King County Metro’s Most Productive Routes

53 Replies to “King County Metro’s Most Productive Routes”

1. 1) It’s a shame the A-Line doesn’t have peak hour service. The C, D & E all have extra service during peak hour where buses run every 5-7 minutes at times.

   2) Is “PMPPM” a good metric to measure the productivity of a route? The 311 (now the 256) outranks 4 RapidRide route but only because it travels a long way to/from Woodinville. I’m sure its ridership comes nowhere near that of the G line.

   3) Would a better metric a ratio between ridership and cost per trip?

   1. If the G Line didn’t exist, a lot of the trips would likely be made by walking, Lime scooters, or nearby bus lines. If the 311/256 didn’t exist, I’d wager most people would just drive instead. So in terms of reducing cars on the road, PMPPM is absolutely a good metric.

      In fact, the G Line is just over 2 miles and has 6000 daily riders vs. Link 1 Line being 33 miles and about 100,000 daily riders. So it has approximately the same ridership per mile as Link. There is no way you can argue that the G Line is a more important route than Link. The trips are so short.

      1. You are ignoring several important factors. First, not every trip involving the G is short. A lot of the trips involving the G are probably transfers. The bus solves an important “last mile” problem. If not for that, some people would most definitely drive.

         Second, the alternative to the RapidRide G for a lot of people is Uber or Lyft. This is definitely driving.

         Speaking of which, good affordable transit encourages urban living. Urban living reduces the overall amount of driving that people do (and other CO2 contributions).

         Measuring rides per platform hour treats all riders the same. It is the better overall measure of the productivity of the route (and the system as a whole). Passenger miles per platform mile favors riders who take long trips over everyone else. It is similar to coverage routes (which unfortunately are not so easily measured).

      2. “If the G Line didn’t exist, a lot of the trips would likely be made by walking,”

         Have you ever tried to walk Madison Street? It’s steep. If there were no service on Madison Street, people would take the routes on Seneca Street, Pine Street, and James/Jefferson Street and walk from there. The 12 was the pre-G service on Madison Street. Its ridership was stunted by its 30-minute frequency weekends and evenings (dropping to 60 minutes during recessions).

   2. 1) It’s a shame the A-Line doesn’t have peak hour service. The C, D & E all have extra service during peak hour where buses run every 5-7 minutes at times.

      I’m guessing that the C, D and E are all crowded during peak. As a result, Metro runs the buses more often than they would like. I think the biggest issue is that the A Line doesn’t have 7.5 service as the baseline (midday) frequency. This may be because Seattle is paying a bit extra for service on the E Line. While it may not be this explicit, that may be the thinking. For Seattle (and Shoreline) running the bus every 7.5 minutes instead of 10 minutes is worth it. But for Federal Way/Tukwila, there may be better values.

      3) Would a better metric a ratio between ridership and cost per trip?

      That is pretty much what ridership per service hour is.

      1. What values would be better than increasing the A-Line frequency? Whether it be during peak or off-peak, the need is still there (from my experience of crowded and late buses)

      2. “I’m guessing that the C, D and E are all crowded during peak. As a result, Metro runs the buses more often than they would like.”

         High ridership is a problem? Metro “would like” to run buses less often? None of them are running every 2-3 minutes during peak, so they aren’t extreme like Vancouver’s Broadway situation. If Metro doesn’t want to run bus routes every 5-7 minutes, it needs to get its head adjusted. That should be its goal on all core routes. Short waits make passengers happy and more willing to ride transit and fund it, and make transit more competitive with driving.

      3. Metro “would like” to run buses less often?

         Those particular buses, yes. So they could run other buses more often. The A Line runs every ten minutes. The 7 runs every ten minutes. The 8 runs every fifteen minutes. But the E runs every five minutes (during peak)? As you can see, the E gets fewer riders per hour during peak than midday (and less than the midday ridership of the 7, 8 or A Line). Running the E Line more often during peak only makes sense if you are responding to crowding.

