again from 8 pm to midnight. During peak times, fixed-route buses should be running more frequently. Then, for the same amount of service, more future stations can be served by micro-transit.
I’m also not against micro-transit being funded directly by neighborhoods. Things like higher RPZ stickers and parking meters can find the nearby micro-transit. After all, if we are penalizing driving, should better transit be offered as a reasonable substitute?
Fixed route buses pick up more people during peak time, but that is also when it is more expensive to run them. It is a combination of extra buses and scheduling. The point is, microtransit is never as good a value as fixed route — you inevitably find yourself somewhere on this spreadsheet: https://humantransit.org//srv/htdocs/wp-content/uploads/Mircrotransit-Flow-Chart-02092018.png.
Ride2 did just that in West Seattle. I think the operating hours were 7-10am and 2-7pm, weekdays only.
The rest of the county, outside of Seattle, seems to get by with service coverage on a 2 mile or so grid. I don’t see why folks in Rainier Valley should get better, unless, of course, they are paying for it through a local TBD or through City of Seattle taxes.
It’s to make up For racial inequality. They just don’t say that.
Beware of assuming the current service is the ideal service. Metro’s service in the rest of King County reflects Metro’s budget limitations, not the lack of need. The core routes in south King County should run every 15 minutes including evenings and weekends. Otherwise it’s not reasonable transit and no wonder the overwhelming majority drive.
The use of micro-transit will be popularized once driverless vehicles become viable. Until then, it probably won’t.
Even then it will be a terrible value. When the vans don’t have drivers, the buses won’t have drivers either. Fixed route systems are simply more efficient — they pick up more riders per hour (and its not even close). That doesn’t mean that we won’t have vans, only that they will be fixed route.
I disagree. When you remove the fixed cost of the driver, the marginal cost of operating a larger vehicle vs smaller becomes more significant. Yes, fixed cost will remain the most cost effective option, but in lower density environments, deploying more smaller vehicles becomes more compelling without the fixed costs of drivers. Routes will be psuedo fixed but can be fairly on demand, particularly off peak. You deploy a fleet based upon peak needs, and off peak you can keep that fleet idle but available for negligible cost; with driver’s many vehicles are pulled out of service off peak to save labor costs
Good example: Bart runs long trains at low frequency off peak, because running shorter trains at higher frequency (same capacity) is expensive. With driverless vehicles, you can decrease your deployed fleet while maintaining headways.
“When you remove the fixed cost of the driver, the marginal cost of operating a larger vehicle vs smaller becomes more significant.”
True, but you can still react to that by running small fixed-route buses with higher frequency, as opposed to larger fixed-route buses with lower frequency. There is still a lot of savings in having the route be fixed, since it makes schedules and travel times much more predictable. If the fixed route is running often enough, walking to it and waiting for it will be faster and more predictable and waiting for the door-to-door vehicle, plus cheaper to operate, since fixed-route vehicles require far less deadhead miles than a variable route.
As an example, let’s say you’re over by Seward Park and want to get to Link. If the 50 runs every 10 minutes, walking to it and waiting for it is likely to be faster than waiting for a Via van sitting by the station to drive all the way over to your house, pick you up, and drive all the way back to the station again. Assuming the bus picks up even just 3-4 other passengers along the way, it’s performing much better in terms of passengers/mile than a Via van sent out just for you.
I can see something like Via making sense for people with disabilities who might not be able to walk up or down a one-block steep hill to a regular bus stop, but don’t require a wheelchair or a driver with special training. So, maybe some hybrid system exists where the fixed route buses get dramatically more frequent, while a Via-style system exists for people with disabilities (run with a much smaller number of vans), but with less stringent eligibility criteria than traditional paratransit (maybe people over 80 years old are automatically eligible, for example). But, for mainstream people, the workhorse transit service for corridors like the 50 is going to be a fixed-route bus, like the 50.
As I wrote, vans can be fixed route. That’s not the issue.
The problem is that microtransit doesn’t scale. If you deviate to serve a house, then everyone else on the van suffers. If you operate “on-demand” then you operate inefficiently. You have to have a fleet of vans, stationed throughout the area, waiting to be called. Otherwise, you have long delays, and the wait time is actually worse than running vans on a fixed route.
