• Our new ticketing site is now live! Using either this or the original site (both powered by TrainSplit) helps support the running of the forum with every ticket purchase! Find out more and ask any questions/give us feedback in this thread!

ORR unveils station usage data between April 2022 and March 2023

Status
Not open for further replies.

hwl

Established Member
Joined
5 Feb 2012
Messages
8,289
How does that work when one is changing between, say, the Elizabeth Line and another rail service?

Take Farringdon. That is shown as 31.5 million entries and exits, with just 640,172 interchanges. Whilst it states that the interchanges doesn't include those to the Elizabeth Line from Thameslink and vice versa, does anyone who does interchange there count as an exit and an entry?
See the report comments in on EL station interchange volumes - simple answer ORR state that the interchange numbers are vast underestimates and they are undertaking work to improve the situation.

Station with interchange number underestimates are flagged in columns J & K of the table 1410 spreadsheet.

The ability/need to use or not use (in most cases) the card readers when interchanging will cloud things at Farringdon e.g. GTR Key on TL followed by oyster/ contactless on EL would count as an exit + an entry but not oyster/contactless for both TL and EL.
 
Sponsor Post - registered members do not see these adverts; click here to register, or click here to log in
R

RailUK Forums

BrianW

Established Member
Joined
22 Mar 2017
Messages
2,133
Attached is a box and whisker chart of 22-23 entry and exits as a percentage of 18-19, ie, the last full pre-pandemic year. I've removed new stations or those with data missing for any reason.

It shows a mean average of 77.5% use in 22-23 compared to 18-19. The mean is slightly higher than the median, and this is reflected in that there are more upper outliers than lower ones. The first and third quartiles are 69.1 to 82.4, meaning half the stations fall within those values, a quarter above and a quarter below.

The outliers at the top are mainly the Elizabeth Line and other London stations. I think that some had some re-calculations done, and others in London may be affected by new Elizabeth Line flows, though there are a few at the lower end of the outliers that it's harder to explain eg Metrocentre, Manea (could they reflect when ticket machines were installed?). Pollokshaws West is the largest outlier at the top that I know nothing about (129.6% use compared to 18-19).

Similarly at the bottom end, the outliers are mainly stations that form parts of group where there may have been recalcuations done - Canon Street, Moorgate. I presume that Lake was affected by the various closures on the Isle of Wight line. Does anyone know the story behind Oxford Road in Manchester - is that commuter changes, or something else? Ditto Snow Hill? My hunch is that they may be down to accounting rather than any real change.

The top ten increases are:

Maryland
222.3​
Manor Park
218.7​
Farringdon
208.5​
Abbey Wood
188.9​
Forest Gate
163.8​
Whitechapel
161.7​
London Paddington
155.0​
Southall
141.8​
Queens Park (London)
134.9​
Pollokshaws West
129.6​

The lowest percentage in 22-23 compared to 18-19.

Rhymney
51.3​
Rauceby
51.3​
Balham
51.3​
Old Hill
50.7​
Birmingham Snow Hill
50.7​
Bexleyheath
50.1​
Manchester Oxford Road
50.1​
Lake
48.6​
Moorgate
48.6​
London Cannon Street
32.6​
This is an excellent and helpful analysis and attempt to understand and explain what the stats mean. Thank you Cuccir.
I have the same puzzlement with Liverpool Street - up from 30m to 80m ?.
This is a very big increase to get just from one new line over the existing six tracks into Liverpool Street.

So what were Paddington and Liverpool Street like just before Covid - is it mostly down to a post Covid recovery ?.
If anything I would expect the Elizabeth line to have reduced entries and Exits at Paddington and Liverpool Street with people simply passing through.

My guess with Victoria is that, post Covid, I have come to see Thameslink and the Elizabeth Line via Farringdon as my chief way of entering/exiting London and beyond. That is from Croydon and so Victoria has fallen of my choice with only Clapham Junction featuring in that direction.

Right, so not keen on stats then.

Won't be perfect and always will have loose ends BUT what alternative is there for those who need to know what is being used and by how much ?.
It does serve to show up bad methods of estimating usage at new stations or provoke question as to why there were errors (up or down).

These stats also help to see any recovery from Covid. To be frank right now I think DfT are probably very keen to save money spent on a declining railway. They will have seen costs go up and revenues come down. Without stats showing any sign of recovery the Dft would be very keen to simply look at service reductions and closures.

Has Corby been used more by people wanting nearby destinations on the Midland Mainline while their stations and line has been shut for electrification ?.

