When Better Service Isn’t – Part V: Streetcar Routes

This article continue the series reviewing routes where the TTC alleges that service has improved during 2019. Please refer to the first two parts for introductory information.

This concluding installment in the series reviews the streetcar routes. The comparisons here are different because the roll out of the new Flexity fleet combined with service resiliency changes and the substitution of buses for streetcars on some routes makes a year-over-year comparison only a snapshot of one point in the transition. Instead, this article compares service in January 2015 when the new fleet was fairly small and the network was operated, for the most part with streetcars, to the service in January 2020.

There is now a pervasive problem on streetcar routes with the amount of time allocated to travel plus recovery, to the point that there can be congestion of multiple cars (or buses as in the case of 505 Dundas) at terminals. This represents a waste of equipment by over-allocation of time so that even the worst case trips will not be late. Most are early, and operators get generous breaks as a result.

When contemplating service levels on the streetcar network, remember that all surplus cars were used up in 1997 when the 510 Spadina line opened, and the fleet did not get net new capacity until the Flexitys began to arrive. There is quite a backlog of demand for better service and more capacity as King Street showed.

501 Queen

In 2015, service on 501 Queen still operated with one branch running Neville-Humber and the other Neville-Long Branch. This meant that service levels west of Humber Loop were determined both by the level on the common downtown part of the route and on the lack of reliability in vehicle spacing, not to mention short turns. The Humber-Long Branch segment is now its own route 501L except late evenings, and it is also part of the 10 minute network. This is a very substantial improvement from past years. The other major change riders have seen in 2019 is the replacement of CLRVs by the new Flexitys which have double the capacity.

East of Humber Loop, the service has been evolving. In 2015 it was scheduled for ALRV-sized vehicles, but as these were withdrawn from service, the schedules were not adjusted for the lower capacity of the CLRVs until comparatively recently. Now the route is operated entirely with Flexitys. Tracking the capacity actually provided as the fleet changed is a topic for a separate article, and I will return to that after there are a few months of full operation with new cars as a basis for an “after” comparison.

Note the huge increase in scheduled round trip times between Humber and Neville from 130 minutes (plus 5 for recovery) in 2015 to 164 minutes (plus 12 for recovery) in the AM peak on 2020. The change in the PM peak is even more substantial going from 138+6 minutes to 192+13. Whether all of this additional time is needed is open to debate because some of this is a “resiliency” change intended to eliminate the need for short turns. The question is whether better service could be provided without unduly harming overall operations, and whether the backlogs of streetcars at terminals can be justified on an ongoing basis. (This problem is not unique to 501 Queen.)

502/503 Kingston Road Services

In 2015, both 502 Downtowner and 503 Kingston Road were operating with CLRVs on a combined peak headway of 6’00”. Today, service is provided by the 503 Kingston Road bus operating every 6’30” in the AM peak and every 7’40” in the PM peak. Off peak service is more frequent now in recognition of the smaller sized vehicle.

504 King / 508 Lake Shore

Between 2015 and 2020, the 504 King route went through a major restructuring first with the addition of route 514 Cherry and later with the new 504A/504B services  and the elimination of through runs between Broadview and Dundas West Stations.

In 2015 there were many bus trippers in the peak period, but all service is now provided by Flexity streetcars operating on considerably improved headways on the common portion of the route from Dufferin to Cherry, and with larger cars.

The 508 Lake Shore trippers have returned after an absence due to car shortages, and there are now 5 Flexitys where previously there were 3 CLRVs, a substantial increase.

505 Dundas

The Dundas route has operated with buses for a few years thanks to the streetcar shortage and ongoing watermain reconstruction along the route. It will resume streetcar operation on March29, 2020.

The level of service now operate with buses is more frequent than the CLRV streetcars they replaced in recognition of vehicle size, but also with some addition of capacity in certain periods.

