Last week Commute Seattle came out with its biennial Modesplit Survey, showing more incremental progress towards a culture of walking, biking, and taking transit to work. In follow up discussions both internally and in comment threads, we’ve wondered about the balance between this cultural shift and the way we allocate our right-of-way. As One Center City prepares both short-term mitigation measures and long-term structural changes to our Downtown streets, it’s important that our right-of-way allocation align with the same progress we’re celebrating.

Getting a sense of which modes get proportional allocations isn’t terribly easy. Bus and bike lanes are discontinuous, bus lanes are often peak-only, and street widths vary with curb cuts, etc. (For a citywide perspective, Brock Howell had a great guest post in the Urbanist last year). In an attempt at getting our heads around Downtown, I counted 536 blocks in Center City (bounded in my definition by 1st Avenue, Mercer, I-5, and Jackson St). I then counted the total number of lanes on each of these 536 blocks to come up with the total number of ‘lane segments’ in Center City: 2,296. For simplicity, I counted parking lanes as full lanes, no matter their width, because they are still scarce space continually occupied by cars.

Graphic by the Author

On these 2,296 lane segments, there are 150 lane segments of bike lanes (6.9%) and 160 lane segments of bus lanes (6.5%).  The remaining 1,986 lane segments (86.4%) are taken up by general purpose lanes, on-street parking, and delivery/loading zones.  Off-peak, only 40 lane segments of bus priority remain (1.7%), mostly on Battery, Wall, and Westlake Avenue.

A few caveats are in order. I counted all streets, even those on which you’d never expect transit service (Clay, Eagle, Terry, etc.). I did not count the Downtown Seattle Transit Tunnel (DSTT) in the ROW allocation, only looking at the surface. And I treated ‘blocks’ as a consistent entity, even though their length varies, especially on the shorter blocks of south Downtown.  But the intent is just to depict visually how much raw space to which cars are entitled, and how much we’ve set aside for bikes and transit. I hope it’s helpful as One Center City discussions move forward.

67 Replies to “How Cars Still Dominate Downtown’s Right of Way”

  1. Nice. I’ve been wanting to do the same thing but ran into the same issues you did.

    I hope we can quickly move through the One Center City process. The clock is ticking.

    Option D appears to be the only really viable option. Transit advocates need to make it clear that truncating the 550 and 41 is a non starter of an idea.

    We need additional surface ROW for buses. Your graphic highlights what should be obvious.

    1. Considering the amount of damage it’ll do to put the 41 and the 550 to the surface, I wouldn’t give up on getting a working arrangement for these routes and the Convention Center to stay out of each other’s way.

      But out of basic humanity to bus passengers more than train riders, if transit agencies can’t get bus signaling, communication and training up to the point where the neither of the two modes can force both to slow down, it’s time to return riders to sunlight.

      Also air, as there’s no power ventilation in “Hush Mode”, and windows won’t open.

      Would suggest that Third Avenue be given to former DSTT buses and trolleys only. With training and signals to keep it all moving. Used to be trolleywire on First all the way to Broad. I don’t remember if it’s still there. Wouldn’t hurt to have some bus service through Belltown that isn’t on Third.

      Mark Dublin

    2. It is possible to split route 41 into two routes, one headed to north downtown and one headed to south downtown. It already has the most peak-direction frequency of any transit line — 14 runs during the peak hour — so an even split would have better than 10-minute headway on each new route. Adding transit capacity to allow it to traverse 3rd Ave would still be better, though.

      I don’t see any acceptable solution for route 550 other than to make room for it on 2nd and 4th, or wherever the rest of the I-90 expresses end up. Indeed, having route 550 join the other routes all headed to Mercer Island P&R is the one bright spot in all that is about to make route 550 awful.

      1. For the 41, Brent, all three will work. Worth some serious investigation to see how much service Third can handle. Really does seem like a natural for DSTT service. Especially if we get Third all transit, all the time.

        Except this time, get equip and train for the coordination the DSTT should have gotten. No need whatever for fareboxes- or cash fares at all. Think rules say we can do that without having to paint buses red and yellow.

        Like when were discussing the 591 of the Future, I think easiest one for the 550 is to swing north on I-5, and aim for Westlake Station. Sorry to make something so promising unawful, but this one should be cheap, easy, and work.

