Where Are The Queen Car Riders Going? (Updated)

The coming TTC meeting includes a long report on the status of the Queen car and various strategies to improve its operation.  I will comment on that separately when I have a chance to digest the material.

The report contains a fascinating table in Appendix A, at physical page 7, showing origin-destintation data for the route broken into five segments:

  • Long Branch to Humber
  • Humber to Bathurst
  • Bathurst to Church
  • Church to Kingston Road
  • Kingston Road to Neville

We learn here that riders originating in the Beach travel overwhelmingly to the Church/Bathurst segment, and I suspect even they are concentrated toward the eastern end of that segment.

Riders from east of Yonge overwhelmingly are destined for stops east of Bathurst, and only a tiny number travels to Long Branch.

Conversely, of riders originating on Lake Shore, well over half (52% peak, 63% all day) are bound for another stop on Lake Shore, not for stops on the Queen line itself.  Those who do continue downtown don’t want to go past Yonge Street.

What is fascinating about the report is that it completely ignores these data although they have profound implications for route structure and service.

People do not want to ride from Neville to Long Branch, but to the central area.  Claims that split routes would foul up travel patterns don’t quite line up with the O-D information in this table, provided that an appropriate overlap of east and west end service exists for the busy central section.

The TTC has consistently ignored the fact that the “Long Branch” service has a strong local demand that is abused by the through operation with the Queen service to Neville.  It is worth noting that the all day boardings west of Humber are 5,500.

Just after the 501 and 507 were merged, the count stood at 7,700.  In previous years when Long Branch had its own service, daily boardings ranged from 11,000-14,000.  This is a textbook example of destroying a service and its demand, and refusing for almost two decades to acknowledge the mistake.

Even in its weakened state, the demand remains over half local for the rather obvious reason that anyone going downtown has much faster ways to get there.  Part of this lies with congestion problems, but a lot has to do with the unreliable service.

Much work has focussed on fixing service to the Beach where, intriguingly, the all day boardings are less than on Lake Shore even though it gets twice as much service (on paper anyhow).  To be fair, the Long Branch segment is roughly twice the length of the Beach segment and the density of demand on the west end is lower than the east, but the optics are poor.  Assuming that every boarding has a matching return trip (not exactly valid, but close enough for a rough estimate), we are moving mountains for the 7,500 trips to and from the Beach segment every day, but the 11,000 on Long Branch are another matter.

Updated February 15:

Some of the discussion in the comments thread took me back to the original data, and a desire to see numbers of riders, not just percentages.  This information is now available in a consolidated table.

The first part of this table is the data reproduced from the TTC report.  The second part converts the percentages back to passenger counts.  As a double-check, I summed these values, and you can see that some of these do not exactly match the boarding counts no doubt due to rounding errors.  However, this is good enough for discussion.

The third part gives percentages expressed by origin rather than by destination.  For example, 69% of the riders going to the Long Branch section of the route originate there, 14% originate from Bathurst to Humber, 15% originate from Church to Bathurst.

The vagaries of demand surveys show up in the fact that the number of people originating in each segment is not the same as the number of people arriving there.  For example, 5,500 people board on the Long Branch segment, but only 5,034 make it their destination.  Similarly, more people board the Humber-Bathurst segment, 8,750, than travel there, 8,291.  Although it is possible that the 501 is gradually depopulating southern Etobicoke and Parkdale, the more likely answers lie in variations in trip patterns (out one way, back another) and the inevitable inaccuracies of sampling.

Of the folks bound for the Beach, 3,889, only 88 originate west of Bathurst Street.  Conversely, only 101 of the 5,034 travellers west of Humber originate east of Church.  Those among us with long memories might observe that this could be partly due to the long decline in service quality that would drive anyone trying to make a long trip across Queen give up and find another route.  In any event, the O-D pattern is concentrated in the central part of the route from Kingston Road to Humber (and more likely Roncesvalles if the data were more finely divided).

Lost in the mists of time are O-D figures for the era when the Queen car had well over 60,000 boardings per day.  Where did those lost riders come from and where were they going?

28 thoughts on “Where Are The Queen Car Riders Going? (Updated)

  1. I looked at this table this afternoon and I think that the following scheme would work well if the demand is as shown in that table:

    501 branch 1: Long Branch Loop – Church (or Parliament)
    501 branch 2: Humber Loop – Kingston Road loop
    501 branch 3: Wolseley Loop (Bathurst) – Neville Park Loop

    I think that if frequencies are kept the same as they are now the TTC would need roughly the same number of streetcars as are used now.

    Alternatively. branch 2 could be merged with routes 502/503 and operated from Humber Loop to Bingham Loop.



