ORR unveils station usage data between April 2022 and March 2023

[Whiggism]

Elizabeth Line

My bad, I didn’t know Elizabeth Line was included I thought it was strictly National Rail Entries and Exits

 

[DynamicSpirit]

How are those entries and exits worked out and apportioned between national rail and London Underground? Are they intended to be national-rail-only or national-rail+Underground figures? You get to the Elizabeth line through the same gatelines as the Underground which must make it difficult to distinguish now. Plus you must get a fair few people arriving at the NR terminus and promptly heading for the Underground: Those should presumably be excluded from entry/exit figures, but it's not clear how you can accurately do that.

(EDIT: And just to make it even more complicated: Some people will be entering/exiting Liverpool Street Elizabeth line via gatelines at Moorgate)

 

[Nicholas Lewis]

Seems odd to me given a big chunk of Lizzie line traffic was originally routed through Liv St

 

[yorkie]

My bad, I didn’t know Elizabeth Line was included I thought it was strictly National Rail Entries and Exits

An interchange from/to London Underground would still count as a National Rail entry/exit.

How are those entries and exits worked out and apportioned between national rail and London Underground? Are they intended to be national-rail-only or national-rail+Underground figures? You get to the Elizabeth line through the same gatelines as the Underground which must make it difficult to distinguish now. Plus you must get a fair few people arriving at the NR terminus and promptly heading for the Underground: Those should presumably be excluded from entry/exit figures, but it's not clear how you can accurately do that.

Indeed; it must be based on estimates and modelling.

 

[Failed Unit]

My bad, I didn’t know Elizabeth Line was included I thought it was strictly National Rail Entries and Exi5

You can’t tell where the passengers came from, with respect did they get on a Shenfield (which was traditionally Liverpool Street) or Abbey Wood which is new.

Few interesting questions on the numbers, if you leave at the Moorgate end if that still counted as Liverpool Street (as many people coming will use that exit, that have arrived at Moorgate)

Passengers from Reading that used to tube to Liverpool Street, will they now be counted as Liverpool St rather than Paddington.

OF course many new journeys as well.

 

[Richardr]

Isn't an issue that they don't have any numbers for interchanges involving the Elizabeith Line? I would imagine that a reasonably large proportion of people getting on and off the Elizabeth Line are changing from a different National Rail line, so aren't really exiting rail at Liverpool Street but interchanging there, and they count as 2 in the morning and 2 in the evening, double what the same passengers on a similar journey used to account for.

 

[Horizon22]

An interchange from/to London Underground would still count as a National Rail entry/exit.

Although the Elizabeth line is a National Rail product (for the purposes of this data at least). This presents some quirks when you are changing onto a National Rail service such as Paddington/Farringdon/Liverpool Street where there might be another tap in/out where the station isn't truly an origin/destination but instead interchange.


== Doublepost prevention - post automatically merged: 21 Dec 2023 ==

How are those entries and exits worked out and apportioned between national rail and London Underground? Are they intended to be national-rail-only or national-rail+Underground figures? You get to the Elizabeth line through the same gatelines as the Underground which must make it difficult to distinguish now. Plus you must get a fair few people arriving at the NR terminus and promptly heading for the Underground: Those should presumably be excluded from entry/exit figures, but it's not clear how you can accurately do that.

(EDIT: And just to make it even more complicated: Some people will be entering/exiting Liverpool Street Elizabeth line via gatelines at Moorgate)

It will be interesting to see if it standardises next year; for part of this period you had to use two gatelines at both Paddington & Liverpool Street and now (for the vast majority of Elizabeth line services) you only use one.

Although the statistics said this was taken into account, I'm not sure that such a huge leap can be attributed to just passenger numbers rising +250% and I'm sure there's some double counting. Especially when Paddington & Liverpool Street rose so much.

