Bellevue Transfers in 2030

66 Replies to “Bellevue Transfers in 2030”

1. You are calling the transfers at Bellevue Transit Center “good” and the transfers at the adjacent Bellevue Downtown Station “mediocre”. I don’t get it. These transfers are pretty much the same thing.

   1. It’s a block farther on the currently suggested route as it would go down 108th not 110th.

      It’s probably not as bad as ‘mediocre’ though.

   2. I am satisfied with “mediocre” as ST cheaped out by not installing down escalators at this Link station. It’s about 50 steps down (like going from floor 4 to floor 1 in a home or office building).

      So for many types of people they must wait for an elevator. At least ST did install two elevators for each side platform.

      Those of us with arthritis symptoms understand this all too well. It can be more painful to walk down stairs as it is to walk up them.. That’s 1/2 of all seniors and 1/4 of all adult women. It’s a much bigger group of people than merely those in wheelchairs or using walkers.

   3. Another time consideration is how long a transferring rider has to wait at the signal at NE 6th and 110th.

      Are there any observations on how this is currently going? Long wait? Jaywalking? I have only been there on the 2 Line opening day — so that was atypical.

      1. It’s not a long wait… it seems shorter than 108th on the other end of the transit center, although I haven’t timed it.

      2. Exactly. It’s not just distance but the impact of crossing streets especially Bellevue’s notoriously car oriented stroads on foot to make a transfer.

         Run to make it across traffic or wait 30+ minutes for the next one? Traffic engineers are so blind to motorist delay and loss of a couple seconds but couldnt care less about a pedestrian just missing a bus by a few seconds by waiting at a busy light and then having to wait 30 plus minutes for the next one. Next time they’ll drive and help contribute to the congestion.

      3. @poncho, 110th hasn’t felt like a stroad when I’ve been crossing it. There isn’t much traffic in any direction, and the raised intersection really helps.

         On the other hand, one lane has been closed off for the construction north of the TC. Maybe that should just stay closed or become a bus lane?

      4. @ William C:

         I see the end state as either making 110th as a two-way “transit mall” with local traffic access to the parking garages, or a 108th-110th as two one-way streets with bus lanes that create a transit loop.

         Of course, had ST kept a way for pedestrians to cross underneath 110th when they changed the station layout a decade ago, there would be much less of a problem. It’s just one of many Link pedestrian access shortcomings around the region emerging now that so many new stations are opening.

      5. 110th is good now that the raised crosswalk has been built, the pedestrian cycle is fairly frequent. My concerns are other transfer points in Bellevue where busy streets must be crossed.

   4. I timed the transfer walks on opening day. It’s 4:37 minutes from the Bellevue Downtown train door to the 550 stop (in the middle of the transit center). That’s enough to make people think twice about doing the transfer or riding transit — on the most critical transfer in the Eastside. The fact that people are thinking about workaround transfers at South Bellevue, East Main, or Wilburton, shows that the primary transfer has some problems. This should have been ST’s highest priority in the Link alignment: good transfers to Stride S1 and S2 and all the bus routes going everywhere. That’s why we have a transit center.

      ST could have put the station under 110th to cut the transfer walk in half and make it closer to a world-class transit hub. But it couldn’t be bothered.

      1. Did you time it the other way, Mike? (No down escalators)

      2. I just timed this, and I got 2:33 from inside the train to the 550 stop. So maybe I walk faster than you do, but I guess either way this is more than your criteria of two minutes. However, you could have a transfer even between different bays of the transit center (e.g. between bays 1 and 6) that takes longer than two minutes.

         I’m not going to argue that the station location is ideal – it’s clearly worse than putting it further west. But I feel comfortable recommending to anyone that they take Link to transfer to a bus at Bellevue Transit Center. Transferring to buses in Downtown Seattle is usually worse.

      3. “Did you time it the other way, Mike?”

         I took the elevators since it’s hard to go down stairs.

      4. “But I feel comfortable recommending to anyone that they take Link to transfer to a bus at Bellevue Transit Center.”

         Good, that’s what this article and debate is for, to get a consensus view on how good the transfers are and whether they’re good enough to not be a deterrent or turn off occasional riders.

         “Transferring to buses in Downtown Seattle is usually worse.”

         Maybe. You’re not crossing a minimum of one street. I have one of the worst transfers, from Westlake Station to the Pike Street eastbound buses two blocks away. But those are outliers, not all transfers.

      5. “this is more than your criteria of two minutes”

         That’s one of the issues too: are my thresholds reasonable? What do average passengers think when they see the Bellevue Downtown – Bellevue TC transfer, especially those from cities with comprehensive metro networks? Would they say it’s normal for a subarea transfer (the main Eastside transfer), or excessively long?

      6. I suppose that it would be impossible to convince ST to rebuild the transit center to directly above the station in some future project?

