Transit Ridership Statistics & Tracking

[steveintoronto]

One problem is that Presto data is not the complete daily count - there are paper fares that have to be added in.

This idea that Presto is a mature and user-ready fountain of mineable data, just begging to be tapped, is an overexaggeration that is mostly offered by IT vendors and internal empire builders.

I'm reminded of this:
Three quarters of Presto users are satisfied with the fare card, poll finds
Which immediately, even if you accept the flattering case made, indicates 25% aren't satisfied!

Article contines:

But a poll of Torontonians conducted by Forum Research earlier this month found among the 58 per cent of respondents who said they have used Presto, 72 per cent were happy with it.
Thirty-five per cent said they were very satisfied with the experience, while 37 per cent said they were somewhat satisfied.
Just over one quarter said they were not satisfied. That included 16 per cent who said they were not satisfied at all.
While the numbers suggest a majority of riders like Presto once they start using it, there was less support for the TTC replacing more familiar forms of payment with the fare card.

And then the methodology:

Forum conducted the poll of 1,427 Toronto adults Jan. 15 to 17, using an interactive voice response telephone survey.

No comment needed on that...science this ain't.

I'm sure every poster in this string, like myself, initially saw the map/chart just released, and thought "oh yeah!"...only to find that the more you studied it, the less sense it made.

And now many of us are wondering: WTF? Was it a really clumsy attempt at PR? An atrocious attempt? And it makes us all that much more hungry for real figures. Perhaps Yurek himself compiled this? It has the mark of an amateur all over it.

 

[innsertnamehere]

Interesting to see that even the busiest GO station - Oakville - has less ridership than the vast majority of TTC stations. Even union station only qualifies as a mid-level subway station in terms of passenger traffic.

I agree with other users, I use GO whenever it makes more sense than the TTC for in city trips. The time savings are always worth $2 - $3.50 (if you still need to use the TTC afterwards) - a lot of times it works out to about a half hour quicker - thats valuing your time at $4-$7/hour! No thanks, I'll take the GO train. I've done Exhibition - Union, Danforth - Union, Danforth - Exhibition, Bloor - Union, Union - Downsview Park, Union - Kennedy, etc. Mind you I live about a 5 minute bike share ride from Union so it's fairly easy for me to do these types of trips..

 

[crs1026]

Interesting to see that even the busiest GO station - Oakville - has less ridership than the vast majority of TTC stations. Even union station only qualifies as a mid-level subway station in terms of passenger traffic.
.

That comparison is thought provoking, but it’s important to keep things in perspective. Imagine the worst case alternative - a further 75,000 single passenger vehicles a day fighting their way into downtown Toronto. Or, best case, 1500 buses coming into the downtown. Or, 75,000 additional condo units downtown and no transit system.

I wonder what the hypothetical offset to congestion at Bloor-Yonge subway station is.... without GO, would we have another 1000 pph using that station at peak?

A per-station count of a couple hundred riders per day can seem like small potatoes, but over the network it adds up.

I do think a “Regional” system might encourage development especially employment related outside the core, which would lead to a more efficient transportation network - more balanced flows at peak, and shorter commuting distances over all. Rather than build parking garages at GO stations, build office buildings so people don’t have to go downtown at all.

- Paul

 

[Midtown Urbanist]

The mega circle is actually for Dixie GO, not Cooksville. It took me a while to catch that.

Oh my god, that is hilarious.

Alright, I love the cartographic expression of the GO lines as much as anyone, but they are impractical from a data visualization perspective.

And so... I have edited the Metrolinx PDF to make this:

I have half the mind to swap the circle sizes with something that corresponds to ridership levels as opposed to "ridership change".

 

[drum118]

I agree with swapping circle size for real numbers, not increase. Also, isn't Millken under construction as well a few others?

 

[Leo_Chan]

I agree with swapping circle size for real numbers, not increase. Also, isn't Millken under construction as well a few others?

Then would Union Station fill up the entire map?

 

[blaixx]

Wow ... no wonder they've been hiding this. West Harbour has 58 riders a day, compared to 1,400 a day at Hamilton GO Train. Don't they both get the same number of trains per day?

West Harbour has 2 trains per weekday and Hamilton GO has 4. But still, assuming its arrivals + departures, only 29 people commute from WH...

 

[44 North]

Interesting to see that even the busiest GO station - Oakville - has less ridership than the vast majority of TTC stations. Even union station only qualifies as a mid-level subway station in terms of passenger traffic.

