Average Weekday Ridership

Link ridership set another record in June, with an average weekday ridership of 23,396 (Saturday 17,510; Sunday 13,919), and a daily range of 9,827 to 28,820.  This represents a 7.5% increase over May’s average of 21,774, and a 16% increase over April’s 20,129.  June represented the sixth straight month of >5% month-over-month growth.  Even with the usual caveats regarding sampling and modeling error, this is very encouraging news.

Ridership tends to follow a logistic growth model, in which ridership grows exponentially until it slows as it approaches a theoretical saturation point.  We’ll probably see ridership stabilize at its “natural level” well before University Link restarts the growth cycle, but it will be exciting to see just what that level is.  In the meantime, even with seasonal variations, we are likely to see continued year-over-year growth.  (Just for fun, a continued 7% rate of growth would yield ridership of 35,000 by the end of the year.)

139 Replies to “Link Ridership Up 7.5% in June”

  1. Just out of curiosity, what level of ridership would be enough to trigger an increase in service frequency or train length? Would it be reasonable to assume that peak frequency is constrained by the shared operations in the downtown tunnel, and that lengthening trains is the only possibility for increased capacity until University Link arrives and the buses are kicked out?

    1. Longer trains aren’t an option right now either. There isn’t enough room in the Pine Street Stub Tunnel for them to turn around.

      1. Ah. I forgot about that. So basically, there’s no possibility of any increase until 2011, and even then, any increase would reduce bus capacity in the tunnel considerably.

      2. Increase in train size? I don’t think so. Someone correct me if I am wrong but the block the train occupies is the same size whether it is a 2-car train or a 4-car one. So if they were running 4-car trains right now, the bus capacity would still be the same.

      3. Yeah but while the train would be twice as long and thus double capacity, it would also be too long for the current Pine Street Stub tunnel, where the trains turn around. They are using the length of the tunnel that would be required to turn around a three- or four-car train for construction of University Link.

      4. Correct. Except in the PST, where a 3 or 4 car train won’t fit today, the trains already occupy the same space as a 4 car train – at least as far as the signals go. In other words, when I’m driving behind a train, I never get close enough to occupy space where a 3rd or 4th car would go, or even the buffer needed for safety.

        There may be logistical difficulties with having all of those passengers on the platform at the same time. I’d deal with that by going to a 3 car train when needed. By that time, many bus routes (41, 71, 72, 73, 76?, 77?, 316?, 301?) will be truncated with transfers available at Link stations further north. The bus platform would be the current “A” and “C” bays while the Link platform would be towards the back bay far enough to allow for the 3rd car. Off the top of my head, I think the current configuration is like this, but I’m not 100% sure.

        I’m hoping to see train frequency not increase, except in the off-peak hours where necessary, until bus route truncations with reliable transfers are available. It makes the whole system more cost effective and reliable. At 7.5 minute headways, that extra 1.5 minute wait really isn’t that big of a deal. Waiting for a bus that is stuck in traffic because it’s been pushed to the surface is.

      5. Yeah I don’t think there will be any need for frequency increases until U Link, ridership on the current line will plateau real soon, with only modest increases to account for densification and increased gas prices after that.

      6. ST may shift to four-car trains in 2016 when the UW Stadium station opens; until then, the Pine Street stub tunnel limits the train size.

      7. Hi Matt

        I am A Operator at king county that operates some of the Link Trains They could Add up to a 3 car train that will fit in the stub, we will soon be getting some new trains in for testing sometime soon.

    2. I don’t think it is feasible to increase frequency too because of the contract on tunnel usage between Metro and ST. I think we have to wait until the contact expires (I think it was the end of 2011). If we are to choose between increasing frequency or running longer trains, I am definitely for frequency. Longer trains look cool because they look more metropolitan, but for users, higher frequency improves convenience as it reduces wait time.

      1. But the thing is, any frequency increase requires paying more operators lots of money while adding an extra car is a marginal operating cost increase. But the point is moot anyways because trains can’t be extended until U Link.

      2. Define “lots”? Operators make a decent wage, but it’s not like all the stories you hear about $100k+ bus drivers represent that majority. To earn “lots” of money, you have to work a LOT of overtime. One of the Link operators can comment here but last I heard, overtime is not as plentiful for Link operators.

      3. By more frequency are you talking about 6 minutes vs. 7.5? Or are you looking at off-peak frequency? Obviously, off-peak frequency can be increased, where necessary, without pushing buses out of the tunnel. Increasing train frequency because it’s “cool” or because it makes our city look “more metropolitan” is not a way to run a transit agency. Balancing costs vs. convenience is. [cue Norman and his anti-Link / pro BRT diatribe here]

        An extra 1.5 minutes of wait time is not a big deal. I use Link to transfer between IDS and Stadium station when the weather is crummy. During rush hour I’ll wait for a Link train without even thinking about it. Sure many buses go to the same spot, but hey… The trains are “cool”. (Well, that and they are more roomy and easier to get on and off of during rush hour. If a 41 comes, however, I’ll grab that – it stops right in front of the base – My own personal 1 seat ride)

  2. are these number based on the passenger counters in the LRVs? by tickets purchased? by ORCA card scans? or a mixture of all these possible metrics?

    1. There are passenger counting devices in some trains. Ridership is extrapolated from these measurements.

      1. yes I know that … but that isn’t enough data to determine the actual ridership

      2. That’s how the numbers are calculated. It isn’t an exact figure, it’s an extrapolation from the measurements.

  3. They come from the automatic passenger counters on LRVs 101 through 110.

  4. Link trains are not even coming close to “capacity”. Why are people talking about increasing frequency? Have you ever seen a Link car with more than 100 people in it, at most (excluding to or from events at the stadiums)?

    I find this hilarious, as you might expect. You think Link cars are already “full” when they have 90 to 100 people in them? Many people on this blog still try to contend the the actual “capacity” of each Link car is 200 passengers! I can tell you beyond a shadow of a doubt, that Link cars are not carrying anything close to 200 people per car at any time of the day. If you think that the real capacity of one Link car is 200 people, then Link trains are never at more than about 50% capacity at this time (again, excluding to and from stadium events).

    So, which is it? Do Link cars really have a “capacity” of 200 passengers per car, or not? And if that is the true capacity, does anyone here think that Link has reached its full capacity already with 23,400 boarding per weekday? lol

    Also, even if ST did want to increase capacity, I don’t think they have enough cars to operate more trains right now, anyway. Or, have they taken delivery of more cars since Link started operating?

    1. No one is talking about increasing frequency — someone asked a question “out of curiosity” regarding future service changes. No one has used the word “full” in this thread, besides you. You’re the only one who seems obsessed about exact capacity figures. Meanwhile, you don’t compare the capacity figures to anything else and really offer no insight at all. Your shtick of inventing straw men to tear down is getting old.

