[Important note: this is not an editorial advocating anything. It is an attempt to quickly analyze some planning tradeoffs.]
When Central Link opened, Metro eliminated some redundant lines (194, 42X), but the bulk of I-5 routes from South King County continue to run into Downtown Seattle, at great cost and with a parallel Link line able to carry those people at nearly zero marginal cost. The best place to make a transfer from I-5 is at Rainier Beach Station, although that location has serious problems. What are the tradeoffs?
First, Google Maps pegs the driving time from Exit 157 (MLK) to Sodo Station as 10 minutes, vs. 3 minutes to MLK & Henderson. Link takes 15 minutes from there to Sodo, meaning that in the baseline scenario the transfer is 8 minutes slower. If you like, subtract a minute for buses being slower than cars.
To this time loss, one must add walking time. If Metro did the user-friendly thing and put the stop on the south side of Henderson, there’s a crossing of two not-crowded lanes. Experience at other stations suggest an easy jaywalk that might take a minute, more for those that insist on a signal. Moreover, there will be wait time: for bus to train, the average varies from 4 to 8 minutes depending on time of day; train to bus, typically 8 to 15 minutes.
On the other hand, the Rainier Beach run is essentially uncongested, while I-5 is often not. The gap shrinks with any delay on I-5. Also, the further north the destination, the smaller the gap is. Going north from Sodo, trains have signal priority and limited-stop advantages over buses, and in some cases a tunnel advantage as well. This adds up to between zero and 6 minutes savings to Westlake in the schedules.
Although in some cases time may be a wash, in general trips with this transfer will be slower and this will cost riders. Unlike the 194/Link tradeoff, all of these trips will still require a bus and won’t attract those that refuse to get on one.
On the other hand, by my count Metro runs 295 trips every weekday via I-5 and the busway, plus 179 on Saturday and 117 on Sunday. With a half-hour of savings per trip, that comes out to around 44,000 service hours, of which 26,000 are charged to the South subarea. For comparison, a draft service cut plan last year took about 45,000 service hours out of the South, and that’s before union concessions saved another 20% of threatened service hours. Alternatively, those hours could be redeployed elsewhere, even by roughly halving headways on routes like the 150.
Other fringe benefits include much improved connectivity between Southcenter and its transit-dependent customers in the Rainier Valley. At some operating cost, buses may terminate at Rainier and Henderson, where there is more bus layover space.
Restored service on the low-ridership southern end of MLK is canceled out by reduced connectivity at the Spokane St. busway stop.
Such a shift also frees up capacity in the busway and DSTT, which can provide a new route to West Seattle post-viaduct and improve reliability on Link.
There are also 198 weekday and 68 weekend trips on Sound Transit buses on the busway, all of which are charged to the Pierce County subarea. Sound Transit and Metro also run 84 and 6 weekday trips, respectively, and 62 weekend ST trips, that bypass the busway to go right into Downtown Seattle. I’ve omitted those from the analysis above.
Whether this kind of service change is worthwhile depends ultimately on how you comparatively value intra-South County service with respect to the quickest possible connections to downtown. Everything has an opportunity cost, and resources can be spent on expresses into Seattle or into reducing headways between suburbs. High capacity transit services like Link provide an option, albeit imperfectly, to avoid gutting one to emphasize the other.
“A bird in the hand is worth two in the bush”, says Gieco, or just ask any Metro transit planner, only the phrase goes something like this – “A bus at the zone is worth two on the schedule”
RBS is a horrible idea for routes originating in Kent, Auburn, or Federal Way. There is NOTHING in it for the riders. No time savings. No superior ride, No reduced cost. That’s going to reduce the number of daily riders, which is counter productive to getting people out of their SOV’s.
Show me the benefit for the riders, then quantify it, and maybe I’ll be convinced.
Also ask a transit planner what the 2/3 rule is. Roughly it says that once a trip is 2/3rds complete, riders have a very high resistance to a transfer. RBS is at least 2/3rd the way into downtown from the origins I just mentioned.
Also, staging enough buses for the reverse commute home is problematic at RBS, assuming you plan to do the same for the PM trips.
I’m hoping Martin will have a “Let’s Fix Link” posting for 1/1/11. There’s a lot of good ideas floating around to get more butts in the seats. Let’s all put on our thinking caps, as Martin has done.
I can see some transfers at RBS, but the I-5 trips are among my least favorite.
In fairness to Martin, I know this is a planning exercise, and not your personal proposal. These are the kind of discussions that NEED to take place to take full advantage of the sunk cost in Link.
Mike, the benefit to riders is that other services aren’t cut or even improved. As an isolated example, you could truncate the 150 and halve its headway. Then the time difference would be smaller and you’d have much better service between Southcenter and Kent.
The 1/3 – 2/3 rule is all well and good, but can have silly results. It would tell you that transfers at Tukwila are better, although in fact they’d be slower.
My hypothetical response would be this. TIBS is a horible place to truncate routes off I-5 for the very same reason that RBS is. There both too damn far off the direct line people are used to traveling.
IF Link had a station at South Center, with easy HOV direct access to/from the bus bays, and lot’s of layover space, I’d say truncate them all. That’s the difference.
I’m not sure Metro would redeploy the hours anywhere, as the whole point would be to balance the budget, so the real result is a loss of service, with no benefit, or maybe a high productivity route being replaced by one with a lower productivity.
So, since we are brainstorming here, it doesn’t seem that it would be too difficult to put in a Link spur line to Southcenter to do what is being proposed here. In relative terms, a smaller investment, as it would connect to the current line for the trip Downtown.
Of course, everything is difficult. But in a perfect world….
there is something in it for the riders. More frequent bus service, and/or more bus routes.
I think the way to go here is the same as with 520 service. Keep the peak single-seat ride service and use Link transfers for midday/evening/weekend service. Sure it means a longer trip if you are riding during those times, but forcing the transfer means greater service frequencies can be offered. Which would you prefer an every-30 minute all-day route to Kent and Southcenter that requires a transfer at RBS or an every 60 minute one-seat ride? The latter scenario is the future if we don’t get away from the single-seat ride mentality.
For Kent, surely the peak single-seat ride should be provided by Sounder?…. just brainstorming here….
Martin’s concept would gain more traction if 5 minutes could be shaved off Link’s schedule between RBS and Westlake (20 instead of 25). I believe it is technically possible once ST/SDOT/Metro gain more experience with joint tunnel operations and MLK signalling priority. Quickening Link will also make it a more attractive commute option when the line extends further south.
There is really only one minute to gain from getting the train into the tunnel faster, and might come at the expense of making loaded buses wait if a train is coming in the next couple minutes.
The light signalization is already working quite well. The current travel time is when there are no red lights.
The best opportunity for travel-time decrease is if Link can be allowed to run 55 mph at-grade. However, without physical crossing barriers, that would be a rather unsafe idea. Last I heard, ST has no plans to install crossing barriers.
So, here’s the conundrum: If the accident rate drops to nil, Link will be stuck with 30 mph at-grade in perpetuity. If the accidents continue, crossing gates might still happen, which could lead (with neighborhood and legislative concurrence) to Link being allowed to go 55 at-grade if it is separated from traffic.
those i5 express buses from the Southeast, should replace their highway link with Sounder from Kent Station.
southwest federal way will soon have link service directly there.
“If Metro did the user-friendly thing and put the stop on the south ”
What kind of parallel universe are you living in? Look at the transfers at Mount Baker and Husky Stadium (under construction).
Touche. There are, however, fewer obstacles to doing the right thing here than in those other places, especially if you make the turn east to RB proper.
I’d like to see a “Let’s fix Metro” post for 1/1/11 (get to work! now! :-). Link’s immediate future from Northgate to S 200th St is mostly determined. Once you decide where to put the tracks and what your headways will be, there’s not much left to optimize. You can, however, optimize the connecting service, and get more transfers that way. Moreover, Metro will soon be FORCED to optimize its service by budget constraints. When that time comes, it would be nice to have well-thought-out ideas to advocate for.
Reading back through STB posts, there was a lively discussion on rationalizing Ranier Valley service (notably including Route 7), and there were talks of reorganizing routes 3 & 4.
“There is NOTHING in it for the riders”
While I agree that terminating I-5 routes at RBS is not optimal, I’d hardly say there is “NOTHING” in it for riders. The reduced costs could either prevent large service cuts in the future, improve headways, or both. In the context of looming service cuts, developing a plan is prudent.
There is also the connection with the airport that would instantly be added to ALL of those routes at that location – Possibly a small silver lining?
Velo, if you’re talking about riders in general, as a group, then yeah, maybe some benefit. If you’re talking about the guy riding from Federal Way to work, and propose kicking him off the bus at RBS, in the rain, in the dark, in the cold, and ask him to walk to the Link Platform to wait for the next train, then I’d say you just lost a rider. Especially when his bus would be arriving in downtown, about the time he’s looking for a seat on Link.
Another factor to throw in the hopper. Metro would have to share all their bus fares on that route with ST as part of their division of revenue for transferees. I have no idea what the formula is, but lost revenue is lost revenue, and maybe it tilts the idea in one direction or the other.
Yup, like I said, “Not optimal”.
As to lost revenue, for the system as a whole, the revenue is unchanged – assuming everybody uses an ORCA card to get transfer credit. But ST ends up with more revenue and Metro ends up with less. That said, virtually every Metro route runs at a loss so it might be close to a wash.