 

[DynamicSpirit]

Isn't an issue that they don't have any numbers for interchanges involving the Elizabeith Line? I would imagine that a reasonably large proportion of people getting on and off the Elizabeth Line are changing from a different National Rail line, so aren't really exiting rail at Liverpool Street but interchanging there, and they count as 2 in the morning and 2 in the evening, double what the same passengers on a similar journey used to account for.

Yes quite a few will be. Though I wonder what the stats are on people coming from the Chelmsford/Billericay/etc. direction changing at Liverpool Street vs. Stratford? Stratford gives a much shorter walk but also less frequent Elizabeth line trains. And for anyone wanting West End destinations served equally by Elizabeth or Central line, Stratford allows delaying the choice of which line to use until you're at the platform and can see how long you'd have to wait.

On the other hand, I guess on the return journey there's a greater incentive to change at Liverpool Street because you're more likely to get a seat on the NR train.

 

[Stephen42]

Isn't an issue that they don't have any numbers for interchanges involving the Elizabeith Line? I would imagine that a reasonably large proportion of people getting on and off the Elizabeth Line are changing from a different National Rail line, so aren't really exiting rail at Liverpool Street but interchanging there, and they count as 2 in the morning and 2 in the evening, double what the same passengers on a similar journey used to account for.

The interchanges are calculated at the final stage after the Origin and Destinations for entries/exits have been worked out with journey planning used to infer the proportion of people changing where. It's fully possible the lack of Elizabeth line journey patterns in some of the earlier stages may have led to artificial splits being created. The methodology documents don't really give enough info in the adjustments to be able to tell. Once the full Origin Destination Matrix is published it should be easier to see what's been done.

 

[NSEWonderer]

Seems to be double counting in some areas as someone mentioned above it’s quite a drastic change even with Elizabeth Line cos running through now.

The LU services are accessed from the same gates the Elizabeth line is so i feel they‘re counting passengers from the central, circle, hammersmith and co as one.

 

[Richardr]

The interchanges are calculated at the final stage after the Origin and Destinations for entries/exits have been worked out with journey planning used to infer the proportion of people changing where. It's fully possible the lack of Elizabeth line journey patterns in some of the earlier stages may have led to artificial splits being created. The methodology documents don't really give enough info in the adjustments to be able to tell. Once the full Origin Destination Matrix is published it should be easier to see what's been done.

There are no adjustments. The data states that (for Liverpool Street and other stations affected)

Interchanges figure a known underestimate. It does not include interchanges to or from Elizabeth line services

I take it from that that there is no estimate from the Elizabeth Line data, and so every interchange involving the Elizabeth Line is double counted.

 

[norbitonflyer]

My bad, I didn’t know Elizabeth Line was included I thought it was strictly National Rail Entries and Exits

The Elizabeth Line, like the Overground, is part of the National Rail system

 

[Adrian1980uk]

Also don't underestimate the overground and GA influence on this as we know overground has recovered well and so has GA

 

[Magdalia]

The Office for Rail and Road News Release provides partial information on how the 80 million exit and entries in 2023 for Liverpool Street is made up.

https://dataportal.orr.gov.uk/media/axnd1tyj/station-usage-2022-23-statistical-release.pdf

Estimates of station usage April 2022 to March 2023 14 December 2023

The most used station in the latest year was London Liverpool Street with an estimated 80.4 million entries and exits. It was the fourth most used station in the previous year (April 2021 to March 2022), but the opening of the central section of the Elizabeth line has seen a large increase its in entries and exits

Figure 3.1 shows that the top three flows between stations all involve Liverpool Street.

Liverpool Street-Tottenham Court Road 5.8m
Liverpool Street-Stratford 5.4m
Liverpool Street-Stansted Airport 5.1m

Table 2.2 analyses new stations and Liverpool Street appears 5 times in the table of busiest flows to/from the new stations

Liverpool Street-Tottenham Court Road 5.8m
Liverpool Street-Bond Street 3.1m
Liverpool Street-Canary Wharf 1.5m
Liverpool Street-Woolwich 1.3m
Liverpool Street-Custom House 0.9m

Note that Bond Street only opened in October 2022.