         Keep in mind, Metro has dramatically cut back service in the last few years. This is why the 7 only runs every ten minutes (instead of every 7.5). If they could run the E every 7.5 minutes all-day long they would (and put the money into running other buses more often) but there is too much crowding.

         What values would be better than increasing the A-Line frequency?

         Probably running one of the half-hour buses every fifteen minutes. It costs the same (you are adding two additional runs every hour) but it makes a lot more difference for those riders. Ridership increases with frequency but the relationship is not linear. You get diminishing returns. Going from 30 to 15 will get you a lot more new riders than going from 10 to 7.5. That doesn’t mean every half hour bus would gain more riders but there are some “edge case” buses. For example the 165 gets over thirty riders an hour — that is quite good, especially for a half hour bus. It could also be spread around. The 187 performs surprisingly well for being so infrequent. It wouldn’t cost much to increase ridership (and improve the rider experience) on that route (given it runs every hour sometimes).

         (Crowding is a different issue. Maybe they should run the A Line more often to deal with crowding.)

   3. Maybe A Line doesn’t need more frequent service during peak hours. Maybe it is doing so well on data because it gets as many riders during off-peak as it does during peak hours?

      A Line doesn’t serve same type of area as C/D/E/H do, so I would imagine it has more diverse trip purpose than C/D/E/H do.

      1. Yes I agree with your observation.

         Bus ridership in Downtown seems more oriented to peak commute jobs. The peak oriented work hours are especially true for government employees — but also for jobs in financial services, and doctor and dentist offices. So buses will be more crowded at the peak commuting times going into and out of Downtown Seattle. .

         RR-A doesn’t really have that pronounced of a peak commute time in the corridor. Many jobs are in categories that don’t have a heavy peak commute start and end time.

2. I see you’re showing “Weighted Average Rides per Platform Hour”; what weighting are you using?

   1. The report from Metro breaks down the RPPH into weekday peak, weekday off-peak, weekday night, Saturday, and Sunday. To get the overall RPPH for a route, I weighted the RPPH for a time period by how many hours that time period is used in a week. For example: weekday peak hours are 5-9am and 3-7pm, M-F. Therefore the weekday peak RPPH has a weight of 40hrs/168hrs (0.24). The same process was done for each time period.

3. I was undervthevimpression is getting killed as part of east link or maybe stride? Maybe I’m wrong..

   1. What?

4. I’d like to see the rest of the 11th-30th RRPH.

   1. Full list as a .csv: https://seattletransitblog.com/wp-content/uploads/2025/11/2024kcmProductivity.csv

      Note that only routes that run during all time periods are considered. Peak only routes or routes without Sunday service are not on that list. Also, DART routes do not have a Peak Passenger Miles per Platform Mile metric because their platform miles vary each run.

      1. It is worth noting that peak-only routes tend to perform poorly in terms of riders per hour (even during peak). From what I can tell, the highest performing peak-only route is the 17. It gets 20.4 riders per hour. There are about fifty routes that do better than it during peak.

         In contrast a few of the peak-only buses do really well in terms of Passenger Miles Per Platform Mile. The 102 is 7th, the 218 is 10th, the 311 is 12th, etc.

      2. “There are about fifty routes that do better than it during peak.”

         I guess part of it is that 17 only runs every 40 minutes and those other 50~ routes probably run more frequent than that?
         I’ve seen several time at 3rd Ave stops that when 17 arrives, it empties the entire platform. There are good amount of people using it although I am not sure if there will be more people using it if it runs more frequent or add service.
         For those whom are going to Ballard along NW Market St, 17X is such a decent option to avoid 40.

      3. I think the routes that only run during peak tend to just be underperforming routes. That is why it only runs during peak. If it performed better during peak then they would probably run it midday as well.

      4. “ I think the routes that only run during peak tend to just be underperforming routes. That is why it only runs during peak.”