      7. I think perception is important. People hate waiting, and are (generally) OK with a short walk. I’m no different. Before One Bus Away I would often just walk, only to be passed by the bus right before the last stop. Even now, Google will steer me towards waiting but I will take Link instead. For example, from Roosevelt to Pinehurst I can take the 73 or I can take Link (followed by another bus). I typically check One Bus Way (or the reader board) and then decide. This one time Google suggested I wait for the bus but I was having none of that. I took Link and (by using One Bus Away) realized that Google was right. Yet I felt like this was the faster trip. I’m sure I spent a fair amount of time walking from platform to platform. I also spent a fair amount of time waiting (since I waited twice). But each wait wasn’t that bad and the walks weren’t so long that it felt worse than waiting.

         The point being, I don’t think it will have much of an impact on ridership. I think frequency and timing are a much bigger deal. If you just miss the train you have to wait ten minutes. That sucks. If you miss the bus you have to wait fifteen minutes. That sucks worse. People do adjust and get a rough idea of how long it takes to get from one place to another (which includes the extra couple minutes of walking) but I think most people will feel it is either good enough, or takes too long.

         There are exceptions of course. It is a different world for those with mobility challenges. Some transfers are annoying, in part because they feel more stressful, like maybe you are doing it wrong (and will miss your connection as a result). Monorail to Link is like that. I can’t say I’ve mastered it. But regular users figure out the tricks (just like I used to know the best way to walk through Frederick and Nelson’s from Westlake Station).

   5. “Another time consideration is how long a transferring rider has to wait at the signal at NE 6th and 110th.”

      My timing includes that. For my two instances the signal wait was medium, maybe a minute or two.

2. The hard thing about assessing the Bellevue Transit Center is about how much curb space the center requires. All these buses cannot stop at the same place, and if it lays over it probably needs its own space.

3. I mentioned this on the Rapid ride K thread, but I’ll say it again here, since it directly relates to transfers in Bellevue…

   In the 2030 network, there will be two ways to get from Kirkland to downtown Seattle, as currently planned – 255 to Link at Husky Stadium or K to Link at downtown Bellevue. This has me speculating whether Metro will even continue to the 255 long term as an all-day route. After all, compared to the K->2 line combo, the 255 is essentially just a shortcut (plus a tiny bit of coverage on Lake Washington Blvd. if the K is rerouted to 108th), and shortcut routes tend to have difficulty remaining around unless they get very good ridership. If the 255 were cut back to peak only some day, Metro would likely try to ease the pain by adding the detour on the 270 into South Kirkland park and ride, which would at least connect the U district route with the K line and allow people who drive to the park and ride to continue to get a once seat ride to the UW. Which would, of course, make Bellevue to U district rides take longer.

   I would personally *not* be enthusiastic about such a shift, but I can also easily imagine a world where a recession happens in the 2030’s, somebody’s got to lose service, and the above restructure (which, while not great from a rider’s perspective, does save a lot of service hours) ending up being the best choice for the greater good that minimizes the pain (with the alternative being running the existing buses in the area half as often in the evenings and on weekends).

   The 542 is another shortcut route that could eventually see itself cut back to peak only, should a recession hit, as the alternative of sitting on Link for 50 minutes would make the route feel superfluous.

   If the long term trend continues of per-service-hour costs getting higher and higher, outpacing inflation, we could eventually see a situation where, by 2040, SR-520 has no buses on it except for the 270, outside of peak commuting hours.

   1. I don’t think Metro will delete the 255. The long-term plan is for Link to replace the 545 and 550 and to truncate the I-90 routes, but to strengthen 520 bus service to UW with the 255 and 542. I could see Metro reducing the 255 to half-hourly or hourly if it has to reduce all Kirkland routes to 1980s levels, but not deleting it.

      But your feedback as a Kirkland constituent is a significant factor in what Metro should do, and might carry some weight with Metro (ha ha). Do we keep the K and 255 as the most frequent routes in the northwestern Eastside? Do we double down on the K and deemphasize the 255? Would a K+270 be an acceptable way to get to Seattle? Or is that asking too much in travel time and transfers? Is K+2 Line a reasonable way to get from Kirkland to downtown Seattle? It it a reasonable way to get to south Seattle or the Pacific Highway corridor?

      1. I think that 255 to Line 1/2 is the best route for most destinations in Seattle. It takes the same amount of time from the South Kirkland Park and Ride to Bellevue TC (via the 250) as it does to Husky Stadium (via the 255). And the frequency at Husky Stadium will be double that of Bellevue TC

         Even to Rainier Valley, taking the 255 to Line 1 will probably be comparable to K -> Line 2 -> Line 1

   2. I mentioned this in the other thread as well:

      I’m not sure about this; the Totem Lake – Juanita – Kirkland leg of the route gets a decent amount of ridership, and Kirkland – UW is fairly busy when school is in session

      Also, I think there will always be a need for a Redmond-UW connection over the 520. It is so much faster than taking Link; something like 20-30 minutes via the 520 versus 50-70 minutes taking Line 2

   3. > In the 2030 network, there will be two ways to get from Kirkland to downtown Seattle, as currently planned

      If you are heading to say cid or pioneer square it’s not too bad to use line 2 but for westlake or especially ud/uw it takes an extra 20 minutes to loop down south.