Running a van from Sunset Hill to Ballard every five minutes would be way more efficient than microtransit, and work out better for the vast majority of riders. You would use fewer vehicles, and there would be less wear and tear on the vehicles. Likewise with service in Rainier Valley. Run a bus on Orcas and Graham, etc., along with running the 50 a lot more often. It is pretty easy to “plug the holes” within the system that way, and get ridership much higher than the best microtransit system in the world.
The dynamic really doesn’t change. Driverless vehicles will simply extend the value of transit (dramatically). This means that pretty much all of Seattle could have frequent transit, if they are willing to walk to an arterial. Those that can’t walk will have on-demand service. There will be very low density areas that still wouldn’t work with fixed route service — areas where a van running every five minutes would pick up only a handful of people each day. But serving those people will always be more expensive, no matter what type of system you use. That means you have the same trade-off. Do you focus on coverage or ridership?
“ The problem is that microtransit doesn’t scale. If you deviate to serve a house, then everyone else on the van suffers. If you operate “on-demand” then you operate inefficiently. You have to have a fleet of vans, stationed throughout the area, waiting to be called. Otherwise, you have long delays, and the wait time is actually worse than running vans on a fixed route.”
The current industry trials are for low-speed autonomous vehicles that are small. At a capacity of 6 seats or 12 riders, the distraction of deviation lessens considerably as most runs would have just 2-4 people on the vehicle. The biggest negative of these are the lower speeds of 20 mph — making a ride over 1.5 or 2 miles unattractive.
There are trials all over the place in these small driverless shuttles — in the US and other countries. I think there are at least four manufacturing consortiums trying to get these small vehicles to be viable. It’s only a matter of time.
Surely, these could be assigned a fixed route. It could be that the system has designated stops that a service could reach in random order or routing. The big advantage of these vehicles is that they can operate on narrower local streets. As guided vehicles, they could even fit into the new, wider protected bicycle lanes because they don’t need a wider street width (if deemed to be appropriate).
“The problem is that microtransit doesn’t scale”.
This is routinely over-stated. Sure it doesn’t scale to the capacity of a 40- or 60- foot bus. If you have a busload of people wanting to go somewhere, run a bus.
But it’s scaling just fine to the target market. Cost per rider is down from about $13 when Via launched to $10 today and still headed lower as ridership grows faster than costs. It’s plenty competitive against fixed route alternatives in that market. There are economies of scale right there if people would just look.
The time-waste of these deviations gets smaller as the vehicle’s service area gets smaller. If the district is all of east and southeast Seattle, then of course it will take twenty minutes for a car to come from Madrona to Seward Park. But if the districts are 1×1 or 2×2 miles, then it only takes a minute to drop off two people five blocks from each other, and three more minutes to get back to the station for another.
“There are trials all over the place in these small driverless shuttles”
There’s Uber too. Just because it exists doesn’t mean it’s efficient or the most cost-effective transit. It just means that corporations are excited about rideshare and driverless shuttles now. Some of that is the dollar signs in selling all these vehicles, because selling ten small cars is more profitable than selling one bus. Some of it is excitement with the suburban vision of everybody going around in microvehicles.
They need to expand Via to serve Capitol Hill station and dad the central district. Getting from cherry hill area to cap hill station either requires two bus routes or walking several blocks. Down hill is fine but the uphill walk sucks. Service running north and south are und 15-th & 18th streets is nonexisant.
There’s the 11, the 2, and the Streetcar all fairly close between those two points. All of them one seat rides. There’s no reason for Via to connect those points. DART exists to assist the disabled.
We need less Via in the region, not more. Using it to replace one seat rides is an awful idea.
I’ve suggested that at least one route needs to connect Mt Baker and Judkins Park stations with Cherry Hill, First Hill and Capitol Hill Stations. Route 8 skirts the areas, and doesn’t go by Judkins Park Station, Cherry Hill or First Hill.
Metro suffered when they proposed restructuring in 2015-16 in a number of ways. I suspect the opening of Judkins Park Station will initiate new discussions. Still, the early planning that Metro has done still hasn’t included one route to tie these places together.
The solutions are not easy, and will probably result in pretty complex route restructuring. I could not only see a Routes 3 and 4 affected, but also Routes 48, 27, 2, 8 and 14 — and maybe even Routes 7, 9X and 106.