         Their productivity is also often hampered by having a higher proportion of the platform hours being when the driver isn’t picking up passengers. Single-direction express bus runs are particularly known for this. If a bus has to return from a route that ends a long way away from a bus yard/ base and it’s only making a few round trips (maybe just one or two because it is peak-only), the hours that it operates with no riders can be quite high.

5. I can see a new Fix the L8 campaign sign: “The L8 is #3”.

   1. With actual improvements to speed and reliability I bet the 8 would be #1 in RPPH

      Although the fix the L8 campaign may now be unnecessary now that one of its biggest supporters will be mayor.

      1. I partly suspect that fixing the L8 was Wilson’s original motivation to run for mayor, like those stories of software engineers who take the job just so that they can fix a bug that’s been annoying them for years “If I was mayor this bus wouldn’t be so damn slow” 🤣

6. I’m surprised the C is below the B, F, and H. I thought the lowest-ridership line was the F, and second-lowest the B. The high H ranking could show its popularity in Delridge and White Center, and the lack of an express from Burien to downtown Seattle (the H is the least bad way so people use it).

   1. All the routes are down from what they were from before the pandemic. The 2018 report (https://cdn.kingcounty.gov/-/media/king-county/depts/metro/documents/about/data-and-reports/2018/system-evaluation.pdf) was the last one before the pandemic that included all the RapidRide routes (for some reason a few were left out of the 2019 and 2020 report). Here are some comparisons of ridership per hour. The order is peak old/new, off-peak old/new:

      A: 54.8/42.6, 58.4/42.7
      B: 42.5/28.1, 34.1/30.3
      C: 47.9/28.8, 35.6/25.5
      D: 63.7/40.2, 53.3/41.7
      E: 58.1/38.3, 56.9/39.0
      F: 32.2/22.7, 33.5/29.4
      H: –/29.9, –/29.6

      The C was never great off-peak — it barely edged above the F and B before the pandemic. It lost more (off-peak) than the other two. It also lost a lot during peak (but so did all of these routes).

      1. The H didn’t exist during the pandemic, unless that data is for the 120.

      2. That is why there are dashes there. The dash is for old and the numbers are for new. A comparison to the 120 is worthy — good point. Here are the numbers from that perspective:

         120/H: 39.6/29.9, 39.3/29.6

         Like above it is: peak old/new, off-peak old/new.

         Thus the bus performed better when it was a 120 (not RapidRide). That just shows how much ridership still lags. I have no doubt the RapidRide H Line is better than the 120 but it is swimming against the tide.

   2. The RossB evaluation is correct. It has to do with the portion of the route’s ridership related to office work. It was higher on the C line than on lines A, D, E, and H. Office work declined significantly. That took riders from Sounder and many one-way peak-only routes oriented to downtown Seattle. Many are suspended.

7. It would be nice to see a map of ridership rather than static numbers. For instance, according to the report, route 3 gets an average of 23 riders per hour – which meets productivity standards if measured by riders per hour. But it fails in productivity if measured by passengers per hour. Knowing by firsthand experience, the bulk of the riders are riding between First Hill and downtown and a peak hour spike between Belmont and downtown. It would be nice to see where riders are boarding/exiting for all routes.

   1. We’ve gathered up the ridership data for the routes. Micheal will periodically do a report on a particular route. Here is a list: https://seattletransitblog.com/2025/11/03/route-ridership-index/. He never broke down the 3 or 4. I can get the data for you for those (or the entire set that we got from the county). Let me know and I’ll email it to you (or put it out on the web somewhere). (As an editor I know what your email is, so there is no need to include it if you want me to email you.)

      1. Yes, please.

8. I moved from Phoenix Az back home to Seattle and I will tell you that the transit here is way better here in Seattle than in Phoenix. Yes Phoenix has bus service and light rail but no commuter rail and not that many express bus service too. That’s why their traffic is worse than here in Seattle. Other than that keep up the good work in Seattle 😃

9. Will there be a companion post about least productive routes?

   1. Good idea. The first one I wanna axe is the 79 and the revived 3-Belleve Ave. They were eliminated for good reason in the first place.