      I’m pretty sure they’ll keep the 255 the more likely one to get cut is probably 544.

      There’s 3 main routes across at 520:

      Kirkland to uw via 255
      Redmond to uw with 544 (currently 545 is to dt seattle is the main one)
      Bellevue to uw via 271/270

      1. Do you mean the 542 from Redmond to UW? There is no 544. The 545 will definitely be deleted because Link is predicated on that. ST added trips to the 542 to prebuild ridership and get people used to it. But more people probably still take the 545 because it goes downtown and is more frequent. The one time I had a choice, one afternoon at Redmond Tech, the 545 came first.

      2. > Do you mean the 542 from Redmond to UW? There is no 544.

         Thanks, yes I meant the 542. And the 544 is being partially restored, I got the two of them flipped.

      3. What’s the 544; I’ve never heard of it.

         There was a 540 from Kirkland to the U-District.

      4. The 544 that used to run Overlake Village – South Kirkland – SLU? If that’s being restored, I’ll be interested to see how it’ll do.

      5. Route 544 came from the North Eastside project of March 2020; it was suspended and never implemented. Route 544 is in the ELC network. It seems a very weak route; it and Route 256 should be reconsidered.

         See: https://cdn.kingcounty.gov/-/media/king-county/depts/metro/documents/projects/east-link-connections/routes/544.pdf?rev=d7a225fac1244fd682ad002ce674451c&hash=872C8A0205C35662CA6FAD6D51871E6F

         https://cdn.kingcounty.gov/-/media/king-county/depts/metro/documents/projects/east-link-connections/routes/256.pdf?rev=1e123caa55ac4bdcb8fd7bacd1342a16&hash=15B7DACA277EB2000D741E67BFDAF66B

      6. @eddiew, thanks!

         The 544 I’m not sure about; I’ll need to see how it does. If SLU doesn’t work, maybe Capitol Hill to duplicate that one AM stop on the 545?

         The 256 does look weak, but maybe it’d do better also pointed to Capitol Hill or SLU?

   4. To be clear, I do want the 255 to exist long term, as it saves considerable time over other options. And, I think, at least during rush hour, the long term future of the 255 is safe.

      But, I can see the bean counters at Metro someday arguing otherwise. Housing in Kirkland is only going to get more expensive with time, and the more expensive it gets, the lower it scores on Metro’s equity metrics. So, you have a gradual shift in service hours out of east king entirely, into other parts of the county. What service hours continue to exist in Kirkland, the K will get priority because it’s RapidRide, and RapidRide routes are usually protected during recessions, at the expense of other routes. At the same, I’m imagining the cost per service hour continuing to increase faster than revenue, forcing slow, gradual cuts in bus hours, system wide.

      Also, if the 255 does get cut back to peak only, it won’t happen all at once, but gradually. In particular, I’m imagining frequency getting cut, causing riders to drive to a park and ride served by a more frequent route, which causes ridership to drop further, leading to more service cuts.

      Also, for those fortunate enough to have a car to drive to south Kirkland park and ride, a detour on the 270 would make the loss of the 255 almost unnoticeable. It would only be people who don’t have cars that would be stuck with an extra transfer. The question is, will there enough of them for Metro to care, after another 10-20 years of housing inflation?

      1. Does the cost of housing really figure into transit service decisions? I’ve never heard of this before. Recently there’s been a shift of service away from the Eastside and to South King County, but I thought that was just because more people were riding transit down there.

      2. @asdf2

         I don’t see 255 getting cut much further than it already has. It might drop down in priority if Kirkland freezes any new construction, but as it stands Kirkland is doing an okay job of building infill housing, and I expect ridership to continue to grow as it densifies, especially in downtown and Totem Lake. Also, the 255 will be cheaper to run once the 520-Montlake exit is complete, and even cheaper if/when it moves to Lake Wash Blvd.

         There are plenty of low-equity routes that have good service because they serve a lot of people. Routes 40 and 44 come to mind.

         @Christopher Cramer

         Higher housing prices is likely strongly correllated with reduced service, but I think the correlation is in the other direction: housing is typically expensive because it is big and takes a lot of land, meaning less density to serve, meaning lower ridership.

         KCM does give an equity score to routes, which influences service. But as far as I can tell ridership is a stronger signal than equity.

