Those Vanishing Streetcar Stops

Readers who follow me on Twitter will know that the question of which streetcar stops are being removed has been a simmering issue for some time. The question has become less “what is the list” than “why is it impossible to get the list”.

A related matter is the degree of consultation, or not, that preceded implementation of the changes.

Several changes for The Beach (Queen Street East and Kingston Road) were announced in an email newsletter from Councillor McMahon, and the format of the list, complete with stop numbers, made it clear that this was a TTC document.

TTC will proceed with the following streetcar stop relocations on May 13 to support the deployment of new streetcars:

On Kingston Road:

  • Move the westbound stops #2786 (Malvern Avenue) and #2799 (Walter Street) to a new stop at the midblock pedestrian signal at Glen Manor Dr
  • Remove the farside westbound stop #2801 at Woodbine Avenue to a new stop nearside of the same intersection

On Queen Street:

  • Move the stops at Kent Road, and Woodward Avenue, to new stops at the pedestrian crossover at Woodfield Road
  • Move the eastbound stop #3055 at Laing Street to a nearside location at Alton Avenue
  • Move the eastbound stop #6807 at Kippendavie Avenue east to the signalized intersection at Elmer Avenue
  • Move the eastbound stop #6815 at Scarboro Beach Boulevard and the eastbound stop #6812 to the signalized intersection at Glen Manor Drive
  • Move the stops at the unsignalized intersections of Lee Avenue and Waverley Road to the signalized intersection at Bellefair Avenue
  • Move the westbound stop #6818 at Sprucehill Road closer to the pedestrian crossover at Beech Avenue

Courtesy of the fact that the TTC’s own website contains out of date information about stop locations while the list in NextBus is current, it did not take long to track down the remaining changes, but the bizarre part of this is that repeated attempts to simply get a list from the TTC ran aground.

Today, I took an inspection tour of the affected locations to verify what has happened, and here is my list:

On King Street:

  • Stops both ways at Trinity Street removed
  • Eastbound stop at Fraser replaced by a new stop at the signal at Joe Shuster Way where there is already a westbound stop.

On Queen Street (in addition to the above):

  • Stops both ways at Connaught removed. (How will operators ever change cars without a transit stop?)
  • Westbound stop at Simcoe replaced by a new stop at the signal at St. Patrick. Now if only the TTC would put an eastbound stop there to replace the one they dropped in the last round at McCaul, and thereby break up the long gap from John to University.
  • Eastbound stop at Gladstone farside replaced by nearside stop. [Thanks to a reader for spotting this.]
  • Westbound stop at Beaconsfield shifted east a short distance to align with the new traffic signal at Abell St.
  • Eastbound stop at Wilson Park shifted west one block to Triller where there is a crosswalk and an existing westbound stop.

On The Queensway:

  • As a result of the restoration of streetcar service to Humber Loop, the stop at Parkside is back in service. This is reflected on NextBus but not on the TTC’s own site.

On Dundas Street:

  • Westbound stop at Crawford shifted one block to Shaw Street where there is a traffic signal and an existing eastbound stop.

On College Street:

  • Stops both ways at Clinton removed. (Thanks to readers who pointed this out in the comments.) [Updated May 18, 2018]

Now that wasn’t hard at all, was it?

(There may be more that I have missed, and if anybody spots one, leave a comment and I will update the article.)

What is not clear is the degree to which local councillors or residents were consulted about this change. This gets us into a rather murky bit of TTC management bafflegab. When the original proposal was before the TTC board in May 2014, there were motions amending the staff recommendation including:

Chair Augimeri moved that the Board:

1. authorize staff to proceed with the recommended changes to the stops in the staff report where consensus has been reached; and

2. refer the remaining stops identified in the staff report back to staff for further consultation with local Councillors and for report back to the next meeting.

The motion by Chair Augimeri carried. [Minutes of May 28, 2014 Board Meeting, Item 14]

It is quite clear that the Board intended that the proposals in the report had to be accepted by those affected. (For the record, there never was a follow up report provided by staff.)

The current round of changes includes several stops that were not part of the original list. When I pressed TTC management on what appeared to be a lack of notice of the change, not even bringing the scheme to the Board for approval, I was told that the 2014 motion was by an old Board and the staff were no longer bound by it.

Say what? Management can simply make up whatever policy they want when the Board is replaced in a new term of Council?

This is not a question of a nerdish railfan wanting to track the locations of stops, but of a much larger issue that will affect many parts of the City when the TTC turns it attention to bus routes. Some of the stop spacings on bus routes are embarrassingly short, and if the same principles are followed as for streetcars, a lot of buses won’t stop as often, or as conveniently as they do today.