      1. There never was service on 75th before the 79. People had been asking for east-west transit there for decades. In college in the 80s I visited somebody who lived there and was surprised there was no service on 75th. The 79 was to address this request.

         You may not approve of the 3-Bellevue, but deleting it would be controversial to say the least. In the G restructure Metro asked the public whether to restore Summit service or put the hours into surrounding routes, and the majority feedback was to restore Summit. Metro isn’t funding it; Seattle’s TBD is. You’re not going to get political consensus to delete it now, so it’s not worth the effort; we have much more important things to press Metro and SDOT on.

      2. The new 3 serving Summit was probably the only good thing to come out of the last restructure in the area. It was quite clever — I didn’t think of it. It should run every fifteen minutes.

         The problem is what surrounds it. The 49 should keep going straight on Broadway. The 10 and 12 should branch at Thomas — if they branch at all. Then you have a nice set of complementary routes. The 3 goes downtown via Summit/Bellevue. The 10/12 goes downtown via 15th. If you are in between there (on Broadway) you catch Link.

         This is just basic consolidation. Right now, unfortunately, it is the opposite. The 3 runs downtown but is infrequent. You can walk over and catch the 49 but it is infrequent as well. Same goes for the 11. The 10 is infrequent anywhere north of Pine. While each bus runs infrequently that is a lot of combined service running very close to each other but not combining for good headways until Pine. The only good option in the area is Link. Speaking of which the 11 and 49 manage to run right by the Link station but still go downtown. Thus you have a lack of consolidation *and* a lot of needless waste.

      3. There never was service on 75th before the 79.

         Sort of. The 71 used to serve part of 75th (in the View Ridge area). It just didn’t go all the way across. The 74 used to serve 55th. Here is an old map showing that: https://seattletransitmap.com/version/1603/SeattleTransitMap_v1603_web.pdf

         Thus they combined them. The bus is about average during peak and below average outside of peak — at least for an “urban” bus. It would be about average for a “suburban” bus. I think in this context “urban” means it serves downtown or the UW while “suburban” does not. It provides coverage and does so reasonably well in my opinion. It really doesn’t strike me as flawed.

         There are two areas that look really flawed to me. The first is the greater Central Area (basically everything east of I-5 between the ship canal and I-90). I’ve written about that before. The other is Shoreline. It is a tough place to cover but the last restructure was a big failure. I’ll probably write something about it the coming weeks and months (but it isn’t hard to see how things have gotten worse despite general improvements overall). The RapidRide E is better than the previous year (mainly because they left it alone).

      4. Mike Orr, former Route 68 did serve NE 75th Street between Roosevelt Way NE and 25th Avenue NE; it connected NTC and the UW. In the old network, Mapleleaf had odd service; the arterial with more business and multifamily housing had an infrequent Route 68 and more service was provided on 5th (routes 66-67 and 242) and 15th avenues NE (routes 73, 77, and 373).

         Route 79 seems to have two segments. The south part is the former Route 74 (or 8 or 30); the north part is NE 75th Street. Note that it duplicates Route 62 on the east slope of View Ridge. Route 79 was one of those routes from which many trips were suspended in fall 2023 to address the operator shortage.

   2. Discussions around the least productive routes require a broader look at the goal of public transit. Some of the least productive routes are the only option for those that use them. Route 208 travels between Issaquah and North Bend. It ranks near the bottom for route productivity but it is the only route that connects people in North Bend with the rest of the network. Public transit is an essential service and some routes don’t need to be productive to be important.

      There are certainly some routes that are less productive because they are redundant or have a poor routing. We do have plans to discuss these routes, and potential improvements, in future posts. These posts will focus more on a sub-region of routes, rather than just sharing a list of the least productive routes.