Metro wouldn’t necessarily lose revenue on the truncated routes because they would have higher passenger turnover than the former express routes. A bus route whose frequency is doubled using the same number of revenue hours has twice the opportunity to pick up fare-paying customers. Freeway express route are the worst performers becuase they have no opportunity to generate revenue along the express portion of the route.
And freeway express routes have worse on-time performance than even local Seattle routes, according to Metro.
If the transfer was limited to the all-day service then the guy riding in from Federal Way may very well still have the option of a single seat ride for his commute.
You know, there is that extra track at Rainier Beach, in between the two running lines. I wonder if it’d make things more palatable to throw a couple extra trains in the mix, that ran as ‘Express to Rainier Beach’. They could pull into that siding and turn around, instead of running all the way to the airport. You probably couldn’t have them be a true express and run without any stops all the way from downtown to Rainier Beach, because they’d run up onto the train ahead of them, but it might be a way to add some capacity/frequency between RBS and downtown. I dunno, just a thought.
I doubt this is possible, given the tight spacing of trains at rush hour and the limited number of trains allowed through the MLK portion of the alignment.
I asked one of the ST people way back when about Express, or ‘Skip-Stop’ service.
He stated the concern about the ‘kill-zone’, when people are in places they shouldn’t be.
What Express service does at local stations is extend that zone.
In other words, people quite possibly would step in front of a moving train, making the assumption that it will be stopping at the station. Unless they read the sign on the vehicle, they would be SOL.
If there isn’t a way to make it obvious that “This train isn’t stopping!”, this seems to be a non-starter. At least it was in the eyes of the engineer I spoke with.
The express run would defeat part of the reason for connecting at RBS: connectivity to major Rainier Valley destinations (including those on Rainier Ave, especially once the 7 gets its trolley wire and reaches RBS).
However, a train waiting for the 101 or other terminating routes to pull up could have potential, if the regular train schedule can’t be timed to these buses.
One further twist on the express train comes to mind: Have some low-use stations have stops only in one direction during certain periods.
For example, during morning peak hour, imagine having all the northbound trains skip certain stations, like Othello, Columbia City, SODO, and Stadium. Passengers could still get the southbound train from those stations and then transfer at the next peak 2-way station (or, in Rainier Valley, just catch the 8 north). If skipping a station saves a minute, that makes Link much more competitive for choosy commuters.
Or do the reverse, and have southbound stops that have a history of few boardings/alightings get skipped.
I’m not advocating a particular plan. But I just wanted to throw another set of ideas out there for people to chew on and bend.
That’s an interesting idea. The RV stations show a very asymetric demand with (not surprisingly) 3X more people getting on northbound and off south bound. At best though you’d only save maybe 5 minutes and likely less than that. I don’t know if the faster trip time will bring in as many people as you’d lose from eliminating stops (back of the napkine calc = lose ~6% of total ridership and need an increase of ~10% at the “big six” to break even). Gut reaction is that you’d end up with a pretty big hit to ridership because the peak direction there would be no increase in speed and only ~1 minute reduction in headways. Another consideration though is the reduced operational costs. Accelerating even a “light” rail vehicle up to speed consumes a lot of power. It would be interesting to compare the cost savings to that of shortened trains.
To make up for Othello and Columbia City Stations being skipped one direction or the other during peak, the frequency on the 8 (or at least the portion of it from RBS to MBS) could be amped up during peak hours.
If that were combined with crossing gates, 55 mph on MLK, mothballing SODO Station, and not allowing stops at Stadium Station in peak-direction runs, plus smoothing the tunnel algorithm so the train doesn’t sit and wait at the station entrance, the travel time from RBS to IDS could be reduced from 19 minutes to roughly 13 minutes. That would hopefully dissolve the arguments against truncating the 101.
Except that if you skip Othello and Columbia in the NB direction AM and SB PM then you lose more than 10% of your ridership. If you do it in the opposite direction you’d lose only about 3% ridership but you don’t speed up the train in the direction that matters. Maybe with significantly beefed up bus service you could get away with skipping Rainer and Columbia entirely and make all stops at Othello (or some other station, which one has the most development currently underway?). If crossing gates are all that’s required for 55mph running then that’s probably a very worth while investment. Seems like Sodo and/or Stadium should definitely be on the block.
Interesting idea but the whole idea about building a regional rail system is that you don’t have oddities like this. You get on a train and it stops everywhere that the line map says, no variation.
I’ll throw out an odd ball and probably dangerous idea. What if say what’s now the NB track became and express stopping only at SEA TIB ID and W. Lake. SB track would be the local making all stops. I already know someone’s going to point out that this would mean 1 hr headways if only one train was running the local route so you have to have multiple trains and do some fancy switching. What can go wrong, it’s all computerized, right :-0
To provide an example from another metropolitan area: Portland’s MAX Yellow Line light rail opened in May 2004, and included a stop at “Vanport Transit Center,” right off of I-5, with the idea (as I understand it) that Clark County C-TRAN express bus routes would stop there instead of going to Downtown Portland.
Less than a year later, during a service cut process (that was later avoided when voters passed a ballot measure on Sept. 20, 2005), an overwhelming majority of C-TRAN express route riders said they would no longer ride public transportation if they did not have a single seat transferless ride to Downtown Portland. So, express service continued, albeit under a new fare structure. ($3 instead of $1.75, no transfers issued/accepted; monthly passes increased to $105 from $52. It should be noted fares and the transfer policy have since been revised.) It appeared to be successful. Later on, in C-TRAN’s 2007 CAFR, it was revealed that although the new fare paid for 74% of the costs of operating the service at the time, ridership on those routes decreased by 40%.
C-TRAN service to MAX at Vanport TC didn’t start until 2007, and consists of its highest ridership (as far as I know) local route, and three “limited” routes, two of which operate one inbound AM trip and one outbound PM trip. Express bus service that ‘parallels’ the light rail continues to this day.
Jason,
I use the C-Tran Express system regularly to commute from North Vancouver to Nike (catch the Max from downtown and Nike provides a shuttle to campus from Beaverton Creek — it’s also not too bad a walk). It typically takes about 40 minutes to drive in the rush morning hour, and only an hour to do the two vehicle transit trip because of the Robertson Tunnel.
The people on the bus are about 60% middle-class office working women with the rest middle-class office working men. They are not going to ride the Yellow Line. Period. End of Story.
I know because I sometimes have to work later at Nike than the expresses run back north in the evening since I’m a database contractor. It takes from two to two and a half hours and four vehicles to get back to my car at the 99th Street P’n’R and costs only 90 cents less. A very bad deal.
When I talk about my journey home the express riders shake their heads. They simply will not transfer.
Now, you can say, “Well you live with a bunch of selfish Republicans” and Clark County is certainly a Republican sort of place. But at least they voluntarily ride a transit system that has a pretty estimable fare recovery. The fares rose 8% (from $3.00 each way to $3.25) last September and ridership didn’t blink. I think it could probably go to $5 each way and folks would still ride it. It’s still much cheaper than parking and driving.
But they wouldn’t do it for $1.25 (the base fare to get to Delta Park) if they had to change to the Yellow Line. They. Would. Not. Ride.
The same is likely true of the express users in King County. Don’t mess with a good thing.
I think it really depends on the route in question. For peak Express service from the South or across the lake you are likely right. However with the line to the north even with a transfer to link at one of the stations from Montlake to Lynnwood it is likely the trip times will be signifigantly reduced due to the rather horrible peak congestion on I-5 North of Downtown. 13 minutes between Northgate TC and Westlake is hard to beat even when there is no traffic much less during the peak of the AM commute. While the projected travel times to Lynnwood TC aren’t nearly as good they are still very comparable to current express service, even one of the SR-99 alignments is chosen.
I wonder if it would be different in Vancouver/Portland if the Yellow Line extended across the river. I would assume that the river crossing is a time bottleneck and that getting the train across the river could change the comparative runtimes meaningfully….
Nathanael,
It wouldn’t really help. The Yellow line simply makes too many stops. There are eleven between Pioneer Square and Delta Park where the interchange to Clark County buses occurs. If the train crossed the river it would still stop at Expo Center, plus it would have a new stop at Jantzen Beach and then three stops through downtown Vancouver and a final one at the Clark College terminus where bus transfers are supposed to occur.
Right now it takes exactly one half hour to get from Pioneer Square to Expo Center. Add all those other stops and it’s probably going to be 42 to 45 minutes between the college and the Square.
The I-5 expresses are scheduled at 25 minutes from 99th street to Alder (one block north of the Square) except the four most congested runs which are scheduled at 35 minutes.
Plus, the Yellow line will not cross the river unless it’s paid for as a part of the CRC, which will provide all-day HOV access removing the worst (though certainly not the only) choke point for the buses.
All in all, the EB’s are better for riders, though they certainly need to charge somewhat higher percentage of their operating costs.
What this painfully points out is the poor choice of routing Link through the RV. It was designed to promote urban rejevnation of the RV and not to move people into DT from south of Seattle. Eventually south sub area riders are going to be forced to transfer. So if direct access from I-5 HOV lanes isn’t feasible to TIB Station then the only “fix” I can see is building out the Boeing access road station complete with transit only ramps from I-5 or shuffle riders off to the South 200th Street station when completed (which is probably what’s going to happen since it maximizes bus hours/miles saved and avoids the I-5/405 bottleneck). Either of these is going to make it really obvious how stupid it is to meander through the RV and Beacon Hill when the overwhelming majority of the ridership is just passing through. Link stops are too far apart to be of use to most people in the RV. It would need to be more streetcar like. But forget about infill stations when the problem is the train already adding 10-15 minutes to the commute of the majority of the ridership.