  2. I mentioned this in a previous comment but what about this:

    501A: Long Branch Loop-Roncesvalles
    501B: Humber Loop-somewhere east of Yonge where there is a loop
    501C: Neville Park Loop-overlap with the 501B

    Are there any way for the Queen Streetcar to turn around after Yonge Street and before Broadview? I’ve seen tracks on Victoria Street around Yonge-Dundas Square, how far down Victoria do they go?
    What about if a branch of the 504 turns west on Queen (is it still called Queen Street west of Roncesvalles?) to Long Branch Loop?

    Think of two routes from LBL? just like the 504 & 505 both are Broadview-Dundas West but two seperate routes through most of it’s “central” parts.

    Steve: The only place on the existing track layout to turn back eastbound cars is via Church, Richmond and Victoria. This is a very messy turnback location because of traffic congestion, but it’s been used for years. Further east gets tricky, and options all involve large on-street loops that would take cars off route for quite a while.

    Options include:

    Add curves in the southwest quadrant at Parliament and reactivate Parliament Loop (at King) so that an eastbound car could turn south on Parliament and use the loop at King as the layover point. This is a very expensive undertaking for any sort of trial operation.

    Loop via Broadview, Dundas, Parliament, possibly with clockwise operation until early afternoon, counter-clockwise thereafter. This can be done with existing track, but there would not be any place for cars to layover at the east end of their trips.

    A Long Branch to Broadview Station route is almost as bad as one going to Neville (it would be roughly equivalent to a service ending at Woodbine Loop). Any “Long Branch” car needs to stay mainly in the west end with forrays downtown only when there is enough demand.

    Turning back at Roncesvalles makes for tricky connections unless the Queen service itself goes through to Humber to provide the overlap. Looping around Roncesvalles Carhouse is somewhat time consuming, and subject to blockage by other movements in the yard. One reason I advocated going to Dundas West was to supplement the 504 service which tends to get short-turned a lot and to provide a direct link to the BD subway. There are many design options here and we could chew up megabytes of posts with each person advocating their own flavours.

    West of Ronces, Queen becomes The Queensway.


  3. The origin-destination survey was done back in 1996. With the rows of new condos, things may have changed a bit since then. On the other hand, moving the loop to Park Lawn essentially drops much of condoland into the new Humber-Bathurst section.

    To be used for route-splitting purposes, it seems to me that the matrix is a start, but more analysis is required at least.

    For example, another way to look at it would be to see what percentage of riders wish to cross from one sector to another. The effect is cumulative.

    So looking at eastbound AM peak riders:
    Crossing Humber: 0.48*1340 = 697
    Crossing Bathurst: 0.34*1340 + 0.81*2020 = 2092
    Crossing Church: 0.05*1340 + 0.09*2020 + .048*940 = 294
    Crossing Kingston Rd: about 100

    This does show that eastbound, with the data we have a split at Church will inconvenience the least number of people. I don’t have the time to examine westbound travel or all-day travel, but hopefully I can open the file at home on my ancient Windows 95 computer.

    The route is in effect split right now. To travel westbound beyond Humber, half the time you have to either wait for (hopefully) the following streetcar, or switch at Humber. Eastbound when west of Humber, the route isn’t split per se, but of course instead of skipping the Humber car, you just don’t get a car at all.

    Steve: The report states that the data are derived from the 2006 Transportation Tomorrow Survey. This is current information.


  4. Oops, I did mean to type “2006”. Still, what was true for 2006 may not be all that true for 2009, and if we’re planning for the future then the differences may trend to increase over time.

    Also, I did a bit of searching on “2006 Transportation Tomorrow Survey”. If this is the survey that’s mentioned at


    then looking at http://www.dmg.utoronto.ca/pdf/tts/2006/dataguide2006_v1.pdf

    they apparently learned of 34,346 rides on the Queen car.

    That may be enough info to go on, but what *really* needs to be done is a dedicated Queen line survey. Hmm, that’s exactly what the report recommends.

    By the way, at the Urban Affairs library I looked through the 1984 Queen streetcar report. Some of the headway graphs looked frighteningly similar to the ones you did a few years ago, horrible outliers and all.

    Steve: The 1984 report is available on my site. Nice to know that the Urban Affairs copy is still available. I expect to receive data for December 2008 and January 2009 from the TTC soon, and will be able to review the line’s operation under a variety of conditions and management strategies.


  5. Yes!

    If the Queen cars could use existing track to link up with the BD subway and have that service supplemented with a services bridging the gap between the two!


    Long Branch Loop to Dundas West Station
    Neville Park Loop to (Coxwell) Broadview Station
    Sunnyside Loop (Outside of Yard) to Coffee Time Loop (Kingston/Queen)

    Steve: It’s called “Woodbine Loop” after the original name of the racetrack that used to be across the street. When “New Woodbine” was opened out by the airport, the track became “Greenwood’ although that street is several blocks west. There are far too many Coffee Times to make that loop name unique!

    I don’t think the folks in the beach will love a forced transfer on their trips downtown, and Broadview Station is already overflowing with King and Dundas cars, most of the time. They have to wait on Broadview to get into the station.