    2. “Have you ever seen a Link car with more than 100 people in it”

      Yes, every day at rush hour. Last thursday I estimate that close to 300 people boarded my train at Westlake alone. My car had over 150 people on it, I counted. It was an amazing site to see the platform at Westlake clear out as hundreds of people boarded the train smoothly and efficiently in under 30 seconds.

      1. This is complete B.S. Please take a picture next time you are in a Link car with 150 people in it during a normal commute, so we can see it. I don’t believe you for a second.

        Of course, at Westlake a lot of people can board Link quickly, because trains are entering Westlake station completely empty, so there is no waiting for passengers to deboard before others can board. But, I don’t believe anywhere near 150 boarded one Link car at Westlake, or that they could do so in 30 seconds.

        By the way, on that Link car which you claim 150 people boarded at Westlake alone, how many other passengers boarded at University, Pioneer Square and International Stations? lol What a crock.

      2. This ammounts to an ad-hominem attack on Zach – would the STB administrators please relieve us of the burden of having to read this drivel?

      3. Oh give me a break — how is it an attack? Anyone who has seen the chaos on tunnel platforms when there are only 100 or so people waiting knows that there is no way 300 people boarded at Westlake. It’s a legitimate question that you don’t want the true answer to, so you hide behind claiming this is an “attack”. A lot of whiners on here looking for mommy to save them….

      4. How can I possibly go on living knowing that Norman doesn’t believe me! I’m crushed! I don’t know how I’ll find the will to get out of bed in the morning.

      5. That’s pretty rich, Norman. You’re king of the anecdote. For any transit system in this city, you’re sure to let us know if you’ve ever seen it run empty, and generalize it into low ridership. Then you actually ask if anyone has ever seen a full train — leaving out sporting events, because those people mysteriously don’t count — and when someone answers you automatically disbelieve them.

      6. The extra capacity is so that Norman can fit his piano and sofa in the train whenever he moves.

      7. I can attest to two rush hour (5pm ish) trips I took last week that were full standing room only, one of which required some polite shoving to get more people on at Pioneer Square station. It wasn’t quite Japan style crazyness, but I think it met the Seattle definition of crush load.

      8. “Standing room only” can mean 75 people on one Link car, since there are only 74 seats on each Link car. This does not tell us how many people were on that car.

        Why don’t some of you people who claim to have seen 150 or more people on one Link car at the same time, or a “crush load” on a Link car, take some pictures to show us. And I don’t mean to or from events at stadiums, when I do believe that Link cars are sometimes fairly crammed with sports fans.

      9. Probably because no one besides you obsesses over exactly how many people are on a particular car—the ridership numbers are the ridership numbers. Also because no one cares whether or not you believe them.

      10. However, Norman, because you asked nicely, I’ll make sure to take a picture next time. Why not.

      11. afterwards, you know he’s just gonna say “this only happened once! and i see someone with a sounders scarf so it’s probably on a game day!”

        there is no pleasing him

      12. Norman, you really have a way with words. This forum really wouldn’t be the same without you.

        Anyway, like I said last month, 25,000 by the end of summer!

        @Alex, I would have asked nicely. Would be cool to see just for the sheer Norman shock factor. Heck, you might as well videotape it from the platform and timestamp it.

      13. I did ask nicely.

        Maybe Oran has a picture of 200 people on one Link train, or a video of 300 people boarding one Link train at Westlake station. He’s taken a lot of pictures and videos of Link.

    3. “Or, have they taken delivery of more cars since Link started operating?”

      Yes, 27 Kinkisharyo LRV’s have been ordered for University Link. The first new car should be delivered this August.

    4. You need to design a system that has enough capacity to be comfortable for daily commuters but that also has the capacity to accommodate large sudden surges in ridership from ballgames and such. If our system were normally anywhere near the full capacity, then there would be no room left when extra events were going on, and commuters would become upset and stop riding because of the overcrowded conditions. So Link cars each have the ability to hold 200 people, but it is certainly not optimal to have 200 people in every Link car.

      1. “If our system were normally anywhere near the full capacity, then there would be no room left when extra events were going on”

        …Try riding Boston’s Green Line after games at Fenway! Fire marshals be damned, we CAN fit 400 people on a trainset…

    5. Norman, I assure you that many Link trains during rush hour are relatively “full”. There are a few seats available and there is isle space for more passengers, but there are also a LOT of standees. I see them pass by when I’m waiting at the gates. I’d take a picture, but then I’d probably be disciplined for using a PED and we can’t have that, can we?

      1. I ride Link often, and actually count passengers, so I believe I have a far better idea of how many people are actually on those trains than you do. And ST’s ridership estimates bear out my observations. I have never counted more than 93 passengers on one Link car at one time during a normal commute. There are probably more than that going to and from events at the stadiums, but that is not a “normal commute.”

        Is anyone going to use the June ridership figures to estimate how many boardings per Link car there are during peak hours, or not? I can’t believe nobody here actually wants to know that information! hehe

      2. Ninety people easily fit on one articluated bus, correct. about 60 seated and 30 standing. You disagree?

        I’m not saying 93 is the capacity of one Link car. [deleted, lying]. The capacity of an articluated bus is 90. I am saying I’ve never ridden on a Link car with more than 93 people on it, other than on opening weekend. I do believe that they get a lot more than 93 people on one Link car to and from events at stadiums. Maybe that is what you have seen from your bus.

        But, even with 93 people on one Link car, that is at least 19 people standing on that one car, because there are only 74 seats per Link car. It is probably a lot more than 19 standees, because usually quite a few seats are taken with luggage, purses, etc., so there are never actually 74 passengers seated. With 93 people on one Link car, there may be about 63 seated and 30 standing, which could look pretty “full” from outside the train looking into the windows.

      3. That’s what I thought. My point is that if you tried to fit 93 people onto an articulated bus, you *can* do it. But that bus has only 2 doors, each of which is narrower than Link’s 4 doors per car. Add to all of that Link’s more open floorplan and you get an easier flow of passengers when the car/bus is completely full. Getting passengers on and off a bus that full takes a lot of time. Link trains that full board and deboard much more smoothly.

        I still object to your 137 number that you keep pulling out of a hat – Crush load on a bus is somewhere around 100 vs. Link’s 200 – per car.

        We’ve been through all of this before. I’m simply adding this comment to point out that you’re making numbers up. [again]

      4. A DE60LF with 90 people on it is pretty damn packed. Now there are some routes where this is regularly exceeded (71/72/73/74 weekdays between the U-District and Downtown for much of the day). That gets very uncomfortable and loading/unloading takes forever. It is especially fun when a driver is being strict about “front door only” after 7 or when outbound from the CBD.

      5. The 137 number that you cling to has been shown to be wrong before. The capacity estimates in the North Link DEIS were written before ST had chosen and LRV design.