      1. Yeah, we need to consider coverage as well (https://humantransit.org/basics-the-ridership-coverage-tradeoff).

   3. The least productive routes/least essential coverage were deleted between 2009 and 2014 over a decade ago.

10. Just a note about the term “platform hour” because it’s not defined in the original post:

    It’s an industry jargon term for the total hours a bus is being driven, including layover/ driver break time as well as the time it takes to get between the yard and the beginning or end of the actual route. A “platform” is when someone is paid to be behind the wheel.

    It’s slightly different than when a bus is in actual service and serving fare-paying passengers.

    I thought it was useful to mention this when considering route productivity stats. Routes further from their yards have their productivity stats dampened a bit because of this. Express routes are often not taking riders in one direction so that really dampens their productivity stats. Conversely, when a route starts near a bus base yard, it helps improve that route’s productivity stats slightly.

    It’s probably the best stat to assess productivity. It does have some minor biases in using it though. And for someone pondering productivity for the first time, I think it helps to understand what the basic term means.

    1. Thanks for sharing that context Al – I should have defined it in the article.

11. While it makes sense to look at a route in totality, it should be noted that the productivity may not apply to an individual segment. The most productive routes tend to have lots of riders along its entire length. Less productive routes may be that way because a segment of the route just doesn’t attract many riders. For example, Routes 2, 14 and 27 don’t have the demand east of MLK that they do west of MLK so they don’t show the route productivity as high as they could.

    That’s not a reason to advocate for service reductions or eliminations for low-use segments. It instead is to highlight why some routes appear to be more or less productive than others.

    1. Speed makes a difference as well. For example the 24 doesn’t pick up that many riders on the western half of Magnolia. But it covers those areas quite quickly. The 3/4 pick up a lot of riders but they are slowed down considerably on James.

12. This are two serious question; I’m not “trolling” or advocating ‘yea” or “nay” to the various routes’ existence.

    1) “What are we to make of these statistics?” That is, “What specifically do they say about the value of the service provided versus its cost to provide it?” Is there a discernible relationship, or is it a random matrix?

    And following that, “if there is a discernable relationship, should that determine in whole or in part the allocation of bus hours?”

    I think the second is fundamentally a “values” political question and as a result has no one “right answer”.

    The first, however, needs to be answered accurately in order for Metro to manage its assets. Is there some sort of relationship among the many variables — housing density, street flow, presence of “destinations” and probably many more that can be used to group routes? I get that that is what people are talking about, but mostly from narrative, not data.

    1. To be clear, I understand that the data presented is certainly valid and is for sure “data”. I have no antagonism toward it. But it is essentially measuring the output of the interactions among the variables alluded to above, and NOT necessarily indicators of excellence or failure. Each net vector has to be evaluated against those of a given line’s “peers”, which is much harder than doing a division between two linear variables.

    2. Metro’s service guidelines call for it to review the routes every year, and consider shifting hours from the 25% lowest-performing to the 25% highest-performing routes or underserved corridors. That doesn’t mean it has to delete bottom-end routes, just that it should consider whether their coverage value is still worth it.

       To me, value is independent of cost. If an arterial exists, and has so many people and jobs, then it should have an optimal amount of transit so people can get around conveniently and there isn’t overcrowding. That generally means a route at least every 15 minutes full time (ideally 10 or 5 minutes), and additional runs to relieve crowding). That applies to all but the most rural areas (e.g., maybe Snoqualmie and Black Diamond are rural enough that 30-minute service is OK). These needs don’t change when costs change, because people still need to get around, and we want to incentivize non-car modes for economic, equity, and environmental reasons. If the county is concerned about high costs, it should address land use (infill development, mixed use, straighten out corridors, pedestrian access) to make it less expensive to serve those areas.

       The riders per platform hour automatically shows which routes are more cost-efficient (because more riders per hour) and more valuable (because more absolute riders).

       The top-performing routes are self-funded through fares, so their operational subsidy is zero. I don’t know how far down the rankings that goes, but the 7 has long been considered a self-funded route, and I think the pre-Link 49, and probably the E and others in the top ten. So if you want to clap at low-cost routes, you can clap at those.