If/when we build a West Seattle line, would it be logistically feasible to make it basically an express through to S 200th St? Route would be something like: Downtown, Alaska Junction, White Center, Burien, S 200th, and maybe Renton? To be faster than the Ranier Valley it would need grade separation all the way; is that possible in that area? I don’t know it well.
I would think at this point the West Seattle line is about 15 years out
@groan I know :-( Which makes me sad, because now that I think about it, it seems to make lots of sense.
Given that Central Link headways will be limited Downtown by the interlining with East Link, any extension south of S 200th St would have spare capacity. If it were possible to make it so that “West Link” trains could join Central Link tracks and head south, that would provide a regional benefit making it justifiable as part of ST3 or ST4.
I’m sure I’m not the first person to have had this idea… I’m curious what the more knowledgeable here people think of the practicality of it. Could we get the ROW? There’s no point doing this if you have to run 35mpg at-grade half the time.
Express trains require 4 track ROW, almost always. I’d rather have more route miles for those $ than 4 track ROW.
@Lloyd I should have put Express in quotes. I’m suggesting a two-track line with three stops between Downtown and S 200th St. It would be the quick way to get downtown from South King, “Express” relative to the current Ranier Valley alignment.
West Seattle is not going to get a light rail connection to DT Seattle. The entire population of W. Seattle is about 60,000 and it’s a dead end meaning the line has nowhere to extend and draw ridership. It would be even more silly to route a line from S. 200th to DT via W. Seattle than it is via RV. There’s 90,000 in the City of Federal Way, 114,000 in Kent, 130,000 in Bellevue, 50,000 in Redmond. These are all large employment center verses a bedroom community (virtually nobody commutes from DT to West Seattle) and light rail ridership is still marginal at best for Central and East Link.
White Center? Burien?
Again, nobody wants to go from DT to Burien or White via West Seattle. The only people that travel through West Seattle are from Vashon. The employment population ratio is about 60%. So there’s a workforce of about 36,000 people in West Seattle. Let’s be generous and say half work DT and even more generous and assume a third would ride the train. That’s only 6,000 people making the commute. Add in a handful of shoppers, UW commuters, etc. and it’s pretty apparent you’d never even get to Central Links current miserable ridership.
https://seattletransitblog.wpcomstaging.com/wp-content/uploads/2010/01/spr_2009_ridership.jpg
All I can say is that Metro sure runs a lot of busses out there. Weekday ridership (approx)… 120: 7k, 21+22: 3k, 55: 2.5k, 54: 3.5k, 56: 1.5k.
Even if only half those people take the train, we’re way above 6k. Those Metro hours could be reallocated to provide a feeder service that could drive ridership up. I’m not buying your pessimism on this, although you seem down on Link generally, so I wonder if you are simply projecting that?
6,000 commuters is 12,000 daily ridership. Even if every single person currently using the bus routes you mention rode the train you’re only at 10,500 daily ridership. Double that and you’re at today’s Central Link ridership. That doesn’t cut it when you’re talking about the cost of a high level bridge, substantial elevated sections and expensive ROW acquisitions. It doesn’t make sense to run all the feeder routes and force a transfer just so people can travel the last four miles by train.
I’ll admit that I can’t rebut your numbers with the data at my disposal, and you might be right about the population and ridership for the next few years in West Seattle/White Center/Burien. However, lots of people who’s opinions I generally trust seem keen on the idea; I’d like to think they weren’t all wrong.
I take issue with it being a bad idea to force transfers. Pretty much our entire strategy with Link is to have fast, frequent trunk service with frequent feeder routes (or Park & Rides.) I admit it’s marginal for people at Alaska Junction but it seems to be a clear win at White Center (which is nearly 8 miles) and Burien (11 miles.)
Why do you keep bringing up Burien and White Center? If you want these people to transfer from a bus to a train terminate at TIB station. Build enough ridership and maybe a direct line from the airport to DT might get built and the RV portion can then get the number of infill stations it needs to be actually useful for everyone in the RV. The whole point of Martin’s post was to try and drum up ridership for a line that’s already built and failing to meet projections. Nobody wants to go from Burien to DT via West Seattle. 509 is a direct shot, 5.5 miles to DT. Instead you want them to take a bus to a train that then takes longer to get where they’re going? West Seattle is an established neighborhood that’s not going to significantly grow over the next 10, 20 or even 50 years. And, because it’s geographically isolated it shouldn’t. If fixed guideway is ever to be viable it would need to be much more like a streetcar than light rail.
“Nobody wants to go from Burien to DT via West Seattle.”
You must not be familiar with Route 120: DT Seattle to Burien via
Delridge and White Center. Runs every 15 minutes all day. I rode it
a few times and see a lot of people use it to get to White Center
from downtown. I’ve heard it was considered a candidate for RapidRide.
I don’t know why anybody would want to ride the 120 end to end and take 46 minutes when the 121 does it in 22-29 minutes via SR-509. Let’s put it this way, if there were an express (which there doesn’t seem to be enough demand for) which route would it take? The 120 has decent ridership because it stops every few blocks.
ST did an Issue Paper on Burien to North Renton (via Southcenter and Tukwilla Sounder/Amtrak) light rail. I expect this to show up in ST3 or ST4. If a West Seattle light rail line is done I would expect it to eventually run all the way to Burien via White Center, Morgan, the junction, etc. I’m not sure if it will all ultimately be the same line, but that would be the logical routing based on past planning. I also suspect W Seattle might show higher ridership than expected due to the limited road routes to the neighborhood.
Put an express bus (not RabidRide) with stop spacing aproximating light rail. Run it on 10-15 minute headways 18 hours a day. When do you think ridership will exceed that of an articulated bus? There is no node of population and employment (like SEA, DT Bellevue, Microsoft, UW) or place for it to develop. and it’s a dead end! It’s only slightly bigger than Kirkland which has far more potential to grow so even if there were money for it in the next 20 years there are many many extensions to Link that have more merit and cost far less. Like extending East Link to Redmond proper. And Woodinville/Bothell from there is on the Redmond spur is dirt cheap.
If you have stops at Spokane & Delridge, Fauntleroy & Avalon, and Alaska & California, you can run very frequent, RapidRide-style bus service all the way up and down California, 35th, Delridge, and 16th (the latter three terminating at the stations) and get a very significant number of riders from transfers. For the few riders you will lose because they don’t want to transfer, you will gain a ton of riders for the frequent service.
You would also get a ton of riders from the Junction and Triangle station areas. Thousands and thousands of people are already riding the bus just from the Junction and the Triangle to Downtown and SODO, and vice versa. When you add in all the extra people who ride light rail but refuse to ride the bus (and there are a lot of those, as the area is mostly upper-middle class), you get a pretty significant ridership number. But that’s all assuming that the areas density stays the same, which it won’t; the Junction/Triangle area is growing like wildfire, and if they extend the light rail down there they could designate the Junction an Urban Center rather than Urban Village and raise the height limit dramatically.
Finally, that wouldn’t be the long-term end of the line. If extended down to Morgan Junction, White Center, and Burien, all grade-separated of course, it could eliminate the quite large number of express routes going down there and stimulate a lot of growth.
A Link line down to West Seattle from Downtown makes a lot of sense. It makes sense in the present, with the large number of existing riders at the Junction and on the corridors extending north and south from possible stations, and it makes sense in the future, with a lot of potential for densification around stations, and potential for future extensions.
I think you may be on to something with the RR type of service. That doesn’t come close to scaling up to light rail though. Piece meal development projects don’t add up to DT Bellevue, Microsoft or UW which is the minimum you need to start even thinking about light rail; especially when you consider the cost of this line (high level bridge, expensive ROW aquisition). Frequent express bus service maybe someday but even the 550 isn’t enough to consider a light rail replacement if it weren’t for the transfers at M.I. plus Microsoft. Even with all that it’s sketchy. There’s no vast tracts of unused land in West Seattle like there is at say Totem Lake. But closer to home (or DT actually); if development is your gig then why not the vast industrial area from George Town to Tukwila? Jobs, large tracts of land to build on, it’s not a peninsula that requires an expensive bridge to be built and developers would likely provide the ROW and build the stations for the privilege of being on the rail line.
Given the fight over transit lanes for the current Rapid Ride line to West Seattle I think the chances of gettting the ROW needed to provide proper BRT service that doesn’t get stuck in traffic are slim without spending as much as light rail to the same area would cost. Without the ROW Rabid Ride is the best that can be hoped for.
With sub-area equity or City of Seattle funding a West Seattle line won’t be compteting with one to Redmond, Issaquah, Kirkland, Tacoma, or Everett for funding.
Bernie: ST at one point tried to run along the Interurban ROW into S.Center, then back up the hill.
It got torpedoed from both directions. Seattle didn’t want to share tax revenue from Northgate to people going to the ‘other’ mall, and Tukilla didn’t want all those ‘trouble makers’ from S.Seattle hanging out in their mall.
So it was never meant to be, according to someone who attended the meetings.
Boeing Access with flyover ramps for buses, wouldn’t be too bad a place to do the deed.
I’ve never understood how light rail from DT to SEA would ever been able to have a stop a Southcenter Mall. DT and SEA are both on the west side of I-5 and that’s where a rail line between the two points should have stayed in order to serve the bulk of the people trying to access DT from the south. The number coming from RV and Beacon Hill will always be a tiny fraction of those coming from Kent, Burien, Des Moise Federal Way, etc. It was dumb to try and make Link perform two fuctions that are at odds with each other. You end up stuck with a system that does both poorly.