Running times for the bus service are in some cases longer, some shorter, and some about the same compared to streetcar services five years ago. The planned service for mid-February 2020 will restore some of the extra running time used during construction as a resiliency adjustment, and this implies that streetcar scheduled times will be longer than they were in 2015 when rail service returns to Dundas. The TTC has not yet announced service levels for streetcar service, but has indicated that as on some other routes, streetcars will be supplemented with peak period bus trippers.

506 Carlton

In 2015, the Carlton route operated entirely with CLRVs, and together with Bathurst this continued late into 2019. The schedule for 2020 shows a marked widening of peak headways, with lesser increases at other times. This provides riders with less frequent service, but with more capacity. The challenge is that on wider headways, gaps and short turns have a greater effect and it is not clear that the route will be better behaved with the new cars.

As on other routes, resiliency adjustments have created longer travel times and terminal layovers.

509 Harbourfront

Between 2015 and 2020, Harbourfront converted from CLRV to Flexity operation. Demand has been strong on this route, and the capacity has generally increased thanks to larger cars. The changes in headways up or down vary by time period.

510 Spadina

In 2015, Spadina had just begun its conversion to new cars, and the schedules were still based on CLRV capacity. The Charlotte Street loop was still in scheduled use at King Street with one third of the service turning back there during many periods. By 2020, all service was scheduled to run south to Queens Quay reflecting the demand south of King Street.

The ratio of cars/hour with the Flexity schedules to CLRV is generally better than the change in the service frequencies, almost all of which are less frequent with the new cars.

In the AM peak the service went from 2’40” to every 4’23”. In the midday, PM peak and early evening, headways doubles or worse representing a net loss of service. Other periods saw wider headways, but not as marked a change.

Spadina has received longer travel and recovery times for resiliency, but this poses a challenge as it is not possible for cars to take long layovers at either terminal. Instead, operators take breaks at Bremner Boulevard which is of little use to them (e.g. for a washroom stop) and the stop delays passengers.

511 Bathurst

Bathurst was the last route to formally convert from CLRV to Flexity operation. Headways on the new schedules are generally wider than they were for the CLRVs with the biggest effect coming during peak periods.

The travel and recovery times on Bathurst have been lengthened quite substantially for “resiliency” with total peak period times in 2020 that are 50% longer than in 2015. I plan to review actual operations and the time required for this route in a future article.

512 St. Clair

As on other routes, the replacement of CLRVs by Flexitys brought wider headways notably during peak periods. What is quite striking, however, is that a route operating on its own reserved lanes also had substantial changes for resiliency with longer travel and recovery times. Service is now slower than it was before reconstruction of St. Clair occurred.

8 thoughts on “When Better Service Isn’t – Part V: Streetcar Routes

  1. I’d love to see an estimate of what all this extra run time costs the city. Roughly ball parking Queen, we have 27 vehicles taking an extra 40 minutes to do a round trip. Since service levels vary throughout the day I’ll pick 12 hours as a typical day and assume 250 weekdays per year. If we pay drivers $100 / hour including benefits then we have 2/3*27*12*250*100=$5.4 million per year in additional operating cost. Presumably depreciation, wear and tear, power draw, etc. are fixed costs.

    This doesn’t begin to account for the cost of additional run time on passengers, which the Texas Transportation Institute loves to do to account for driver delay. They claim congestion costs the US economy $166 billion per year in lost productivity. Using the same methods, slowing streetcars down must cost Toronto millions of dollars.

    Can the city afford to keep padding the run time?

    Steve: As long as Rick Leary keeps patting himself on the back for reducing short turns, and builds schedules so generous that they are almost totally unnecessary, we will have this crap. It’s the sort of technical nuance that the pols don’t like to get embroiled in. The opposite end of the scale is a push for “efficiency” that trims everything to the point the service falls apart.


  2. Steve said: What is quite striking, however, is that a route operating on its own reserved lanes also had substantial changes for resiliency with longer travel and recovery times. Service is now slower than it was before reconstruction of St. Clair occurred.

    This is not a very good argument (!!) for the benefits of streetcars (or buses) in their own ROW which seems to be the standard TTC answer to suggestions for extending routes. As having a ROW would seem to be a useful idea, what have the TTC done to screw it up?