        Why don’t you quote the capacity numbers from the 2008 operating plan that you linked to just yesterday? It’s on page 35 if you can’t find it.

        Since I know you won’t, I’ll do it myself.

        “6.3 Load Factors

        For midday levels of service, a load factor of 1.0 is used to
        represent a seat for every rider. A load factor of 2.0 has been established as a standard for Central Line peak period service. The 2.7 load factor produces a maximum assumed vehicle capacity of 200 persons for the 95-foot articulated light rail vehicle, equivalent to 800 persons for a four-car train.

        Peak Hour Capacity for Planning: Seats 74; Load Factor 2.0; Passengers per car 148.
        Maximum Capacity: Seats 74; Load Factor 2.7; Passengers per car 200. ”

        “6.11 Future Line Capacity

        After the light rail line is extended north and train operations require sole possession of the DSTT, the signal
        system will enable trains to operate at 2-minute headways. At that time four-car trains will also be able to operate in the DSTT tunnel. Under these circumstances, the line will be capable of carrying nearly 18,000 riders per hour in each direction, as summarized below. This volume is likely to be experienced in the central core of the system after several extensions to the Initial Segment.”

      6. I stand corrected. You’re not making up the 137 number. I’d guess that’s an average capacity while the “200” is a crush number, but that’s just speculation on my part so for the purposes of this discussion, I’ll take the 137 number.

        The fact remains that you are comparing a single train car’s capacity of 137 to a single articulated bus at 90 or 100. A bus, with 2 doors and on-board payment is not going to operate nearly as smoothly as a Link Train car with 4 wider doors, 74 seats, off-train payment systems, and a more open floorplan.

        If I seem a little touchy about this here’s why: Operating a completely full bus per Metro’s policy is a maddening affair. If you only open the front door on pay as you leave routes, people scramble all over each other to get off the bus which takes a long time. If violate policy and open both doors, some people get off without paying while others “flash” their ORCA cards (do I *look* like a computer?), and if they do tap their ORCA cards, that slows things down even more. BRT features you add, such as 3 doors, Swift-like off-bus payment, more open floorplans, etc… will improve the bus performance, but it will never reach the same capacity as a Link train.

        Every time you compare 90 bus passengers to Link’s 137 as if they are somehow equal, I just want to scream at you, because I *know* they are not compatible based on almost daily professional experience. Do you get it?

      7. I’ve heard the official crush load capacity of an articulated bus is 80, although once or twice after Garfield got out I have seen 100 squeeze on the 48. Link’s published official crush load capacity is 200, so I would suspect that it can actually hold 250 in each car if people really squeeze like in Japan.

  5. I don’t have time now to do this, but someone on this blog should do it.

    ST has a formula for how many riders they expect during each period of the day: early morning; morning peak; late morning, early afternoon, non-peak; afternoon peak; early evening non-peak; and late night. There is a certain percentage of total boardings they expect during those periods. I think they use the non-peak hours as a base (x), and the the peak hours are expected to carry 2x, and the early morning and late night .5x, or something like that.

    If someone can find ST’s actual formula, and then use the total ridership estimates for June, they could estimate how many boardings per Link train there are during each peak hour. Then we could get some idea of how close to “capacity” Link trains are during peak hours right now.

    I wonder if anyone on this blog really wants to know that information, or not.

    The highest number of riders I have recorded on any one Link car at any given time in June was 93 on a train that left Westlake at 5:30 pm heading south. This peak load occurred at SODO station, after which more people got off at each station than got on, so the number of people on the Link car gradually fell as we got farther and farther south of downtown.

    1. I find myself not really caring what “peak capacity” looks like. Frequency is lower than originally planned – it’s amazing that given that and the recession, we’re this close to projections.

      1. Well, since Link was sold to the public as having enormous capacity (“one light rail line has the carrying capacity of a 12-lane highway”) which was a blatant lie from the very beginning, it is very big of Ben to say that he really doesn’t care what the “peak capacity” of Link is. No wonder, since it is a fraction of what was advertised.

        “We’re this close to projection.” I guess that depends on which projection you are speaking about.


        On this SDOT webpage from sometime after 2007, they were projecting 40,000 weekday riders on Link by 2009: “By 2009, Downtown Seattle will connect to the airport and carry about 40,000 riders a day.”

        (As an aside, at the bottom of this webpage is a “slideshow” showing a street with supposedly the same number of people in cars, on bikes, or in one bus, to illustrate how much less room one bus takes on a street than the same number of people in cars. Someone on this blog once asked where they could find this, so here it is.)


        And on this ST Central Link Operations Plan from July, 2008, on page 33 of 61 (page 38 on acrobat reader), Table 5-1 shows that ST projected year 2010 daily boardings of 32,600 in year 2010 from Westlake to the airport.

        Of course, in typical ST fashion, ST has kept “moving the goalposts”, and lowering their projections for Central Link ridersihp as Link got closer to opening, so that it would be easier to meet the projections.

        Link’s actual weekday average of 23,396 in June, 2010, is not all that close to the 40,000 Greg Nickels said would be achieved in 2009, or even to the 32,600 ST was projecting two years ago.

      2. So I guess you missed the whole part about ‘recession’ and ‘frequency less than projected.’

      3. “one light rail line has the carrying capacity of a 12-lane highway”

        One line, not one car. Trains travel in two directions, and multiple trains travel in each direction.

      4. 4 car trains can carry 800 riders at peak loads. If trains run at 2.5 minute headways (24 trains/hour/direction) that is 19,200 riders/hour/direction.

        If a highway lane handles 1200 vehicles/hour (that’s 20 vehicles/minute or a vehicle every 3 seconds) with 1.3 passengers, that’s 1560 people/hour/lane.

        So yes, one track of Link light rail has the peak capacity of 12 freeway lanes. And freeway capacity actually decreases during periods of congestion and peak traffic.

      5. If you go by the Washington State Drivers Manual and leave 4 seconds between you and the person in front of you, The capacity a Freeway lane is supposed to have is 900 vehicles per hour, obviously we don’t, nevertheless, and I’d really like to see a freeway in Washington state carry 2,200 cars per lane per hour. That would be downright impressive. So tell me which one and at which time a day so I can go see this.

      6. No, I am not laughing.

        Earlier in this thread you stated Link capacity at 200.

        Federal Highway standards start at 1600 vehicles/hour and then adjust it downward for factors like interchanges, trucks, lane width, shoulder width. 1200 vehicles/hour is much more realistic, and that is before congestion or any accidents. If you subscribe to WSDOT incident reports you will notice that there are at least 20 incidents of accidents and disabled vehicles in each peak.

        The average car carries less than 1.3 riders – I was generous. If we would restrict the freeways to cars carrying 4 or more riders, then we would triple our freeway capacity, but we would exclude the 90% of vehicles with only a single occupant.