I hope it is caused by the improved service and it then goes to show what the latent demand was.
Actually I'm quite keen on stats Peter, though skepical, about their cynical abuse to support this or that 'conclusion'. This thread shows that many Forumites, like you, seek to understand what they say, and what if any decisions might be made as a result. I note and sympathise with, your puzzlements, guesses, surmises and hopes.
 

Horizon22

Established Member
Associate Staff
Jobs & Careers
Joined
8 Sep 2019
Messages
11,028
Location
London
Lies, damned lies and stats ... comparing apples and oranges. What is more useful is information; eg WHY has usage at X gone up or down by this or that %? As identified upthread the increase in WFH and reduction in Monday-Friday commuting is likely a strong factor (in some places, eg 'Stockbroker belt to Waterloo). Long Covid effects, strikes and their chilling effect, short-forming, crush-loading, surly service, new routes eg facilitated by Lizzie Line, very multifactorial; thus hard to relate effect / cause.

Anticipated? Quoted? 'Fiddled' Inflated to improve 'business case'? Politics? Covid?

Will it ever be possible to attain a 'steady state'/ 'normal' baseline?

What value are these figures to planners- some I hope or why bother- not good enough on their own to justify the expenditure. The 'so what' question?

At some point you’d have to include the numbers though - most (not all as there is induced ‘new’ demand) of the increase is passengers who have moved from one dataset not included in this data (LU passengers) to this dataset (rail passengers). You can’t just pretend they don’t exist! Also note the ORR are aware their figures in C.London are definitely not as reliable as TfL’s so there’s going to be some recalibration going forward.

Such dramatic changes are common in data & statistics when a new input is added or a factor changes and it just needs to be noted. It won’t be so dramatic next year.
 
Last edited:

northwichcat

Veteran Member
Joined
23 Jan 2009
Messages
32,692
Location
Northwich
TfL get far more accurate passenger flow data than the mainline as their station wifi tracks phones and devices (even if not connected) to establish passenger flows. It was a known issue before the Elizabeth Line that there are multiple routes between the same tube stations.

They can't track phones that have WiFi turned off or are turned off altogether. I don't leave everything running when it's not in use e.g. WiFi, mobile data, Bluetooth, NFC, hotpsot etc. as it means the battery lasts much longer. In fact my phone has a setting that automatically turns off some of those if they aren't in use to preserve battery life. I'm also skeptical about what you say. While a lot of people accept T&Cs that permit data sharing without reading them, TfL can't just track people without consent - that would be illegal. And if they can legally track say 40% of passengers, then they're making presumptions about the other 60% based on having data for 4 out of 10 passengers. That may work for the example I gave when I was making the exact same journey as work colleagues, but wouldn't work for other journeys.
 

Acfb

Member
Joined
12 Aug 2018
Messages
616
No, initial proposals tend to underestimate usage - see Okehampton, Borders, Ebbw Vale etc.

Borders is quite interesting as I would argue that the three Borders Stations are the 'success' which is what I expected given the good existing bus service in Midlothian (contrary to @tbtc) (with the village of Stow getting 2tph post COVID as well while the bus service has been cut to 1p2h) with the Midlothian stations underperforming despite the amount of new housing.That said I'm fairly surprised Eskbank hasn't recovered closer to precovid levels even taking into account the dense bus service (as it was heaving with people going to Edinburgh on a Saturday in August last year when I used it go to Tweedbank).
 
Last edited:

Peter0124

Established Member
Joined
20 Nov 2016
Messages
2,753
Location
Glasgow
It looks like Edinburgh Waverley (18.2m) is closing the gap with Glasgow Central (20.8m), atleast compared to pre-covid. Its a lot tighter now than it used to be.

Could we see Waverley overtake Central in years to come, and become Scotland's busiest?

I also wonder if the increase at St Pancras has anything to do with interchanges from TL to the Elizabeth at Farringdon
 

MikeWM

Established Member
Joined
26 Mar 2010
Messages
4,963
Location
Ely
Cambridge might be the nearest big city but Ely is a decent size place, in terms of shopping and entertainment.

Shopping yes (and also the staff for the shops), but I'm less sure about entertainment :) There are a few decent pubs, but otherwise there's only really the Leisure Park, and that's far enough away from the station and with no particular bus link, that I'd expect that very few will travel to it by public transport.

Would there be school/college traffic?

I wouldn't have thought much. Soham has its own secondary school, but nowhere to take A-levels so some may travel to Ely for that (again though, Ely College is quite a long way from the station).

If Soham is being used just to get to Ely then sounds like they have used a load of cash and capacity to replace the bus....

