Sheffield To Brighton Direct

[BigB]

So to summarise the above, this is a non starter as it stands.
However, should the MML to Sheffield be electrified, it is possible for a 700 to reach there from Brighton, but would introduce many potential issues for the core (see any discussion on extending Elizabeth line services).
So to come from a slightly different angle, how far up the MML could a Brighton service run without impacting core reliability, accepting that they already run to Cambridge??

 

[Bald Rick]

So to summarise the above, this is a non starter as it stands.
However, should the MML to Sheffield be electrified, it is possible for a 700 to reach there from Brighton, but would introduce many potential issues for the core (see any discussion on extending Elizabeth line services).
So to come from a slightly different angle, how far up the MML could a Brighton service run without impacting core reliability, accepting that they already run to Cambridge??

As far as the wires go, assuming there are appropriate levels of timetable allowances and sensible resourcing arrangements.

I’m not so sure that the customers of the top end of the MML woudl be too happy though.

 

[Topological]

As far as the wires go, assuming there are appropriate levels of timetable allowances and sensible resourcing arrangements.

I’m not so sure that the customers of the top end of the MML woudl be too happy though.

This then made me think that there are smaller stations for commuting between Leicester and Nottingham as well. Given people might want direct trains from Nottingham to Luton, or Loughborough to St Albans then we should just get the wires up and extend Thameslink to Nottingham - can 700s do selective door opening for the 2-car platform at Syston?

(In case anyone is in any doubt, I do believe you could run a commuter EMU on the south MML, but do not advocate Syston to Brighton, or Attenborough to Gravesend)

 

[zwk500]

So to come from a slightly different angle, how far up the MML could a Brighton service run without impacting core reliability, accepting that they already run to Cambridge??

Arguably not as far as they currently do!

 

[NCT]

I think there should be a Dunblane to Ventimiglia service via the Thameslink core and RER D.

 

[Magdalia]

So to come from a slightly different angle, how far up the MML could a Brighton service run without impacting core reliability, accepting that they already run to Cambridge??

Bedford, which is where they run to now.

Distance wise, the only Thameslink route that goes more than 60 miles from the core is Peterborough: it is not a good advert for service resilience.

 

[William3000]

So to summarise the above, this is a non starter as it stands.
However, should the MML to Sheffield be electrified, it is possible for a 700 to reach there from Brighton, but would introduce many potential issues for the core (see any discussion on extending Elizabeth line services).
So to come from a slightly different angle, how far up the MML could a Brighton service run without impacting core reliability, accepting that they already run to Cambridge??

Thameslink currently run as far north as Cambridge and Peterborough. It’s been suggested by some (and dismissed by others) that they could run beyond Bedford to Corby. Sheffield does seem a stretch too far / especially on a class 700! I’ve wondered if Thameslink runs to Cambridge and Peterborough already why Corby would be too far - the slow lines north of Bedford are relatively empty and besides they would simply be replacing the EMR connect service which runs to Corby in any case. It may also enable potential new stations at Clapham/Oakley, Rushden/Irchester and Burton Latimer.

In doing so, could it free up capacity for two more trains to add to the 2 to Sheffield and 2 to Nottingham. Perhaps a third to Sheffield, and slow to Nottingham via Corby, Oakham, and Melton Mowbray.

== Doublepost prevention - post automatically merged: 23 Dec 2025 ==

Bedford, which is where they run to now.

Distance wise, the only Thameslink route that goes more than 60 miles from the core is Peterborough: it is not a good advert for service resilience.

As I understand Cambridge is just 60 miles northbound (over the new chord north of Hitchin) and just under 59 miles southbound to London St Pancras International. Obviously Peterborough is more like 76 miles from London. I think Bedford is 48 miles and Corby 70 miles.

 

[Magdalia]

I’ve wondered if Thameslink runs to Cambridge and Peterborough already why Corby would be too far

Peterborough is too far, as is frequently demonstrated, so Corby is too far too.

From Kings Cross Cambridge is 58 miles and Peterborough 76 miles.

From St Pancras Bedford is 50 miles, Kettering is 72 miles and Corby is 79 miles.

Corby is further away from London than Peterborough. Given the track record of the Peterborough service it would be crazy to repeat the same mistake with Corby.