        You asked the question of how a Link rail line can carry as many passengers as 12 lanes of highway – and the math supports that statement as long as we permit unfettered access to our highway lanes of vehicles with one occupant and before we factor in the fact that some of these vehicles will break down or crash. That’s the real world.

      7. No, I did not say Link capacity was 200 per car. [deleted, continued lying after being repeatedly disproved]

        Highways in this area regularly carry more than 2,000 vehicles per hour per lane. This is accepted by virtually everyone, except you, apparently.

        We are not talking about what the average car carries, we are talking about what the “capacity” of the average car is. It is at least five.

        The average car actually carries about 1.6 passengers. So, how many passengers does the average Link car ACTUALLY CARRY??? Are you claiming that the average Link car actually carries 200 people past a given point, and this continues for a full hour?

        And, by the way, Link is never going to run at 2.5-minute headways. I doubt it will ever have 3-minute headways. Four minutes, maybe.

      8. Norman,

        Unless you prohibit SOVs from the freeways or require SOVs to stop and pick up riders on the on-ramps, the theoretical capacity of your car is irrelevant since taht capacity is not available to others. If 90+% of the cars on the road carry one person, that is how many passengers the lane can handle. That’s private transportation.

        All of Link’s theoretical capacity is available to the public. When a train pulls into the station, people can board right up to it’s theoretical capacity.

        Some rail lines operate 90 second headways. Link could easily operate 2 minute headways if the signaling on MLK allowed it. We all agree that it won’t, in part because it will share the downtown tunnel with East Link, but 4-car LRT technology clearly allows for short headways like that and that is the passenger carrying capacity of one rail track.

        I don’t know of any of our peak hour roadways carrying 2000 car/hour/lane. When I sit alongside WB 520 in the evenings, it hardly carries 600/hour/lane.

      9. “Highways in this area regularly carry more than 2,000 vehicles per hour per lane.”

        Prove it. Show us a video!

      10. I was about to suggest that Norman should count passengers in cars on the freeway as he seems just as concerned about car capacity as he does Link capacity.

      11. Norman has a point. If you’re comparing capacity, then you should compare potential passengers, not actual passengers. BUT, if you’re comparing theoretical capacities, you can make assumptions that make rail look better, too (e.g 2-minute headways).

        The real point here is that Link can grow to its capacity easily (it is easy to get people to take the train) where a freeway lane cannot (converting 520 to all HOV 5 is a non-starter.)

      12. However, freeways do not just carry cars. They also carry buses and vans.

        A bus can carry 90 passengers. A freeway could easily carry several hundred buses per hour in one lane. But let’s just use a very conservative number of 200 buses per hour in one freeway lane. Is anyone going to claim that 200 buses per hour in one lane is any less possible than Link operating 3-dminute headways? Remember, buses do not have stops on freeways — they use exits and fan out to many different streets for the bus stops.

        200 buses per hour gives 18,000 people per hour in one lane on a freeway.

        And you still have room for over 1,000 cars per hour, plus several hundred vans per hour.

        Let’s just use a conservative 200 buses per hour; 200 15-person vans per hour; and 1,000 cars per hour; for a total of 1,400 vehicles per hour in one lane. That is well below vehicle capacity.

        Total capacity of that one highway lane = 18,000 in buses; 3,000 in vans; 5,000 people in cars. That is a total of 26,000 people per hour in one highway lane with 1,400 vehicles per hour.

        Now, even if every car had only one person in it, that still leaves a capacity of 22,000 passengers per hour in that one freeway lane.

        And that is with a very conservative 200 buses per hour.

      13. The peak volume in 2009 on I-5 at Midway was 15450 vehicles in one hour. 15450 * 1.6 people per car = 24720 people per hour. Link could carry the same number of people in one hour with 103 people per car at 2 minute headways, or 155 people per car at 3 minute headways. I-5 at Midway is 10 lanes wide.

        The peak volume in 2009 on I-5 at 185th was 14815 vehicles in one hour. 14815 * 1.6 people per car = 23704 people per hour. Link could carry the same number of people in one hour with only 98 people per car at 2 minute headways, or 148 people per car at 3 minute headways. I-5 at 185th is 9 lanes wide.

        The peak volume in 2009 on SR-520 at 148th was 8614 vehicles in one hour. 8614 * 1.6 people per car = 13782 people per hour. Link could carry the same number of people in one hour with 144 people per car at 5 minute headways, or 114 people per car at 4 minute headways. SR-520 is six lanes wide at 148th.

        Using Norman’s estimate of 1.6 people per passenger car and 93 people per Link car it is easy to see that Link’s people moving ability is indeed equal to that of what many of our freeways currently move. And these are real numbers, not hypothetical Norman Land numbers from freeways filled only with buses or vans.

      14. 93 was the GREATEST number of people I have seen in one Link car — not the average. The GREATEST number I have seen in one car is 9. So, since I have seen one car with 9 people in it, I guess we should use 9 people per car as the “capacity”, right?

        If you want to compare the AVERAGE people per auto, which is about 1.6, then compare that to the AVERAGE number of people in one Link car, at any given time, which is about 25, or 30, I expect. It is certainly not close to 93.

        But, we can compare actual numbers of people using highways to actual numbers of people using Link, which is far more important than theoretical “capacity”.

        The 520 bridge carries about 190,000 people per day. That is in four lanes.

        Central Link averages about 24,000 BOARDINGS per day on two tracks. There is no point along the Link line where it carries anywhere near 24,000 people per day. The peak load is probably a little over half that, or, say, 14,000 people per day past a point, on both tracks combined.

        That is 190,000 people per day on the 520 bridge; around 14,000 people per day past the peak load point on Central Link.

        Any questions?

      15. So you’re saying that it is impossible for light rail to carry more people than what Central Link is carrying today? That makes almost as much sense as a freeway filled with 15-passenger vans! Did 520 carry 190,000 people per day the first year that it was open? Of course not!

        You said that light rail was sold to voters based on it having the capacity to move as many people as what a freeway does, a real freeway, not an imaginary freeway filled with buses and vans. It does indeed have the capacity to move as many people as what our freeways currently move.

        And who cares anyways? I doubt if a single person voted for light rail because of its ultimate carrying capacity. They voted for it because riding a train is a hell of a lot better than being stuck on a freeway sucking exhaust fumes and a hell of a lot better than riding the bus.

      16. So Norman, where did you see 200 buses in one lane on I-5? That that number is absurdly outlandish, so let’s break down some actual numbers, and not just numbers that “everybody” “accepts”:

        I-5 is the largest freeway in the state. Seattle is the largest city along I-5. Seattle is the largest employment center in the state, and thus has the likelihood of having more buses on the freeway than anywhere else. Metro, Sound Transit, and Community Transit all run bus service to and from downtown.