      3. I doubt they will kill off or even reduce the 255. It performs fairly well, and covers a relatively dense area. Much of it is upscale, but there is plenty that is not. I see it complementing the K. The K is useless for getting to the UW (and doesn’t serve Juanita).

      4. I hope you’re right. But not super optimistic.

      5. This is all unfounded speculation, that Metro would reduce the 255 to peak-only, or divert the 270 to South Kirkland P&R. The 255 is the primary route between the northwest Eastside and Seattle. That’s the highest-ridership corridor in the northwest Eastside. The 250’s ridership between Kirkland and Bellevue is pitiful. The K is hoping to change that, but that’s uncertain. You wouldn’t cut off the northwest Eastside from the entire west side based on a hunch that someday Kirkland-Bellevue ridership might overtake it. That would be like reducing the 101 and 150, when they’re the highest-ridership routes in Renton and Kent.

         Equity is historically underserved areas: lower-income, BIPOC, mostly non 9-5 non-downtown service workers. Those are the same areas where ridership didn’t decrease much during the lockdowns, and where transit service has historically been less.

         Metro has a map of equity-priority areas. It’s mostly what you’d expect: south King County, southeast Seattle, eastern West Seattle. But it has some surprising pockets too, like parts of Lake City, Bitter Lake, Broadview, somewhere north of Northgate Mall, Issaquah, and Snoqualmie. The county councilmembers know that parts of north Seattle are on the list but they sometimes seem to forget.

         Housing prices are rising regionwide. That shows the fallacy of saying that replacing single-family houses with multifamily or gentrification raises prices. Prices are rising everywhere, and even if those multifamily/gentrifying buildings weren’t built, prices would rise on the existing buildings anyway. It’s a regionwide housing shortage.

         However, lower-income people have the biggest crisis because housing prices keep rising faster than wages, and all the below-median units are already filled so only above-median units are available — and those are the ones they can’t afford. Lower-income people disproportionally live in equity-emphasis areas that nobody else wants, have non 9-5 jobs, ride transit more, and don’t have a car. So it all goes together.

         Of course, the transit network should be comprehensive: 5-15 minute service everywhere full-time, in both Skyway and Kirkland, like cities worldwide typically have. But the counties aren’t ready for such a commitment, so we have to split the pie fairly, recognize that lower-income people have more difficulty and fewer choices, and address historic imbalances that neglected transit in lower-income areas.

         The 255 and K would be the last routes in the northwest Eastside that Metro would cut. If those get reduced to half-hourly or hourly or peak-only, then everything else would be hourly or suspended.

      6. The 255 is emblematic of many eastside buses. I could see how housing costs would affect transit use.

         Whether Kirkland allows more density or not housing prices are going up. It isn’t like the multi-family housing in downtown Kirkland is remotely affordable, and the height of the buildings means each unit uses a small area of the building footprint but still is wildly unaffordable, way more than in downtown Seattle or CH. Totem Lake is multi-family and dense but not very affordable and has terrible transit ridership per resident but a good local mall.

         IMHO for routes like the 255 ridership and transfers will depend on whether the trip is into Seattle or within the eastside, and whether it is for work/school or discretionary.

         When I used transit for work on the eastside I wouldn’t do a transit to transit transfer but would drive to a park and ride for a one seat trip. Even going into Seattle I don’t like a bus to bus transfer so prefer a park and ride and one seat ride.

         Seattle and the UW are just denser and much more difficult to park. When you get there you are “there” more often. So from the eastside most eastsiders will not want to transfer again in Seattle, and many see a transfer downtown or on The Ave. as too dangerous, which is the reason for the inefficient long one seat suburban buses to First Hill. It isn’t as if the transit agency doesn’t know those peak runs are not expensive per rider (but cheaper than many S. King Co. buses per rider).

         They want whatever transit they take from the eastside to get them to their Seattle destination. That is mostly downtown, UW and maybe SLU. Transit to transit transfers for trips within the Eastside are not popular when almost everyone has a car and there is so much park and ride space. Transit to transit transfers in Seattle are dangerous or also not popular.

         I think wealthier eastsiders will be more willing to take Link than a bus to Seattle although those trips will be more and more discretionary so although housing prices will continue to soar that won’t matter as much compared to whether they want to go into Seattle and Link goes to where they want to go without a transfer in Seattle.

         What really matters is whether those discretionary transit users think there is anything in Seattle other than work that is not available in Kirkland, Bellevue, or other areas on the eastside. Today there are just a lot fewer people going from the eastside into Seattle. Before Covid we always went into Seattle to party. Much less so today probably due to habit and Seattle has less to offer to make taking a bus there worthwhile. I would think that will be the main determination whether Metro keeps routes like the 255, although eastside transit service and money is getting shifted to other areas a lot.