        No need to get anywhere near Fifth and Jackson. No pressure on Third Avenue. Cross-the-sidewalk connection to whole length of LINK. And all buses on Third and Fourth, and South Lake Union Streetcar. Unless we terminate 550 in South Lake Union, same as the 591.

        So don’t coddle the Seattle culture. The Inevitably Awful is a more shameful enabler of inaction than “We’re doomed because we didn’t do ‘Forward Thrust!”


      2. The 550 could be split as well, in much the same manner, as Alex suggested: I like all those combinations, and I think they work reasonably well. Some riders come out ahead and some come out behind, but that is better than truncating at the edge of downtown — or suddenly skipping downtown — where the majority would lose out.

        I put the SR 520 bus truncations in a different category, since the time penalty is relatively minor, the UW is a popular destination and the service savings would be huge. In that case you would likely come out ahead as long as you add service to various routes.

      3. I don’t think many people on here are metro 255 riders. The truncation of this route to the University would be as disruptive as truncating the 550 in the ID is to many riders. I catch the 255 every day and see 1-5 people per ride get on/off at the Montlake stop, everyone else is heading to or from downtown. We all know that increasing the number of transfers on a commute makes transit less convenient and increases the number of people who will switch to driving.

  2. Just to make the comparison fair, I think that the tunnel should be inlcuded. I assume it will still get the point across but won’t be as vulnerable to complaints of bias.

    In addition, it would be interesting to see the off-peak modeshare breakdown as well, since you have the off-peak lane share.

    1. I agree the tunnel should be included – and the viaduct and I-5, too.

      And maybe the Great Northern Tunnel as well?

      1. Sidewalks should also be included. In Center City, more transit space is needed, but I’m guessing that if similar data and graphics were collected for the U District, they would support converting some car space (such as parking on the Ave) to pedestrian space. The fact that people are allowed to monopolize so much space for cars that are not even being used while pedestrians are crammed together on too-narrow sidewalks is just as aggravating as how limited transit right-of-way is in the Center City.

      2. Eric is right. ROW or right-of-way is wider than the street pavement and it includes sidewalks. ROW goes pretty much from building front to building front in Downtown Seattle. The correct calculation should include sidewalks if you want to call it ROW. Otherwise, you should just call it pavement width.

      3. and while we’re requesting a magic data pony, we should also demand the full analysis of sidewalk percentage less driveways. should it also subtract sidewalk space consumed with newspaper boxes?
        The graphic is excellent, and useful as is.

      4. The final graphic includes pedestrians and trains while the infrastructure to support them isn’t included in the first two graphics. Adding in the DSTT and sidewalks using the same basic block/lane methodology would make sense.

        However, it doesn’t need to be expanded into a graduate thesis to be useful for discussion. :)

      5. I agree with William, if you include the transit tunnel, then you have to include the freeways (the viaduct, I-5 and I-90) and my guess is things look pretty much as when you started (although with a smaller percentage of bike lanes).

        As for the sidewalks, who cares? The only reason that mode is shown on the last graphic is because it is meant to be accurate and detailed (you could show “other”, but why bother). No one is arguing for bigger sidewalks. The point is transit is getting short changed, and that is obvious.

        About the only quibble I have is not separating out parking. Unfortunately, that simply complicates things more. Some of the parking is “load/unload only” or taxi pickup, and that serves a very specific purpose. So to get a meaningful assessment, you would need to sub-categorize even more (by type of parking). If parking lanes turn out to be a small percentage (10% or so) then it doesn’t really matter. It makes the case for transit very strong. It is one thing to say you don’t want to add transit lanes because you need to make deliveries, it is another thing to say you want to protect a mode that is very small, and one we shouldn’t be encouraging anyway.

    2. I didn’t include the tunnel because I only wanted to look at rights-of-way that face competitive allocation. The DSTT is separate and translt-only of course, but even if I had added it, under this method it would have been a single ‘lane segment’ in each direction for 19 blocks. It would have increased the peak transit allocation to 8%. Not much of a difference, so I stuck to surface streets only.

      1. The issue with not including the tunnel this is that your last graphic (the commuting behavior one) does include passengers using the tunnel. So by including the tunnel in one data set and excluding it in the other you are creating an apples to oranges situation. This would be much less misleading if the commuting behavior graphic was a different style or otherwise disconnected from the lane share graphics.

      2. The graphs err in the favor of undercounting car space by similarly not including the freeways, which, BTW, ought to have some transit priority.