I can't find figures for Liverpool Street-Abbey Wood or Liverpool Street-Whitechapel.

Figure 3.2 shows flows between stations where at least one is outside London. Obviously Liverpool Street-Stansted Airport is top but Liverpool Street-Chelmsford at 2.6m also comes in the top 10.

From this information we can make some rough estimates of the impact of some of the changes:

• the Tottenham Court Road and Bond Street figures suggest that opening the central section of the Elizabeth Line adds about 10m new journeys
• the Canary Wharf, Woolwich and Custom House figures suggest that opening to Abbey Wood adds about 5m new journeys
• taking these together, and adding in the Shenfield line stations, will show that the Elizabeth Line is now a very big proportion of all exits and entrances at Liverpool Street
• longer distance travel makes a significant contribution with Stansted Airport and Chelmsford together about 10% of all exits and entrances

 

[etr221]

Looking at the data, the most common origin/destination for journeys to/from Liverpool Street is Tottenham Court Road. The same for Paddington...

 

[JW4]

Birmingham New Street had 30,726,280 journeys
Here’s what I found for the number of journeys to/from stations where New Street was the most popular destination (or for Euston New Street’s most popular destination):
London Euston - 2,530,404
Coventry - 2,146,702
Wolverhampton - 1,718,454
Selly Oak - 1,489,510
Birmingham International - 1,271,054
University - 1,018,966
Five Ways - 616,256
Walsall - 587,768
Derby - 434,080
Bournville - 430,242
Sutton Coldfield - 428,826
Kings Norton - 418,992
Tame Bridge Parkway - 389,708
Redditch - 386,662
Longbridge - 368,528
Marston Green - 366,136
Sandwell & Dudley - 365,656
Aston - 351,274
Tamworth - 340,184
Chester Road - 332,774
Four Oaks - 330,560
Smethwick Rolfe Street - 314,816
Northfield - 308,858
Duddeston - 308,734
Smethwick Galton Bridge - 299,952
Lichfield City - 295,190
Bromsgrove - 290,666
Coseley - 284,348
Telford Central - 257,386
Lea Hall - 250,092
Wylde Green - 240,160
Tile Hill - 239,082
Erdington - 227,138
Burton-on-Trent - 222,704
Shrewsbury - 219,668
Nuneaton - 219,316
Gravelly Hill - 215,926
Stechford - 211,360
Dudley Port - 184,170
Worcester Foregate Street - 172,918
Tipton - 162,814
Canley - 156,466
Blake Street - 146,912
Barnt Green - 138,670
Cannock - 136,764
Hednesford - 125,244
Berkswell - 119,180
Hamstead - 111,722
Colehsill Parkway - 110,340
Adderley Park - 96,564
Witton - 94,000
Worcestershire Parkway - 92,476
Bescot Stadium - 91,358
Hinckley - 83,834
Landywood - 81,850
Penkridge - 78,772
Butlers Lane - 71,786
Hampton-in-Arden - 70,716
Rugeley Town - 69,134
Alvechurch - 66,950
Wilnecote - 58,818
Bloxwich North - 56,098
Perry Barr - 52,066
Bloxwich - 46,898
Codsall - 38,376
Shifnal - 33,074
Stratford-upon-Avon Parkway - 26,146
Water Orton - 20,498
Albrighton - 18,916
Cosford - 12,086
Willington - 9,314
Bordesley - 962

 

[Llandudno]

Birmingham New Street - Derby/Burton/Tamworth over 1million in total, most of the passengers suffering overcrowding on hopelessly inadequate Cross Country service and capacity provision.

 

[The Ham]

Birmingham New Street - Derby/Burton/Tamworth over 1million in total, most of the passengers suffering overcrowding on hopelessly inadequate Cross Country service and capacity provision.