I suspect there is some truth in that, but given the future of the bus was seriously in question very recently, the people in Soham may be happy at having a rather more permanent option now.
 

dk1

Veteran Member
Joined
2 Oct 2009
Messages
19,887
Location
East Anglia
I wouldn't have thought much. Soham has its own secondary school, but nowhere to take A-levels so some may travel to Ely for that (again though, Ely College is quite a long way from the station).
Lots of schoolchildren travel to Ely from Cambridge for education. I think it must be a very well respected school.
 

Bartsimho

Member
Joined
17 Jan 2023
Messages
623
Location
Chesterfield
Have they changed some of the reporting areas? Looking on the Power BI visualisation: https://app.powerbi.com/view?r=eyJr...J9&pageName=ReportSection196136ade60361f5e617
I don't remember Kings Sutton being classed as the East Mids even though it technically is because of it being in Northants. Although I might just have been ignoring it before with it being so out of the way.

Interestingly Derby's Origin/Destination station is New Street rather than St Pancras like Nottingham and Leicester. Suggests that despite XC being short formed people really want to make that journey. Although I can sort of see 4tph between the pair compared to 2tph to London. Do we just have the Main Origin/Destinations or a full list for stations (I'm guessing not).

For untapped demand surely Mansfield as it in spades with only 302,000+111,000 entries and exits compared to it's population of 108,000. This will be driven by the poor service along the line and the lack of destinations. The main issue being the Single Track section although it could be worked around with Pye Bridge. I could also suggest a lessening of services at Newstead (30,000) to facilitate quicker Nottingham trains (also cite that Hucknall and Bulwell have the trams)
 

hwl

Established Member
Joined
5 Feb 2012
Messages
8,289
They can't track phones that have WiFi turned off or are turned off altogether. I don't leave everything running when it's not in use e.g. WiFi, mobile data, Bluetooth, NFC, hotpsot etc. as it means the battery lasts much longer. In fact my phone has a setting that automatically turns off some of those if they aren't in use to preserve battery life. I'm also skeptical about what you say. While a lot of people accept T&Cs that permit data sharing without reading them, TfL can't just track people without consent - that would be illegal. And if they can legally track say 40% of passengers, then they're making presumptions about the other 60% based on having data for 4 out of 10 passengers. That may work for the example I gave when I was making the exact same journey as work colleagues, but wouldn't work for other journeys.
 

yorksrob

Veteran Member
Joined
6 Aug 2009
Messages
44,414
Location
Yorks
Borders is quite interesting as I would argue that the three Borders Stations are the 'success' which is what I expected given the good existing bus service in Midlothian (contrary to @tbtc) (with the village of Stow getting 2tph post COVID as well while the bus service has been cut to 1p2h) with the Midlothian stations underperforming despite the amount of new housing.That said I'm fairly surprised Eskbank hasn't recovered closer to precovid levels even taking into account the dense bus service (as it was heaving with people going to Edinburgh on a Saturday in August last year when I used it go to Tweedbank).

Yes, it's interesting that the outer stations on the line have been exceeding expectations, with the ones nearer to Edinburgh not so much.

It does illustrate the vital role that rail plays in connecting more rural areas (in spite of that being talked down by many on here).
 

MikeWM

Established Member
Joined
26 Mar 2010
Messages
4,963
Location
Ely
Lots of schoolchildren travel to Ely from Cambridge for education. I think it must be a very well respected school.

There's King's Ely, and that is indeed near the station, but that's a very expensive private school. Maybe there's some traffic from Soham to there, but I'm not convinced it is going to be much, I'd imagine most of it is from very rich people that are more likely to be living in somewhere like central Cambridge.

But I was referring to Ely College, which seems to have improved but wasn't terribly good not so long ago. Can't imagine too many people from Cambridge going there when there are a number of similar places (with a better reputation) in Cambridge itself.
 

Bartsimho

Member
Joined
17 Jan 2023
Messages
623
Location
Chesterfield
1 train a day each direction and none on the weekends.
Even worse than I thought. That puts the passengers per train at about half a person but the actual service has been worse with strikes. It really is a case of running it into the ground and than using the lack of people to justify closure (suspension of service most likely) it seems. Northern have done an incredible job at killing the station
 

dk1

Veteran Member
Joined
2 Oct 2009
Messages
19,887
Location
East Anglia
There's King's Ely, and that is indeed near the station, but that's a very expensive private school. Maybe there's some traffic from Soham to there, but I'm not convinced it is going to be much, I'd imagine most of it is from very rich people that are more likely to be living in somewhere like central Cambridge.

But I was referring to Ely College, which seems to have improved but wasn't terribly good not so long ago. Can't imagine too many people from Cambridge going there when there are a number of similar places (with a better reputation) in Cambridge itself.