What needs to happen is for Peterborough to be taken out of Thameslink, not for more long distance journeys to be added in, but taking Peterborough out of Thameslink is difficult.

the slow lines north of Bedford are relatively empty

The slow lines are also slow, much of Bedford-Wellingborough (15 miles) is 50mph maximum, so using the slow line would have a significant impact on journey times, as would adding extra stops.

 

[William3000]

Peterborough is too far, as is frequently demonstrated, so Corby is too far too.

From Kings Cross Cambridge is 58 miles and Peterborough 76 miles.

From St Pancras Bedford is 50 miles, Kettering is 72 miles and Corby is 79 miles.

Corby is further away from London than Peterborough. Given the track record of the Peterborough service it would be crazy to repeat the same mistake with Corby.


The slow lines are also slow, much of Bedford-Wellingborough (15 miles) is 50mph maximum, so using the slow line would have a significant impact on journey times, as would adding extra stops.

London King’s Cross is further than that northbound as explained as a result of the chord.

That’s not my understanding - the figures I have are Bedford 48 miles; Kettering 64 miles, and Corby 70 miles but regardless my point isn’t so much whether Corby is suitable, simply that if Peterborough and Cambridge are it suggests that Corby might be - appreciating that line speed increases may be necessary - speeding up the slow lines from 50mph to 70mph in common with the ECML would be a start.

But given that the EWR proposals have limited turn back facilities for Thameslink trains at Bedford and given that on the ECML trains go considerably further north than Bedford AND given there are some large settlements comparable with places like Arlesey, Baldock, Sandy etc north of Bedford which don’t have stations, one has to look strategically and long term about these things.

Irchester/Rushden could benefit from a station, as could Burton Latimer and Thameslink achieves these so I disagree that Corby wouldn’t be an appropriate location to which to extend Thameslink - if one is considering it in the context of Peterborough and Cambridge.

While I appreciate there are significantly reliability issues through the core the limited scope of terminating platforms at London St Pancras International means the core needs to work even harder and as I understand from meetings I’ve attended there is scope for a few additional services through the core.

If Huntingdon, St Neots etc are served entirely by Thameslink trains I don’t see why Wellingborough and Corby needs something fundamentally different. Yes Corby is a larger town but when you factor in that Huntingdon serves Alconbury, Brampton, Godmanchester, and St Ives it’s a fairly big catchment area, born out in the fact it’s has more than triple the passenger numbers of Corby.

== Doublepost prevention - post automatically merged: 23 Dec 2025 ==

Peterborough is too far, as is frequently demonstrated, so Corby is too far too.

From Kings Cross Cambridge is 58 miles and Peterborough 76 miles.

From St Pancras Bedford is 50 miles, Kettering is 72 miles and Corby is 79 miles.

Corby is further away from London than Peterborough. Given the track record of the Peterborough service it would be crazy to repeat the same mistake with Corby.

What needs to happen is for Peterborough to be taken out of Thameslink, not for more long distance journeys to be added in, but taking Peterborough out of Thameslink is difficult.


The slow lines are also slow, much of Bedford-Wellingborough (15 miles) is 50mph maximum, so using the slow line would have a significant impact on journey times, as would adding extra stops.

If Peterborough is too far then arguably Cambridge is too. But they are logical places at which to change trains. Without travelling as far north as Peterborough, passengers at intermediate stations would need some way of travelling north.

If the Peterborough to Horsham trains were amended so that Horsham ran into London Blackfriars and Peterborough ran into London King’s Crosd would there be capacity at those stations? There doesn’t seem much scope for swapping over services as the only ones with potential would be the commuter services to Hertford North/Stevenage, Welwyn Garden City, and possibly extending to Letchworth Garden City. But these run into Moorgate and Moorgate would be inappropriate for longer distance services.

== Doublepost prevention - post automatically merged: 23 Dec 2025 ==

Peterborough is too far, as is frequently demonstrated, so Corby is too far too.

From Kings Cross Cambridge is 58 miles and Peterborough 76 miles.

From St Pancras Bedford is 50 miles, Kettering is 72 miles and Corby is 79 miles.

Corby is further away from London than Peterborough. Given the track record of the Peterborough service it would be crazy to repeat the same mistake with Corby.

What needs to happen is for Peterborough to be taken out of Thameslink, not for more long distance journeys to be added in, but taking Peterborough out of Thameslink is difficult.