Many of the changes are quite reasonable and take into account the fact that there are now both crosswalks and traffic signals at locations where they did not exist when the transit stops were first installed. This type of change has less to do with new streetcars than simply reflecting the updated street design.

Another justification for elimination of stops in the 2014 round was that this would speed service. In fact, the effects were minimal because many stops that were dropped were not at traffic signals, and they did not represent much delay to streetcar service. This time around, most changes are relocations.

Memo to Councillors with bus routes: Pay attention to what the TTC is up to in your ward.

15 thoughts on “Those Vanishing Streetcar Stops

  1. I live near Queen / Woodfield and can confirm that area residents received a notice from the TTC that the Connaught and Kent / Woodward stops were to be removed several years ago before they started putting the sidewalk cuts in. I will attempt to find the document and post here if successful.

    Steve: Thanks for confirming this. That would have been from the original 2014 report, and I’m not surprised that you received notice given the Board’s motion at the time. The issue going forward is that such notice be given, including consultation, as the TTC moves to consolidate/restructure its bus stops.


  2. Steve said: “Some of the stop spacings on bus routes are embarrassingly short, and if the same principles are followed as for streetcars, a lot of buses won’t stop as often, or as conveniently as they do today.”

    Are not buses fundamentally different from streetcars in this sense since they stop only on demand (if someone presses the button or if there is a person on the stop)? Perhaps this is theoretically true for streetcars as well (is it?), but in practice due to demand streetcars stop at each stop (I don’t ride streetcars that often, mostly subway and bus, but I’ve never been in a streetcar that has missed a stop).

    So, those closely-spaced stops are not used all the time, but they do provide convenience for a non-negligible number of people. Furthermore, is not the whole point of express buses to provide faster service on very busy routes, by avoiding those very same closely-spaced stops? Is this not basically a solved problem?

    It’s especially critical in the suburbs, where stops may be closely spaced from the perspective of the trunk road which has bus service, but for the pedestrian part of the journey from one’s house it might make a significant difference.

    Steve: Your argument in part parallels my own on the subject of lesser-used stops. Streetcars do NOT stop at every stop unless the demand is there, and this reduces the walking distance for riders.


  3. The stops at College & Clinton were removed without notice this week. Grace/College & Euclid/College are in operation. TTC claimed it’s to improve timing, but it’s a lighted intersection and a heavy traffic intersection most days and evenings. Even the bus drivers on the route were surprised by it.

    Steve: Thanks to both who flagged this location. I missed it in my scan through the routes and thought Carlton had emerged unscathed on this round. It was not to be.


  4. Steve said: “Your argument in part parallels my own on the subject of lesser-used stops. Streetcars do NOT stop at every stop unless the demand is there, and this reduces the walking distance for riders.”

    The question becomes balancing the need to stop for a single or pair of riders – versus the effort required for them to walk whatever the distance would be to the next logical stop. This really in my mind comes down to number at the stop, and distance to next stop. If it is 1 or 2 riders at a stop, stopping a full car and this saved the riders perhaps 100 meters of walking – I struggle – when it is 6 or 7 stopping a half full car, to save them 250 meters, well, that seems a different thing. It really does become a question of stop spacing, and density of use. This like LRT, a stop every 200 meters seems excessive, especially if particular stop use is light, a stop every 2,000 meters, well, makes it onerous to use. Downtown a stop every 300-400 meters does not seem to create either excessive walking or stops, but that depends on having to cross streets etc. You should not have to cross a major cross street to catch the Street car either where there is heavy traffic.


  5. Andre S – take the Dundas Streetcar from Dundas west station going eastbound around 10pm on a weeknight – I do it every night and many many times we will cruise without stopping from Sorauren to Dufferin, and maybe only stop 4 or 5 times all the way to Spadina. #themoreyouknow


  6. There seems to be a lot of confusion about the Parkside 501 stop. First week the streetcar was back in service, the automated announcements were skipping Parkside. Some drivers interpreted that as meaning they didn’t have to stop there, others had a bit of initiative and stopped anyway if someone requested it or if there were riders waiting. Different drivers had different opinions on the official status of the stop. I contacted customer service via both email and Twitter requesting clarification, letting them know that both drivers and riders were confused. The email respondent said the stop was no longer in service, while TTC Helps told me the stop WAS in service. All further requests for clarification were ignored.

    A week later, paper notices went up in the stairway leading up to the bridge saying the stop was out of service. So, it took them 2-3 weeks to provide any official information that wasn’t contradictory, and then about 2 weeks later, the notices were removed and the stop was seemingly back in service. However, and this is grimly predictable, the automated announcer STILL doesn’t announce it, so every time the stop is requested, riders get anxious because we have no idea whether this particular driver will be stopping at Parkside or at the next one that’s been announced.