  6. I think Miroslav’s comment on Dundas Square is actually interesting… it requires a new switch at either Victoria or Church along Queen though, as it can only loop 506 cars currently (or 508 cars if they wanted to push its origin point north along Church), but Dundas Square cannot loop Queen cars in its as-is layout (it’s also under construction at the moment).

    I also agree with Mike’s comment about the Kingston Road service playing a role in this mix, but with reservations (survey data suggests Kingston Road service should focus mostly on Kingston Road, not to downtown).

    I was intrigued by the O-D data much like Steve. I do think that service should extend beyond Bathurst [Wolseley] from Neville for a wider overlap, and so I’m thinking Sunnyside, as the loop is existing and avoids going through the carhouse proper (but still passes by the front of it).

    501 Queen: Neville – Sunnyside
    503 Kingston Road: Bingham [Scarborough Village when the extension is built] – Woodbine
    507 Lake Shore-Queen West: Long Branch – Queen/Church – Richmond – Victoria
    508 Lake Shore-King West: Long Branch – King/Church – Wellington – York

    I’d leave the 502 route as is (I love McCaul Loop, it’s unique in Toronto)… just adjust its schedule, make the 503 the main service on Kingston Road instead (currently 502 is the all-day route), and improve reliability.

    There are obvious problems with fleet availability at the moment to implement such service combinations, and I do agree with the report about the planned construction on Church St., but besides that, this kind of blended service should balance demand and supply rather well according to the available O-D data. Service would be high between Long Branch and Bingham, with moderate service in the Beaches.

    Steve: Any loop using the downtown tracks on Victoria, Church, etc., is going to be a nightmare to operate. Victoria Street is always congested and cars cannot lay over anywhere for recovery or headway adjustment. Any new loop needs to be well east of Yonge to be clear of this mess.


  7. Steve: Any loop using the downtown tracks on Victoria, Church, etc., is going to be a nightmare to operate. Victoria Street is always congested and cars cannot lay over anywhere for recovery or headway adjustment. Any new loop needs to be well east of Yonge to be clear of this mess.

    I frequently see cars (501, 504) laying over on Richmond between Church and Victoria. I know that this means that they impede traffic, but that could be an option.

    Also, could a turn be built from Queen onto Berti and from Berti to Richmond. This would require Berti to be 1-way southbound instead of northbound. The cars could then layover on Berti. The only thing I am not sure about is whether the corner of Queen and Berti could accomodate a streetcar turn.

    Steve: There’s actually another reason for wanting a downtown terminal well east of Yonge. If a car is late and has to be short turned, then Church is available as a short-turn point. Cars would still actually serve Yonge westbound. If Church itself is the terminal, then the short turns would occur at McCaul and cars would go west empty. We already have problems with 502s short turning westbound at Church and leaving downtown empty eastbound from Victoria at the height of the pm peak, while also creating a very wide gap eastbound at Yonge.


  8. As we all turn our minds to how to make the ‘501’ function better…..
    I have a couple of questions, Steve.

    1) I saw in a post/article/comment somewhere this weekend that supposedly the TTC was considering a loop through the parking lot at Broadview & Queen.

    I assume, though it was not specified in the article/comment, they meant the Green P lot just north of Queen. This is the first I have heard of this. Is this true? Does it even make sense?

    Steve: There’s a small problem with that parking lot — it doesn’t go anywhere — and would only make sense as a loop back onto Broadview, a rather tight one at that, comparable to, say, Neville. Alternately, they could build an on-street loop via Thompson (behind the Broadview House) and Hamilton. As I’ve said before, if they want to run this as a trial service, then just loop via Broadview, Dundas and Parliament. No construction required.

    2) I have long thought that the Victoria to James St section of Queen should be a transit mall. I know, in general these ideas don’t work out well in North America. But in this case, it strikes me as having several benefits and few drawbacks.

    The benefit is not so much in ROW/no competing traffic (its only 2 blocks), though that doesn’t hurt.

    But rather that this is a heavily used transit and pedestrian area.

    Currently there is no room for a significant surface transit/shelter/facility for passengers, and w/no islands this is a relatively un-safe transit stop, which also causes motorist no end of head aches. So it should be win/win there by taking cars out of the equation, you can provide full street car stations, and faster boarding/alighting.

    Also, I though that this would be the perfect place for a subway style signal that manages headway. If the blocked off area can hold 2 or 3 ALRVs then it would be possible to simply hold them (without backing up traffic) to manage the headway.

    Finally, benefits wise, you might be able to build 1 or 2 layover tracks within a transit mall or keep spares at the ready mid-line.

    From the perspective of opposition, there is no on-street parking in this section, and no loading docks that I’m aware of.

    Any thoughts?

    Steve: I like the idea, sort of, but the last thing we need is subway signals. CIS control is already able to communicate with the streetcars via their onboard displays.