The slow lines are also slow, much of Bedford-Wellingborough (15 miles) is 50mph maximum, so using the slow line would have a significant impact on journey times, as would adding extra stops.

Could a solution to the Peterborough issue be to run Peterborough to London King’s Cross, and Horsham to London Blackfriars. Do both have the capacity or could the slow to Cambridge also be taken through the core if there isn’t capacity for both that AND the Peterborough to London semi fast? Could it impact on the resilience on the two track Hitchin to Shepreth branch junction. Would there be a need for proper passing loops at Letchworth and just south-west of Royston?

 

[Magdalia]

London King’s Cross is further than that northbound as explained as a result of the chord.

The Hitchin flyover adds 52 chains, it is not significant.

That’s not my understanding - the figures I have are Bedford 48 miles; Kettering 64 miles, and Corby 70 miles

My distances come from working timetables, and Real Time Trains schedules for St Pancras-Corby trains.

If Huntingdon, St Neots etc are served entirely by Thameslink trains I don’t see why Wellingborough and Corby needs something fundamentally different.

Have you asked the residents of Huntingdon and St Neots, what they would prefer, a reliable service to Kings Cross, which is what they had up to 2018, or the unreliable Thameslink service that they get now? Various discussions on this forum will give you the answer.

Wellingborough, Kettering and Corby need to be saved from the same disaster.

if one is considering it in the context of Peterborough and Cambridge.

The Cambridge context is very different because it has the biggest passenger flows across the core between East of England and South East stations. Of all of the East of England stations (Herts, Beds and Cambs), Cambridge has the biggest flows to both Gatwick and Brighton, and that's what makes keeping Cambridge in Thameslink worthwhile.

While I appreciate there are significantly reliability issues through the core the limited scope of terminating platforms at London St Pancras International means the core needs to work even harder

The Peterborough context is that it was included in Thameslink for the reason you give here, to release London terminal platform capacity. Time has shown that including Peterborough in Thameslink was a mistake, but now the platform capacity at Kings Cross has gone. Including Corby in Thameslink would be making the same mistake.

one has to look strategically and long term about these things.

Strategically and long term the priority for Thameslink is to improve reliability and resilience, by simplification, and particularly by removing routes that add lots of risk for little reward. Taking Peterborough out of Thameslink would be moving in the right direction, but adding Corby to Thameslink would be the opposite: adding lots of risk for very little reward.

 

[zwk500]

Peterborough is too far, as is frequently demonstrated, so Corby is too far too.

If Peterborough is too far then arguably Cambridge is too. But they are logical places at which to change trains. Without travelling as far north as Peterborough, passengers at intermediate stations would need some way of travelling north.

It's not about distance alone, or even journey time alone, but also where the services sit in the overall service pattern structure, the potential for contra-direction flows, and the available infrastructure.

Cambridge and Peterborough are both important junctions for further rail services and significant destinations in their own right. Wellingborough, Kettering and Corby are not destinations for any significant traffic, and only Kettering offers meaningful interchange (that is replicated to some extent at Luton Airport Parkway and Bedford).
Thameslink forms the semi-fast services to London, running fast in from Welwyn as the WGC starters take up the slow pattern. The Corby trains now provide Bedford with the equivalent of LNER for Stevenage or the King's Lynn-King's Cross services for Cambridge.

 

[William3000]

The Hitchin flyover adds 52 chains, it is not significant.


My distances come from working timetables, and Real Time Trains schedules for St Pancras-Corby trains.


Have you asked the residents of Huntingdon and St Neots, what they would prefer, a reliable service to Kings Cross, which is what they had up to 2018, or the unreliable Thameslink service that they get now? Various discussions on this forum will give you the answer.

Wellingborough, Kettering and Corby need to be saved from the same disaster.


The Cambridge context is very different because it has the biggest passenger flows across the core between East of England and South East stations. Of all of the East of England stations (Herts, Beds and Cambs), Cambridge has the biggest flows to both Gatwick and Brighton, and that's what makes keeping Cambridge in Thameslink worthwhile.



The Peterborough context is that it was included in Thameslink for the reason you give here, to release London terminal platform capacity. Time has shown that including Peterborough in Thameslink was a mistake, but now the platform capacity at Kings Cross has gone. Including Corby in Thameslink would be making the same mistake.