        Metro’s fleet is 1,443. Sound Transit’s is somewhere around 250. CT’s is about 290. Grand total for three agencies is just under 1,990. Sound Transit has a spare ratio of 25%, and I would expect CT and Metro to be about the same. So a theoretical 1,492 buses on the road at any given time for ALL routes that those three agencies operate. Many of those never enter downtown. During peak commuting hours, at any given minute there will be at least one bus will be deadheading into or through downtown. That bus then has a theoretical capacity of 90 (or less depending on the model), but an actual used capacity of 1.

      17. Where did you ever see 4-car Link trains every 2.5 minutes?

        People are claiming that ST COULD run 4-car trains every 2.5 minutes.

        I am claiming that I-5 easily COULD carry 200 buses per hour in one lane.

        However, if you want to compare what is actually happening, the 520 bridge is actually carrying 190,000 people per day past one point, while Central Link is actually carrying around 14,000 people per day past one point, at its peak load. That is the reality of highways vs Link right now.

        If you want to talk about what COULD happen in the future, then certainly, I-5 or 520 COULD carry a lot more buses than they do right now. In fact, I believe there is a plan to increase bus service over the 520 bridge in the next few years.

      18. Right, Link was built for its future capacity. Just like I-5 and 520 were built for their future capacity 40 years ago, a capacity that they’ve now reached. I’m glad we finally agree on something.

      19. Norman, a lane of traffic on 520 or I-5 could not carry 200 buses per hour because those lanes are already gridlocked at rush hour. Link’s tracks could run more trains because they are not gridlocked — Link operates in its own right of way.

        There is no political will to remove a GP lane for a transit lane — there is political will to build light rail, which is why the 2008 vote defeated won despite your arguments. The people who most strongly support massive bus infrastructure do not spend their time or political efforts to accomplish that goal; they spend their time counting the passengers on Link trains and rehashing the same arguments from the past about maximum capacity.

      20. Norman,
        Buses do stop on the freeway. Yarrow Point and Evergreen Point are two that come to mind on SR-520.

      21. “Let’s just use a conservative 200 buses per hour; 200 15-person vans per hour; and 1,000 cars per hour; for a total of 1,400 vehicles per hour in one lane.”

        A 60 foot coach requires a 6 second following distance in perfect weather. Let’s be “conservative” and give a 2 second following distance for a car and van:

        6×200 =1200 seconds (buses)
        + 2×200 = 400 seconds (vans)
        + 2×1000=2000 seconds (cars)
        3600 seconds or one hour

        Hmmm… Norman, did you forget something? Like the length of all vehicles involved? How about weather conditions? Merging? Loading? Unloading? Hardly “conservative”.

        Once the Pine Street Stub tunnel is long enough to accept 4 car trains and the buses are out of the tunnel, a train every 2.5 minutes is well within the parameters of reality. It will require dedicated and well-trained operators. But at least you’re only dealing with about 60-100 professional operators. Not the *thousands* of drivers of varying skill levels and driving records required to achieve your scenario.

        Ok, I’m done poking holes in Norman’s arguments. I don’t know why I feel compelled to respond to his lunacy. Some day maybe I’ll learn to just let it go…

      22. VeloBusDriver,
        You respond to Norman for the mental exercise. Think of it as training for those dinner parties when uninformed indiviuals make statements that are typical views of the general public, but generally naive, and how you can frame responses.
        As you discuss them with Norman, you refine your arguments, so that you can meet others head on when they say the same things. Just tune up your sound-bite reply.

        Norman’s responsibility is to not go too far off the deep end with his numbers, or to mix up opinions with facts, lest he lose credibility when he actually has a correct argument.

      23. While I can understand a lot of questioning of “facts” here, there is one that should not be questioned.

        A lane of freeway can carry over 2000 cars per hour and do so on a regular basis.

        The assumed maximum capacity used to be set at 2,000 vehicles per hour per lane, but has actually been adjusted upward.

        Freeway speeds are typically 45 mph when this happens and any minor incident sends the freeway capacity from up to 2,200 vehicles per hour per lane to much, much less.

        For those of you that don’t believe this, look up the in the Highway Capacity Manual.

      24. During the I-405 program, we used 2100 vphpl. Even then, the arguments that are made have to be thought out as to what really is an equal comparison.

        If you argue using the highest lane capacity possible, then you have to use the highest LR capacity possible with the system that has been built.

        You also get into the issue of per-hour capacity vs. all day, which means, are you going to compare 50yr, mature system (highways in P/S region), with the 1 year ridership numbers from Link (or Norman’s observations)?

        Transit supporters should shy away from the practise above, that is,… ‘cherry picking’ numbers for a favorable outcome, since using the best apples-apples comparsion already puts transit in a favorable light.

      25. Buses take cars off roads — people who were driving cars, switch to taking the bus when bus service is increased. This has been proven when Metro increases bus service. Ergo, to say that highways are too congested to add buses is just ignorant: buses eliminate many cars from the roads, thus freeing up space for the addition buses. Is this really too difficult a concept to comprehend?

        To say that at some future date ST “could” operated trains at 2.5-minutes headways, even though there is no plan now to ever do so, but that area highways could never, at any point in the future, have 200 buses per hour on one lane each direction, is just absurd. Of course, highways can handle 200 buses per hour in one lane. And nobody here has given even any hint of an argument that this is not possible, yet some keep pretending that it could never happen.

      26. It will never happen because no one wants to pay for the army of bus drivers that it would require to run 200 buses per hour up and down I-5. Link is real, the expansions to Link that are in planning and under construction are real. Link adds real transit capacity. Your pipe dream of running an armada of buses on I-5 is not real, it’s a delusion. Understand the difference?

      27. Buses on 20 second headways is a real stretch but you don’t need anything close to 200 buses an hour to equal Central Link. That would be close to 200,000 people a day which Link won’t be moving any time in the foreseeable future; most likely never. Plus you’re already paying for a good percentage of the bus service because Link requires shadow service and connecting routes. Then you have to factor in that bus service is much easier to curtail off peak. And there’s no need to push everyone onto buses to solve the peak capacity problem with I-5. Just getting 1 out of 5 SOV drivers to carpool (better yet 5 people to vanpool) would be the equivalent of building an extra lane. And if you don’t believe, higher gas taxes, tolling, parking fees and employer incentives can get people to car pool/van pool then forget about them moving within walking distance of the train or doubling their comute time with a bus transfer.

      28. “Just getting 1 out of 5 SOV drivers to carpool (better yet 5 people to vanpool) would be the equivalent of building an extra lane.”

        Sure. But….
        “And if you don’t believe, higher gas taxes, tolling, parking fees”
        The moment you manage to change a currently free lane on I-5 into a toll lane, I will take my hat off to you. Until then, a rail tunnel is more politicially plausible.

      29. Central Link can never operate at shorter than 5-minute headways because it runs on the surface down Rainier Valley. It might not be possible to have headways shorter than 6 minutes, in reality.