         But I don’t know how many people will be riding or transferring. I was just in Bellevue yesterday at noon dropping off some stuff to be reupholstered. 130th is reduced to one lane one way at 20th for construction so I had to drive down 132nd to Spring and then over to 130th to get to the upholsterer. While waiting for the light at 132nd and Spring to go right the cross bars came down, and the train crossed through the intersection at five miles/hour, and I swear to God there was not one rider on the train. At the same time right next to the station is a gravel pit, and like most of Bellevue a lot of parking lots and suburban street layouts.

         So at this point I agree with asdf2. I see the 255 getting cut back to peak because of low ridership. Some think it is unfair to shift taxes and transit service from the Eastside to other equity areas but Metro has shown on peak routes to Seattle it will do that if ridership is low. Metro just doesn’t have the money to rely on induced demand.

      7. @Frank K

         The 255 isn’t a standout route but it does not do poorly. I ride it fairly regularly (a few times a week) and it is reasonably busy, even at non-peak hours. Metro’s numbers back up that observation, and ridership continues to grow (albeit at a slower pace than Seattle-side)

         I think you have a skewed perspective based on your own transit usage. Ridership on the eastside is clearly weaker than in Seattle, and that is likely for the reasons you mentioned, but ridership is not really that bad. Every single frequent route on the eastside does reasonably well and is unlikely to get cut. From my perspective poorly performing infrequent coverage routes and peak-only routes are much more likely to get cut.

      8. I agree John. I also think the 255 will get a boost in ridership as the 520 work raps up. I would imagine a lot of people just gave up on the route. Soon it will be better than ever.

      9. The end of the construction will help, but at least in my case, I really need the evening service to get back to every 15 minutes before I ride it as often as I used to, something which I don’t have much faith will happen.

         It doesn’t matter how fast the is getting into Seattle if I’m stuck waiting at the bus stop for a long time to get back home.

4. The Link platform is much closer to 112th than it is to 110th. But there is no bus proposed on 112th at the station. Is this a missed opportunity?

5. One other issue with the Bellevue Transit Center is the added time for signal waits and pedestrians need more time to walk across wider streets. Every turn generally adds time. That’s especially true for a wide street like NE 8th where signals take quite awhile to cycle through. It’s also so what an issue on NE 4th.

   So going from Nordstrom in Bellevue to Whole Foods on NE 8th isn’t a direct trip. Neither is going from Old Bellevue to Main and 116th.

6. I timed the walk signals on either side of Bellevue Transit Center. On 110th, on the east side between the transit center and the train station, it was 1:44. Halving that for an average wait gives you 52 seconds. On 112th, on the west side between Bay 1 and the other bays, it was 1:08. Halving that gives you 34 seconds.

   1. Halving the time is not the correct assumption. You are thinking like a driver.

      The pedestrian countdown starts about five to ten seconds after the walk sign pops on. Sometimes cities make it longer — but generally there is at most about 30 seconds of every cycle when it’s advisable to cross as a pedestrian. So if you miss a cycle you’re waiting a long time.

      This is a reason why overly wide streets with left turn phases are so awful. There is only a small window of time in each cycle when pedestrians can cross with some sort of protection. .

      Of course, jaywalking is common when pedestrians don’t want to wait that extra minute plus. They will see the punt form but cross anyway. So it’s kind of a fuzzy opportunity percentage — but it’s certainly not half unless jaywalking is assumed.

   2. And if you don’t notice the green in the 10-second window you have to wait another cycle.

      1. And if you have to press a beg button to ask permission to cross the street, the time required to cross doubles. You have to wait for the light to complete an entire cycle before it has any effect.

   3. Average wait time is only one aspect of it. The likelihood that you will have to wait makes a big difference. Drivers routinely brag about “making every light” or whine about the opposite. Same thing happens when walking.

      It gets complicated. Measuring the cycle is the first step, but you also have to measure how long your phase lasts. Then you have to calculate how long you can cross within that phase. With every phase there is basically “waste” as well. There is a brief period when a light is red for everyone (right before the light turns green for someone). There is also the time the light is yellow. For example a car might have a 20 second phase, but 3 of that is the yellow light. If you arrive during that phase, you can’t cross. With walking it is worse (which is what Al is getting at). In some cases you only have a few seconds to cross. A bigger portion of your phase is wasted.

      So imagine a light cycle that takes a 60 second to go through every phase. 40 seconds are spent on other phases. 20 seconds on the crossing phase, but if you arrive in the last 10 seconds, you can’t cross. Thus:

      0:00-0:40 — wait
      0:40-0:50 — cross
      0:50-0:60 — wait (flashing “Don’t Walk” followed by solid “Don’t Walk”)

      Thus there is a one in six chance that you don’t wait at all. If you do wait, you could be waiting up to 50 seconds.