      3. Yeah, what Brent said. If you include the freeways then it probably swings the numbers back to where they began. Only a small part of the I-5 express lanes and I-90 are bus only, if I remember right. Nothing for the viaduct. That means a lot of lanes for general purpose automobiles and very little for transit. About all that would do is make the space allocated for biking seem smaller.

        About the only argument for including the bus tunnel was actually stated by Zach. Even if you include them, only a small segment of downtown surface area is designated for transit, even though the vast majority of people heading there are using it.

  3. In general, I am a bit weary of proportional-allocation arguments because they have been used time and time again in other parts of the city to argue for devoting the street exclusively to cars and not bother to accommodate other modes. Witness the outcry over the Nickerson and 125th St. road diets as examples.

    There is a difference here, of course, in that more car lanes on a road makes the road much more dangerous for bikes and pedestrians, which is why these road diets were done in the first place, whereas, more bus lanes downtown does not make the road more dangerous for cars or any other road users.

    1. It’s a good point, and I’ve often countered to car types who complain that way that if we’d followed their logic, we would have never built the freeway system because everyone traveled by train. Rather than say modes are ‘deserving’ of something based on current usage, I think the more impactful argument is to say that we have a whole suite of policy reasons to move away from cars and towards transit/walk/bike, and given that, we should see proactive movement towards proportionality that reflects those values.

      1. Unfortunately, best comeback can only have its full effect if delivered on-site to a motorist going zero or less for an hour, in which time ten trains, each one same capacity of a mile of traffic, go by just uphill at 60.

        Text? “And that would be a bad idea because??????” But I remember the early days of our seventy-year shift to a cars-only world. So I think one of the things that encouraged so much mistaken thinking was that up through about 1960, transit was still good enough to take enough load off car traffic that things could actually move.

        Speaking of which, I think I’ve suddenly thought of an acceptable use for at least an automation mode for cars. Which would be a lot more effective address to distracted driving than fines and crashes. Have cars go automated automatically when brake gets depressed more than ten times a second for five minutes.

        Though since hands need to stay free in case jam suddenly lifts, create hands-free twitter. And everybody select a tone-of-voice-word-choice icon. Contests like seeing who can use the phrases “Beautiful Wall” and “Dishonest Press”. and win by having their message appear on FOX news! Here goes:

        “Looking at the Scots Highland’s history of terrorism, whoever let this world of violence, theft (usually sheep!), and religious fanaticism in here should be very, very ashamed about Timothy McVey. Even worse than fact that nobody’s wall ever worked in keeping out Scots is that there’s no way to get them to pay for anything!”

        Check the screen! Did I ace out Steve Bannon?


    2. It’s one piece of information, but it’s only a minor guide because there has been no study or comparison that could definitively say what the percentage should be or what the impacts of changing it to another percentage would be. That would require a citywide inventory of all the neighborhoods, and comparisons with other cities that have different percentages, if they even have that data. But it’s still something that’s good to know and think about.

      I wonder if we should also look at potential-bus streets and see how the percentage changes. I’m not sure if including every little non-bus street skews what we should be thinking about. The neighborhoods have a lot of two-lane residential streets that never have traffic so cars and bikes and peds coexist well, and things like bus streets and greenways are really more appropriate every ten blocks, not on every single block.

      1. I’m pretty sure peak bus flow, peak car flow, peak bike flow, and peak pedestrian flow are roughly simultaneous. The gimmicks that allow off-peak access to bus lanes don’t tend to have engineering basis, but still assuage some special interests. But those peak/off-peak workarounds lead to confusion, and cars using the transit-only ROW when they aren’t supposed to be.

        Simplifying the transit lanes (especially all of 3rd Ave) by making them 24/7, and painting the street red, should be the low-hanging fruit of mode reallocation.

      2. @Brent — I wonder how much of that is for freight though. For deliveries, there may not be a very practical alternative. If traffic is minor, there isn’t much of a problem. I’m not sure how best to accommodate both uses. Maybe a “delivery only” rule, with drivers required to have a permit. At a minimum, tighten up the timing. On 3rd Avenue you can drive on that road from 9:00 AM to 3:00 PM (then again at 6:30 PM). How about a sign that says “Delivery only — 10:00 AM to 2:00 PM” (with the same “Do Not Enter” symbol that exists today). That is pretty clear cut (I won’t drive down that road). It would be pretty hard to talk your way out of that, really. Either you are delivering goods, or you are cheating, and it would be pretty obvious (unlike today). The folks most likely to be cheating would be truck drivers who are running late (a little after 2:00 PM) and those folks aren’t likely to cause much of a problem.