        Also, once East Link opens, it will have to alternate with Central Link trains in the downtown tunnel, again meaning it will be physically impossible for either Central Link or East Link to operate at anything close to 2-minute headways.

        So, if you want to talk about actual possibilities, instead of fantasies, please stop claiming 2-minute, or 2.5-minute headways will ever be possible on Central Link.

        Thank you.

      30. I may just not understand here, but how would it not be possible to run trains constantly in the tunnel, CTA runs about 6 trains on .5 miles at a time in the loop (1 in each station, 1 waiting outside each station) Effectively running on 30 second headways if I’m not mistaken.
        Now, I am aware that El Trains regularly run red lights(as I watch the lights as we go by when I’m bored), and such, but how would it not be possible for us to do the same here, and just run one train on the heels of another through the DSTT?

      31. I’m not sure what you’re referring to when you use the term Central Link, but it’s important to note that the junction for East Link will be located south of the International District station. The tunnel stations will contain both the trains going to the airport and the trains going to Bellevue/Overlake. Thus, if each branch has 6 minute headways, the tunnel will have 3 minute headways.

      32. “Central Link can never operate at shorter than 5-minute headways because it runs on the surface down Rainier Valley. It might not be possible to have headways shorter than 6 minutes, in reality.”

        When/if we run into capacity problems, we can build another line through Aurora and SODO. (Separate from the Ballard-West Seattle line, which is too far away to make a major difference to the central line’s ridership.)

        Five minutes all day would be a dream — and equivalent to SkyTrain. But noise-hating neighbors would have some complaints.

        “Buses do stop on the freeway. Yarrow Point and Evergreen Point are two that come to mind on SR-520.”

        The bus has to get out of its lane and then merge back into traffic. Yarrow Point, Evergreen Point, Rainier, and Montlake are the best examples of freeway stops. In other places, the bus goes onto a side road with turns and traffic lights, which adds to travel time.

      33. The time to run the numbers is when U-Link opens. The capacity of Link is officially *not needed* until U-Link opens anyway and is therefore irrelevant until U-Link opens.

    2. How can you see all the doors to do counts when there are that many people standing?

  6. According to the Journal of Law and Economics in February 1915, 518 private jitneys in Seattle carried 49,000 passengers daily. Link has a ways to go.

    1. What was the average number of dedicated parking space per residential or retail unit at the time? The average rate of automobile ownership? Note by the way that Link is carrying its ridership with just 35 cars (effectively 17 trains).

  7. Link numbers are heading in the right direction. Only time will tell how successful it will be, but looking at current performance measures and ST selected peer systems, it’s going to be a long uphill climb. U-Link will undoubtedly improve performance measures, but that’s still 6 years away.
    Cost per boarding for the 6 peer systems (MIN, STL, NJ, PDX, SJC, DEN)ranged from $2.02 to 5.31 with the average being $2.94 Link averaged $9.94 last year and is budgeted to be $5.63 this year. If growth continues on the present level, cost per boarding will come in at $6.78 (7.1 mil riders for 2010)
    Peer systems riders per revenue hour were between 52 and 91 with the average being 71. Link averaged 38 for 2009 and is budgeted to achieve 59 for 2010, which is a long way from being ‘overcrowded’.
    (Source: Nat’l Transit Database – 2008 , ST 2010 budget)
    Current Metro bus cost per boarding is $3.67
    As can be seen, replacing bus service with rail service is generally a good deal, but as the data shows, is not a given.

    1. Are the numbers you are using to compare Link to current, or their first year of operation?

      1. Peer system data is for 2008, not their startup year. The point I was making is that Link has a way to go before reaching “profitability” as they say in the business world.
        It would be an interesting exercise to dredge up all those systems startup years, then adjust for inflation. Go for it. I’m not sure the comparison would be all that insightful.

      2. I’m quite tired, so maybe I’m missing something, but wouldn’t first-year to first-year comparisons be far more insightful than the comparisons you made?

      3. Possibly, but FTA accounting practices and reporting has evolved over the decades. You would have to adjust for that. Some of those peer groups are quite mature now, and startup sceneries were all over the place (segments, times, stations)so it would really be an apples to oranges comparison.
        ST’s performance guidelines correctly looks at industry ‘norms’ when setting service and budgets against similar agency operations. I’ve just listed the systems they selected and pulled up their performance numbers. Now we have a basis for knowing how far the journey may be.
        Example, if you want to cut your cost per boarding in half, you need the same cost of service and double the riders. Or keep your riders, and cram them onto half as many trains running.

      4. Well, they DID build a branch line before building the trunk line — a classic mistake in railroad planning. If U-Link weren’t under construction already, Link might be in for a very bad future. Luckily U-Link *is* under construction already. It will be fine and it will cost less to operate than buses.

      5. I’m curious, why did they build Airport Link before U-Link? I’m sure there’s a perfectly good reason for it (geology/expense/etc.), but I’m curious what it is.

      6. Money, and it took a long time for the Port of Seattle to come to an agreement with Sound Transit on station location and who was going to pay to relocate the airport drives. Sound Transit decided to make it a separate project so that it wouldn’t delay the rest of Central Link.

      7. Sorry, I think you misread my question. I’m wondering why they built the segment in the Rainier Valley before the segment in Capitol Hill / North Seattle, given that the latter is expected to have much greater ridership. That shouldn’t have anything to do with the Port, no?

      8. Sorry, Airport Link was the name of the project that built the extension from Tukwila to the airport that opened in December. I didn’t realize you were talking about the entire light rail line.

        The original 1996 plan was to build a line from the U-District to S. 200th Street all at once, but the cost was much higher than what they anticipated, mostly because of the cost of tunneling, and tax revenues were less than anticipated, so they split the line in 2. The southern half was cheaper to build, and included the maintenance base which is required for operation. The FTA ruled that Central Link was the “minimum operable segment” and therefore eligible for grants. It would have also been political suicide to delay construction of light rail in the south in favor of building in the north, due to a long history of political malfeasance directed at the minority neighborhoods of South Seattle.

      9. zed,
        thanks for second attempt at answering the aleks question. the ST board chose to build south-first in early 2001 after if was found that tunneling under Portage Bay would be much more costly than originally estimated. ST was under political attack. They may have felt they needed to build something, or anything, to avoid death as an agency. another option would have been to make the initial segment between Capitol Hill and Mt. Baker, using the DSTT in both directions, while studying the Montlake Cut alignment that was effectually chosen. then the 1996 measure could have funded Link between NE 45th Street and Mt. Baker. the south KC funds could have been redeployed on non-Link improvements (e.g., center access ramps at South Spokane Street and frequent service on a 194 type bus service). the north alignments have always been forecast to attract much higher ridership. the south first decision in 2001 has had significant fallout. but the 2008 measure will allow the north line to be opened in about 2020.