    Storage tracks would be a problem in the short blocks because the curves into and out of them would occupy a good chunk of the available space.

    Queen Street isn’t just Queen and Yonge, however, and the real challenge is in the blocks further west where there is a lot of day-long, year-round activity.


  9. Unfortunately the Parliament Loop (King and Parliament) is no longer available as it was leased to the Porsche dealer who is supposed to be building a new dealership there (and moving off the First Parliament site. He has several permit applications pending for the new building.

    It does seem to me to be a good idea to add the missing curve(s) at Queen/Parliament because having Queen cars go eastbound to Parliament does help to avoid unnecessary transfers and there is now much more potential business in the ‘east downtown’ area (Yonge to Parliament). In an earlier post you noted that the Queen-Parliament curves are scheduled to be rebuilt in 2010 so I hope the TTC look at this BEFORE they rebuild the track!

    I actually wrote to the TTC a month or so ago suggesting that they should look at making a track connection between King and Queen at River Street (the streets are less than 100 yards apart there) when Waterfront Toronto reconstructs River Street (2010?) and brings it south of King and into the West Don Lands. Bill Dawson responded and said ‘thanks but no thanks’. I still think it’s a good idea with VERY little extra track being required and it would allow a ‘direction change’ for both King and Queen eastbound cars – presumably they would get back to their own streets by using Parliament.

    Steve: You presume too much. King and Parliament do not cross at right angles, and curves in the northeast quadrant would be very difficult, especially a west to north. If the River Street connection existed, it would make an easy loop for a King car (via River, Queen, Parliament), but the reverse for a Queen would be trickier.


  10. If people want to suggest River St., I think only a single track southbound between King and Queen, accessible from either direction on Queen and turning on to either direction on King (just have to get rid of the traffic island in the way) is needed. Combine this with a new “U-track” from eastbound on King to westbound on Queen (the opposite direction for such a “U” is impossible) where the two streets meet just before the bridge across the Don River, and a bi-directional “River St. Loop” can be implemented for use by both King and Queen services, if they want a new loop in this area at reasonable cost.

    It’d be nice to have the connection to Bayview from River removed though, perhaps in exchange for a connection to the Richmond and Adelaide ramps to/from the DVP as part of the project that is supposed to revamp those ramps anyway.

    Steve: What you propose would be rather challenging on the approach ramps to the Don Bridge and would definitely require special traffic signals to shepherd the streetcars through the loop.

    Can we please stop dwelling on schemes to add bits and pieces of track that may produce short turn locations, but not termini where things like a rest break, a crew change, etc. can take place without totally blocking traffic?

    This conversation is getting tiring, and has diverted the question of demand patterns on Queen into an exercise in drawing track maps.


  11. Why can the Long Branch car not be turned at McCaul? The connections are already there, and it is close to the University line? Although some cars would/chould short turn at Humber/Park Lawn.

    Steve: “Close” is not the same thing as “there”. It’s a fair walk from McCaul over to University (I know, I do it quite regularly) and the last time I looked, we were supposed to be making our system accessible and easy to use, not designing routes for the convenience of wherever we happen to have tracks. University itself is the west edge of the business district, and you are asking people who want a Long Branch car to hoof it all the way over to McCaul to get one, or to get on whatever shows up and transfer enroute.

    There is a job for you in a service planning department somewhere.


  12. I am trying to say that there is another option other then Church Street, which would make a new 507 useless as it goes too far east. If it were up to me, the car would run to Roncesvalles as I would make sure a loop would be placed in the area. East of Roncesvalles would be a subway. But that is asking too much of the TTC. So would be my alternative:

    507 Long Branch to Humber/Park Lawn to connect with the 501.
    513 Long Branch to York, south to Richmond, west to Univeristy, north to Queen and west to Long Branch.

    The 507 portion is the easiest to bring back, and implement. The 513 (an entirely new route) would not as new tracks would have to be built. The 513 would complement the 507.

    Steve: Have you noticed that York is one way northbound and extremely unlikely to change to suit your plan. Also a stop northbound at Queen and University would be challenging given the road geometry and traffic patterns.

    Once again I ask, please stop drawing track maps or I will close this thread down. It is especially annoying to have proposals where I can’t help feeling the writers have not actually stood on the ground and looked at what they might propose. Maybe I should come out to the 905 and rearrange all of their routes to suit my perverse sense of where I think the locals want to travel.


  13. Well, then make it south along University, west along Richmond, north along Duncan. The point is not the street specifically (I was thinking York as it is not always that busy) but that it connects with the subway.

    The reason why I prefer GO over the 501 is for the simple reason that GO is generally far more reliable. I generally use the 501 over the former 507 portion of the route which does make me somewhat bias to making sure that there is reliable service on the Lakeshore. However, I do see know traffic is downtown, and if a restored 507 spent too much time downtown, it will still see delay problems due to congestion.