Assuming 360 days, 15 services a day, in each direction that's about 100 people per service, given there's only 200 seats in a 4 coach 22x that's about half the train.

Even if it's 2tph (30 trains a day) that's still 50 people.

 

[Llandudno]

Assuming 360 days, 15 services a day, in each direction that's about 100 people per service, given there's only 200 seats in a 4 coach 22x that's about half the train.

Even if it's 2tph (30 trains a day) that's still 50 people.

Indeed, if only the loadings were spread evenly across the 18 hour operating day!

 

[bspahh]

Just for interest copying this here, it follows a discussion on Portsmouth-Cardiff line and is extract of ORR station data, which shows how growth is uneven, this is in the Portsmouth-Cardiff thread

Picked 2000-01 as round number about 20 years ago
Picked 2018-19 as last full year before any covid effects

Rank (national, station name, 2023-23 entance & exits
2000-01, 2018-19, % latest vs 2000-01, % latest vs 2018-19

37 Cardiff Central, 10,185,022, 6,707,944, 12,934,304, 151.8%, 78.7%
40 Bristol Temple Meads, 9,291,680, 4,184,675, 11,367,652, 222.0%, 81.7%
72 Southampton Central, 5,495,672, 4,072,160, 6,664,714, 135.0%, 82.5%
73 Bath Spa, 5,468,466, 3,009,640, 6,538,056, 181.7%, 83.6%
203, Newport (South Wales), 2,339,566, 1,466,693, 2,745,064, 159.5%, 85.2%
273 Fratton, 1,775,756, 994,274, 1,735,300, 178.6%, 102.3%
278 Portsmouth Harbour, 1,746,574, 499,012, 2,100,528, 350.0%, 83.1%
281 Portsmouth and Southsea, 1,707,210, 2,542,049, 2,053,186, 67.2%, 83.1%
287 Salisbury, 1,621,562, 1,317,775, 1,979,880, 123.1%, 81.9%,
323 Fareham, 1,415,156 , 1,103,362, 1,701,386, 128.3%, 83.2%
568 Cosham, 765,158, 588,460, 925,066, 130.0%, 82.7%
586 Trowbridge, 734,768, 404,998, 933,894, 181.4%, 78.7%
752 Filton Abbey Wood, 521,474, 336,183, 901,872, 155.1%, 57.8%
757 Westbury, 518,996, 231,093, 548,720, 224.6%, 94.6%
842 Bradford-On-Avon, 426,700, 211,968, 534,086, 201.3%, 79.9%
848 Keynsham, 418,586, 118,047, 511,642, 354.6%, 81.8%
886 Romsey, 384,366, 286,468, 520,856, 134.2%, 73.8%
1020 Oldfield Park, 297,810, 127,783, 332,654, 233.1%, 89.5%
1026 Warminster, 295,452, 247,665, 348,658, 119.3%, 84.7%
1360 Severn Tunnel Junction, 165,992, 90,946, 229,362, 182.5%, 72.4%
1751 Patchway, 77,776, 11,035, 104,078, 704.8%, 74.7%
1990 Freshford, 44,094, 19,549, 53,368, 225.6%, 82.6%
2179 Avoncliff, 23,402, 8,866, 24,396, 264.0%, 95.9%
2300 Dilton Marsh, 12,076, 10,119, 17,004, 119.3%, 71.0%

Average 2018-19 was 160% of 2000-01
Average 2022-23 is 82% of 2018-19

So although at 82% of pre pandemic, still 60% above 22 years ago

This is the table with some text formatting:

Rank Station Name                    entances & exits             % change
                               2023-23    2000-01     2018-19 23vs2000 23vs2018
  37 Cardiff Central        10,185,022  6,707,944  12,934,304  151.8%  78.7%
  40 Bristol Temple Meads    9,291,680  4,184,675  11,367,652  222.0%  81.7%
  72 Southampton Central     5,495,672  4,072,160   6,664,714  135.0%  82.5%
  73 Bath Spa                5,468,466  3,009,640   6,538,056  181.7%  83.6%
 203 Newport (South Wales)   2,339,566  1,466,693   2,745,064  159.5%  85.2%
 273 Fratton                 1,775,756    994,274   1,735,300  178.6% 102.3%
 278 Portsmouth Harbour      1,746,574    499,012   2,100,528  350.0%  83.1%
 281 Portsmouth and Southsea 1,707,210  2,542,049   2,053,186   67.2%  83.1%
 287 Salisbury               1,621,562  1,317,775   1,979,880  123.1%  81.9%
 323 Fareham                 1,415,156  1,103,362   1,701,386  128.3%  83.2%
 568 Cosham                    765,158    588,460     925,066  130.0%  82.7%
 586 Trowbridge                734,768    404,998     933,894  181.4%  78.7%
 752 Filton Abbey Wood         521,474    336,183     901,872  155.1%  57.8%
 757 Westbury                  518,996    231,093     548,720  224.6%  94.6%
 842 Bradford-On-Avon          426,700    211,968     534,086  201.3%  79.9%
 848 Keynsham                  418,586    118,047     511,642  354.6%  81.8%
 886 Romsey                    384,366    286,468     520,856  134.2%  73.8%
1020 Oldfield Park             297,810    127,783     332,654  233.1%  89.5%
1026 Warminster                295,452    247,665     348,658  119.3%  84.7%
1360 Severn Tunnel Junction    165,992     90,946     229,362  182.5%  72.4%
1751 Patchway                   77,776     11,035     104,078  704.8%  74.7%
1990 Freshford                  44,094     19,549      53,368  225.6%  82.6%
2179 Avoncliff                  23,402      8,866      24,396  264.0%  95.9%
2300 Dilton Marsh               12,076     10,119      17,004  119.3%  71.0%

 

[DynamicSpirit]

Looking at the data, the most common origin/destination for journeys to/from Liverpool Street is Tottenham Court Road. The same for Paddington...

I don't for a second believe that 5.8m/80.4m = a full 7% of journeys starting/ending at Liverpool Street were journeys exclusively between Liverpool Street and Tottenham Court Road (even allowing for that that may include people changing at TCR for the Northern line, which presumably means they are seen as exiting the national rail network at that point).

On the other hand I can well believe that nearly 5.8m passengers arrived at Liverpool Street on other services and transferred to the Elizabeth line to get to Tottenham Court Road.

I'm therefore inclined to suspect that the figures attribute to entry/exit at Liverpool Street a lot of journeys that should actually be attributed to interchange.

 

[HamworthyGoods]

My bad, I didn’t know Elizabeth Line was included I thought it was strictly National Rail Entries and Exits

Elizabeth Line is National Rail!

 

[The Ham]

Indeed, if only the loadings were spread evenly across the 18 hour operating day!

Indeed, it would be interesting to know the data for 2019/20 and therefore why there was no justification for more capacity for XC across their core given just how many more people would have been using these services.

Especially given once you reach certain passenger loadings you discourage people from using the railways.

 

[70014IronDuke]

Assuming 360 days, 15 services a day, in each direction that's about 100 people per service, given there's only 200 seats in a 4 coach 22x that's about half the train.

Even if it's 2tph (30 trains a day) that's still 50 people.

But surely many of these passengers will be using the Cardiff - Nottingham service too, won't they?

 

[takno]

Assuming 360 days, 15 services a day, in each direction that's about 100 people per service, given there's only 200 seats in a 4 coach 22x that's about half the train.

Even if it's 2tph (30 trains a day) that's still 50 people.

There's 3 trains an hour most hours. The Cardiff-Nottingham and New Street-Nottingham have knocking on 200 seats each and carry mostly local traffic over the segment, and the Voyager is mostly 5 coaches or doubled-up.