When I see the lowish (relative to TTC) numbers I think it helps make a case for more GO rail expansion. Feel like with recent RER promises people have got to thinking that commuter rail investments going forward must have high-use subway-like stations, ignoring that trickling in riders from far pockets is the bread and butter of GO and has been from its inception. Let's see what stations have to offer in Orangeville, central Durham, Brantford, and Cambridge and stack them up against Licolnville, Acton, West Harbour, etc.

Oh my god, that is hilarious.

Alright, I love the cartographic expression of the GO lines as much as anyone, but they are impractical from a data visualization perspective.

And so... I have edited the Metrolinx PDF to make this:

I have half the mind to swap the circle sizes with something that corresponds to ridership levels as opposed to "ridership change".

Very nice, and good layout. Maybe give a try doing the ridership circle sizes. Reaperexpress has made it easy by making a great database, just gotta figure out a style. Say a base circle size, then ten pixels every 250 riders?

 

[robmausser]

Oh my god, that is hilarious.

Alright, I love the cartographic expression of the GO lines as much as anyone, but they are impractical from a data visualization perspective.

And so... I have edited the Metrolinx PDF to make this:

I have half the mind to swap the circle sizes with something that corresponds to ridership levels as opposed to "ridership change".

It would be a nice touch to have two circles, a darker one to show the previous ridership level, and a lighter halo that shows the increase around it. So wouldnt just compare ridership levels between stations, but how much if any, it increased. Maybe a red circle inside the dark circle if it decreased.

 

[Krypto98]

I sorted the Riderships by Highest Daily and highest Daily

 

[Midtown Urbanist]

I think people are being a little overly critical on West Harbour, and even the Kitchener and Guelph stops. It takes time for ridership to build out, new travel patterns to emerge, and so forth.

But that underlays how important it is to be tracking these ridership changes over time.

 

[reaperexpress]

I sorted the Riderships by Highest Daily and highest Daily

Unfortunately I discovered that the Daily ridership figures are invalid to compare because they aren't a consistent metric. Some stations use a 'weekday' ridership figure while other stations use an 'weekday+weekend average' figure. The ranking of stations by daily ridership is actually as follows (link):

(...)

This is simply total ridership divided by the number of days (244), so stations without weekend service have a big disadvantage.

And that's another crucial point. Some of those station have no service outside of weekday peak. Linearity of comparison is problematic in so many ways. Some posters have made excellent attempts to add meaningful dimension, but someone is going to have to come up with an 'interactive' map that displays the many criteria by clicking on selection button, as per the filters posters have described.

I suspect it's a matter of time until someone does...and should, since it seems it's beyond Metrolinx themselves to do it. Or there might be sinister reasons ML doesn't!

Ideally we would have the ridership by weekday and weekend separately so we could compare average weekday ridership between the stations, but there's no way to get that other than Metrolinx giving it to us.

I think people are being a little overly critical on West Harbour, and even the Kitchener and Guelph stops. It takes time for ridership to build out, new travel patterns to emerge, and so forth.

But that underlays how important it is to be tracking these ridership changes over time.

The ridership figures are also heavily affected by the service actually run to the station. See for example on the Barrie Line how King City station outranks every station without all-day service. The ridership trickling in and out on the all-day every-day service at King City actually outweighs the powerhouse peak-period commuter demand from Barrie, Bradford, East Gwillimbury or Newmarket.

And of course the Niagara stations will be at the bottom, when they are served by less than one train per day on average. In fact, the Niagara Falls station ridership is very impressive when you consider that the average daily ridership over the 244-day period includes 200 days (82%) where the ridership was zero because there was not a single GO train arrival or departure from the station (maths on second page in link).

On top of having few departures per day, the remote stations also have departure times that are totally impractical for their local areas, being timed solely for people who work in an office in Toronto a silly distance away. Kitchener has five departures per day, with the last one leaving at 7:10 AM, before I even wake up in the morning. West Harbour has only two departures per day, with the last one departing at 6:39 AM. And of course the absurdest of all is the new Niagara commuter service, though it's not yet reflected in this 2018 data, with the last GO train of the day departing at a nonsensical 5:19 AM.

If these stations actually offered train service at times practical for closer destinations where people in those cities actually work, of course their ridership would increase drastically. Relatively few people in Guelph or St. Catharines work in downtown Toronto, but many people in those cities work in Kitchener or Hamilton, respectively, and might commute by train if there were a schedule that allowed them to do so.