As I recall, Tukwilla WANTED Link to serve Southcenter. The reason that alignment was dropped was an issue of cost and travel time. The costs weren’t within the South King sub-area budget for ST1 and the increase in travel time was enough to put Federal grants at risk.
Tukwilla did fight the later SR-99 alignment which is why Link follows 599, I-5, and 518 instead.
There was no conspiracy involving the City of Seattle, Northgate, or Downtown Merchants.
Chris, sometimes the public dialog is at odds with the private dialog for face saving reasons. Yes, it was a cost v. time issue to the public.
If I remember from my hazy memory of those scoping meetings, (although I do have the documentation in boxes at home… somewhere), the cost of that alignment was high enough that even though there was an increase in ridership, it wasn’t enough to justify as big of an expense as it was.
I seem to remember it was something to the effect of “It costs 4 times as much for that segment (Boeing Access Rd and Marginal way to the current location of the Tukwila station) to go to Southcenter, than ST’s preferred alingment up Hwy 99, but with only double the ridership for that segment.”
Don’t take these figures verbatim, though. They’d have to be researched, but the ’tilt’ was too great in the wrong direction.
Bernie,
Stadium Station is used all day by operators going to/from the bases to their exchange points (for bus routes that do their operator exchanges by Link stations). This saves Metro some costs in extra supervisor vehicles being used to do those field exchanges.
SODO Station ought to be an ideal connection point between West Seattle buses and Link to southeast Seattle, but it isn’t being used as such. If Metro doesn’t plan to make use of SODO Station, then mothballing it and saving a minute or two on travel time might reduce costs a little and draw a few more riders. Then, if TOD takes off around SODO Station, re-open it later. But draw in more riders with that saved minute or two in the meantime.
Maybe Stadium Station could be treated similarly to the rail base, with stops only made to pick or drop transit personnel, except around games.
Between fixing the tunnel algorithm, and mothballing those two stations for the most part, travel time could drop enough to reduce the number of trains operating for the vast majority of service hours.
Plus, it might save enough travel time to make truncating the 101 closer to a dead heat with its direct trip.
If getting ridership is of concern to ST, they need to look at all options.
Bernie, the “overwhelming majority” isn’t “passing through.” About half of all Link trips begin or end between Beacon Hill and RB.
That may change when more of South Link opens, or buses get truncated, but I suspect it will never be overwhelming.
Meanwhile we’d have 12,000 or so riders on Link and all the garbage about “cost per ride” would be twice as bad as currently.
But if you push ridership up significantly by forcing transfers from all points south that changes. That’s why it’s such a tough sell. I think a far better approach is to force the transfers on routes that currently use Rainer, and Beacon Ave so that the ridership is coming from the area near where the line was routed. Ancedotally, I only see Link in the RV on weekends but on the half full half size trains virtually nobody is getting on/off at Othello, Rainer Beach or Columbia (Can’t see the platform at Mount Baker).
A) You’re making the assumption that there wouldn’t be any stations contributing to ridership on west side alignment (Burien, Boeing Field, George Town maybe). B) There wouldn’t be this resentment to truncating routes if the train was time competitive. And C) the line would be longer sans the expense of the tunnel and station under Beacon Hill (albeit, Beacon Hill seems to be the only “winner” of the stations between SoDo and TIB). Although it still wouldn’t likely be any better in ridership numbers you would have at least replace more Metro and ST bus service hours and be in a much better position to extend the line and remain time competitive with express buses.
I’m with Martin on this. There simply isn’t enough between TIBS and SODO to generate much in the way of ridership. It might be offset some by increased ridership from the far South end of the line, but given population and employment density I don’t see South King County being able to make up for the lost Rainier Valley ridership.
I 100% agree that fixing service in the SE is more important. That’s why the RV was an editorial and this explicitly is not.
We have the station data, there’s no need to rely on weekend anecdotes. In fact it’s Sodo and Stadium that are the dogs. I’m on the 124 fairly often and I’m very, very skeptical that Boeing Field and Georgetown would generate squat. They also don’t have nearly the development potential, as they’re hemmed in by an airfield and a river that happens to be a superfund site.
I can’t really speculate what the ridership would be from South County. You save money on construction but probably lose federal $ for lower ridership. I guess the question is are you building this more for people in Seattle or people in Federal Way. I’m biased in that regard but I think potential for density and car-free living is better in the former.
Weekdays yes but Sodo and Stadium are the big winners for weekends which account for almost a 1/3 of the week (I know, it doesn’t seem that way when your working). But, as I’ve opined before, one station in this area should have been enough. Columbia and Othello are still dogs and on weekends they’re more like snails. We can argue the effectiveness of public money on private benefactors of pro sports teams but demand intensive nodes like this are exactly what justifies the argument for public transit.
I’m on the 124 fairly often and I’m very, very skeptical that Boeing Field and Georgetown would generate squat. They also don’t have nearly the development potential, as they’re hemmed in by an airfield and a river that happens to be a superfund site.
I can’t really speculate what the ridership would be from South County. You save money on construction but probably lose federal $ for lower ridership. I guess the question is are you building this more for people in Seattle or people in Federal Way. I’m biased in that regard but I think potential for density and car-free living is better in the former.
OK, time out. Any chance of an edit function before light rail service to W. Seatle?
This can’t be emphasized enough. You don’t build a train line where there aren’t any people!
I prefer cat-like stations, myself :)
But seriously: based on the data from spring 2010 which STB previously posted, your claim just isn’t true. On weekdays, Columbia City and Othello each see about 1,000 boardings — 80% of Pioneer Square, almost as much as Beacon Hill, but *way* more than Stadium and SODO. A routing along Marginal Way would have run through districts that are similar to SODO, and so I think it’s reasonable to expect that ridership would have been similar.
On Saturdays, there are four strong stations (Westlake, ID, TIBS, Seatac) and three weak stations (Pioneer Square, SODO, and Rainier Beach). On Sundays, Stadium is weak as well. The other stations are all in the middle, and I doubt that the differences in their performance are statistically significant.
You could argue whether or not the absolute performance of the RV stations is up to snuff, but by any measure, it’s clear that the RV stations (Rainier Beach excepted) perform better than Stadium and SODO. On Sunday, SODO sees 60% of Othello’s boardings. If Othello’s a snail, then SODO’s a snail facing a headwind.
(An interesting anecdote about Stadium’s performance is that, on weekdays, far more people seem to get off there than get on. ID, the other station near the stadiums, sees the same thing. I wonder if there are a bunch of people who take the train to the game from work, and then take a cab home. Or even simpler, do games ever end after Link stops running?)
Stadium and SODO I understand are there because of crush capacity due to sporting events. There can be more people at Quest Field than live in West Seatte ;-) But it does seem crazy to make all of these stops when there’s nothing going on. How far is it from Stadium Station to Pioneer Square? ID sees more Link traffic than Pioneer Square but I suspect that if you add in bus ridership Pioneer Square would come out ahead. Othello and Columbia each contribute about 5% of the total ridership. But when you add up the end points (DT + TIB & SEA) it’s 75% of the ridership. The five stations spread out over the 8 or 9 miles inbetween only produce 25% of the ridership. So, even if you pull all of that out we still be at 15,000 weekday ridership without RV and Beacon Hill. You’d see some boost of end to enders just because travel time would be faster. A significant boost if Bus to Link transfers were time competetive. Ridership from stations between SODO and Tukwila likely would less than RV but besides bus transfers from points south it would also have made for a viable bus transfer point from West Seattle which would put the numbers right back to where they are now. I’m not buying into the arguement that development is limited by a river and an airfield. Both of those are huge job creators (shipping, maintenance and travel). Thousands, if not tens of thousands of jobs have been lost from the region. There’s acres and acres of flat empty parking lots all zoned commercial just beggging to be brought back to life.
Stadium and/or SODO should just be special stops for events at Quest and Safeco. Othello mid-day Saturday gets 392 boardings over six hours or 65 people an hour so ~22 people per train. I’ve never seen even five people waiting mid-day at Othello. I’m really having a hard time believing these numbers.
“Stadium and SODO I understand are there because of crush capacity due to sporting events. There can be more people at Quest Field than live in West Seatte ;-)”
Is SODO really useful for stadium access? I’ve never gotten off there and explored the area, but it seems like it’s a bit distant from Safeco Field and even farther from Qwest Field. Frankly, for Qwest Field, ID station seems to be the most convenient.
Some people park down there near SODO then take Link to the stadiums. The few on-street spots near the station go fast.
I’m not sure where you got your “392 people over 6 hours” number from. The page I linked to only has time breakdowns for weekdays. Northbound, Othello sees 302 passengers from 9 to 3. That’s 50 passengers an hour, or 8 passengers per train (since there are 6 trains per hour). And that’s a weekday.
Anyway, as usual, I think it’s silly to spend time arguing over what should have been. We have Central Link, and it goes through the RV. The only questions are: should we turn any stations into part-time service (or suspend them entirely), and when we have the money, should we build a bypass?
For the first, I think it would be a great idea to do what Brent says, and only use Stadium station for Metro personnel and/or when there’s a game.
It might also be worth exploring skip-stop service, though that would require more service (if every train was skip-stop, you’d need 2x the service to maintain effective headways). But then the question becomes, is this really a good use of our money, when (for much less) we can just keep running express buses?