 

[Richardr]

I don't for a second believe that 5.8m/80.4m = a full 7% of journeys starting/ending at Liverpool Street were journeys exclusively between Liverpool Street and Tottenham Court Road (even allowing for that that may include people changing at TCR for the Northern line, which presumably means they are seen as exiting the national rail network at that point).

On the other hand I can well believe that nearly 5.8m passengers arrived at Liverpool Street on other services and transferred to the Elizabeth line to get to Tottenham Court Road.

I'm therefore inclined to suspect that the figures attribute to entry/exit at Liverpool Street a lot of journeys that should actually be attributed to interchange.

Another example is Farringdon, where the most popular journey is to and from Liverpool Street. No one will change from the tube to the Elizabeth Line to go to Liverpool Street at Farringdon, and there is little to gain changing from the Central Line at Liverpool Street instead of continuing to Chancery Lane. Virtually no one lives near Farringdon, and few live around Liverpool Street. I suspect near enough all of those will be commuters changing either from Thameslink at Farringdon or changing from another National Rail service to the Elizabeth Line at Liverpool Street.

 

[Horizon22]

I don't for a second believe that 5.8m/80.4m = a full 7% of journeys starting/ending at Liverpool Street were journeys exclusively between Liverpool Street and Tottenham Court Road (even allowing for that that may include people changing at TCR for the Northern line, which presumably means they are seen as exiting the national rail network at that point).

On the other hand I can well believe that nearly 5.8m passengers arrived at Liverpool Street on other services and transferred to the Elizabeth line to get to Tottenham Court Road.

I'm therefore inclined to suspect that the figures attribute to entry/exit at Liverpool Street a lot of journeys that should actually be attributed to interchange.

Nor can I believe that over 5 million are doing just Stratford - Liverpool Street; most of these people would surely have been on through trains from elsewhere.

 

[Adrian1980uk]

I notice across the statistic announcements the comparison with pre-pandemic is quietly being dropped, I agree it's getting less relevant but as the passenger numbers and revenue increases I'm getting the feeling it's more to do with the fact it's generally getting somewhere over 90% and if the increase continues it will get over 100 and justification for cuts will disappear

 

[Snow1964]

I notice across the statistic announcements the comparison with pre-pandemic is quietly being dropped, I agree it's getting less relevant but as the passenger numbers and revenue increases I'm getting the feeling it's more to do with the fact it's generally getting somewhere over 90% and if the increase continues it will get over 100 and justification for cuts will disappear

The restrictions ended officially about March 2022, coming up to 2 years ago.

Just had the third half year timetable change since then, and people want to get back to normal, not be reminded of temporary bad times. 21 months to not be able to return to normal operating looks rather bad.

 

[Horizon22]

As I stated the law requires consent to collect data. You were implying TfL has an accurate record of who travels which way through WiFi data. I was disputing that, as there's no feasible legal way of collecting that. You only collect the data from some passengers, and the collection sample may not be representative of all passengers. For example, would someone without a phone be more or less likely to take an alternative route due to the TfL website saying there are delays on one specific route?



I don't walk around with WiFi turned on when I'm not connected to a network and not trying to connect. That just results in the battery running down quicker. Therefore, you have no record of my journeys other than by where I tapped in and out. I'm not exercising any right to opt out as the result of signage, I'm choosing not to opt in by leaving it turned off.

When you go to connect to WiFi, you normally have a little tickbox you have to select. Pretty standard for most public WiFi. This essentially gives consent.

It’s obviously all completely anonymised and stripped of individual information once analysed.

== Doublepost prevention - post automatically merged: 23 Dec 2023 ==

The restrictions ended officially about March 2022, coming up to 2 years ago.

Just had the third half year timetable change since then, and people want to get back to normal, not be reminded of temporary bad times. 21 months to not be able to return to normal operating looks rather bad.

As stated countless times, passenger numbers might be back but generally they are travelling on cheaper tickets, so there’s still a gap to plug somehow. Arguably this could be done with extra subsidy, but that’s not how the government sees it.