    3. But we can’t lose sight of the fact that certain modes are simply better than others. From an environmental and health standpoint, transit is better than driving alone. From a practicality standpoint, it scales better. Simply put, transit is better for society, so unless you have that same argument (i. e. folks aren’t using it) there is no reason not to allocate more space for it.

      It reminds me of vending machines in schools. Do you sell candy, or apples? Apples are clearly better for everyone, so maybe that is the way to go. But if you can’t sell apples (and you can sell candy) then maybe you just put up with a little obesity and tooth decay to get new history books. However, in this case, apples are actually *more* popular than candy. There is no “tough choice” here. It is simply a matter of taking space from cars and allocating it to the buses.

      About the only problem I have in general with this graphic is that include biking. Unfortunately, biking is a lot more complicated. For one thing, a lot of people won’t bike unless the route is relatively level. But more to the point, for many, biking requires a constant, complete set of streets that are comfortable for the rider. So having a street that works well for bikes is great, but only if you can get to that street. As many bikers have mentioned, we lack that for downtown. As a result, while the numbers look good for biking infrastructure (relative to ridership) that simply isn’t the case on the ground.

      So while this set of graphics is quite accurate, simply showing single occupancy driving versus transit (both in percentage of users and space allocated) might be better. As it is, though, it is great. I think the obvious take away is that our system is very much outdated. We need to allocate more space for transit.

      1. The issue with biking downtown is partially topography but we have a lot of infrastructure that just terminates without connecting to anything useful. Why doesn’t 2nd Ave. need bike lanes North of Pike/Pine? It’s not like the bikes just levitate there or other traffic disappears.

      2. @R — Yeah, exactly. More so than bus lanes, bike lanes need a complete path. You can have bus lanes for have the journey, and the bus might actually be much faster than a car. But if you only have bike lanes for half the trip, and the other half is very dangerous, most bike riders won’t make the trip. You need a complete network of safe streets for bikes (a combination of bike lanes and quiet side streets).

  4. Great research.

    Is the pedestrian pushing a beg button … or looking at their phone? I read it as a phone at first. Which might not be the right message.

    I think depicting a beg button is a good idea as it reminds that pedestrians do actually have to travel on the roadway some of the time, in addition to using the sidewalk.

    1. Looking at a phone. I thought I should be as accurate as possible, and be a little cheeky. That’s what I see most pedestrians doing. :)

    2. Jonathan, cautionary example of how wrong words lead to mistaken thinking. Take a leaf from those speakers that go “Wait! Wait! Wait!”. Which NEVER beg!

      Have giant speakers, like laser-focused boom-boxes, give same message to cars when you push the button. If windshields start to break, volume only needs to be raised ’till the rest of the windows disintegrate too.

      Even more than a button-pushing hour (I’ve always suspected these things are sadistic UW psych experiments to see how many times the subject has to push the button before they start using a rock)(or the guy from Candid Camera puts his arm around you and point to the camera, and you win a refrigerator) no Nation has ever stayed free once it starts taking orders from a light post.

      Cars? Since being stopped is their customary rush hour condition, they shouldn’t even need a tail-light broken to obey like a happy dog. And the owners get out and roll them over and pat their oil pan.


  5. And yet we are talking about truncating bus routes to remove buses from downtown while doing nothing to restrict SOVs

    1. A case can be made for truncating the routes irrespective of how they compete with SOV traffic.

      Metro/ST essentially have a budget of the number of bus service hours they can pay for. The idea of truncation is that you trade service that is duplicative for service that expands the system either by adding new routes, extending routes further at their distant ends, or increasing the frequency of service.

      Even with the transfer it is possible that a truncated bus line and subway could provide faster and more reliable service than a bus which travels into the city center. The city center portion of a trip is often the slowest and most disruption prone portion of a route.

      1. That tradeoff is a proper one when the transfer is well-designed and efficient, and when there is an increase in frequency. However when evening and weekend service remains at 30 minute headways (and 60 minutes weekend evenings) and there is no frequency increase… and the transfer isn’t efficient, and there won’t be many service hours saved… it serves to make driving more attractive as an alternative.

        But my point stands – we’re being told to remove buses from downtown while there is nothing similar for SOVs. And really strangely we’re being told to run more express buses downtown that duplicate Sounder while priced lower.