 

[Streety McCarface]

Then why it still below the 3,050 reported for 2015? The change is so small though, I can't imagine it's statistically significant.

I can't imagine for a minute that Sheppard West ridership really increased by almost a thousand riders from 40,640 to 41,600 at the former terminus when they extended the line. With 20,000 alone riding the 196 York University Rocket that suggests that the remaining ridership doubled! It's quite clear that someone inserted 2017 pre-extension numbers into that 2018 table!

Feel like it's better to move this chat over here, but these numbers are really interesting.

It's impossible for Sheppard west ridership numbers to be included from previous years since the extension opened before 2018. I've set up a spreadsheet that has all data from the past decade together. Unfortunately, UrbanToronto is stupid in that it doesn't let you share excel files, so this PDF will have to do. If anyone wants the full spreadsheet, I'll upload it to google drive or something and share a link if there's interest. Nevertheless, into the basic analysis:

The Good

Ridership across most lines with the exception of 3 are up for this year
Unsurprisingly, Line 1 has seen the most growth with around a 10.3% increase in ridership from fiscal 2016. However, if we are to compare this to the line median, we get somewhere around a 5.89% increase. While this number means nothing because it doesn't necessarily account for the added length of the TYSSE and I'm too lazy to do some integral calculus, I'm just going to assume that said line saw the most growth.
This is followed by line 4, which saw a nearly 5% increase in ridership. This is very similar to line 2's 4.9% increase.
What's hard to believe is that Line 3's ridership decreased despite all stations except Kennedy seeing an increase in ridership. This likely means that there has been an increase (slight to none, cannot make any determinations) in inter-scarborough transportation.

The following stations saw the greatest increase in ridership from 2016 (Not included: TYSSE)
York Mills Station — 36.18%, 7880 passenger increase
Warden Station — 35.53%, 10480 passenger increase
Eglinton West — 32.20%, 5220 passenger increase
Rosedale Station — 29.28%, 1760 passenger increase
Wilson Station — 24.46%, 5750 passenger increase
Sherborne Station — 23.97%, 6000 passenger increase
Museum Station — 22.31$, 2160 passenger increase
Union Station — 21.27%, 25190 passenger increase
Bay Station — 20.67%, 5600 passenger increase
As well, Union Station followed by Kennedy Stations (Line 2) saw the greatest increase in traffic for the year

Line 4 ridership is at its second all time high, Line 2 ridership has recovered and is also at its second all time high. Line 1 is at its all time high (Likely due to TYSSE)

These statistics clearly do not indicate any bias due to the new counting system, meaning these platform counts are not negatively affected by surface route usage.
It also is interesting to note that ridership at most major bus terminals has increased significantly, especially Wilson, York Mills, Eglinton West, Don Mills

The Bad
Stations that saw the greatest ridership decrease from 2016
Dupont Station — (42.45%), -6270 passengers. I have no idea why this station saw such a decrease
Yorkdale Station — (18.15%), -5070 passengers. Likely due to 2016 being an outlier year
Christie Station — (15.13%), -2230 passengers.
Chester Station — (13.30$), -890 passengers
North York Centre — (12.84%), -3740 passengers. Not sure why, likely sampling error
Summerhill Station — (9.08%) -570 passengers.

As you can tell, most of these stations are already the stations with some of the lowest riderships in the system. The fact that they're falling further is telling. We need old town riders.

We now have a new worst! Downsview Park is now the station with the lowest ridership in the system.

The Ugly
Bloor Station's ridership seems to only be sampled on a bi-annual basis. This is bad news for statisticians.

Leslie ridership decreased despite plenty of new buildings going up at CPP in the near vicinity. Are these riders using Oriole instead? Oriole's ridership has increased by 12.5% from the previous sampling period, however, only 405 passengers use the station on the average day (it's actually closer to 565 passengers due to Metrolinx' shitty statistics). Nevertheless, this accounts for about 70 daily rides, or 35 new customers for metrolinx, not nearly enough new riders for 2 new large condos, even on Sheppard. It doesn't even make up for the 105 customers lost from Leslie station.

I have no explanation for Warden or Union's sudden rise in station usage. While GO has increased frequency, actual ridership did not increase by much over the past year, so there's no way to justify a sudden increase of 25K passengers (12.5K daily users).

Any other thoughts?

 

[Midtown Urbanist]

Unfortunately, UrbanToronto is stupid in that it doesn't let you share excel files, so this PDF will have to do.

I also encountered that issue with the image I posted above, I had wanted to attach a downloadable PDF.