Of course, no one thought to design a bus (or any other vehicle) turnaround on 23rd or on Rainier at the Judkins Park Station; unlike other RV Link stations this one won’t have adjacent small blocks for turning around. That makes something like Via even harder to implement.
Cherry Hill (as a hill, since no distinct building has been mentioned) is less than 10 blocks away from 23rd. The 48 basically serves it now. Granted it goes to UW Station and not Capitol Hill Station, but we’re still talking about some of the best grid in Metro’s system today and Link is more than adequate to get people to Capitol Hill. There is neither need nor reason for any restructuring for quite some time. Let the rest of the city improve their mass transit options to within a shadow of this area’s before getting greedy and asking for more.
It is not about the Central Area being “greedy”, it is about building a better transit network for the most densely populated part of the city. The area is long overdue for a restructure. They nibbled around the edges when Link went through, but the sparsity of stations (only one between the UW and downtown) made it very difficult. Likewise the station at Judkins Park won’t change things dramatically. But Madison BRT will. Partly because it forces it — some routes will have to change. But also because it will have a lot of stops and extremely good frequency, making transfers a lot more appealing.
There should be a grid. Making things more complicated, there needs to be essentially two grids — one that follows the cardinal directions, and one that follows the 45 degree angle of streets like Madison. The biggest missing connection in the area is a bus along Boren, connecting First Hill with South Lake Union. But there are other big holes in the grid, and Cherry Hill is one of them. To get from Cherry Hill Swedish to Kaiser takes forever on a bus (https://goo.gl/maps/Mj84Zgt8cW3TWBRF8). Solving that problem wouldn’t be easy. You can run a bus on 15th/14th — making a grid. But that misses CHS by quite a ways. You could send the bus to Link once it reaches John, but that gets pretty messy, pretty fast (you end up with low frequency buses). You can not even worry about it, and just extend the 4 down past Judkins Park. But that mainly works as a connection to the East Side. If I’m headed downtown, I’m taking the 3/4. If I’m headed to any of the Rainier Valley stops (or places south, like the airport), this doesn’t do me any good. If I’m headed north, then the 12 minutes it takes to get from Judkins Park back up to CHS are enough to send me scrambling for a better option. I’m not saying that extending the 4 past Judkins is a bad idea — I’m just saying it doesn’t solve the problem that Eric mentioned. Getting from say, Northgate to Cherry Hill will likely always be a bit of a pain, requiring a three seat ride.
But that isn’t the end of the world. We are going to have to get used to three seat rides. I’m sure there are a lot of people who go between Lake City and Belltown, and that will require a three seat ride (or a lot of walking). In this case, Northgate to Cherry Hill will require a three seat ride as well. But Northgate is far enough north to where Link makes up for it. Lake City to Cherry Hill is worse, because it would likely involve taking the 48 (or a four seat ride). UW to Cherry Hill is a two seat ride (again, involving the 48). But that is just the way it goes — there is no cheap and easy way to solve the problem. The best we can hope for is something enables a lot more fast connections, even if some areas (like Cherry Hill) remain challenging.
By the way, this is yet another argument for a Metro 8 subway. Build that, and a lot of these issues vanish.
Absurd. An area with 4 routes within a 14 minute walk away is in no way “long overdue for a restructure”, and demanding such an area get a restructure is the height of attention seeking behavior. Senpai should ignore you.
This is adjacent to downtown and one of the densest parts of Seattle. A 15-minute walk in Broadview corresponds to a 5-minute walk here. Since it is dense, it can fill up closely-spaced buses more easily than outer neighborhoods.
Metro has a kind of solution in mind. Its 2025 plan wants to turn the 60 into a Broadway-John-12th route (going north to the U-District and south to Othello). Then there will be a route to Capitol Hill Station at the bottom of First Hill, a 5-block walk from Swedish Cherry Hill.
I wholeheartedly reject your temporal revisionism. 15 minutes is 15 minutes, no matter your location. Time and space do not magically warp with population density.
A 15 minute walk to a bus/light rail stop is not some insurmountable burden with mass transit. For those incapable of such an act, we have DART. For everyone else, we have two feet.
I have to laugh about how many posters are not concerned about Cherry Hill not being very accessible, but so passionately declare that West Seattle Junction or Ballard Link stations are too far from where they need to be. The issues are about the same walking distances and the costs to change a few bus routes (maybe not even increasing net service hours) pales in comparison to hundreds of millions in subway construction.