    Multiply this one instance of incompetence by the number of stop changes you’ve listed above, and you’ve got a lot of confused riders across the city.

    Steve: I was there on May 17, and the stop was announced westbound and was in service.


  7. Steve said:

    On The Queensway:

    As a result of the restoration of streetcar service to Humber Loop, the stop at Parkside is back in service. This is reflected on NextBus but not on the TTC’s own site.

    Not only that, the announcement for stops on the streetcars has NOT been restored to include Parkside Drive which had been eliminated when buses took over since there was no place for a bus to stop.

    I was on an eastbound car and announcement was “Next stop: St.Joseph’s hospital” HOWEVER the car stopped at Parkside Drive and people got off thinking it was the hospital since the operator did not bother to make an announcement to correct the automated system.

    Liked by 1 person

  8. Here are the new stop spacings (using google earth) in the beaches …. note the TTC still does not have an accurate list up. The all stops on the 501 route page is actually missing stops.

    Woodbine to Elmer – 176m
    Elmer to Bellefair – 307m
    Bellefair to Wineva – 334m
    Wineva to Glen Manor – 253m
    Glen Manor to Beech – 438m
    Beech to Silver Birch – 200m
    Silver Birch to Neville Park – 230m.

    For Glen Manor to Beech … there is a wheelchair cutout at the midway point at Balsam, however it is only on the eastbound side.

    Steve: Do the new stop locations have cutouts, or is this something else they missed?


  9. I think the eastbound Queen at Gladstone stop has been moved to be nearside. At least there was a stop pole there when I passed by westbound.

    Steve: That makes sense. NextBus shows a stop at Sudbury which is the southern continuation of Gladstone.

    Parkside Drive has Schrödigner’s stop. It’s half there and half not there. The quantum function only collapses when the streetcar stops there, or does not stop there.

    Steve: But if there is nobody on the platform, is there anyone present to observe the stop?


  10. As much as I’d rather it not happen, I’ve been expecting this to eventually happen on Broadview (specific stop modifications or removal) Avenue on the north-south 504/505 merged route (at least north of Dundas, the two routes share railway here), but it hasn’t happened yet. The southbound stop at Wolfrey was moved about 40 yards north earlier this year, although the tram shelter right at the crosswalk stayed put where it’s been for most of my lifetime. Being that the old, longtime stop was right next to a crosswalk that, sadly, is often abused by automobile drivers, I’m a little more comfortable with the streetcars stopping well enough back that cars have a better feel for where they (hopefully) will stop (and if they’re on the right and following the rules, they’ll be well behind the streetcar doors).

    I’m a little disappointed that the Kingston/Woodbine westbound stop (502/503, with the side route terminating at the Bingham loop where the Coxwell 22A bus and Kingston 12 bus stops) has been removed. That was often where one of my only good friends in junior high and high school, with me in tow, would get on or off on our way home from school, and it had a lot of good memories for me.

    Steve: The TTC tried to get rid of the stops at Mountstephen Street, but the local Councillor (Fletcher) objected, and so they remain.


  11. My understanding of King / Trinity is that Pam McConnell was a strong proponent of this stop. When the issue of removing it (along with many other stops in front of churches) came up in 2014, she argued against it and won. She’s no longer with us, and so neither is the stop.

    The notification was posted at least 5 days before the stop was removed.

    Liked by 1 person

  12. I am definitely concerned that the TTC people who did this will next turn their attention to bus routes. If the rationale is “stops should be at pedestrian crossings”, then there are long stretches of suburban roads that have neither. And as far as I know, the traditional crossings with the pointing finger are being deprecated as undesirable and unsafe on faster streets.

    As a simple example, Horner Ave. would have only two stops between Brown’s Line and Kipling: at Beta St. and at Thirtieth St. This despite the fact that the blocks are oriented N-S in this area, so there are many residential streets running out to Horner. And those intermediate stops do get used.


  13. In theory, it should be possible to roughly calculate whether to keep a stop, eliminate it, or add a new stop. The last item is particularly interesting. In spite of major growth in Toronto’s population, with many new residential buildings and areas, we rarely hear about the TTC studying adding new stops.

    The question can be expressed mathematically. Let R be the extra ride time for all the people on the bus incurred by a particular stop. Let S be the extra walk/cycle time incurred by the users of a particular stop, if that stop is eliminated, to get to the next stop.

    Then R = N*T*P

    Where N is the average number of people riding the bus when it gets to the stop;
    T is the extra trip time incurred by decelerating, stopping and accelerating again; and,
    P is the probability that the bus will stop at this particular stop.