Strategically and long term the priority for Thameslink is to improve reliability and resilience, by simplification, and particularly by removing routes that add lots of risk for little reward. Taking Peterborough out of Thameslink would be moving in the right direction, but adding Corby to Thameslink would be the opposite: adding lots of risk for very little reward.

I regularly use Huntingdon and have many friends in the area. I don’t think they have necessarily put two and two together linking the delays to Thameslink travelling through the core but you’re correct it’s an issue. Most people I speak to in St Ives/Huntingdon who use Huntingdon (but who don’t work in the industry) love the direct link through London enabling them to avoid the underground to get to the Elizabeth Line and Gatwick Airport in particular without problems either carrying luggage on tubes BUT they also don’t like the regular unreliability of the service.

Comments on this forum probably aren’t necessarily an accurate reflection of widespread public opinion because it’s self-selecting with many of us working in engineering, transport planning, or on the railways more generally, or at least many with an active interest in the railways.

So I think there are pros and cons. As somebody who uses Thameslink trains regularly I value the direct link to the Elizabeth Line via Farringdon, the ability to reach central London and the River Thames without using the tube, and being able to travel to Gatwick AirPort directly. If however stations from Arlesey northwards lost this service then as long as interchange facilities at places like Hitchin, Stevenage, or Finsbury Park were provided without significant time delays I think it wouldn’t be too much of an inconvenience.

== Doublepost prevention - post automatically merged: 23 Dec 2025 ==

It's not about distance alone, or even journey time alone, but also where the services sit in the overall service pattern structure, the potential for contra-direction flows, and the available infrastructure.

Cambridge and Peterborough are both important junctions for further rail services and significant destinations in their own right. Wellingborough, Kettering and Corby are not destinations for any significant traffic, and only Kettering offers meaningful interchange (that is replicated to some extent at Luton Airport Parkway and Bedford).
Thameslink forms the semi-fast services to London, running fast in from Welwyn as the WGC starters take up the slow pattern. The Corby trains now provide Bedford with the equivalent of LNER for Stevenage or the King's Lynn-King's Cross services for Cambridge.

Agreed - Kettering is interesting though. Regardless of whether Corby is included within Thameslink or EMR, I certainly think a slight rejig of the long distance EMR trains is necessary. EWR/Universal Studios mean Bedford will and should become a much more important station - more long distance trains should stop there - whether places like Beeston and Long Eaton need to be served by long distance trains is a point of discussion to perhaps compensate. Also East Midlands Parkway continues to be something of a white elephant station. Kettering needs at least one long distance trains should stop to stop there so passengers can travel north and Wellingborough passengers can change to travel north.

Luton Airport could probably do with the semi fast Nottingham stopping there too but as long as most trains stop at Bedford, there are lots of opportunities to get to Luton Airport from there

 

[zwk500]

Have you asked the residents of Huntingdon and St Neots, what they would prefer, a reliable service to Kings Cross, which is what they had up to 2018, or the unreliable Thameslink service that they get now? Various discussions on this forum will give you the answer.

I suspect the answer from most people not on the forums would be 'we like Thameslink, we just want it to work.'

 

[martin butler]

Why would anyone want to have through trains though from Sheffield to Brighton, What's the market? and there's a lot of problems to be solved,

 

[Magdalia]

I suspect the answer from most people not on the forums would be 'we like Thameslink, we just want it to work.'

The data from Huntingdon and St Neots suggests that this is prioritising the few to the disadvantage of the many. Most Huntingdon and St Neots passengers do not use Thameslink to go beyond St Pancras.

In both cases the biggest London flows are still to Kings Cross and St Pancras, not into the core (110730 and 158649 for Huntingdon, 84649 and 81479 for St Neots).

And the next biggest are local: for Huntingdon. Peterborough (128589) is bigger than all of Farringdon-London Bridge (108742), and for St Neots, Huntingdon+Peterborough (108215) is bigger than all of Farringdon-London Bridge (96041).

The local traffic is especially disadvantaged by Thameslink disruption, often caused by events far away.

Between Bedford and Corby the local flows are important too, the biggest are:

57321 Bedford-Wellingborough
39413 Wellingborough-Kettering
34151 Bedford-Kettering
31923 Corby-Leicester
26656 Bedford-Corby
25400 Kettering-Corby

Those passengers are not going to be impressed if they were relying on a Thameslink service that disintegrated frequently, especially those trying to travel between Corby and Leicester.