    Steve: Part of the problem has been with operators taking breaks at terminals, part is an absence of headway management. The TTC’s standard fix is to extend travel and recovery times, but this just makes the problem worse.


  3. What does this look like in wasted dollars? By that I mean, what is the cost if we ran enough vehicles to make up for the bloated run times? I’m assuming we’re at least in the tens of millions per year. This is the sort of thing that should cross partisan and ideological lines on Council and engender a strong demand for a fix. To think, we could quite possibly achieve a larger service increase in percentage terms than is proposed over the next several years, simply from cutting run time bloat! (with proper line management!)


  4. I frequently see (mid-day) 3 to 5 St.Clair cars coming/going at Gunn’s Loop with minimal riders and hit and miss supervision (now and again). All this while waiting for ONE bus to continue westward.

    What is needed is full-time supervision at all times. Cost can be nil by just cutting off one of those near-empty cars. Money for an additional bus on St.Clair West between Gunns and Jane would likewise be easy to come by with a more balanced vehicle count on 512. End result would be happy customers!

    Steve: There was a point where a route supervisor would stand at the west end of the line and actually dispatch service. Looking at the tracking data, I could tell when he arrived, and when he left, the problem was that transparent. I fear that there is a whole generation of “management” who do not know how to manage.


  5. I am totally baffled that we don’t have route supervision in the form of automated systems…the cost is minuscule to the savings…an excel spreadsheet could do this for most routes…tracking can be done on a daily basis per driver with the GPS data…and it can and should be obvious to them and all riders when they are failing in the form of flashing lights, buzzers, gates, text messages, apps, paintball guns and yes even their supervisors and union reps…the whole thing is ridiculous, if McDonalds cashiers can accurately and reliably cook French fries then we should be able to get streetcars to leave on time…the goal should be 0 seconds of “delay from scheduled or headway” after appropriate time for breaks (and not including technical delays)…this should be easy to schedule and track, and all drivers should be compensated for hitting this mark…this would save money in operations and make transit way more efficient and appealing.


  6. Steve wrote: “I fear that there is a whole generation of “management” who do not know how to manage.”

    All too true. Management incompetence has lead to lousy service and wasted money. What we need is strong leadership at the top to demand and get performance at all levels. I see that Andy Byford has resigned his position in New York. Perhaps we can get him to come back here and (once again) turn things around.


  7. The money wasted by having unproductive TTC vehicles, either running in bunches or laying over, has already been pointed out by many commentators — including me in the past. It still bears repeating. During the Queen St. bus replacement, out by Long Branch loop there would be 10 to 15 buses laying over….they actually had to double-park. If you figure the headway was about four or five minutes, you can see that every bus got a nice little rest, doing nothing. And when they left to head eastbound, it was not unusual for them to run in pairs or triplets.

    Another thing, that particularly affects streetcars, is the removal of the stops at Bloor or Danforth, just before the station. I know the stop on Bathurst at Bloor is gone; I’m not sure about Dundas or Broadview. So when the streetcars are sitting, waiting for the station to clear of yet more streetcars, passengers can’t get off at the intersection. Some may actually want Bloor Street (just as other riders get off at Dundas, or Queen), while anyone willing to walk to the station will probably beat the streetcar. And with Presto’s two hour transfer, it’s not a big deal to get off and walk.

    Steve: There is no stop northbound on Broadview at Danforth except for the night car. There is still a stop northbound on Dundas at Bloor because it has a safety island.


  8. If there’s extra driver time to be had, what’s the excuse for not running on headways rather than schedules and moving to step back crewing at terminals? On frequent service routes (pretty well every streetcar route at peak but the 502/503) riders care about headways, not schedules anyway, and it would mitigate the issue of backed up vehicles at terminals.

    Steve: The TTC and ATU 113 really need to come up with a totally new approach to service and crewing so that terminal breaks do not require so many vehicles to sit around awaiting their operator’s return. Different approaches would be needed depending on the length and geography of each route, but this should be possible.


Comments are closed.