      2. I’d like to see more dedicated transit lanes at chokepoints downtown. Buses can’t deviate from their routes but the SOVs always find their most efficient routes, so dedicating some lanes to transit is a good tradeoff

      3. There is no plan to run more express buses downtown, except those currently in the tunnel. My plan Friday, which you got totally wrong, is to move trips to where the ridership is actually going.

      4. I can’t speak for the north line, but there are very few buses that duplicate Sounder in the south. The 578 doesn’t run when Sounder is running. The 590, 592, and 594 are the main routes between Lakewood or Tacoma and Seattle. Sounder, which sometimes takes longer than these buses, duplicates them only incidentally because it serves both the West Pierce-Green River Valley and Valley-Seattle markets. The 580 only duplicates Sounder on trips for which the paired train reaches Lakewood.

      5. Yeah, what Carl said. The transfer is not likely to be well designed or efficient. The transit tunnel was designed as a bus tunnel, and worked really well for that purpose. But it is a deep bore tunnel, and transfers from buses will always suffer from that penalty, along with the awkwardness of the surrounding streets.

        But more to the point, the service inferiority of surface transit through downtown is largely the result of lack of investment in downtown transit infrastructure. Widen the transit tunnel (to accommodate the train and buses) add the WSTT and we aren’t having this conversation. All the buses go into tunnels, and they all move without interruption. There is a penalty for riders — time spent on (broken) escalators — but as shown forty years ago, a small price to pay for fast, reliable service. Even today, a large percentage of riders take the train *just within downtown* despite the extra time spent getting to and from the platform.

        No one is suggesting that (better to have a stupid SR 99 tunnel and light rail to Fife) but if you simply allocate space for buses on the surface, then the buses aren’t stuck slogging through downtown, and everyone wins. Riders get to where they want to go much faster, and the transit agencies aren’t spending a bundle to have drivers staring at cars blocking their way.

  6. Zach, it’s stunningly obvious that you should either have excluded the riders using the DSTT or added the “block-lanes” that carry them. Since there are twenty-two blocks under which the tunnel passes and two lanes per block, plus the center “passing lane” in three stations each two blocks long, that is a total of fifty “lane blocks”. If you add those to the total you derived for surface streets, it raises the denominator of your percentages by 2.18% to 2,348. That’s a trivial increase.

    However, it also adds those same fifty “lane blocks” to the two numerators of the transit “lane share”, raising them to 200 for peak hours (8.51%) and 90 (3.83%) for off-peak hours.

    As a result the sizes of the auto- walk- and bike-reserved blocks would each have shrunk (ever so slightly for the cars) while the transit-reserved would have increased noticeably. but the difference would still have been enormous. You just gave the Sam’s of the world ammunition to claim that you’re biased.

    Clearly, nearly all of us who post or comment here are “biased” in that we favor good transit solutions. But like good progressives we pride ourselves on being “data-driven”. You really should correct the post, which I understand is probably not a trivial task.

    1. I see that you already answered Stephen making the same point.

      The clear answer to your reply is then “Remove the transit ridership which uses the dedicated right of way.” That is far more damaging to transit mode-share, though. That’s exactly why it’s there.

  7. Second Brent’s comment above about making Third a 24/7 bus-only street, at least between Bell and Washington where the congestion is most serious.

    I’ve been thinking in vague terms about a similar calculation of how many people each north/south avenue downtown moves during peak hour. Even back-of-the-napkin math suggests that Third and the tunnel each move many more people than any of the SOV-dominated avenues, and that Second and Fourth are next best only because of the buses they host. Thinking in terms of efficient mobility for people, the single best thing we can do is to make Third work better. And that requires getting the cars off of it. When turning, even legally, they block buses. Even when going straight, they get in buses’ way.

  8. Now, about that square footage allocation of non-public right of way, such as parking lots and structures…..

    1. Yeah, as long as we continue to allow parking garages and car rental facilities to be built downtown, while not adding to transit ROW, the spatial imbalance will get worse. Please tell me we don’t have mandatory parking minima downtown at least.

      1. Correct – I believe there is no parking minimum in any of the urban hubs or villages, which would include all of the downtown core. All the new parking being build in SLU is at the discretion of the developers.