    2. Skehan-

      The 2009 numbers are not relevant; the line is longer now and is continuing to grow in riders every month. As others have pointed out, you did not evaluate systems in their start-up years which I think is important. I really only think the metrics you’re speaking about will be useful until 2011 or 2012 when Link ridership begins to settle to more normal growth rates. But if every month is 5-7% growth then using snapshots doesn’t characterize the system accurately.

      These snapshots would be looking at the income of a 21 year old and saying that he has a “long hill to climb” to make the income of his “peer earners” who happen to be ten years older. It ain’t a peer system if it’s been around for much longer and has already had its ridership settle comfortably.

      1. Jensen-
        I didn’t even try to sort out data for the peer systems that Sound Transit has selected to measure itself against, for the reasons stated. If you think it is so important, then be my guest, I’ll be curious to see how you adjusted the data to make a fair comparison.
        As far as looking at current and projected metrics that ST considers important, it’s like comparing your speed running the mile to the 4-minute mile (now considered the standard for middle distance runners {wikipedia)).
        You have to keep in mind what’s normal and excellent in order to gage how you’re doing.
        All this nonsense about somebody counted this many on a car, and somebody else saw something else last month is anecdotal and totally bogus for evaluation of the system. Look at the real numbers each year and compare them to what is expected for a growing system. That was my entire point. I don’t recall making any judgments other than it’s a long climb to get Links cost per boarding under that of comparable bus service, which is the whole point of ‘mass transit’ – lowering your unit cost over time to justify the investment made to achieve that.

      2. “the whole point of ‘mass transit’ – lowering your unit cost over time to justify the investment ”

        Not necessarily. It may be one goal but there is a score of other reasons for buildings good transportation on dedicated ROW.

      3. Yes, there are lots of ancillary reasons to build mass transit, but the original driving force is to lower unit cost. That’s the difference between transit and mass transit – lots more bodies filling larger/faster/more reliable vehicles. If your just replacing a 3 dollar bus ride with a 6 dollar train ride, then your failing the primary objective, which is to reduce unit cost where more patrons can be served using the same resources.
        Burger King and McDonalds seem to have figured all this out.
        If social engineering tops your list of reasons for building mass transit, then you need to examine other ways to achieve that. A couple of billion to promote TOD along MLK is not that strong an argument for mass transit.

      4. You look at the dollars and cents but don’t look at other costs: congestion, travel time, etc. The 70-series of buses might have lower trip costs at the moment, but they’re also packed and slow compared.

        Yeah, all investments should eventually have benefits that exceed the costs, but those benefits and costs are not nearly all price-related.

      5. Mike, the point is that the data is useful to log and measure, but not useful to form conclusions from. I have no problem with the raw data, but you are implying that we should make conclusions along the longs “Link is performing worse than peer systems.” That is not a fair conclusion to reach given where Link is in its lifecycle, and especially given where Link was in 2009.

        This fairness in how data would be interpreted is also why I don’t run around saying that Link is the fastest growing light rail network in the country, even if it is.

      6. That’s a fair point. I would be amazed if Link were performing close to mature peer systems at this point. Decision and policy makers have a tough time balancing budgets with wise allocation of resources tempered with growing pains. Knowing what is “normal” is helpful. That’s where the comparison takes place.

  8. I would expect some of these increases during the summer to be due to increased tourism around this time, but still, I think plenty of it is based on natural ridership growth. Once we get to the fall we’ll probably start getting just a few percent increase per month but it’s not nothing. Does anyone know if there are any more Link-oriented service changes coming up in September (or early October, whichever one it is this year)? Also, next year the first couple TODs ( Tamarack Place and Station at Othello Park) will open. That should bring small but not insignificant ridership increases. I’m in San Francisco right now, and it’s great to see how many people take the Muni, especially the Muni Metro. Hopefully the same will happen in Seattle, with people of all income levels taking the Muni because it is simply cheaper and more convenient than driving.

    1. I don’t think tourism comes close to offsetting the number of people who leave on vacation once school is out. We got an email at work today reminding people that tapping in and out are required. That combined with revenue from cash fares should give a pretty good indication of who’s going where. Does anyone know if the automagic counters measure boarding and alighting or how they can distiguish between the two? How do they come up with average trip length?

    2. Rapid Ride A (Tukwila-Federal Way) starts in October. That could draw a few people into taking Link + A. It won’t be a dramatic difference because it’ll still be faster to take ST Express if your’re going all the way to Federal Way or Tacoma. And DP might point out that eliminating some of the stops on Pacific Highway may lose other riders. So it may be a wash.

  9. Looks as though after a bunch of hand-wringing and sniping Central Link will be right around its ridership estimates, and could easily surpass them. Kudos, especially given the economy and the fact that—as Ben mentioned above—ridership estimates were made assuming free rides in the tunnel and shorter headways.

    1. ridership estimates were made assuming free rides in the tunnel

      Really, I thought Metro explicitly ignores rides within the FRZ when reporting ridership. It seems pretty ridiculous to put FRZ trips in the same category as revenue fares, especially when it’s a duplication of existing bus service. Link should never be free. In fact I think the FRZ should be restricted to only surface buses if it’s retained at all. Off peak it’s not a big deal but during rush hour if you’re looking at standing room only loads the free loaders significantly slow down the boarding process.

      1. Not sure about Metro, but ST has been pointing it out since late-winter/early spring. I think it’s mentioned on the first page of either the first or second quarter ST ridership report.

        Completely concur that Link shouldn’t be free, by the way, just pointing out that since it isn’t it loses potential riders downtown.

      2. ST asked the public last June whether to make Link free in the ride-free area, in exchange for a 25c increase in the base fare. People rightly said no. Why make every commute trip more expensive to serve intra-downtown patrons who already have several alternatives.

  10. You know, if people just ignored Norman we wouldn’t have to have THE EXACT SAME GO ROUND over and over again.

    It is not quite obvious that he has his conclusions and doesn’t care about the facts?

  11. Norman earlier said that the capacity of Link cars is 137. He cited an EIS, which does not actually say what he thinks it does.

    Documents from Sound Transit, including this one (pp. 35), make clear that the maximum capacity is 200 passengers per car. The capacity that ST uses for planning — i.e. what they would expect people to regularly put up with and not just for special events — is 148 passengers per car (see the same document). The 137 number does not represent the maximum capacity of cars nor the capacity that ST uses for service planning.

    Norman, you seem intent on arguing about the capacity talking points of the 2008 Mass Transit Now vote even though the region overwhelmingly rejected your case. No one besides you is arguing about maximum system capacity — and why should they? We don’t plan our road, bus, nor rail infrastructure based on crush-load traffic. 200 people per car would never be comfortable or a regular solution, but nor would be a requirement that all cars carry their full capacity in passengers.