    My suggestion was (and is) made on the information the TTC provided, and which you posted above, that over half of the passengers from the Lake Shore do not head downtown or past Yonge Street if they do. Thus I was (and am) suggesting a route based on what passengers desire. It is also based on your comment about “not designing routes for the convenience of wherever we happen to have tracks.” Then I am used as an example of when I do just that.

    Steve, this site is worthwhile but please realize that people are trying to provide alternatives. Most people who post do appear to use transit (and the TTC), so their thoughts, opinions, suggestions, etc. are of benefit to us all. We may day dream at times, but that’s what makes reading the comments to your posts just as interesting as your posts. Please keep up the hard work.


  14. In regards to the relative volume of passengers for the East and West based on the AM peak shows more riders in the West, the row for the ‘All day’ destinations shows more of the demand is to go East (36% + 15% = 51%) rather than to the West (5% + 26% = 31%). Interesting – but not sure what it means.

    Steve: The row you are looking at is for passengers boarding between Bathurst and Church, bothways, but you have omitted the trips that are local to that segment (the 18% not included in your totals above). What these numbers tell us is that more riders boarding between Bathurst and Church are going somewhere east of Church than west of Bathurst. Please see updated text in the main post regarding the absolute numbers of riders rather than percentages.

    I’m not going to suggest any track building, route splits or anything. You mused about the resources and effort (i.e mountains) being devoted to satisfying us whiners in the East. Keep in mind, the complaints from the Beach were about the all-day transit SERVICE on Queen here – because that’s what we know about.

    It’s the TTC choice in how to respond to the complaints and deliver service. The fact that this is being looked at with one rather large constraint is not necessarily at our request. The numbers in the attached report show that service to both the East and West could be easily handled with that constraint being removed. (And we wouldn’t have to worry about people posting track layout suggestions.)


  15. You may well not publish this but I think what people are trying to say in your blog (and, of course, it is YOURS) is to raise legitimate suggestions as to how the downtown streetcar network could be improved to make it more appropriate for the needs of the 21st century. Not everyone using cars on King or Queen wants to end or start their trip at Church or Yonge. Having to rely on the, flawed, Parliament Street looping possibilities for turning east-bound cars back to the west makes providing service where it’s needed – King and Queen from Parliament to Yonge – rather difficult. River Street may well not work but perhaps there is some other possibility – adding curves at Parliament and Queen during the 2010 reconstruction of that corner would seem to be fairly easy and would surely be ‘better than nothing’.

    Steve: What was becoming tiresome was suggestions that implied the authors’ familiarity with the area in question might extend no further than Google Maps. Also, in the short term at least, any proposal has to be based on the existing track layout because that’s what we have to work with for any trial. Finally, a basic requirement is that services from both sides of the city need to cross at least Yonge Street.

    Schemes to terminate route segments short of Yonge (on either side) ignore the principal destination for inbound riders and a major transfer point for outbound riders. That pesky traffic is something we just have to put up with, although that’s where ideas such as a transit mall come into play.


  16. Steve, here is what you said about the figures the TTC released:

    Conversely, of riders originating on Lake Shore, well over half (52% peak, 63% all day) are bound for another stop on Lake Shore, not for stops on the Queen line itself.

    That would suggest to me that a route for the Lake Shore (i.e. a restored 507) does not have to go as far as Yonge Street. Thus a “scheme” to terminate a restored 507 route along the Lake Shore would not have to cross Yonge. That’s my point: people on the Lake Shore do not require a slow running streetcar to Yonge Street. They need service on the Lake Shore.

    Please do not take me the wrong way, but I am using the figures you are generously providing us with as a reason for what I am trying to say. Yes it means that a person who chooses to take the streetcar from Yonge Street to some point west of Humber may have to change cars, but the figures tend to suggest that those people would be in the minority.

    Steve: I also said that I wondered what the O-D data would look like from the days when the routes were separate and even Queen alone carried over 60,000 per day. We may find that people going downtown from Lake Shore gave up years ago and turned to the comparatively frequent north-south services to the BD subway.

    However, if service is going to enter downtown from the west, it will pick up riders east of Humber and a lot of them will want to get to Yonge.


  17. I’m amazed no one has suggested resurrecting Mutual St Loop (east side of Mutual just above Queen) which was opened about the same time as McCaul when routes were split in late 1928 (partly to ameliorate late QUEEN & BEACH cars on the crosstown segment). It was the eastern terminus for the new LAKE SHORE route from Long Branch from January 1929 to 1937, and then used occasionally after that until an accident there during the big December 1944 snow storm caused it to be abandoned almost immediately.

    Of course someone may have built on the lot but if that’s the case there’s a huge parking lot on the north side of Queen between Dalhousie & Mutual that can accommodate multiple layover trackage, and could even be used by short turn 502s and 504s. Keeps all the routes off the street, allows short or long layovers, just needs three or four side-by-side tracks.