      2. If we had adequate transit and transit ROW there, maybe they wouldn’t have felt the need to build parking.

      3. Easy one, Brent. Just have the Mayor order the Police Chief to go on the radio and tell the public that since the entire SPD is directing traffic in and out of parking garages in South Lake Union 24-7-365, either drive there and sit in your cars four stories under Amazon, or go get robbed and murdered any place of your choice.

        Should definitely keep a huge number of cars, and drivers who’s choice of travel modes suggest they will believe it, off the streets. Especially if you also put tow-truck hooks on the bumpers of every bus they see as they drive in.


  9. I counted 536 blocks in Center City (bounded in my definition by 1st Avenue, Mercer, I-5, and Jackson St). I then counted the total number of lanes on each of these 536 blocks to come up with the total number of ‘lane segments’ in Center City: 2,296.


    Something must be off here. So youre saying on average the streets are more than 4 lanes wide. Even with all the narrow streets counted.

      1. Right – outside of some tiny streets around the waterfront, I think every street is at least 4 lanes wide, 2 parking and 2 GP. The 6~8 lane avenues outweigh those small side streets.

  10. Commute Seattle has a definition that extends to Broadway (middle of Capitol Hill/First Hill), north of Mercer Street (edge of LQA) and Dearborn Street (edge of SODO). Is that the boundary of Zach’s other calculations? The One Center City boundaries are much smaller, by the way.

    Some other important technical points on these data:

    Not all trips Downtown are work trips. I realize that this is the only survey data that is available for Downtown destinations, but there are Downtown deliveries, meeting/dentist/doctor appointments, errand runners, paratransit vans, taxis and people picking up spouses, for example. The delivery and drop-off issue is particularly important because the term “parking lane” is not for just parking, and is often restricted to deliveries or loading/unloading people in many places.

    Downtown streets are also used to get from one part of Seattle to another (perhaps some data about how many vehicles on Downtown streets that are going through Downtown versus going to Downtown would help). I would guess that between 10 to 30 percent of Downtown traffic (20 to 40 percent of the actual Downtown core traffic) is neither starting nor stopping in Downtown depending on the hour.

    Not all streets are the same. The traffic lane capacity of a side street is much lower than that of a freeway or even a street with synchronized signals. Streets without synchronized signals are less advantageous for either transit or other vehicles and maybe even bicyclists. Finally, many parking lanes are narrower than a vehicle lane, so how those are counted is debatable — especially if there are intersection pedestrian bulbs. I would suggest that unless a parking and delivery and loading/unloading “lane” can be converted to a vehicle lane without removing a pedestrian bulb, it should be removed from the calculation — and other parking-only and delivery-only and loading/unloading “lanes” should be counted as a half-lane in your calculation.

    Finally, buses are using a number of other Downtown lanes, even though they aren’t called bus lanes in this chart. While most of Downtown buses and streetcars are already in transit-only lanes in the Downtown core, other streets in the broad definition of Downtown have buses in mixed-flow traffic.

    In sum, the data is illustrative but something that it doesn’t tell the primary story. Travel time and convenience measurements matter much more to me because those are things that more directly affect people, and space allocation does not.

    1. Good points. But keep in mind, the whole argument here is that more surface space should be allocated for transit, given the relative demand. I think even without the relative demand there is a strong argument for more space given over to transit — this simply strengthens the argument. As to your points:

      1) Not all trips are work related trips. Of course not. But does that matter? If you start “taking lanes” and “traffic is suddenly much worse”, who cares if you aren’t trying to commute. If you are attending an evening concert at Benaroya Hall, and want to park, go ahead. But traffic on the way home won’t be an issue, so what difference does it make if you have three lanes or two? Same with the midday dentist appointment or weekend visit to downtown.

      2) Yes, downtown streets are used to get from one part of town to another. The same is true for transit. Given the huge number of transfers that are required downtown, it is probably a bigger number. The commuting number serves as a general guide, and a very good one in my estimation.

      3) No, not all streets are the same. So what? A quick look at the transit map ( show that buses go just about everywhere downtown. No matter how you got the numbers close to the number of users, you are bound to make a substantial improvement in transit speed and reliability. Speed up Battery, for example, and now the 26, 28 and E Line are faster. Wall? The 5. Blanchard? The 40 and C. Fairview, Westlake, on and on it goes, to say nothing of the avenues, which is mainly what people are asking for. In other words, subtract the parking, and simply grant buses an extra lane where they go and you probably get very close to the ratio that matches ridership.