I stay out of West Seattle Junction or Ballard Link station arguments. They don’t impact me. I would object to Via serving either one, however.
People’s expectations are different in denser areas. Link’s stop spacing is 1-2 miles outside downtown and 0.25 miles inside it. San Francisco has a denser bus grid north of Market than in the western and southern quarters of the city, and all of those are denser than the grid in Oakland or the suburbs. In downtown-adjacent SF people expect to walk five blocks to an always-frequent line. In outer neighborhoods they may walk 10-15 blocks to a less-frequent line. So the issue is not the concept that denser neighborhoods should have a denser transit network, but what level is appropriate for the western Central District.
I’d rather err on overestimating than underestimating, because the entire problem with our postwar transit-network design is we keep underestimating what is needed. That depresses ridership and the usefulness of the network. We need to optimize it. You wouldn’t tell somebody in SF’s Chinatown to walk 15 minutes to a bus stop; they’d think that’s inappropriate for such an urban neighborhood. And they’d think that’s “so American”.
“People’s expectations are different in denser areas.”
Peoples? Or yours? Have you done or seen any study supporting this position, or is this just “I think this, therefore it must be true for the majority.”?
“Getting from cherry hill area to cap hill station either requires two bus routes or walking several blocks.”
“There’s the 11, the 2, and the Streetcar all fairly close between those two points. All of them one seat rides.”
None of those routes goes to the 17th & Jefferson area. There’s a general transit hole running northwest-southeast from SLU to First Hill to Cherry Hill, and Cherry Hill to Capitol Hill is also missing, and southwest Capitol Hill to the Harborview area. I live at Pine & Bellevue and occasionally have appointments at Swedish Central or Swedish Cherry Hill. There’s no reasonable bus route. You can take the 49 to Broadway and walk, or take the 11 to 17th and walk. Essentially, there’s transit southwest-northeast thanks to Madison Street and the 12, but there is nothing for trips northwest-southeast. And while it’s reasonable to transfer between Lake City and Fremont because it’s a long distance, it does not make sense when the distance on each leg is a mile or less, especially given that you might wait 15 minutes each for a transfer, for a total of 30 minutes if you just miss all the buses. There are arguments that the street grid doesn’t allow a northwest-southeast route, but at least acknowledge that it’s a hole in the transit network, in the highest-density area after downtown, and an optimal solution would have grid routes and routes on both diagonals, and we should try to make the network better.
I too have doctors at Swedish Cherry Hill. That’s why I know the mass transit around it is robust. You’re less than 10 blocks away from the 48. The Streetcar is 12 minutes away according to Google, and the average person can walk faster than Google’s approximation.
There is no transit hole at Cherry Hill. It is served by numerous routes going in many directions. Most of Seattle would do anything to get its kevel of service.
on weekdays, routes 3 and 4 provide eight trips per hour per direction on Jefferson Street, so the average wait is 7.5/2 or 3.75 minutes. that makes a one-transfer trip pretty reasonable. Routes 3 and 4 have transfer points with Link on 3rd Avenue, with Route 60 at 9th Avenue, and with the FHSC on Broadway.
“ Absurd. An area with 4 routes within a 14 minute walk away is in no way “long overdue for a restructure”, and demanding such an area get a restructure is the height of attention seeking behavior. ”
Given the recent significant densification of Capitol Hill, the CD and First Hill combined with new Link Stations (Capitol Hill and Judkins Park) a restructuring is more than warranted. Even if routes don’t change, the concept deserves study. It’s not like many other areas of Seattle that have only had modest infill and no Link stations opened or under construction.
Yes. It’s still not clear what “the concept” is, but at minimum we should study whether all parts of east Seattle and north Rainier have adequate access to Capitol Hill Station and/or Judkins Park Station, and whether all trip corridors in central Seattle are adequately served, and whether parts of the CD not near RapidRide G should have better access to a downtown Link station, or whether directing them to Capitol Hill or Judkins Park is sufficient.
“Some customers told Metro that Ride2 service connecting to the water taxi, which sails every 35-40 minutes, was inefficient compared to transit that operates more often, including the minibuses on Routes 773 and 775.” -Seattle Times