    Similarly, S = E*A

    Where E is the average extra walk/cycle time per person to get to the next stop if this particular one is eliminated; and,
    A is the average number of people (this number may be fractional) that arrive at the stop during the bus headway.

    If R>S, or N*T*P>E*A, then the stop should be eliminated. Conversely, if S>R, or E*A>N*T*P, then the stop should be retained. And the exact same logic applies to adding new stops at a proposed new stop location.

    Let us consider a real life example, the Sunnybrook 124 bus. This is interesting because the vast majority of the demand is on the ends of the route. Most people get on/off at either Lawrence subway station or else at Glendon/Sunnybrook. However, there are a large number of little-used stops in between. So everyone on a packed bus waits while one person gets on or off at one of these stops.

    Suppose we pick the stop at Lawrence and Wanless Crescent East. Let us try to estimate values for R and S. Based upon the totally unscientific method of “what I see when I am on the bus,” I will estimate the average number of riders as 40 people. Various people have studied values of average stop time. See this article.

    But I will estimate it as 15 seconds. And the probability of stopping is fairly low. I will guess that P = 0.15

    This gives R = N*T*P = 1.5 minutes. How about S?

    Since the stop spacing here is about 80 metres, I estimate that E is 5 minutes. And we have a fractional value of A at 0.15

    In this case, S=E*A = 0.75 minutes. In this example, R>S so we should eliminate the stop at Lawrence and Wanless Crescent East.

    My problem with what happens in reality is that these decisions are usually not data-driven. Instead it becomes an exercise of “who can scream the loudest and/or has most political clout to save their favourite stop.”

    Steve: This is a perfect example of how one can become so lost in methodology that one loses sight of the actual problem. You construct two values:

    One is the time lost by riders already on the bus, and the other is the extra time required for passengers who accumulate at the stop awaiting the bus’ arrival.

    Several factors skew your result:

    • You presume that the bus is full. This is not always the case, and the route you selected is unusual in having a very strong end-to-end demand. On a more important corridor this would call for a mix of express and local service.
    • You make no allowance for the time that might be saved by anyone alighting from the bus, only by a passenger boarding.
    • You equate the probability of stopping with the number of passengers using the stop. Both values are .15. In other words, about one time in seven a bus will stop for one passenger.
    • You make no allowance for the variable nature of stop delay that is affected both by demand (ons and offs) and more importantly by the presence or absence of a traffic signal. Stops at signals incur longer delays because the bus often will fall out of the traffic wave and catch a red light. TSP might fix this, but then we could also get into a discussion of how the delays to cross-street motorists fit into the equation.

    Most importantly, you do not perform a sensitivity test, the most basic part of evaluating any formula. In effect, what happens if we start “turning the knobs”.

    It is important to note that because the bus does not stop all of the time, the delay is felt only by riders on the bus that actually does stop. In your example, 15% of riders are delayed, but the other 85% are not. Meanwhile, the would-be users of the stop face longer access times. It is well documented in transit literature that in-vehicle time is perceived to be less of a burden to demand than access and waiting time.

    A simple example here is that your time lost by riders is 1.5 minutes while the time saved by one would-be rider is .75 minutes.

    You have assumed an arrival rate that already has the headway built in so that it takes about 7 headways simply to accumulate one rider. The most frequent peak service on Sunnybrook is every 6 minutes, and off peak service is every 20. That’s one rider every 42 minutes peak, or one every 2 hours and 20 minutes off peak. For all practical purposes, there is no demand here at all.

    If the bus is only half full, or if there is more than one passenger using the stop, the relative values of your two factors shift substantially. You have used the worst-case situation.

    Your formula justifies removal of stops more on heavily travelled routes where more riders would experience “delay”, and it ignores the disincentive to transit use of making access more difficult. This is a last-mile problem in miniature.

    The assumption that stops are kept for political reasons shows that your outlook already presumes that inconvenience to users of stops is of lower value to you, and can be dismissed as a “political” concern. No. People just don’t want to walk further.

    Please do not start a complex debate on the validity, or not, of your formula or my analysis.


  14. Eliminating stops to make journeys a few seconds faster deserves closer scrutiny.

    Has anyone polled riders to determine if their priority is fast service?
    Or is frequent service more important? Or proximity to stops?

    The TTC’s ideal seems to be a block-long streetcar or bus that runs every half hour.
    I’d rather take a shorter vehicle that stops more frequently and runs more often.

    As Uber/Lyft etc. start to eat into fare revenue for transit systems, why is the TTC seemingly bent on opening the market for them?


Comments are closed.