The same goes for the bypass. Could it be worthwhile? Maybe, especially if it’s cheap enough. But I’d much rather use that money to build West Link, or to open a station at 15th and Mercer, or any of lots of other things.
Ah, my mistake. The 392 is weekday. Reduced by the overall difference for Saturday that would be 300 people over 6 hours, 50 people an hour. Given the asymetric nature; about 4 southbound and 12 northbound boarding every train. Given that there is no schedule you think that randomly driving by you’d see at least four people standing around or someone about every minute walking up. The vast majority of the time though I don’t see a single person on either platform.
Ah, so you’re saying it’s 392 total, not just a given direction. Okay, but we have the directional numbers — it’s 302 north, 89 south. (So 391 total, but close enough.) Northbound, you should expect to see about 8 people per train on a weekday; southbound, about 2-3. Both numbers are of course smaller on a Saturday.
Now, if you never see anyone standing there, then I don’t know what to say. But I’m inclined to believe numbers over anecdotal evidence…
Yes both directions since I can see both platforms. I scaled back the weekday total 75% to account for Saturday so 300/6 is 50/hr (someone approaching almost every minute). Spread over 6 trains gets you 8 which would be 2 SB and 6 NB per train (not 4 and 12 Somewhere I got the numbers doubled). I’d still expect to almost always see people coming and going and on the platforms. I didn’t say there’s never anyone there but just more often than not (i.e. more than half the time) and usually that’s just a single person. I’m not even counting the alightings since unless you happen by right after the train you can’t tell how many got off. Someone that rides Link regularly on the weekends would be a much better source.
To add some facts for Bernie, the population of West Seattle is about 80,000(2000 census, surely more now), the same population as SE Seattle, where Link currently runs and generates 12,000 riders/day.
From http://www.westseattle.com/
I had a more definitive link when I searched for the info earlier today and should have documented it. I believe it was 2009 projections or possibly even 2010 census data. Doesn’t really matter really. It’s about half that of Bellevue with no concentrated development of job center like Bellevue has and light rail to Bellevue is marginal even with the addition of Microsoft and the use or a bridge that’s already built. Why is it so hard for the light rail “cult” to recognize a dog? There are so many other projects with more potential, less cost and unfunded. From the City of Issaquah Police Dept. Home page:
Which means more people commute to Issaquah than the total bus ridership of West Seattle and it’s growing exponentially faster. I’m betting there’s as many or more that commute from Issaquah meaning you have some potential for a balanced load versus West Seattle which is a bedroom community without an employment center or any hope of ever having one.
Bernie,
Exactly. A no-stop express line from I-5 and BAR to the Maintenance Facility loop track using the fly-under to rejoin the main line is a must once service moves south of the airport. It would be cheap because it could be nearly ALL at grade along the east side of Airport Way. It would have to bridge over the railroad tracks at the Airport Way overpass of course, and probably shouldn’t come down north of there. So it’s by no means free, but by no means exorbitant, either.
Unfortunately the SK sales tax revenues have fallen most of all the Sub-Areas so things aren’t going very far south of the airport any time soon.
Do any of the connecting buses along MLK have *timed* transfers? Meaning that if you take a certain Link train from downtown Seattle, you know that it will connect with your 30-minute headway bus (which will wait until Link arrives and passengers can get to the bus)? Maybe not during rush hours, but mid-day, evening and weekends?
Oh, wait, that would mean Link would have to have a schedule, and Metro would need to coordinate with Link’s schedule. Maybe an idea for 1/1/11 column on helping Link. Timed transfers from rail to bus work throughout Europe.
By the way, I don’t mean the 8, I mean the east-west buses at the Link stations.
Some Metro timetables list “partial service on Link” and show connecting train times.
connecting train times for a service that doesn’t publish them hmm…
I guess you could start off by having northbound busses on I-5 divert to rainier beach whenever there is too much traffic on I-5.
Then you could move to have all southbound rush hour busses leaving rainier beach station
Finally, as a temporary cost cutting measure, you could have all other buses from the south terminate at sodo station, and have southbound link trains announce weather buses were at this point leaving from sodo, or rainier beach.
Just a thought.
Complexity in spades. I like it!
Getting agreement between ST & Metro for emergency reroutes to Link would be key. Riders who don’t have ORCA card would need to be assured that they don’t have to pay an additional fare. Ideally everybody would have ORCA cards but we don’t live in that world yet.
Getting this system up and running wouldn’t cost much and would improve reliability for bus riders headed south. The key is to make it as easy as possible for riders (And bus drivers since they will have to explain what to do about 20 or 30 times per trip, even when we announce it over the PA)
You could start small, just have northbound peak trips stop and Rainier Beach. You’d just have to figure out the price issue.
In the absence of a forced transfer using Airport Way with BAT lanes through Georgetown would provide a more reliable schedule, too. True a slower schedule when the freeway is running freely, but a much more reliable schedule. Industrial Way would be a great way to get to the bus mall without having to bounce along Spokane.
This exercise is one way to improve Link’s reliability and productivity and save Metro’s costs. Most of the delays occur in the downtown tunnel. 12 buses during the peak 4 pm hour, all heading towards Renton, plus another 4 buses per hour to Kent.
So some ridership may be lost from forced transfers but ridership is also lost when Metro raises fares (as it has been continually for the last few years) and cuts service. Which is the less of many necessary evils?
And of course, the 101 should truncate at RBS, and the 106 should be surfaced. Am I right in thinking that the 106 is the only all-local service bus in the DSTT?
It is now after Metro surfaced the former 174, now 124. Except the 70-series routes to the U-District which run local on Eastlake in the evenings and all day on Sunday.
Yeah there needs to be fewer buses in the tunnel. It’s painfully slow to use the tunnel now, especially at the entrances.
The optimum location for transfers to Link from buses from the south would be a station near S. 133 in Tukwila adjacent to SR-599. If Sound Transit and Metro were serious about providing seemless bus transfers to the south they would have built a station here. SR-599 offers direct access to I-5 and 405 and Interurban Ave. offers a quick route to Southcenter, Tukwila and Renton. A station there would have offered painless connections to freeway express routes and would get the buses off of I-5 before they hit the traffic jams that always seem to start around Boeing Field.
He he, I began writing a post about that after I saw Martin’s post.
I think you should. This would be a much better connection than what Martin suggested, although it certainly would require a significant capital expenditure. You would have to build a new areal station, a I-5 NB to NB SR-599 HOV ramp, and preferably an elevated bus terminal that feeds directly into the mezzanine level of the station. It would also be good to build a full I-405 NB to I-5 NB ramp (current HOV ramp doesn’t feed into the HOV lane).
This would really effectively collect service off of I-5, I-405, SR-167, and SR-518 and create a really strong transfer point.
Either a station at SR-599/133rd St or Boeing Access Rd station would work in this instance. The 133rd St station would have to wait for ST3 since it was not included in ST1 or ST2 and the Boeing Access Rd station is up in the air and dependant on BNSF and allowing more room for a platform.
Whatever the case is, both these stations are needed not only so buses could be fed into this station and also to add infill stations between Rainier Beach and Tukwila/Int’l Blvd stations.
Frankly I never understood the Boeing Access Road station. It seemed like a solution looking for a problem. Who would ever transfer between Sounder and Link?
The beauty of a station on SR-599 is that buses would have direct access to the southbound I-5 HOV lanes and it would be easy to build a direct access ramp from the northbound I-5 HOV lanes. It would also be a logical truncation point for the 150, whose frequency could nearly be doubled just by eliminating the freeway portion. The 150 takes ~30 minutes to travel from the Tukwila P&R to University Street Station, Link would take about 25 minutes from S. 133rd to University Street, so the travel time would be a wash. But the frequency of the 150 could be increased to every 7.5 minutes instead of its current 15, decreasing overall trip time.
About that Boeing Access Road station – My thought is to have frequent service on a sounder line perhaps using a light weight diesel or diesel/electric locomotive similar to what is used on some BritRail lines. I’m envisioning service head ways of 20 minutes to Auburn and hourly to Puyallup/Tacoma. With this type of train, it may make sense to electrify parts of the routes if it doesn’t interfere with BNSF operations. I’d also add a station at Georgetown. There could be a savings of up to 15 minutes from Boeing Access to downtown versus Link. Truncating buses and feeding them there and at stations in Tukwilla, Kent, and Auburn might also save significant VMT’s and service hours.
As for Zed’s question about who would ever transfer between Link and Sounder? if Sounder service were frequent enough, I think anyone coming from Sea-Tac or TIB or from Rapid Ride, might consider a transfer if it saves them 15 minutes and far fewer stops (2 Sounder stops versus 8 Link). This would be not unlike the transfer between the Airport shuttle train and MTA at JFK. Further I think the Boeing Access Station would have the potential of attracting significant Boeing personnel especially if local bus and shuttles were available.
This goes against the principle of eliminating fossil fuels in proposing a diesel train but for that particular route it makes the most sense. And if it eliminates dozens of bus trips into downtown then it is a net plus.
The downside that I can see are that if your intended destination is above King Street then an additional transfer or trek to ID Station is involved and your time savings might evaporate but if your destination is near southern downtown then this route might be just the thing.
I’d rather have the money spent to build 200th St Station and Des Moines Station faster. For far south routes connecting to Link, they should be better than an infill station.