Ahh fair enough. Wasn’t sure name of school or anything just notice Ely station gets mobbed mid-afternoons.
 

Peterthegreat

Established Member
Joined
22 Feb 2021
Messages
1,802
Location
South Yorkshire
Even worse than I thought. That puts the passengers per train at about half a person but the actual service has been worse with strikes. It really is a case of running it into the ground and than using the lack of people to justify closure (suspension of service most likely) it seems. Northern have done an incredible job at killing the station
For the period in question the service was only on Saturdays and I believe was suspended for most of it.
 

MikeWM

Established Member
Joined
26 Mar 2010
Messages
4,963
Location
Ely
Interesting to look at Hatton and Lapworth - they certainly punch above their weight when it comes to season ticket entries and exits, driven by the annual season ticket being the cheapest Gold Card.

Hatton has 78,188 entries and exits with 57,508 of these being season ticket entries/exits
Lapworth has 77,008 extries and exits with 60,528 of these being season ticket entries/exits

The FAQs tell us that an annual season ticket assume 480 single journeys and we do need to realise that there might be some genuine season ticket holders at these stations we can take a guess at the number of people who buy a Hatton to Lapworth purely for the Gold Card benefits. Suspect it's around 100.

Somewhat interesting too to look at the Lichfield Stations:

Lichfield TV 688,278 with 205,228 season tickets.
Lichfield City 633,292 with 164,576 season tickets.

I'm not sure we can deduce in this case how many people have a corresponding season for the Gold Card (though I know it is at least 1 ;)
 

bib

Member
Joined
15 Nov 2021
Messages
249
Location
East Midlands
Stansted-Liverpool Street winning the prize for busiest airport route, accounting for about 60% of journeys to stansted airport station. 7.9m train entries and exits at the station compared to about 23m flight passengers in 2022. I wonder which airport has the highest access by public transport?
 

BrianW

Established Member
Joined
22 Mar 2017
Messages
2,133
At some point you’d have to include the numbers though - most (not all as there is induced ‘new’ demand) of the increase is passengers who have moved from one dataset not included in this data (LU passengers) to this dataset (rail passengers). You can’t just pretend they don’t exist! Also note the ORR are aware their figures in C.London are definitely not as reliable as TfL’s so there’s going to be some recalibration going forward.

Such dramatic changes are common in data & statistics when a new input is added or a factor changes and it just needs to be noted. It won’t be so dramatic next year.
My bold- You're right- that's to say I agree with you- it needs to be taken into account. Time-series stats suffer 'step changes' with changes in service pattern and/or recalibration. E.g do Farringdon connections warrant redesign; should Brigg be closed, or promoted; Gold card- like schemes be spread more widely; car parks (and/or charges), air travel be increased (or abolished?). More 'so what'/'what if' planners and more option analysis needed? I've had my say; I'm 'with you'. Thanks.
 

Peter Sarf

Established Member
Joined
12 Oct 2010
Messages
9,754
Location
Croydon
May have something to do with the EL offering faster times towards the west via Paddington.

You'll notice Waterloo brimming whenever Paddington has serious delays towards reading and Co like the incident on Thursday of last week.

Only way to explain how Paddington has risen so much as it used to be a nightmare to get to from South east and Co having to use the Southeastern, Jubliee and Bakerloo to just get into Paddington alone.
Of course. Waterloo is on the Bakerloo line so avoids a change to another tube if wanting Paddington.

For me Croydon - Victoria - Oxford Circus - Paddington was my standard route.
Thinking about it now Croydon - Clapham Junction - Waterloo - Paddington would have been nicer as only one tube.
Now blown out of the water by Thameslik - Farringdon then Elizabeth line for so many opportunities.

But why Paddington and Liverpool Street have gained I do not know ?. I would have expected their figures to have moved to the EL core. I wonder if Liverpool Street has gained at the expense of Stratford. Changing off a GA straight on to the Elizabeth line misses out the Central Line !.
 

bluenoxid

Established Member
Joined
9 Feb 2008
Messages
2,663
I wonder which airport has the highest access by public transport?
There are statistics published for this. I am sure that I’ve seen all airports in the past but the focus of those that I have seen on Google is the top 5. Stansted appears to be the best. Manchester appears to be the worst of the top 5 (hands up, I don’t help these stats). Heathrow has fewer private car journeys but taxis shift a larger percentage instead.
 