      Beg buttons complicate it even further. Some beg buttons for mid-block crossings are on their own phase. Some of these are very responsive. As soon as you press them the phase is initiated. The other lights turn yellow, and a few seconds later you have a walk light. Thus you are guaranteed to wait, but the wait is just a few seconds. Unfortunately, even some of these make you wait longer. Sometimes they are designed such that if someone crossed just recently, you have to wait a while.

      Other beg buttons are timed to a traffic cycle. The most common type is a beg button that corresponds with a green light (for cars). The beg button can serve two purposes. One is to lengthen the phase (since it often takes longer for a pedestrian to cross than a car). The other is to initiate the phase in the first place. For crossings where cars are rare there are detectors in the ground. The beg button does the same thing. There are some beg buttons that are only detectors. In other words, if a car pulls up and the light turns green, then the walk signal is on as well. But there are others where the light will turn green and if no one has pressed the beg button, there is no walk signal. These basically guarantee waiting.

      If in that example above there is a beg button then things change. There is no ten second “lucky” period (where there is on waiting at all). Your wait time ranges from about a second (you press the beg button just in time) to 60 seconds (you pressed the button a moment later). Of course that assumes that no one else pressed the beg button.

      Looking at the streets, it appears that there are beg buttons. But it still isn’t clear if these are just for notification (similar to the detectors in the street) or if you have to press the button to get a walk signal. Even if they are required for the walk signal, it is likely that much of the time, someone else has pressed the button. Thus in the middle of the day a typical pedestrian has a good shot at crossing the street. Late at night they might be the only pedestrian, but it is also possible that they will initiate the phase (and it works like the mid-block example I gave).

      Hopefully these buttons are for notification only. If they aren’t, then they should be. Otherwise Bellevue is a joke. Beg buttons to get a walk signal downtown? How bush league.

      1. The City of Bellevue should be lobbied hard to remove the beg buttons at NE 6th and 110th, now that it’s literally between Link and the BTC as of earlier this year. It’s beyond irresponsible to leave them now that crossing 110th is a big proportion of the transit transfers.

         Ideally the light would stay green for NE 6th as the default.

      2. @Al S., the ped crossing is actually its own scramble phase, which does make sense given the TC platform in what’s effectively the center of NE 6th. Maybe that should be the default?

      3. It would be an interesting approach, William C. Crosswalks on walk until a car shows up and triggers the green. A vehicle “beg button”!

         Regardless, the situation should have been discussed and resolved before 2 Line opened. Are the pedestrian beg buttons still there after the recent changes?

         I’m not sure if ST could be a catalyst for making more changes as Stride comes on line. The Grand Connection effort is not yet fully designed either. The City brags about merely raising the crossing to eliminate the curb but it strikes me as ancillary to the bigger challenge.

         https://bellevuewa.gov/city-government/departments/transportation/projects/transportation-capital-projects/bellevue-transit-center-raised-intersections

      4. “But there are others where the light will turn green and if no one has pressed the beg button, there is no walk signal. These basically guarantee waiting. ”

         Unless they cross anyway thinking the signal is broken. If it’s green for the adjacent car lane going the same direction, it should be green for pedestrians.

      5. Unless they cross anyway thinking the signal is broken.

         Yes, I forgot to mention jaywalking.

         If it’s green for the adjacent car lane going the same direction, it should be green for pedestrians.

         Agreed. That is pretty standard in most of Seattle. I’m not sure that is the case here. That doesn’t mean that beg buttons have no purpose. They can be notification buttons. In other words, every time the light turns green (whether it was triggered by a car or just consistent time phases) there is a corresponding walk signal. But on streets that have very few cars (or a time a day when there are very few cars) a car has to trigger the signal or a pedestrian has to signal it (by pushing the beg button).

      6. I guess Seattle did the work during the pandemic: https://sdotblog.seattle.gov/2020/05/07/seattle-takes-new-steps-to-fine-tune-traffic-signals-for-people-walking-and-rolling-during-covid-19-health-crisis/.

         Nearly 75% of walk signals in Urban Centers and Hub Urban Villages will now display a walk signal every time there is a green light for cars.

         The city of Bellevue should certainly adopt this policy for the downtown area.

      7. A vehicle “beg button”!

         There are vehicle beg buttons. You can see them on the street. It is annoying for bike riders, because they don’t trigger them. Sometimes the easiest thing for a cyclists to do is go on the sidewalk and press the pedestrian beg button. Or just run the light.

         Anyway, the often look like little circles: https://maps.app.goo.gl/1cW8sMUDjm4FG9hQ7. This is a good example, since it is for the left turn signal. If cars don’t trigger it, then that phase is skipped.

      8. @Al S., the beg buttons are there, but I don’t know what happens if they aren’t pressed – whenever I’ve been there, someone’s already pressed them.