       You’d need a whole series of articles to fully define “value” before you can say whether the routes are well-matched to their costs and how hours should maybe be reallocated.

       In the absence of that, I go by how valuable a route is to the residents and visitors who do or want to live/work/go to all areas in the city or county. I know what’s personally valuable to me as a non-driver for the places I go to. I can extrapolate that to what other people living in different neighborhoods with different jobs and responsibilities and preferences would want, based on what their counterparts in cities with excellent transit do. That gives an ideal list of what we should have. But then it has to go through the political process, nimby challenges, and dividing limited current resources to implement any of it.

       If you want to collaborate on an article to define value, you can reach the editors at the contact address.

       1. I can extrapolate that to what other people living in different neighborhoods with different jobs and responsibilities and preferences would want, based on what their counterparts in cities with excellent transit do.

          No, you can’t. That is exactly what I mean by “narrative”. It’s an emotional preferences thing. I love good transit, too, but most of the people who live around me don’t give a rip about it, unless a bus is broken down and blocking a lane. And that’s true all around the country except in, perhaps, two dozen fortunate cities.

          Yes, that is lamentable in oh, so many ways, but it’s Quixotic to try opposing it as a public policy position. There will be no more buses that people will agree to subsidize, a few self-funding routes notwithstanding. So willy-nilly adding service that isn’t used is a dumb use of public resources. The buses mentioned above which have more frequent service now than they did in 2018 but are carrying fewer people are certainly not “proof” that “more frequent service” is everywhere and always a good thing. Seattle is a bus-riding city; Federal Way is not; Auburn is not; Issaquah is not; even BRT-loving Lynnwood is not. At least, not to the degree necessary to result in lasting changes.

       2. > That is exactly what I mean by “narrative”. It’s an emotional preferences thing.

          If you’re honestly asking the question ““What specifically do [these statistics] say about the value of the service provided versus its cost to provide it?” (emphasis mine), you need to define “value”.

          (Hint: you have the definition in your original comment: “I think the second is fundamentally a ‘values’ political question and as a result has no one ‘right answer'”. Economic value is just as arbitrary as moral value.)

          Your first request is, really, a demand for a multi-variate statistical analysis of every possible geospatial urban statistic against the efficiency statistic discussed in the blog post. This analysis would be meaningless without an equally-statistical (and inherently equally arbitrary) definition of value.

          I think what you’re actually looking for is a one-number-rules-all statistic which would give indelible justification to modify service. You could use all sorts of calculus to come up with such a number, but again, the weights and measures would be fundamentally arbitrary.

          You demean the concept of a “narrative” as an “emotional preferences thing”, but that’s exactly backwards. The service of the emotional preferences of the served (and the yet-unserved!) is the only true value any social service can offer – and public transit is most definitely a social service.

       3. “I can extrapolate that to what other people living in different neighborhoods with different jobs and responsibilities and preferences would want, based on what their counterparts in cities with excellent transit do.” — “No, you can’t. That is exactly what I mean by “narrative”. It’s an emotional preferences thing.”

          When there’s no available data and we can’t feasibly commission the missing studies, it’s extrapolations or nothing. I’d argue that a reasoned if imperfect extrapolation is better than the blind default assumption that latent ridersip demand is zero.

          And I was talking about the kinds of trips that other people make for their own reasons, both those who would ride transit if it were feasible for their trip, and for those at the margin who might be willing to take transit if given a small incentive (good service, fare concession, marketing promotion, etc).

          “I love good transit, too, but most of the people who live around me don’t give a rip about it”

          Now you’re making an emotional assumption. Americans want more transit and are more willing to use it than the current route networks and pro-car propaganda suggest. Americans aren’t that different from Europeans or Canadians, it’s just that the don’t get the opportunity to show it because they have much fewer choices in transit, housing, and walkability. Non-American cites go full out with frequent buses, transit-priority lanes, medium density, walkable access, and bike corridors, and ridership follows suit. To a greater or lesser extent, but it shows a serious commitment to it. American cities build something half-assed and pretend it’s a complete solution, or wring their hands saying “it won’t work here” so it doesn’t get built.