    Steve: I believe that land is the proposed site for a condo development that may be on hold thanks to current economic conditions. However, the TTC might get away with taking a bite out of the parkland on the northwest corner at Sherbourne and Queen.


  18. Steve, you complained repeatedly how the TTC is using the Humber Loop even when it is deserted with nothing for miles around.

    Given that the Lakeshore GO Line is right above it, why not build a GO station right there and integrate it with the loop? The streetcar will remain to provide local service and funnel riders into the Humber GO Station, while GO itself will be improved to 15 minute frequencies.

    Steve: This would add riders into GO at its peak point, not necessarily the most comfortable transfer for TTC riders. Also, of course, there’s the little matter of fare integration. Long Branch and Mimico Stations are already used by some on the Lake Shore although, obviously, not as conveniently. That embankment, by the way, would be a tricky place to build a station for 12-car trains.


  19. It isn’t that there is nothing there, as there are residential building south of the CN tracks and Gardiner. The problem is the long dark tunnel you have to walk through (yes, there is a sidewalk there.) I have used the sidewalk during the day, as do other people, but I do understand why people would not like using it.

    As Steve points out, people on the Lake Shore do use GO – Long Branch GO station is located within a two minute walk from Long Branch loop – the only thing separating them is the parking lot for the GO station, and a side street for about 10 homes. People are not going to use a slow moving streetcar east of Humber if they can take the faster moving GO train – that is just common sense. A new GO station at Humber would likely add demand there as it would be easier to get to from the Lake Shore then Mimico would be. It would likely add demand to GO, while decreasing demand on the TTC for the eastbound 501.

    I hope that the new Park Lawn Loop (or the “new” Humber Loop as I like call it for simplicity) will allow for the same types of operation as the current Humber Loop – i.e. the TTC will still be able to turn back cars on the Lake Shore. That way a restored 507 car would still be possible.

    Steve, I wonder what you would say about my following suggestion: keep the 501 car running (according to the “schedule”) between Neville Park and Humber with “every” alternating car going to Long Branch. However, the TTC would use two to three CLRVs to restore the 507. That way those who wish to use the streetcar for local service on the Lake Shore could do just that. The restored 507 would run from Long Branch Loop to either the current Humber Loop, or the loop at Park Lawn. This would be a compromise solution that would benefit most people in my opinion.

    Steve: The big challenge would be to ensure that the “507” cars actually ran on a proper headway rather than making up their own schedules and shadowing a 501 whenever possible. Park Lawn to Long Branch is about 40 minutes round trip, certainly in the off peak, and three cars would provide a 14-15 minute headway. Frankly I would prefer to see better.

    The question that would remain is how many Queen cars would get west of Sunnyside Loop. There could be a big underserved area through the new condo strip from Park Lawn to High Park. That’s another reason for taking the “507” further east such as Dundas West.


  20. At least 15 minute service is better than waht the Lake Shore gets at some point. That’s also part of the problem, we can’t keep too many streetcars downtown which is partly why the extreme ends of the route receive poor service – the cars simply do no get there. At the moment, the TTC calls for 11 minute service west of Humber and that’s with the 50-50 ratio that is supposed to happen.

    If a restored 507 car were to be implemented, I would not complain too much about Dundas West as a terminus, but would if it were downtown. Cars need to be available west of Park Lawn for those who wish to use the car locally. Perhaps a similar process as the 501 is required – run every second 507 car to Dundas West station, with the other cars going as far as Park Lawn/Humber. Of course that would require more than four cars to operate, which is why I am suggesting only three cars (or four maybe) for a 507 from Long Branch to Park Lawn/Humber as any more would mena less cars and drivers for other routes.

    If four cars are used, perhaps more 501 cars could be short turned at Park Lawn/Humber, thus increasing service between Park Lawn and Sunnyside.


  21. To answer your titular question, this is one (former) Queen car rider who moved the heck away five years ago partly because of the effed up service. My current home is now within walking distance of a subway station and I’m so thankful to not have to put up with that abuse anymore.


  22. So…

    Another report from Dr. Soberman… I can’t help but wonder how much of the report is his original thinking/analysis and how much was rewritten by TTC to buttress their apparent anti-trolley bias? Trolley’s make economic sense in Vancouver, even without BC Hydro (Electric Power) running Greater Vancouver transit (with the integrated fares of my youth)… why not Toronto?

    It would seem to me the most compelling case for running trolley buses would be to add them as the supplemental vehicles on existing overwhelmed streetcar routes, where capacity is already constrained, and the electrical wire and substations already exist. It would make a lot more sense than adding diesel buses to streetcar routes or totally new power/wire infrastructure to spike the business case for trolleys.