I wonder also about truncations to Sounder. There’s alot of duplicate service to untangle here…
From Tukwila, Sounder takes 12-15 minutes to King St. (scheduled for 20 mins as padding), whereas both Route 161 (East Hill-Seattle, 5 peak trips) and Route 102 (Fairwood-Seattle, 7 peak trips) take 30-40 minutes to travel an equivalent distance.
From Auburn, Sounder takes 35 minutes to King St. Route 152 (Auburn Sounder to Seattle via Star Lake, 6 peak trips) takes 1 hour. Star Lake commuters already have connections to Kent Sounder (183, every 30 minutes), one-seat rides to downtown Seattle (190, 192, ten peak trips), and access to Link via SeaTac (ST 574). What’s the purpose of the 152?
Lots more I’m sure…
The 152 exists so I can walk a mile and catch a bus to downtown.
Realistically though, it provides service to the Auburn P&R (the one next to the airport). I’m not sure why it continues to Auburn Station–possibly to tighten the headways of 919 or 180 between AS & AP&R.
Seems to me like some thought should have been put, back in the day, into locating park-and-rides next to rail lines and planned rail lines, but apparently none was….
I’ve skimmed the comments here and I’ve noticed a trend: there’s a lot of YITBYs (Yes, In Their Backyards). Guestimating from what I’ve read from previous posts, a lot of the commenters are north end/east side and are just saying “yeah sure, these efficiencies look great on paper, let’s put them in!”
As someone who actually lives in South King County, I’ll give you a resounding NO. My main routes are 577 and 578. I’ve defended time after time that these routes should NOT connect to Link. Google Maps clocks the time from FWTC to 4th & University as 29-40 minutes. Scheduled time peaks at 35. On time performance is very good, though I don’t have numbers to back that up.
TIBS to University Street Station is 32 minutes. On time performance is about what you’d expect. FWTC to TIBS is a minimum of 16 minutes. Even assuming only 60 seconds is lost in a transfer means the trip increases by 13 minutes. A time increase of over 40% PLUS a transfer that didn’t exist before and all I get out of it is shorter headways on the bus? If it ain’t broke, don’t fix it.
Or, let’s look at RBS. FWTC to RBS by bus is about 21 minutes. Giving it the best chance of success and assuming only 60 seconds to make a transfer, we add that to Link’s 23 minutes to get from RBS to USS, you’re looking at a total time of no less than 45 minutes, which while better than TIBS, still sucks.
This whole “redundancy” is a bunch of BS. Just because a handfull of routes all travel along the same path doesn’t mean it’s a bad choice, it just means that all of them are using the best choice.
I-5 isn’t the only place that has redundancy. Look at 3rd Ave. We have a whole ton of routes traveling on it that split off going the same place. Why not just run a shuttle at 30 second headways that dump people at either end of the RFA? You’d keep the RFA and get pay as you board on all routes since all “downtown routes” would terminate at a transfer point just outside the RFA.
I think I asked about the 578/577 once in the past and someone said that it’s highly likely to stay even when Link is fully built out for precisely the reason that you mention, it’s just faster when the roads are flowing well. The numbers may come out differently for other routes in the Rainier Valley and Renton.
If people up north are more friendly to two seat rides, it might be because some of the ones up north, like the 7x busses, may be faster as Link transfers because Link will be much faster than busses between Downtown and and the U-District.
Your comparison to 3rd Ave doesn’t make any sense. Most of the busses down 3rd are low-headway urban routes that either pass through downtown with the same number or continue with a different number. Your shuttle idea would cost MORE money for less convenient service. At least most of the ideas floated here would save money.
Thank you Tim for making my point. He’s the guy I was talking about that would say “screw you, I’m back to driving”.
I live in South King County. It’s difficult to get some places without driving or making the trip 2-3 times longer.
Downtown is about the same or less. Why anyone would drive to Downtown is beyond me.
Believe it or not, this sentence perfectly captures everything that I dislike about Metro’s system planning.
I happen to agree with you: taking a bus to downtown is so darn convenient that driving there just seems silly. But the real question we should be asking is, why isn’t this true for *every* bus trip?
The example I like to use is getting from Ballard to Ravenna. With a real grid system — e.g. highly connecting service with few overlapping routes, and service that was actually frequent (5-8 minute headways) — getting from arbitrary points in Seattle to downtown wouldn’t be any harder, but getting from arbitrary points in Seattle to other points in Seattle would become far easier.
I have no problem with maintaining the 577/578 as real express buses — i.e. premium service that costs more (to reflect its higher operating cost) and only runs during periods of high demand. But off-peak, I’m less convinced. In particular, I think it’s a mistake not to factor in headways.
Let’s assume that you’re equally likely to want to go to downtown at any particular time. On weekdays, the off-peak headway is about 30 minutes, so the expected wait time is 15 minutes. Let’s say that rerouting to Link would allow the buses to become 3x as frequent, so the expected wait time was now 5 minutes. In the old system, the expected door-to-door time is 15 minutes (waiting) plus 37 minutes (the scheduled off-peak trip time), for a total of 52 minutes. In the new system (assuming a transfer at RBS), you have 44 minutes of travel time, plus an average of 10 minutes waiting (5 minutes for the 578, and 5 minutes for Link), for a total of 54 minutes. Not better than today, but not much worse either.
On Sundays, it’s even more dramatic. The bus runs every hour, so the expected wait time is 30 minutes. With the assumptions from the previous paragraph, Link would shave off 15 minutes of wait time, which more than makes up for the extra 7 minutes of travel time.
And, of course, this all neglects the fact that the 577 only goes downtown. If you want to go to Columbia City, or Beacon Hill, or the CD via Mt Baker, then the 577/578 don’t help you at all. So, for these riders, access to Link would be much better than the alternative.
I’m not saying that this change should be made immediately — just that, when a decision is made, it should be based on these sorts of numbers. For example, let’s say my assumption is wrong, and truncating the routes would only allow you to double the frequency, rather than triple it. In that case, this story becomes a lot less compelling. On the other hand, it might turn out that it’s worthwhile to do this truncation on Sunday, but not on other days. But either way, simply measuring total route times will not lead to the right conclusion.
By the way, as far as your YITBY comment, I should note that, as a Capitol Hill resident, I have repeatedly argued for exactly these types of changes in my neighborhood. I’d like to see the 43 eliminated, the 14 truncated at downtown, the 49 rerouted down Broadway along the 9 route, the 10 rerouted from downtown to 12th Ave in First Hill, and the 8 split at 22nd and John. I’d like the savings to be used to get 8-10 minute or better service, all day, on the 8 (Capitol Hill segment), the 11, the 9/49, and the 48. And I’d like all of this to happen even without Link, even though I’d be losing a one-seat ride to downtown (though I admit that I’d only be a 4-block walk away from the 11).
I totally agree about the Capitol Hill routes! Especially once Link opens, they need to do that. Also, the 12 should be eliminated along 19th. It’s ridiculous to have a frequent-service bus on 19th when there are also frequent-service buses on 15th and on 23rd, as well as on John. If you eliminated it, you could also make the 11 frequent service to keep the frequent service up Madison that the 12 provides. The only issue I have is that it seems unfair to completely eliminate the 43 without giving the people up along 23rd between John and Aloha a good alternative to get to Downtown or even to a light rail station.
Yes, I forgot about the 12. Such a strange bus. It should definitely go, same as the 14. In a perfect world, it would be nice to have super-frequent buses on every arterial, but realistically, 8-block spacing is totally adequate, especially for mostly-residential streets like Summit and 19th.
As far as getting people from 23rd to light rail: the 48 already brings them to Mount Baker today, and to UW Station in 2016. In the meantime, if the 11 had 5-minute frequency, that would be a totally reasonable way to get downtown. But you’re right; politically speaking, it would probably make more sense to defer that change until after Link opens. (I still think it would have been really nice to have a station at 15th and Mercer…)
See the Rapid Trolley Network which was developed for the Transit portion of the Surface/Transit/I-5 AWV replacement option.
There are some good ideas there for what to do with much of the core Seattle Metro service.
Bernie,
The issue is with the number of service hours Metro needs to eliminate the option is between slightly worse service with a transfer or drasticly reduced service while keeping the one-seat ride.
Many routes are going to see weekend, evening, midday, and even peak service either eliminated or drasticly reduced. If the choice is between every half hour service with a Link transfer and no service at all which is better?
Where are you pulling those numbers from? If I have the schedule, I show up at the transit center five minutes before the bus is scheduled to arrive, not 30. Five minutes gives me enough time to find a spot and take the stairs down to the bay.
Or are you assuming that I’m transferring from one of the feeder routes (I only use the 183 when I don’t have my car since its span of service is so poor).
No, they don’t. Instead I’d hop on the 574 and transfer at SeaTac/Airport station. Transferring downtown makes no sense.
The expected wait time only applies when people arrive at stops by random and not consult a schedule, which is typically when headways are below 15 minutes.
Anyway, if you miss the bus like I have on many occasions, it sucks to have to wait another 30-60 minutes.
The point-to-point non-stop downtown express buses can stay, just charge a higher fare for the privilege. Buses that serve a corridor benefit more from frequent service than expresses do because people use it for trips other than commuting. Which is what the 574 & 577/578 setup does.
The reason I used the expected wait times isn’t because I think people are waiting at the bus stop for 30 minutes, but because, if you spontaneously decide that you want to make an off-peak trip, that’s the average amount of time that you’ll have to wait between your decision and actually getting on the bus. Even if you spend most of that time doing something useful at home, it still means that you reach your destination later than you might otherwise have.