         @Ross, most of the loops I’ve seen around here are set to be triggered by bikes. It’s usually worked out well when I’ve been biking around, though I’ve seen a few annoying places where it hasn’t worked and I haven’t realized it until a cycle’s already gone by.

      9. I’ve seen a few patterns.

         1) Traditional one is a long green phase (20-30 seconds?) and short red-flashing phase. All of them were like this until a decade or so ago.

         2) Countdown: a short green phase (sometimes down to 10 seconds) and a long red-flashing countdown (up to 30 seconds). This lets you know whether there’s enough time to walk or run across while it’s red-flashing. I go across more on this one than the previous one because the red-flashing phase is longer.

         And if there’s more than two lanes and cars are moving, I tend to wait even though New Yorkers wouldn’t, because I’m not sure if I can trust my judgment on whether the car will reach me, and I don’t trust that drivers can brake normally until they slow down enough to prove it, because who knows if they’re not paying attention or drunk or the car doesn’t respond normally on a wet road.

         This pattern has a real problem when the green phase is only 10 seconds and you don’t notice it until it’s gone. If you wait till the next cycle you may miss it again and have to wait for a third cycle. Humans can’t just stare at something for minutes at a time without their thoughts drifting and losing perception of the light. Then they depend on their unconscious change-awareness to notice when it changes, but that’s what fails a lot of the time with walk signs. If you only have a 10-second window to notice it, then you can miss it. Walk lights need to be more usable.

         3) The car lane turns green but the crosswalk doesn’t. They never used to do this, and SDOT just silently changed some of them without explanation. Some suburban ones never turn green unless you press the beg button because pedestrians are such an afterthought. Sometimes there’s a reason for it remaining red, as when there’s a new turn phase with cars coming at you. The first few times I encountered this I thought the walk sign was broken. If the other directions are empty you don’t know whether somebody else has a green. Sometimes I’ve gone but another direction had a green turn and a car that wasn’t there came at me. So now I weigh the likelyhood more. The directions always used to rotate predictably clockwise or counter-clockwise, with maybe a left turn before or after the straight. But now these light are unpredictable and you don’t know which order they’re going in, or whether the crosswalk is red for a legitimate reason or it will never turn green. The walk lights need to be more predictable again so we can tell whether they know what they’re doing.

         I do remember from my one driving course in high school that drivers need the red-flashing phase to turn, otherwise they’re stuck at the intersection while pedestrians hog the entire car-green phase. So I check if a car has been waiting to turn, and if so, I give them the red-flashing phase. This happens in high-pedestrian areas like Pine Street, where there can be a continuous flow of pedestrians for several cycles, including those who don’t allow for car turns.

      10. Then there’s the issue of pressing beg buttons when your hands are full with heavy groceries, or you’re tired or sore and you’d have to turn and detour to get to the button.

      11. “There are vehicle beg buttons. You can see them on the street. It is annoying for bike riders, because they don’t trigger them. ”

          There are actually markers on the road showing where you need to wait on a bike to trigger them. At least in Seattle.

          There is new technology they are using in a Tacoma. It looks bell on a pole above the intersection. The engineer I talked to claimed it could differentiate between a bike, car or pedestrian, and trigger the light appropriately. In my experience it mostly works.

      12. > @Al S., the beg buttons are there, but I don’t know what happens if they aren’t pressed – whenever I’ve been there, someone’s already pressed them.
          > There is new technology they are using in a Tacoma. It looks bell on a pole above the intersection

          There’s the traffic detection cameras, with mass production of smart phones making cameras cheap and photo recognition software (or infrared) some cities have started using those instead as you don’t need the metal loop and makes repaving roads much easier.

          Another pro is as Cam noted it can detect pedestrians and bikes as well. I’ve heard some cases cities are using it to detect buses as well so you don’t need to install special hardware on buses for say the transit signal

          For example https://www.flir.eu/products/traficam-ai/

      13. I always thought they should use lasers. It seems like having two posts (with a laser going across) would be cheaper than something buried in the street. Maybe birds make a mess of it. The photo recognition thing seems like a good idea (as long as it works). It is always helpful if people know how the thing works, so they can help trigger it. Like waving your hands in front of the lasers to wash your hands (which come to think of it is a very good counter-argument for my traffic-laser idea).

      14. Tacoma is using something similar to this high dynamic range camera (maybe the FE3).

          https://www.cubic.com/transportation/vehicle-pedestrian-detection

7. Mike et al,

   I love the notion of measuring transfer quality. Transfers are where the transit trip really slows down; there is waiting; there is walking or rolling between stops. I have a riff on the term seamless; transit cannot be seamless; seams are of time (waiting), distance (walking or rolling), information (complexity of fare structure or network design), or money (fares). If agencies discuss seamless, they are being imprecise at best.

   Please note that STB and its readers still have opportunity to influence Metro on the K Line pathway, the fate of Route 255 in a K Line restructure, and influence the Council on the East Link restructure ordinance.