Last edited:

norbitonflyer

Established Member
Joined
24 Mar 2020
Messages
5,814
Location
SW London
But why Paddington and Liverpool Street have gained I do not know ?. I would have expected their figures to have moved to the EL core.
Paddington and Liverpool Street figures include the Elizabeth line. I would expect the number no longer recorded (because e.g an Ealing- Bond Street journey can now be made without changing) are far outnumbered by people chaging to the EL there (e.g a Newbury - Bond Street) now being recorded twice - beacuse the system reads the paper Newbury-Paddington ticket and Oyster Paddington - Bond Street journeys as separate.
 

Horizon22

Established Member
Associate Staff
Jobs & Careers
Joined
8 Sep 2019
Messages
11,028
Location
London
But why Paddington and Liverpool Street have gained I do not know ?. I would have expected their figures to have moved to the EL core. I wonder if Liverpool Street has gained at the expense of Stratford. Changing off a GA straight on to the Elizabeth line misses out the Central Line !.

But you could (and people do) that at Stratford with no long walk at Liverpool Street. So that is not an explanation.

I reckon as others have suggested there's some data anomalies that haven't been fully realised when people have to tap in/out at the two different Paddington/Liverpool Street gatelines for half of this period. But they have likely just outright grown too as has the rest of the Elizabeth line network.
 

357

Established Member
Joined
12 Nov 2018
Messages
2,284
They can't track phones that have WiFi turned off or are turned off altogether. I don't leave everything running when it's not in use e.g. WiFi, mobile data, Bluetooth, NFC, hotpsot etc. as it means the battery lasts much longer. In fact my phone has a setting that automatically turns off some of those if they aren't in use to preserve battery life. I'm also skeptical about what you say. While a lot of people accept T&Cs that permit data sharing without reading them, TfL can't just track people without consent - that would be illegal. And if they can legally track say 40% of passengers, then they're making presumptions about the other 60% based on having data for 4 out of 10 passengers. That may work for the example I gave when I was making the exact same journey as work colleagues, but wouldn't work for other journeys.
What law do you say TfL is breaking?

Signs have been put up at stations explaining the Wi-Fi data collection that is taking place and how to opt-out.

In summary, we are collecting the location of devices on our network, identifying them by a pseudonymised version of their MAC address.

We are trying to understand patterns of movement across our network and how customers as a whole use it, not how specific individuals use it.

We are not collecting data on browsing activity, cookies, phone numbers or whether the Wi-Fi service is used.

However, if you would like to opt-out, you can do this by turning off Wi-Fi on your device, turning your device off or putting your device into airplane mode while at our stations.

If the device finds a Wi-Fi network that is known to the device, it will automatically connect to that network. If the device finds unknown networks, it will list these in your device settings so you can decide whether to connect to one of them.

When you are near one of our station Wi-Fi access points (installed on TfL privately-owned property) and you have Wi-Fi enabled, your device will send a probing request to connect. This will be received by our Wi-Fi network, even if your device does not subsequently connect.

This data collection is carried out independently by TfL, not via Virgin Media.

We will not be able to identify any individuals from the data collected. We have designed the process to identify patterns and to avoid identifying individuals. We are trying to understand how customers as a whole use the network, not how specific individuals use it.

All data collected is automatically depersonalised, using a one-way pseudonymisation process to ensure TfL is unable to identify any individual. This happens immediately after the data is first collected.

TfL has no plans to match the Wi-Fi connection data to any other data held about individuals (eg Oyster and Contactless data). There is no way to systematically do this if we wanted to.

If the device has been signed up for free Wi-Fi on the London Underground network and the Elizabeth line stations noted above, the device will disclose its genuine MAC address. This is known as an authenticated device.

The following processing only relates to authenticated devices.

We process authenticated device MAC address connections (along with the date and time the device authenticated with the Wi-Fi network and the location of each router the device connected to). This helps us to better understand how customers move through and between stations - we look at how long it took for a device to travel between stations, the routes the device took and waiting times at busy periods.

As a customer enters a station (if their device has Wi-Fi enabled), the device will attempt to connect to the Wi-Fi network. This is recorded in our Wi-Fi data control system. We carry out a process of hashing once a MAC address is collected. Hashing is the process of generating a new value from a string of text (in this instance the MAC address).

We carry out two rounds of hashing to make sure we never hold the original MAC address. One hash is with a single value (pepper), and the second is with another unique random value (salt) for each MAC address. This pseudonymisation process provides continuity to the data without needing to record the MAC address which could identify an individual device. This means that we have to preserve our hashing key to maintain continuity of the pseudonymised data.

We do not collect any other data generated by your device. This includes web browsing data and data from website cookies.


Further interesting information: https://www.ianvisits.co.uk/article...helping-tfl-update-its-journey-planner-34051/
 
Status
Not open for further replies.

Top