    If the Bathurst streetcar south of Bloor was converted to trolley buses in interim, it could free up CLRV’s for Queen/King routes to provide more reliable service on the two highest demand streetcar routes, until the new LRV’s are in service.

    Steve: There will be new LRVs in Toronto long before the TTC gets any TBs, let alone strings overhead for them to run under. This is not a meaningful option for the CLRV availability problem.

    Further, if the TTC really wanted a real cheap alternative to “it will never happen” $450M expansion of Bloor/Yonge subway platforms (really a TTC quid pro quo for Provincial funding of our new TC LRV’s)… how about a return of trolley buses on an express Bay route to siphon passengers away from Bloor/Yonge platform congestion to provide Bay Streeters with an express alternative to their financial district jobs?

    Steve: The problem is that the transfer at Bay Station is not an attractive one. Any time people might save by being diverted that way is eaten up by getting from the train to the street, waiting in whatever weather, and getting on the bus. If the number of people diverted from BY were the equivalent of one train load, we would still be looking at a bus headway of 90 seconds or less.

    The TTC has in the past spiked so many EA’s to their preferred technology (York U 1993/2003, Steeles 1993) Soberman SRT 2005 report, that this should come as no surprise that a report is tabled under the guise of academic independence when it’s wholly dependent and massaged by TTC prior to release to public.

    Where there is a will there is a way… maybe even a better way, if you’re looking for it and not waiting for capital & operating subsidies that will never arrive until there’s a blue moon over Toronto or swan boats on the Don!


  23. Steve, why not merge the King and Queen routes into one downtown loop. It would require some modification to the Queen/King/Queensway/Roncy(?) intersection in the west and new track laid where King and Queen meet in the East to complete the route. However it would allow the ALRV’s to be shared among the 2 streets, and remain, mostly, in the downtown core where there is demand. Eastern and Western segments (Kingston, Long Branch) would be run more locally with diversions into the core when demand requires.

    The only real problem I’d see with this would be that the loop would be constantly affected by traffic along both Queen and King.

    Steve: As I have said in response to other proposals, anything that is going to be done needs first to be possible without building track (that has a lead time of two years — one for design and procurement, one to be part of the construction cycle). No track is needed as Ronces as cars would use the carhouse or Sunnyside Loop as their terminus. To the east, the question would be whether to go as far as Parliament/Broadview to complete the loop.


  24. Steve, I am puzzled by the summary table.

    For Long-Branch/Humber, AM peak riders sum up to 129%, which is a huge error. This means that almost one-third of the AM peak ridership of 1340 comes from the outer-space branch of the Queen car (501 BETELGEUSE). I think this shows that the data are not good enough for a detailed view of the line’s operation.

    Steve: You have to read the tables correctly.

    The first set reads across and show that of 1340 boardings on the LB/H segment, 52% got off in that segment, 15% got of between Humber and Bathurst, etc across the row for a total of 101% (those are TTC numbers, not mine).

    The second set is derived from the first by multiplying the boarding count by the percentages. The row for the AM peak LB/H sums to 1353 (versus 1340 boardings) probably due to rounding error in the percentages (the extra 1% above).

    The third set takes the counts of people arriving in each segment and converts that to percentages. In other words, while the first set tells us where the people boarding in each segment went, the last set tells us for each segment where the riders came from. In this case, the LB/H column reads down with 88% of the AM peak riders to this segment originating within the segment, 10% from Humber to Bathurst, and 2% east of Church.

    The 129% you cite is taken by reading across the third set of figures, but this is meaningless. That’s why the subtitle says “read down”. Sorry if this is confusing.

    It seems to me that the easiest place to terminate an east-end service would be Wolseley loop, based on the heavy ridership I observe between Yonge and roughly Bathurst. Where a west-end service would terminate is a lot trickier. I’m not sure if Church et. al. would be so very bad, given i) the infrequency of service (there won’t be huge lines of streetcars, except of course when they bunch) and ii) “no layovers” could be actually a *good* think if you think about it the right way (the way a rider, waiting for the car to reappear, would think).

    Finally, the interesting thing about the 1984 report was the complete lack of any mention of Lake Shore or 507 service. There was lots of talk about the 502 Kingston Road and 503 Tripper service in the report, and various rejiggings, but no consideration of any sort for anything west of Humber. (The book is a lot longer than the report you have on your site. It includes headway graphs and operational considerations. I just skimmed it looking out for “Long Branch” and anything else that seemed interesting.)

    Steve: We didn’t survey the Kingston Road or Long Branch services due to limitations on the number of volunteers. They stood on street corners, on one day in the rain, for three hours keeping track of the movement of streetcars. We analyzed what we could.


  25. Steve wrote:

    As I have said in response to other proposals, anything that is going to be done needs first to be possible without building track (that has a lead time of two years — one for design and procurement, one to be part of the construction cycle). No track is needed as Ronces as cars would use the carhouse or Sunnyside Loop as their terminus. To the east, the question would be whether to go as far as Parliament/Broadview to complete the loop.