You are right that people want Thameslink to work, and up to now it doesn't work anywhere near frequently enough.

It's not about distance alone, or even journey time alone, but also where the services sit in the overall service pattern structure, the potential for contra-direction flows, and the available infrastructure.

One of the causes of poor performance is that is that the Thameslink network is too complicated, and includes too many routes. The relationship isn't proportional, but every extra mile in the Thameslink network is another mile where the OHL can fall down, a rail can be broken, a signal can fail, a set of points can fail, or, sadly, where someone can go under a train.

One of the ways to make Thameslink work better is to improve resilience by simplifying routes and reducing risk. When in a hole, stop digging, and adding Corby (or Sheffield) to Thameslink would be more digging.

So I think there are pros and cons. As somebody who uses Thameslink trains regularly I value the direct link to the Elizabeth Line via Farringdon, the ability to reach central London and the River Thames without using the tube, and being able to travel to Gatwick AirPort directly. If however stations from Arlesey northwards lost this service then as long as interchange facilities at places like Hitchin, Stevenage, or Finsbury Park were provided without significant time delays I think it wouldn’t be too much of an inconvenience.

You are right that there are pros and cons. Other things being equal, over short distances the pros outweigh the cons, but as distance (or time) gets longer, there is a tipping point where the number of users of Thameslink connectivity has declined, and the performance risk has increased. Experience since 2018 suggests that the tipping point is about 50-60 miles and 60-75 minutes from the core to the extremity, which is Bedford, Cambridge and Brighton. I can't see that the pros outweigh the cons for Corby or Sheffield.

 

[zwk500]

The data from Huntingdon and St Neots suggests that this is prioritising the few to the disadvantage of the many. Most Huntingdon and St Neots passengers do not use Thameslink to go beyond St Pancras.

In both cases the biggest London flows are still to Kings Cross and St Pancras, not into the core (110730 and 158649 for Huntingdon, 84649 and 81479 for St Neots).

There are questions about how 'Thameslink stations' tickets are allocated in those statistics, although many Thameslink passengers are probably on tap-in, tap-out tickets now.

And the next biggest are local: for Huntingdon. Peterborough (128589) is bigger than all of Farringdon-London Bridge (108742), and for St Neots, Huntingdon+Peterborough (108215) is bigger than all of Farringdon-London Bridge (96041).

This does, however, suggest that cutting Thameslink back from Peterborough would then need to be replaced by an additional semi-fast service to pick up this important local connection.

The local traffic is especially disadvantaged by Thameslink disruption, often caused by events far away.

Agreed.

One of the causes of poor performance is that is that the Thameslink network is too complicated, and includes too many routes. The relationship isn't proportional, but every extra mile in the Thameslink network is another mile where the OHL can fall down, a rail can be broken, a signal can fail, a set of points can fail, or, sadly, where someone can go under a train.

Are you suggesting that the performance of Thameslink would meaningfully improve if the route was cut back to Huntingdon and Royston/Letchworth GC?

One of the ways to make Thameslink work better is to improve resilience by simplifying routes and reducing risk. When in a hole, stop digging, and adding Corby (or Sheffield) to Thameslink would be more digging.

Agree with both of these points 100%, and you've hinted at the critical issue with Thameslink resilience which is the level of interaction with different routes (both other Thameslink and other Operator). There are too many conflicting junctions on the Thameslink route, including points where trains switch between fast and slow tracks that are 'paired by use' so they are crossing an oncoming track one way or another. The timetable can carefully 'pair' such movements but then if one train is only a couple of minutes late, the window closes very fast.

 

[JonathanH]

There are too many conflicting junctions on the Thameslink route, including points where trains switch between fast and slow tracks that are 'paired by use' so they are crossing an oncoming track one way or another.

There is also a fair amount of recovery in the schedules to allow for that. It appears that the main issue is generally the train turning up at all, rather than delays which can be recovered.

However, Sheffield to Brighton is definitely not what Thameslink should be about. It always seems to be the Midland line where these proposals are focused on this forum rather than suggesting extensions beyond Peterborough.