    Here’s the fact that matters: the marginal cost of increasing Central Link’s system capacity is running another train, whereas the marginal cost of increasing I-5’s system capacity is demolishing part of downtown Seattle and spending hundreds of millions on a new lane of traffic. So whether Central Link represents the same theoretical capacity of 12 lanes of traffic or “just” 10 lanes of traffic isn’t relevant when there cost of increasing freeway capacity is simply too great to be feasible.

    1. Absolutely wrong. The marginal cost of increasing capacity on I-5 is adding buses. Or van pools. Or car pools. You don’t need more lanes — just more people per vehicle.

      Central Link’s current capacity is 2,192 per hour in each direction in peak hours, with 2-car trains every 7.5 minutes.

      I-5’s capacity is at least 2,000 vehicles per hour per lane. If only 50 of those vehicles per hour in one lane on I-5 is a bus (does anyone want to claim that 50 buses per hour in one lane of I-5 is not “possible”), that gives a capacity of 4,500 people per hour in one lane of I-5 JUST IN THOSE 50 BUSES!

      So, I-5’s current capacity in one lane is 4,500 per hour with just 50 buses per hour in that lane, and no other vehicles whatsoever! That is more than twice the current capacity of Central Link. Add 1,000 cars to that lane, for a total of only 1,050 vehicles per hour in that lane, and you get another 5,000 people per hour capacity, or a total of 9,500 people per hour in one lane of I-5 right now. That is with 50 buses and 1,000 cars per hour in one lane.

      Now, double the number of buses in that lane, and you add another 4,500 people per hour in capacity, without adding any lanes!

      Are you going to argue that that is not possible right now to operate 50 buses plus 1,000 cars in one hour in one lane on I-5? That gives over 4 times the capacity of Central Link right now.

      Highways have enormous unused capacity. They don’t need more lanes — they just need more people/vehicle.

      1. “Are you going to argue that that is not possible right now to operate 50 buses plus 1,000 cars in one hour in one lane on I-5? That gives over 4 times the capacity of Central Link right now.

        Highways have enormous unused capacity. They don’t need more lanes — they just need more people/vehicle.”

        Haven’t you gotten that can of paint yet, Norman?
        The one to change the GP lane to a Transit Only lane?

      2. That won’t fly but changing I-5 to 3+ 7am-7pm and open to GP traffic the remaining 12 hours and weekends might and would be big incentive to use transit or car pool.

      3. Yahoo!!! IIIII LIKE IT!!!!!

        Where would you start having I-5 become this all-carpool corridor?
        The count lines?

      4. to change the GP lane to a Transit Only lane?

        The place to change 2+ TO 3+ would be exactly where it is now. Yahoo!!! Glad you like it.

      5. “changing I-5 to 3+ 7am-7pm and open to GP traffic the remaining 12 hours ”

        That quote says what is says. THAT is the quote I like.

        “change 2+ TO 3+” is a bit better, but not precise.

        On this forum, you can’t leave any wiggle room.

      6. Bernie, there are no carpool lanes on much of I-5 through Seattle so I don’t think your plan is an easy lift.

        Norman, putting another bus into grid-locked, rush-hour traffic does not increase capacity.

      7. Minor tweak: I’d even do 2+ and freight. 3+ would be better, but there’s plenty of capacity if you take the SOVs off the road.

  12. Norman,

    At one point, you said: “If you want to compare the AVERAGE people per auto, which is about 1.6, then compare that to the AVERAGE number of people in one Link car, at any given time, which is about 25, or 30, I expect. It is certainly not close to 93.”

    So I looked into it. I think there are about 119 trains per weekday in each direction, with two cars per train. This might be a couple off, as I just extrapolated from their schedule pretty quickly without thinking about it very much. And we already know the weekday ridership is at 23,396 (unless you think THAT is wrong, too). So that makes just over 49 people per car, and 98 people per train. this includes the late night and early morning and off peak. I think it’s very feasible for trains to have 100 people plus on them at rush hour with these numbers. Does that help you understand anything on this forum any better?

    1. You’re not taking into acount the fact that people don’t ride the entire line. Someone can get on DT and get off at Beacon Hill. Someone else gets on at Beacon Hill and rides to the airport. Two “riders” but only one person on the train at any one time. If the airport rider gets on at Columbia City there is nobody on the train for several miles even though statistically you have two riders.

      1. That’s the same methodology that has WSDOT saying I-405 it carries 800,000 people daily.

        Since the lane counts change along its length, if you said I-405 was 6 gp lanes in each direction for the full 31 miles, and it carried the same load as the Evergreen Point Floating Bridge daily, it would carry 175,000 vehicles per day, times 1.3 person per vehicle. Roughly 225,000 per day.

        The real question now isn’t how much a road carries now, versus what a rail line carries now, it’s a matter of what infrastructure do you want to build for the future.

        Is investing in more highway capacity worth the expense for the small payback over the LONG term.

        The reason I-405 is only adding 4 GP lanes is because they can use up the available greens space in the current ROW, for the most part.

        Building enough lanes to last farther into the future, (where a rail system would be garnering a healthy ridership), drives the costs up because now you have to buy up land to do it.

        Let’s just say that adding 6 GP lanes on I-405 would show the same overall ‘ridership’ equivalent to a rail line on the same corridor, BEYOND the 30 year horizon used for planning. The rail line does come out cheaper, but it doesn’t get the immediate effect the highway lane does.

        If you could say all things being equal, the question becomes.. What do you want the place to look like when you are done?


      2. The line is 15.6 miles and the average rider (according to ST statistics put out in the past couple days) is 7.1 miles. which is slightly less than half. i don’t want to do the math right now, but if it was at 7.8 or higher, then you’d have overlap with all “average” riders, meaning 49 people at peak. so it would be a little lower. obviously there are other factors (westlake station is busier than rainier beach, duh) but since there are no individual station boarding numbers, i have no way of figuring that out. i think this is a good proxy for the discussion that we’re having

      3. ORCA data ought to show how many taps-on/taps-off there have been for each station pairing.

    2. There are 248 Link trains per weekday, or 496 cars. The average comes to 94 BOARDINGS per train, or 47 BOARDINGS per car.

      As the next poster points out, there are not that many people on the average car at any one time, since people get on and off at almost every station along the route, and the average trip is 7.1 miles (according to ST). So, the average number of people on any one Link car at any given time is probably about 25 to 30, as I wrote before.

      Does this help you understand anything on this forum any better?

  13. “Does this help you understand anything on this forum any better?”

    Not with the numbers you qoute.

    But when I first got involved in these issues years ago, comments like this from people with your point of view had me look into the real numbers from the agencies tasked with providing these services and infrastructure.

    At first, since I supported rail systems as a viable alternative, I also held the popular belief that it wasn’t cost effective.

    Now that I’ve been deep into the numbers with my experience on the I-405 Corridord Program Citizens Committee, I’m more convinced than ever that rail systems are the preferred alternative.

Comments are closed.