      4) No one is disputing that buses are using general purpose lanes, but that misses the point. They shouldn’t have to. If you allocated according to use — roughly half the lanes for transit — the buses wouldn’t bother with the general purpose lanes.

      5) Travel time and convenience are reflected in the amount of space allocated for transit only travel. There are plenty of buses downtown — more than enough, actually. That covers the convenience issue. They are stuck going too slow because they are stuck behind general purpose traffic, and that is because there aren’t enough bus lanes for them.

  11. Thanks for taking the time to do this analysis! It looks pretty labor intensive, but it sure is important.

    I think taking the length of the DSTT and breaking it up by an average block length would make sense, given that it is included in the mode share. A related question is how much of the travel in unrestricted lanes is goods versus people. It’s surely in the minority, but it does add value not seen in this analysis.

      1. Dang. Here I was hoping to come across a reason to create dedicated kayak commuter docking.

  12. I assume others have ridden buses recently on Seattle city streets. It’s a helluva noisy, bouncy ride on all the broken pavement on most streets. Any word on when the City’s street fixers will be going out to patch some of the worst problem areas? It’s beyond trying to report potholes; the problem is way to large to report them one by one.

  13. Are you counting a bike lane as the same width as the other types of lanes? Did your method count 2nd Ave PBL as two lanes for bike? For a rough comparison it would seem fair to count a bike lane as 50% of the other types of lanes, it would be a great day when bikes get 7% of downtown ROW

  14. I think this is a great post as is. But I would modify the charts just a bit, to address some of the criticism:

    1) Get rid of the bike lane reference. The point of these charts is to show how few streets are available tor buses versus cars, not to talk about other modes. You don’t mention sidewalks, so I see no need to mention bike lanes (which can be thought of as sidewalks in this discussion).

    2) Remove bikes, walking and “other” from the commuter data (again, because they don’t use the main streets). Focus on what is important — motorized transport. You can see the ratios just fine right now, but if you focus on just transit and cars then the numbers are even more striking (54% by transit). You would need another asterisk (“does not include biking, walking, or telecommuting and ‘other’ as noted by survey participants”).

    3) Include the transit tunnel and the freeway lanes (including SR 99). There is no reason to omit something that is important for transportation (and the transit tunnels and freeways are important) especially since if you include them, it doesn’t alter the numbers much at all. Omitting them implies that it would.

    4) Break out the parking and loading zones. If this turns out to be relatively small, then it bolsters the case that what needs to be done is to take regular lanes. If it turns out to be relatively large, then we should focus our efforts there.

    I don’t feel that strongly about item number 2 (it might not be worth it) but I think the other items would greatly improve the way this information is presented. I think there is a very strong argument here that we need to greatly increase the amount of space allocated for transit, given these numbers. Addressing counter arguments would likely strengthen the argument, in my opinion. My guess is that when you include things like space allocated for parking, loading and unloading, and the transit tunnel, there is still a huge disconnect between the amount of space allocated for moving general purpose vehicles and the amount of space dedicated for transit.

    1. thanks RossB, as puny as 7% is, I think it is a huge exaggeration of how much bike infrastructure there is downtown and is not helpful for those of us who would like to see an increase in that infrastructure and in the share of bike commuting. Many of the bike lanes that do exist are built in extra space that was already on the road, i.e. dividing a 14′ lane into a 9′ lane and 5′ bike lane.

  15. But this is assuming these modes operate at the same overall efficiency. It’s not like we’re cramming more buses than we have cars into 10% of the space that we’ve allocated for single-occupancy vehicles – those buses are operating at many times the “density” or efficiency of a single person in their car. Assuming a bus is about 10 times denser than an SOV (maybe a bus is twice as big, but carrying 20 people instead of 1), those 150 lane segments are equivalent to 1,500 car lane segments. All of a sudden we’ve got pretty comparable amounts of ROW dedicated to these modes. Not perfect, but also not the order of magnitude discrepancy that’s noted in the graphic.

    Don’t get me wrong – I’m all for shifting the balance towards more transit, bikes, peds, carpooling, etc. I just think we should be comparing apples to apples (rather than grapes to grapefruits) as we determine our progress and the amount of additional work cut out for us.

  16. Hi Zach! Cool post & graphic! I have the same questions as many people who’ve already commented. How did you count bike lane segments? Did you adjust them in anyway because they’re half the width of car lanes? Did you count the block segments of the 2nd Ave 2-way PBL as two lane segments or one?

Comments are closed.