 

[Farnborough]

The more I think about Thameslink, the more I come to the conclusion that it needs cutting back... I suspect very few (if any) passengers do an end-to-end? So I'd limit it to Bedford and Stevenage (and no further) to Gatwick (and no further) - plus the St Albans to Sutton loopers.

Everything north of Bedford is EMR (as now); everything north of Stevenage is GN; everything south of Gatwick is Southern, with the other southern branches Southern or SouthEastern as appropriate.

Make Thameslink a reliable semi-fast service through the core... and let the other services fill the bigger picture.

 

[JonathanH]

The more I think about Thameslink, the more I come to the conclusion that it needs cutting back... I suspect very few (if any) passengers do an end-to-end? So I'd limit it to Bedford and Stevenage (and no further) to Gatwick (and no further) - plus the St Albans to Sutton loopers.

Bedford and Stevenage aren't equivalent destinations. Luton and Stevenage maybe?

However, Stevenage doesn't have turnback facilities, or stabling, so that becomes Letchworth.

Bedford, Cambridge and Peterborough are the historical termini of the flows that Thameslink serves, and crucially require 12-car people mover trains for their level of custom, which destinations beyond them don't. Same applies with Brighton on the other end.

 

[Farnborough]

Bedford, Cambridge and Peterborough are the historical termini of the flows that Thameslink serves, and crucially require 12-car people mover trains for their level of custom, which destinations beyond them don't. Same applies with Brighton on the other end.

Sure... but not all trains from Bedford, Cambridge and Peterborough do not need to go to Brighton. The majority can stop in London (or Gatwick).

People going from Brighton to London (and vice versa) do not need a train that goes through to Cambridge.
People going from Bedford, Cambridge and Peterborough to London do not need a train that goes to Brighton.

 

[William3000]

Bedford and Stevenage aren't equivalent destinations. Luton and Stevenage maybe?

However, Stevenage doesn't have turnback facilities, or stabling, so that becomes Letchworth.

Bedford, Cambridge and Peterborough are the historical termini of the flows that Thameslink serves, and crucially require 12-car people mover trains for their level of custom, which destinations beyond them don't. Same applies with Brighton on the other end.

Hitchin, Letchworth GC, Royston, and Cambridge value their links to Gatwick Airport. The only airport with a direct rail link from Hitchin, Letchworth, and Royston, while Cambridge benefits from a direct link to Stansted Airport. Links to airports link Luton and Birmingham are poor by rail and Heathrow benefits from Thameslink direct interchange at Farringdon. Leave it as it is I say.

 

[Magdalia]

The more I think about Thameslink, the more I come to the conclusion that it needs cutting back... I suspect very few (if any) passengers do an end-to-end?

That depends on what you mean by very few.

One of the useful features of the origin destination matrix dataset is the regional tag, which makes it fairly easy to pick out by eye the biggest long distance Thameslink flows. I have identified the following South East-East of England flows bigger than 10k one way journeys:

38775 Gatwick-Cambridge
35426 Gatwick-St Albans
26884 Brighton-Cambridge
26191 Gatwick-Luton Airport Parkway
24694 Gatwick-Bedford
22222 Brighton-St Albans
17119 Brighton-Bedford
15249 Gatwick-Stevenage
13679 Brighton-Luton Airport Parkway
13886 Gatwick-Peterborough
12896 Brighton-Luton
12728 Gatwick-Luton
11375 Gatwick-Harpenden

There is a big asymmetry between the South East, where Gatwick and Brighton are the only really important stations outside London, and the East of England, where lots of stations on all routes have reasonable flows, especially with Gatwick.

Hitchin, Letchworth GC, Royston, and Cambridge value their links to Gatwick Airport.

Gatwick-Hitchin 9130
Gatwick-Letchworth 4292
Gatwick-Royston 5119

This isn't really the place for discussion on how to amend the existing Thameslink service to improve resilience, there is another discussion for that.

Bedford, Cambridge and Peterborough are the historical termini of the flows that Thameslink serves, and crucially require 12-car people mover trains for their level of custom, which destinations beyond them don't. Same applies with Brighton on the other end.

What's important in this discussion is to draw a red line at Bedford, Cambridge and Peterborough which says Thameslink goes no further. Fortunately there is a blue line at Brighton!

 

[William3000]

That depends on what you mean by very few.