   Mike applies the term “meandering” to Route 249; that is valid; but it should also be applied to ELC routes 203 and 240; Route 240 is a pretzel; Route 203 misses Eastgate. Routes just outside the ELC scope also meander; see routes 230 and 231 at KTC, Route 239 inbound at Brickyard, and routes 225 and 239 avoiding 120th Avenue NE. The scope of ELC should be expanded to correct these issues. Routes stuck in regular traffic congestion also suffer minutes of delay; that has a similar cost of meandering; consider ELC Route 256 in the p.m. peak on the ramp between eastbound SR-520 and northbound I-405; or southbound 111 on the east to south ramp between I-90 and I-405. Back to Route 249; ELC preserves the least important part of the route; the Bel-Red and Overlake Link stations make the current pathway slightly revised to be powerful; the part south of BTC could be served by another route and could shift to 108th Avenue SE.

   Consider the K Line pathway recommended by Metro. Since its terminal is inside the TLTC, it cannot serve the NE 128th Street overcrossing with short walk transfers. Should Route 250 of today and the K Line of tomorrow serve the Spring District station? Yes please. Of course riders will want both a K Line and Route 255; should the latter stay on 108th Avenue NE; could the former take over for Route 250?

   Why was ST so cheap or unimaginative at BTC? Could BTC add a stop pair on NE 6th Street just west of 112th Avenue NE so the I-405 bus routes have a short walk transfer with Link? At Mercer Island, how about bus routes serving a stop pair on 77th Avenue SE to reduce that transfer walk; riders would not have to cross North Mercer Way.

   Since hours and minutes are scarce, excessive duplication should be avoided. So, why did ST keep Route 566 from the project? Does it duplicate Link between BTC and RTS? Do ELC routes 111 and 342 duplicate ST service on I-405? Does Route 630 duplicate Link? Would not MI be better off with shorter intra MI waits?

   1. Seamless transfers only matter when the routes get very frequent. Who cares if it takes 5 minutes to walk from one stop to the other if the bus isn’t going to show up for another 20 minutes anyway?

      And, seamless transfers do have tradeoffs. In many areas, the only way to get it is to detour every bus around into a transit center, which means slower trips for everyone, including people who are not even transferring. For instance, Sound Transit could have gotten seamless transfers between the 554 and KC Metro routes at eastgate by detouring the 554 into the transit center bus bays, but they wisely did not.

      1. I didn’t get into frequency because it would have made the issue overwhelmingly complicated, and transfers alone are worth looking at. I’m thinking about a South King County transfers article too, although I don’t know as much about the area or what the transfers at KDM or Federal Way will be like, and it may make sense to wait until a Federal Way restructure proposal comes out. (The process just started.)

         I think most Eastside riders know that the 2 Line runs every 10 minutes; a few routes like the 245, 250, 255, and 271 run every 15 minutes weekdays; but only the B runs every 15 minutes weekends and evenings. The rest are half-hourly, and some are hourly evenings and Sundays. So transfers are only convenient weekday daytime and between the B and Link, and otherwise you only do a 2-seat ride if you have no other choice and must go there.

      2. transit has many seams everywhere. an objective might be to minimize seams subject to the constraints of budget, coaches, and right of way. Yes, initially, Route 554 did serve the Eastgate local bays. But some route should connect those local bays with Link at Mercer Island via I-90. The local bays are about 800 feet and four floors from the freeway station. That is a wide seam.

      3. Seamless transfers only matter when the routes get very frequent.

         I wouldn’t say that. To quote Jarrett Walker:

         transit planners generally observe that the walking distance that most people seem to tolerate — the one beyond which ridership falls off drastically — is about 400m (around 1/4 mi) for a local-stop service, and about 1000m (around 3/5 mi) for a very fast, frequent, and reliable rapid transit service.

         So riders will walk farther to “fast, frequent and reliable” service. It is quite likely that “frequent” is essential here.

         There are other issues. Quite often, service is timed. The longer it takes to make the transfer, the harder it is to time. People don’t all travel at the same speed (https://www.researchgate.net/figure/The-walking-speed-of-pedestrians-is-distributed-normally-with-an-estimated-mean-of-134_fig4_286071735). If the vehicle is infrequent, you don’t want to leave too many riders waiting. So basically aim for half the speed of an average walker. Of course that also means that quite a few people will be waiting once they’ve made the walk. The farther the walk, the longer the wait. So the average person walked there in five minutes but a slow person got there in ten minutes. That means the average person has to wait five minutes (and fast walkers wait longer). Either that, or more people miss the connection.

         Thus they both matter. If the walk if fairly short (e. g. 30 seconds) then you can try and time things (e. g. for a one minute transfer). If both are frequent, then the main drawback is the walk itself.

Comments are closed.