Now, for commuting, none of this matters, since commuters can set up their schedule to match the buses. But that’s why I’m only talking about off-peak service. :)
Exactly. Couldn’t have said it better myself.
“…all I get out of it is shorter headways on the bus?”
Shorter headways and/or longer operating hours isn’t nothing. It could mean significant time savings for someone with a shift starting/ending time that currently isn’t served well. Could also mean significant time savings for someone who is transferring from a Federal Way local route to the FWTC-DSTT express (e.g. someone going from 35th Ave SW & SW 342nd currently has a 20 min layover at FWTC if they want to get downtown by 8am – doubling headways could give this person a 5 min layover instead and erase the time loss from transferring).
As for subarea preferences… I think this same concept would work equally well for an intercept at Northgate and Eastgate once light rail makes it to those spots… the south end is on the table now because it’s the only part of the region to have the high-capacity trunk line.
At best this transfers lost productivity from one person to another. The downside is that, given both riders are using transit you’re a lot more likely to lose the person that’s made transit work well for them than gain in areas where the decision/ablity to use transit is marginal. Making transit more mediocre for everyone isn’t a very good strategy for building ridership.
I’m going to pick on that specific example of 35th Ave SW & SW 342nd–route 179 is a mile away and goes straight to downtown. I don’t care to do a comparison of that route, but I’m sure it’d add at least 10 minutes if 179 kept its south end routing and terminated at TIBS or RBS.
I don’t think that’s a very good case for truncating routes. Sure headways improve some for the outlying connectors, but it costs ALL riders 15 minutes, regardless of whether or not they benefit from the shorter transfer penalty at FWTC.
I’ll also touch on the point that a lot of the commuters at FWTC are “choice” riders. If you include driving to the transit center, parking, and standing at the bay waiting for the bus, the time spent doing this is about equal to driving alone and parking in a garage somewhere downtown. I’d rather not piss off these riders by adding 15 minutes to their commute and making them stand in the cold for an extra 15 minutes.
Note that there is a lot of service from FWTC to downtown; see this graph as a comparison. There’s also a lot of demand at the FWTC; the garage can fill to capacity by mid-day (which sucks when you’re trying to catch the 578, as the stops immediately before and after FWTC are Auburn and Downtown. 577 used to stop right outside the Gateway theaters, but I imagine the landlord got tired of Hide & Ride).
I’d truncate the 255 at UW Station if it means more frequent buses and later service and slightly longer trip.
You waste much more time when you miss a bus that comes only every 30-60 minutes versus a frequent bus to a frequent rail line. And you have to add padding time because of uncertainty. The bus may arrive early at an estimated timepoint and then you’re stuck waiting for another 30 minutes to an hour.
Exactly what I was trying to say. Just because a bus travels a route that is roughly parallel to a pair of steel rails does NOT mean it’s redundant.
In fact, it’s not redundant at all since the South service runs express to Downtown, whereas Link makes stops all the way to Downtown. Different types of service. It’s like route 301 vs 358.
“I’m going to pick on that specific example of 35th Ave SW & SW 342nd–route 179 is a mile away and goes straight to downtown”
A whopping 6 times per day.
How many times a day do you need to go back and forth from work? There’s a reason it only operates in one direction during the rush hour. Mid day demand is negligle. And less than 15 min. headways during rush hour would mostly just cut route productivity.
Did you actually read Yorik’s comment Bernie?
Yes, if you shift the service hours you gain nothing in savings for Metro. You piss off a much larger portion of you ridership than you help. If you disagree with my assesment or does speak to the topic then say why instead of asking a retorical question you seem to have assumed the wrong answer to anyway.
[Grr, ended up posting to the wrong spot in the thread, STB move those “reply” links!]
Bernie,
The issue is with the number of service hours Metro needs to eliminate the option is between slightly worse service with a transfer or drasticly reduced service while keeping the one-seat ride.
Many routes are going to see weekend, evening, midday, and even peak service either eliminated or drasticly reduced. If the choice is between every half hour service with a Link transfer and no service at all which is better?
I can see why there is momentum to do something with Rainier Beach, it really sucks for the most part. I don’t think transferring there makes any sense at all. I tried to commuting to Ballard from there and it took me over a hour and a half one-way. I’d love to see ST try and push this; either quite a few of the planners aren’t riding Link from Rainier or they want ridership numbers to plummet, I’m not sure which.
Getting to downtown alone took way to long alone, waiting for the #18 wasn’t as bad. So expecting users to wade through this is so unfeasible it really is laughable.
Lets face it. Here’s where ST screwed up so royally, lied about the tunnel, now have a light rail line that rides likes garbage, and doesn’t magically revitalize the neighborhood as promised.
Surprise, surprise…….
Not to pile on Martin, but here’s another reason why transferring from bus to rail is not automatically a good deal.
When ST was negotiating with the Feds, there was congressional concern that Central Link provided too few ‘New Riders’, in relation to just converting existing bus riders to a new rail system.
Somehow they got the notion that their money would be going for lots of new transit riders. So I’m not saying truncating routes is not important, but getting new riders into the system is a high priority with the FTA when they pony up 1/2 billion.
At this point I think just trying to bump up ridership regarless if they are “new riders” or not would be a good thing. Both Metro and ST really haven’t taken advantage of using Link as a trunk line. I can only hope a better job of re-aligning bus service is done when U Link and North Link open.
At the same time I’m not really worried about Central Link ridership affecting Sound Transits ablity to compete for New Starts grants for ST2. The first project likely to apply for funding (North Link) is very strong on the merits even if the projections are as far off as they were for Central Link. I seriously doubt this is the case though.
It is also worth remembering the New Starts criteria and scoring have changed a bit from what they were during the Bush Administration. These factors favor additional grants for ST2 projects.
Given that very few buses will stop at UW Station, and most will stop about three minutes’ walk away, Metro/ST don’t seem to want to use UW Station as a transit hub.
As a result, lots of riders, including Capitol Hill and U riders, will have to backtrack to downtown to get to major eastside destinations such as Microsoft.
Perhaps there is still time to get Metro/ST to reconsider this nonsensical plan to have buses avoid UW Station.
I have to admit I’m really disappointed by the transportation and transit plan WSDOT, SDOT, The UW, ST, and Metro have shown so far for the Montlake area. Everything from eliminating the Montlake Flyer stop, to assuming few riders will be transferring at UW station, to not giving transit true priority access through this chokepoint.
At this point the best that can be hoped for is some sane stop relocations and possibly truncating mid-day 520 service to the U District via Montlake.
I will say I disagree with those who think the U-district to downtown routes should be truncated at UW station. I think it is better to wait until North Link opens to start feeding those routes into link.
This isn’t to say Metro and ST routes in the city of Seattle shouldn’t be restructured to more of a grid which would provide much better trunk line service than is currently offered with fewer service hours than are currently used.
Route 7 should be extended to RBS.
Yeah, and get rid of the Prentice Loop.
Some entity will have to pay to extend the trolley wire.
I’d de-energize the Prentice Loop as a start. As hardware (wire, switches, pole mounts) were removed, inspect them, and re-mount on the new RBS loop and layover by line crews during periods they aren’t untangling trolley poles, or doing routine maintenance. Sure it takes longer, and there are other costs involved, but it gets done.
Not sure if they need another sub-station.
Mike, I suspect the labor to string the wire is a far greater portion of the cost of adding new wire than the hardware itself.
I’m guessing that it 50/50 for new construction, but if the crews are just waiting for a service call, then there already being paid. Probably a bad idea, because once they do traffic control for stringing wire, they wouldn’t want to pick up and leave in a big hurry.
Idea withdrawn.
I still think dropping the Prentice Loop and extending the 7 to RBS would be a good move. Even better would be to do what a number of the options for the rapid trolley network call for and make a new ETB route combining the 7 between RBS and Mt Baker TC and the 48 between Mt Baker TC and the U-District. Service between Mt Baker TC and Downtown could be provided by a truncated 7 and the North portion of the 48 replaced with a new route or combined with the tail of the 71. Though in the latter case turnback/layover space would need to be found for the 71 somewhere in the Roosevelt area.
Rainier Beach isn’t a good transfer point for most of the buses coming from the south, and just because it’s a Link station doesn’t mean we should terminate buses there. However, we should restructure our service in general so that instead of having not-very-frequent express routes from everywhere to Downtown, we have quite frequent express routes from just a few points to Downtown, and frequent local service connecting to those points. If they made frequent express routes just from Downtown to Renton TC, from Downtown to Federal Way TC, and from Downtown to Kent and Auburn TCs then eliminated all other Downtown express routes and replaced them with frequent feeder service, you could end up with far higher ridership.
I agree with all of this. An added bonus is (I think) fewer buses competing for limited resources DT. Less buses with more people for about the same (including feeder service) cost; what’s not to like?
I dont know why i write such details for the ‘blog but… Even in the best of conditions this move would not really deliver the expected benefits. Part of the problem comes in rainier beach, of how do you terminate the service, lay it over and get it turned around and heading back again. Not the easiest place to make those arrangements, and it can quickly eat away any cost savings you had by terminating service there. Once you solved that, ST would have to eat some of that cost savings away by adding capasity to LINK to handle the added ridership (even if you lost a good percentage you’d still have to expand LINK to compensate, especally now with single cars on nights/weekend). The major problem with peak express service, is that the routes often have long deadheads, and low turnover. Its going to be a giant pig no matter how you look at it, or who you have operate it. The best ways to help allievate some of it, is charge a premium fare for the service. Also i think more investigation needs to be made into consolidation of I-5 services and the best way to operate them such as, ultra consoldated into two or three routes, with high frequency), or seperate routes with lower frequency, and of course the issue of who operates and pays for them how.