One of the useful features of the origin destination matrix dataset is the regional tag, which makes it fairly easy to pick out by eye the biggest long distance Thameslink flows. I have identified the following South East-East of England flows bigger than 10k one way journeys:

38775 Gatwick-Cambridge
35426 Gatwick-St Albans
26884 Brighton-Cambridge
26191 Gatwick-Luton Airport Parkway
24694 Gatwick-Bedford
22222 Brighton-St Albans
17119 Brighton-Bedford
15249 Gatwick-Stevenage
13679 Brighton-Luton Airport Parkway
13886 Gatwick-Peterborough
12896 Brighton-Luton
12728 Gatwick-Luton
11375 Gatwick-Harpenden

There is a big asymmetry between the South East, where Gatwick and Brighton are the only really important stations outside London, and the East of England, where lots of stations on all routes have reasonable flows, especially with Gatwick.


Gatwick-Hitchin 9130
Gatwick-Letchworth 4292
Gatwick-Royston 5119

This isn't really the place for discussion on how to amend the existing Thameslink service to improve resilience, there is another discussion for that.



What's important in this discussion is to draw a red line at Bedford, Cambridge and Peterborough which says Thameslink goes no further. Fortunately there is a blue line at Brighton!

It’s interesting how many people travel from Cambridge to Gatwick given that of the 6 busiest airports in the UK, 4 are closer to Cambridge than Gatwick. Stansted, Luton, Heathrow, and Birmingham are all closer. But Luton and Birmingham don’t have rail connections from Cambridge which supresses their demand. The other issue is that algorithms when booking flights tend to look at all London airports together so Birmingham gets missed. It will be interesting to see if anything changes when HS2 opens - Birmingham Interchange will likely be quicker to reach from London than Stansted

== Doublepost prevention - post automatically merged: 23 Dec 2025 ==

It’s interesting how many people travel from Cambridge to Gatwick given that of the 6 busiest airports in the UK, 4 are closer to Cambridge than Gatwick. Stansted, Luton, Heathrow, and Birmingham are all closer. But Luton and Birmingham don’t have rail connections from Cambridge which supresses their demand. The other issue is that algorithms when booking flights tend to look at all London airports together so Birmingham gets missed. It will be interesting to see if anything changes when HS2 opens - Birmingham Interchange will likely be quicker to reach from London than Stansted

As somebody living near Cambridge, I used to always use Stansted for shorthall and Birmingham for long haul and occasionally Heathrow. But the direct train through to Gatwick has made a huge difference and the figures bear that out.

 

[Magdalia]

It’s interesting how many people travel from Cambridge to Gatwick given that of the 6 busiest airports in the UK, 4 are closer to Cambridge than Gatwick. Stansted, Luton, Heathrow, and Birmingham are all closer.

It is important to remember that the airport to airport bit is the middle part of a longer journey. I don't fly anymore, but I have lots of neighbours who do, and visiting family abroad is the main reason that they fly. For them, their choice of flight usually prioritises the destination airport, not the starting airport. For example, one of my neighbours has family in northern Spain, and they always fly from Gatwick because that's the only airport that has direct flights to the nearest Spanish airport to their destination. For them, it is much easier to do a long journey from Cambridge to Gatwick in England than it is to do a long journey from airport to destination in Spain.

It is also important to remember that, compared to most other places in the UK, Cambridge has a lot of people arriving from abroad to visit.

But rail journeys between Cambridge and airports is dominated by Stansted: the Cambridge-Stansted one way flow is 215498. and there's another 20360 from Cambridge North.

 

[zwk500]

Sure... but not all trains from Bedford, Cambridge and Peterborough do not need to go to Brighton. The majority can stop in London (or Gatwick).

People going from Brighton to London (and vice versa) do not need a train that goes through to Cambridge.
People going from Bedford, Cambridge and Peterborough to London do not need a train that goes to Brighton.

No, however trains from Brighton do need to run at least as far as St Pancras. The connection to Farringdon and the King's Cross-St Pancras-Euston complex is incredibly important. As there's nowhere on the MML or ECML to terminate a 12-car train immediately north of St Pancras, they must run through.