    I agree that for an immediate fixture, no new track is possible, but for long term fixes new track may be required, if only minor changes.

    The loop problem would depend on how to run the loop. According to my map of the current tracks, it would be possible to go east along Queen, south on Parliament and west along King.

    The question would then be if you wanted cars going in the opposite direction as well – which would make sense. I am not familiar with the Parliament/Queen intersection, and my map of streetcar tracks suggests that it would be possible to go from northbound Parliament to westbound Queen, but it is not totally clear. The other solution would be run as far east as Church.

    Steve jumps in: You have a bad map. There is no east to south surve at Queen & Parliament. Never has been. Nor a north to west while we’re at it.

    A 507 car could use Roncesvalles carhouse or Dundas West Station. Another solution would be to add a connection at Dufferin later on to allow for the cars to interchange via Queen/Dufferin/King and/or King/Dufferin/Queen. Cars from the east could also turn back at Church. This would allow for a total intergration of streetcar lines. The only problem as I understand it is that the TTC would not find using the carhouse as a turning loop as being practical.

    Steve: It is possible to use Dufferin today for west King, north Dufferin, east Queen moves, but not the reverse. A full set of tracks is available further east at Shaw which would stand in for Bathurst as a western boundary for an east end service.

    Cars from the east MUST not turn back at Church or they miss the main stops downtown.

    I think that you need a much more accurate track map.


  26. Every now and then I find myself walking down the street on my way to wait for the Queen Street East streetcar (west bound in the morning). Arriving at my stop I, no less that five out of eight times, then make the decision that walking to Broadview is better than waiting for a car that is still not yet in sight; From Brooklyn street you can see well past the Greenwood yards. Eight out eight times I take the TTC I promise myself that I will buy better rain/snow gear and continue to take my bicycle to work. Just need to be careful of the damn tracks those infrequent streetcars use.


  27. Wotan says:

    “A 507 car could use Roncesvalles carhouse or Dundas West Station. Another solution would be to add a connection at Dufferin later on to allow for the cars to interchange via Queen/Dufferin/King and/or King/Dufferin/Queen. Cars from the east could also turn back at Church. This would allow for a total intergration of streetcar lines. The only problem as I understand it is that the TTC would not find using the carhouse as a turning loop as being practical”.

    The problem with this curve is that it would have a vertical curve from going downhill on Queen to a Horizontal curve to go south on Dufferin. Main line rail vehicles have trouble doing both curves at the same time; I hate to think what a single blade switch would do to a low floor vehicle wheel.

    Steve: A proposal to install curves in the southwest quadrant as part of the intersection work was deleted for just this reason.

    It is a pity that the Mutual street loop is not still available. It would make a perfect east end for the Queen west car. The Beach car could turn at McCaul or Bathurst but I would hate to think what turning all those cars at Woseley would do to the frequent Bathurst service.

    When (if) the WWLRT line gets built then it would seem sensible to run the Long Branch and maybe the Roncesvalles cars along it to the downtown and leave King West as a King West car. While through routing does eliminate the need for down town turn back facilities and reduces the need for overlapping service it does make line management difficult as well as making it difficult to run different headways on each end of the line.

    How accurate is the TTC’s O-D survey and when was it done? It would be nice to have a finer divide for the origins and destinations as Humber to Bathurst and Church to Kingston Road are long stretches How many of the 165 passengers who get on from the old 507 line actually make it past Jarvis or Parliament?

    Steve: The data are based on the Transportation Tomorrow survey, and I suspect they are constrained by the granularity within its mapping of trips. The TTC plans to do its own on-car survey this spring. I have no idea of their target sample size.

    When the TTC opened the Bloor Danforth line they split most of the North South routes at the subway to make it easier to tailor services. Why can’t they do this with the east west lines? I know that turn back facilities do not exist at the moment but it might be worth while looking into the cost and benefits of doing so.

    I am not holding my breath because the TTC finds it difficult to think of adding switches to most intersections when they are doing track work even if doing it makes sense. I know that there are a few exceptions but they were a while back.

    Steve: It’s important to know the OD patterns on each line so that we don’t put an artificial break at Yonge Street just because that happens to be where the subway is. Also, the west ends of Dundas and Carlton, for example, are much shorter than Queen to Long Branch. There’s a point where carving up a route into sections would be counter-productive.


  28. I used to ride the Queen streetcar, I’m likely one of the missing people you allude to. I now ride a bike, drive, or take GO. Anything but the streetcar. One more than one occassion I’ve waited over 1.5hrs. The one time the wait lasted 3 full hours, the next times I gave up and don’t know how long it actually took. So I can’t bother with the whole route anymore, it’s just too darn flakey. I’d use GO more often if it ran more often, but at once an hour bicycling is actually faster from Mimico to Downtown, and I really need the execise.


Comments are closed.