I swear you used to be able to upload files as attachments to posts. Part of the reason I had set up this thread was so that information like this can be preserved on the UrbanToronto database, while information published on government webpages may have their urls changed, files deleted, or information intentionally made hidden. Paging @Edward Skira

 

[Midtown Urbanist]

Feel like it's better to move this chat over here, but these numbers are really interesting.

It's impossible for Sheppard west ridership numbers to be included from previous years since the extension opened before 2018. I've set up a spreadsheet that has all data from the past decade together. Unfortunately, UrbanToronto is stupid in that it doesn't let you share excel files, so this PDF will have to do. If anyone wants the full spreadsheet, I'll upload it to google drive or something and share a link if there's interest. Nevertheless, into the basic analysis:

The Good

Ridership across most lines with the exception of 3 are up for this year
Unsurprisingly, Line 1 has seen the most growth with around a 10.3% increase in ridership from fiscal 2016. However, if we are to compare this to the line median, we get somewhere around a 5.89% increase. While this number means nothing because it doesn't necessarily account for the added length of the TYSSE and I'm too lazy to do some integral calculus, I'm just going to assume that said line saw the most growth.
This is followed by line 4, which saw a nearly 5% increase in ridership. This is very similar to line 2's 4.9% increase.
What's hard to believe is that Line 3's ridership decreased despite all stations except Kennedy seeing an increase in ridership. This likely means that there has been an increase (slight to none, cannot make any determinations) in inter-scarborough transportation.

The following stations saw the greatest increase in ridership from 2016 (Not included: TYSSE)
York Mills Station — 36.18%, 7880 passenger increase
Warden Station — 35.53%, 10480 passenger increase
Eglinton West — 32.20%, 5220 passenger increase
Rosedale Station — 29.28%, 1760 passenger increase
Wilson Station — 24.46%, 5750 passenger increase
Sherborne Station — 23.97%, 6000 passenger increase
Museum Station — 22.31$, 2160 passenger increase
Union Station — 21.27%, 25190 passenger increase
Bay Station — 20.67%, 5600 passenger increase
As well, Union Station followed by Kennedy Stations (Line 2) saw the greatest increase in traffic for the year

Line 4 ridership is at its second all time high, Line 2 ridership has recovered and is also at its second all time high. Line 1 is at its all time high (Likely due to TYSSE)

These statistics clearly do not indicate any bias due to the new counting system, meaning these platform counts are not negatively affected by surface route usage.
It also is interesting to note that ridership at most major bus terminals has increased significantly, especially Wilson, York Mills, Eglinton West, Don Mills

The Bad
Stations that saw the greatest ridership decrease from 2016
Dupont Station — (42.45%), -6270 passengers. I have no idea why this station saw such a decrease
Yorkdale Station — (18.15%), -5070 passengers. Likely due to 2016 being an outlier year
Christie Station — (15.13%), -2230 passengers.
Chester Station — (13.30$), -890 passengers
North York Centre — (12.84%), -3740 passengers. Not sure why, likely sampling error
Summerhill Station — (9.08%) -570 passengers.

As you can tell, most of these stations are already the stations with some of the lowest riderships in the system. The fact that they're falling further is telling. We need old town riders.

We now have a new worst! Downsview Park is now the station with the lowest ridership in the system.

The Ugly
Bloor Station's ridership seems to only be sampled on a bi-annual basis. This is bad news for statisticians.

Leslie ridership decreased despite plenty of new buildings going up at CPP in the near vicinity. Are these riders using Oriole instead? Oriole's ridership has increased by 12.5% from the previous sampling period, however, only 405 passengers use the station on the average day (it's actually closer to 565 passengers due to Metrolinx' shitty statistics). Nevertheless, this accounts for about 70 daily rides, or 35 new customers for metrolinx, not nearly enough new riders for 2 new large condos, even on Sheppard. It doesn't even make up for the 105 customers lost from Leslie station.

I have no explanation for Warden or Union's sudden rise in station usage. While GO has increased frequency, actual ridership did not increase by much over the past year, so there's no way to justify a sudden increase of 25K passengers (12.5K daily users).

Any other thoughts?

Thank you for cross-posting and writing up those highlights. I hadn't realized that the 2018 numbers are up!

Here are the 2018 Subway Ridership Numbers:

• Subway Ridership 2018

It appears that there is also new service summaries:

• Service Summary, February 17 to March 30, 2019
• Service Summary, January 6, 2019 to February 16, 2019