Thameslink's problem is not really the end-to-end journey pairs being too long, but rather that it's got lots of different O-Ds at 2tph rather than concentrating flows on 4 or more tph and letting people simply step off in the core and catch the next train to the other branch.
I don't know if the flat crossing at Copenhagen Jn is a big issue given the much lower level of traffic into King's Cross on the Slows but the number of flat junctions on the Southern half of the route certainly is an issue. The croydon area remodelling (CARS) would have resolved issues at Windmill bridge Jn but there's still Blackfriars, Herne Hill, North Kent East, etc..

 

[Farnborough]

Thameslink's problem is not really the end-to-end journey pairs being too long, but rather that it's got lots of different O-Ds at 2tph rather than concentrating flows on 4 or more tph and letting people simply step off in the core and catch the next train to the other branch.

That's what I was inferring...

A 4 tph service between Gatwick and each of Bedford and Stevenage gives a 7 1⁄2 minute maximum wait - with other inter-changes with Southern at East Croydon etc... if connections are reliable, then there's no issue. At the moment, connections are unreliable because Thameslink is trying to do too much over too far.

Single-train everywhere to everywhere is an infeasible service pattern...

 

[Magdalia]

Single-train everywhere to everywhere is an infeasible service pattern...

Yet this was the origin of this discussion, a proposal for through running Brighton-Sheffield!

Thameslink's problem is not really the end-to-end journey pairs being too long

I do think that Peterborough is too long, in both distance and time, and that Rainham is too long in terms of time, but not distance.

it's got lots of different O-Ds at 2tph rather than concentrating flows on 4 or more tph

A 4 tph service between Gatwick and each of Bedford and Stevenage gives a 7 1⁄2 minute maximum wait

I don't count myself as an expert on third rail land but one of the major difficulties is that the Quarry Lines and the Redhill Lines are effectively two separate routes, and, if the Redhill Line trains call at intermediate stations, then their journey time between East Croydon and Gatwick is significantly longer than via the Quarry Lines. To really sort Thameslink it needs 8tph via Quarry with the Arun Valley services running via Redhill.

 

[zwk500]

I don't count myself as an expert on third rail land but one of the major difficulties is that the Quarry Lines and the Redhill Lines are effectively two separate routes, and, if the Redhill Line trains call at intermediate stations, then their journey time between East Croydon and Gatwick is significantly longer than via the Quarry Lines. To really sort Thameslink it needs 8tph via Quarry with the Southern Coastway services running via Redhill.

Sending the Coastway Services via Redhill to pick up the intermediate services would be a very retrograde step. Those services are often full and standing by Gatwick even outside peak times, they need to run via the Quarry lines. Also consider the track layout - it's paired by use from Balcombe Tunnel Jn, with the Crawley line joining on the slow line. There's only 1 right-handed double-lead junction to get from Slow to Fast NB (at the north end of Gatwick airport) and all the left-handed ladders are single-lead only between Three bridges and Redhill so going from Fast to Slow northbound is even harder.

 

[JonathanH]

Sending the Coastway Services via Redhill to pick up the intermediate services would be a very retrograde step. Those services are often full and standing by Gatwick even outside peak times, they need to run via the Quarry lines. Also consider the track layout - it's paired by use from Balcombe Tunnel Jn, with the Crawley line joining on the slow line. There's only 1 right-handed double-lead junction to get from Slow to Fast NB (at the north end of Gatwick airport) and all the left-handed ladders are single-lead only between Three bridges and Redhill so going from Fast to Slow northbound is even harder.

I think the suggestion was that the Portsmouth / Bognor services off the Arun Valley go via Redhill, but that is not a good idea, for the overcrowding reasons you set out above. Even though they emerge at Three Bridges on the "Redhill" side of the formation, sooner or later they have to cross to the Victoria route, whether at Stoats Nest or Selhurst if they go via Redhill.

Having Redhill as a separate "branch" from everything going further south does manage the London demand fairly well (at the cost of having to connect for destinations beyond Horsham / Three Bridges).

8tph Thameslink down the Quarry Line is certainly not a panacea.

 

[NCT]

Just simplifying Thameslink pairings to ECML outer - Quarry and Bedford - Redhill would help matters. And all MML Slows to Elephant. Rainham needs to go - it was a last minute reactive change as Windmill Bridge needed to spit out 2tph.

The thing with Peterborough is this - the slow line stations between Stevenage and Peterborough are one station bank (economically and operationally). If Arlesey is a suitable Thameslink location then those Thameslink services have to run to/from Peterborough.