          All the things we complain about: Link terminating at Tacoma Dome, buses terminating at Tacoma Dome, 30-minute evening/weekend service, buses caught in traffic, getting off at Ash Way P&R and walking 40 minutes to your house because the 119 is hourly and only goes halfway and you get hit by a car because there are no streetlights or sidewaks and cars speed, no down escalators, convoluted journeys — why aren’t these all being fixed? There should be a plan to fix several of these each year so that many of the problems will be gone in ten or twenty years, but there isn’t. Instead we get these half-assed solutions which are better than nothing but are still half-assed and don’t really solve the problem.

          In city after city in the US over the past decade, as the bus networks expand or contract or get better or worse, ridership goes up and down along with it. If we really had 15-minute all-day suburban transit in a lot more of Pugetopolis, ridership would increase more than many people assume, because people will ride good transit if they have the choice, even if they still want a house and yard and the option of driving. And if they had good transit, their attitudes toward transit and transit taxes would improve.

          So that’s what would happen if we really improve bus transit throughout Seattle’s neighborhoods, and have sensible strategic rail lines and gondolas where they make sense. It wouldn’t be everybody switching to transit, but a lot of people doing so. We can get car mode share down to 50% if we really try.

    3. But it is essentially measuring the output of the interactions among the variables alluded to above, and NOT necessarily indicators of excellence or failure.

       So what? As Micheal explained, no one is trying to make that value judgement.

       But data like this should still influence decision making. Consider the Metro 8 for example. It is quite reasonable for someone to say “Why spend all this time and effort for a bus that only carries about 7,000 riders a day. Even if you made it faster it wouldn’t be a great bus.”

       Yet is you look at the numbers you can see that is clearly not the case. When it comes to ridership per service hour it is one of our best buses (i. e. it performs really well compared to peers) despite being really slow. You don’t need to do a fancy analysis of the land development and travel patterns. This metric alone is enough to make a very good case for making the bus faster.

       Then there are restructures. This gets more complicated. But with some effort you can make some conclusions. For example the 20 was replaced by the 61. Was this a good value? That depends on what your priorities are. From a coverage standpoint it most certainly was not. But from a ridership standpoint it definitely was. More people benefit but some people have to walk longer to get to their bus. It is difficult to determine whether people are saving much time on their trips but presumably that is reflected in ridership.

       If nothing else, this makes for interesting information. Many people (including me) are surprised that the A continues to get more riders per service hour than the E. Sure, it is only one metric but it is damn important one.

    4. Public transit is for the public good, but it’s not financially realistic for an agency to provide service from everywhere to everywhere frequently and directly at all hours of the day and night. There has to be a way to allocate the limited financial resources to be both productive and equitable.

       It is perhaps the most performance-driven public service that local government has. Part of that is a legacy of its origin with private, for-profit businesses. But even the current environment where needs and fairness in public investments is fundamental to good government, productivity stats are needed to decide how to spend limited transit dollars most effectively.

       Another aspect is that transit tends to serve people that are less politically powerful. If we built and operated transit by political lobbying without looking at performance measures, we would have tracks and stations and vehicles serving places where the residents have more time, money and political savviness to get what they want while the system as a whole gets less use. It’s a societal equalizer to political influence.

       Frankly, it’s the lack of looking at productivity hard enough that ends up with political boondoggles. ST3 is a poster child of not only bad cost estimating, but of drawing up plans on a map without adequately analyzing what the ridership and travel time benefits (and maybe overcrowding) would be. So not only does it become unaffordable to build, it becomes much more expensive to operate per rider if there’s much less ridership.

       I salute the concept of performance measures for transit as a government service. I wish that forecasting them was taken more into account in other local government investments and services. While any measure has unique biases, it provides a good quantitative signpost by which to gauge how useful our tax dollars are.

Comments are closed.