The most logical place is the area behind the RBS bus station on the NE corner of the intersection. It appears the land is either already publicly owned or could be acquired and a bus turn around or layover lot created.
In addition, some of this land could be used as temporary P&R space until it can get TODed.
Not to mention low demand on those deadheads. Try riding one of the southbound 577s in the morning. They’re practically empty.
Metro’s policy allows customers to ride on deadheading coaches. There are a couple of reasons why they don’t put them on the schedule though:
1) Low demand
2) You wouldn’t really know what the schedule is, since the deadhead doesn’t start until the incoming trip ends, and often the incoming trip ends with estimated timepoints
3) It’s cheaper to deadhead a coach than to put it on a schedule. Once an operator gets to the end of the run, they just continue on instead of having to wait for the clock to advance. Also, they don’t have to stop the coach at certain points to pick up or discharge passengers.
*There are two exceptions to the allowing customers to ride on deadheading coaches: 1) Any coach signed as “Not In Service” and 2) Safety rules. One of these is coaches headed to north base. There’s nowhere to let you out except inside the base, and passengers are forbidden from entering the base.
Another reason: sub area accounting rules. Peak-only, one-way routes get charged to the originating sub area. The operating hours for two-way routes are jointly paid from both sub areas. 40-40-20 comes in to effect.
Few people know that riding on a deadheading bus is allowed.
577 runs to a park & ride with low employment density. Some routes, like the 252, 255, 257 serve Totem Lake, a large employment center. It would be nice to have express service there. That Metro is increasing service on the 255 all-day, in both directions, suggests demand.
I hear everyone in their comments about leaving the commuters with their expensive one-seat rides.
That still leaves what I’ve been suggesting all along: Truncating the 101 and maybe 150 at RBS. The 101 would stay on MLK and never get on the freeway. South-Renton-P&R-to-downtown commuters would still have the 102.
The 101 is already losing service hours. Truncation would allow service hours to be restored.
Truncating the 101 and extending hours would bring back some riders and hopefully add new ones who were using their cars for local trips.
Don’t forget there are many potential commute patterns outside of downtown Seattle. Many Boeing Jobs in Renton and the Kent Valley. Many light industry jobs. A frequent Sounder train with local feeder service might get significant numbers of people out of their cars to the Boeing Plant #2, Boeing Renton, Boeing Flight Safety, Valley Medical Center, South Center Mall and other area retail. People who could take Sounder to the Kent Valley but don’t because they feel they need their car during their lunch period might reconsider if there were frequent sounder service coupled with frequent feeder and local service.
A friend of mine works just south of Southcenter mall and lives a short ways from the Sumner Sounder Station. While he was getting his engine replaced, he took the Sounder to Kent, since it was easier to transfer to the 150 there than to walk from the Tukwila Sounder Station.
I walked from TSS to his work once (in the rain even) and didn’t understand what the big deal was. Much easier than waiting for the 150.
Great discussion. Too often proposals to truncate routes get shot down with, “People will stop riding if it takes longer or they have to transfer.” Here Martin and some commentators point out, “Sometimes slower is better, if it increases frequency or destinations.” This needs to be heard. The total trip time consists of riding and waiting, and waiting again if you transfer. The reason transit is heavily used in London/Chicago/SF is that nobody needs a schedule ever. You simply take the trip length, add 5-10 minutes per boarding, and that’s it. For both rail and bus. If you want to arrive at a particular time, you just subtract the travel time and wait time, and that’s when you should leave. If you just want to travel now, you go to the stop without worrying that you may have to wait 30-60 minutes.
This region has never had anything like this, a network of routes running at least every 15 minutes from 6am-midnight and 30 minutes night owl. The only corridors that achieve this (non night owl) are Eastlake/Fairview, Rapid Ride A, and Link. Elsewhere — especially off peak — riding transit is more like airlines or Greyhound. Plan your trip around the schedule, or wait 30-60 minutes at the bus stop. That makes the total trip time (from the time you want to go to the time you arrive) an hour or longer, and double that for a round trip. People say, “Screw that, I’ll drive instead.” Those are the riders lost.
Truncating the 101 and 102 off peak, for instance, may lose some riders, but it may gain other riders if it’s running every 15-30 minutes. People who are willing to travel for a little longer in exchange for not waiting as long or not having to worry about the schedule. This must not be lost in the debate about how to make Metro cuts. It has an uphill battle because the general public doesn’t know it’s even possible, and doesn’t realize that a slightly slower but more frequent trip may be attractive to some riders. Because even if the travel time lengthens, the travel+wait time may compensate for it and make other trips feasable (trips that previously weren’t feasable on transit). Especially if part of the trip is on Link, which gives many more destinations than an express bus can reach for almost the same speed.
Limiting truncations to off-peak is also a worthwhile strategy. People are most concerned about their peak-hour commute, and many don’t care what happens to off-peak service. So keep the peak service as-is to avoid a backlash, but restructure the off-peak service and insist that frequency is more important than one-seat rides.
As for Link vs ST Express, it depends on the length of the segment, and whether you’d have to transfer buses. Link is competitive from downtown to SeaTac, but not to Federal Way or Tacoma. Link could be competitive with those places if it eliminated intervening stops and ran through Georgetown, but then it would no longer serve Seattle residents. So it’s a tradeoff. But if you look at non-downtown trips that require bus transfers, such as Rainier-Northgate, Rainier-Bellevue, Rainier-SeaTac, Roosevelt-Lynnwood, Bellevue-Capitol Hill, Bellevue-Shoreline, Link is dramatically faster, sometimes cutting trip times in half. Even with a bus transfer, like Capitol Hill-Sand Point, Capitol Hill-85th/Aurora. And even where Link takes longer like downtown-SeaTac, the short headways compensate for it.
N 45th also has 15-minute service for 20 hours a day.
In general, I completely agree with you. But I think that 15 minutes is a very low bar for “frequent service”, even off-peak. That means that a 2-bus trip has an expected wait time of 15 minutes. Add that to the fact that service patterns heavily favor downtown, and suddenly, getting from Fremont to Capitol Hill takes 45 minutes instead of 20 (my record travel time via bus).
Metro says, if we can’t afford to provide 15-minute headways on this route, we’ll drop it down to 30-minute or 60-minute. I say, if you can’t afford to provide 15-minute headways on a route, and it isn’t providing a lifeline service for transit-dependent populations, get rid of it and redirect the resources somewhere else.
” 8-block spacing is totally adequate, especially for mostly-residential streets like Summit and 19th.”
that area isnt even in king county is it?? you coulndt find an area in king county, the area in which the discussion is referring, to say what is totally adequate???
Wha? I’m talking about two streets in Capitol Hill, namely Summit Ave E and 19th Ave E, i.e. the streets which Metro routes 12 and 14 run down. Was that unclear?
ok..google earth shows summit at a dead end just before 19th. i was searching for the intersection you claimed at “summit and 19th”. the software showed an area north of everett.
i was searching for the intersection you claimed at “summit and 19th”. the software showed an area north of everett. i guess your claim was two separate streets and not and intersection.
nice wide-ranging discussion.
back to original Duke post; the truncations should only be considered after 2016 when Link could have four-car trains rather than the two-car trains of today; the Pine Street stub tunnel is a limiting factor.
Perhaps the South King County markets could be split based on their distance and travel times. Federal Way and its center access ramp could have a more frequent Route 577; routes 101, 106, and 150 could eventually be truncated but might go past a Link station to provide improved connectivity and frequency to new markets.
STB could impact an upcoming ST study on South King County, as they cannot afford Link to South 272nd Street and will look for alternative projects.
How about infill stations at South 110th and 133rd streets and improved bus layover at TIBS? How about improved service frequency on routes 574, 577, and 594?
Itinerary #1
Walk 0.4 mile S from MIRROR LAKE PARK to (5 minutes??)
Depart S 320th St & 8th Ave S At 07:25 AM On Route MT 179 DOWNTOWN SEATTLE
Arrive Pine St & 4th Ave At 08:02 AM
http://tripplanner.kingcounty.gov/cgi-bin/itin.pl
metro website says an intenerary from mirror lake in fed-way to near blanchard in d-town seattle takes about 40 minutes total…on the 179. the online map shows 179 travelling a good way up i-5. 40 minutes for a ~23 mile morning commute doesnt seem bad at all. car competitive even???? is there some bus+rail plan in the works to do away with such i-5 routes??? the above time seems quite fast.
Itinerary #1
Walk 0.1 mile NE from KENT/JAMES STREET P-R to
Depart W James St & Lincoln Ave N At 07:04 AM On Route MT 159 DOWNTOWN SEATTLE
Arrive 4th Ave & University St At 07:44 AM
Transfer to
Depart 4th Ave & University St At 07:50 AM On Route MT 33 E MAGNOLIA
Arrive 4th Ave & Lenora St At 07:54 AM
http://tripplanner.kingcounty.gov/cgi-bin/itin.pl
are these timepoints regularly met???
on the east side of i-5 , from james st in kent to near blanchard d-town is ~21 miles and 50 minutes. i have been on aurora in rush hour and could walk nearly as fast as a car.
are there transit advoactes in king county who are truly calling for busses off of i-5??
A detailed look at not sending buses up I-5.?? did you really do that with your non-editorial???