Major disruption through Retford (16/04/2026) and residual effects over following days

[dodecahedron67]

Fantastic joined-up thinking this morning. TPE refused point-blank to hold the 05:26 Newcastle-Edinburgh, so the passengers of the 20:30 Lumo who were travelling onward to Edinburgh got a bonus 105 minute delay waiting for the delayed 06:16 LNER which has just departed.

Estimated arrival into Edinburgh now 08:30 - a neat 12 hours after departing Kings Cross.

I'm very sanguine about the situation and the delay - but the overall railway's response this morning from those sat comfortably at their desk feels like some sort of crass joke.

Given 1S99 had to use the same platform that the TPE had to leave from because of how many trains were ‘dumped’ at Newcastle at about 4-5am, I don’t think they ‘refused point blank’, I suspect there literally wasn’t the platform capacity to hold that train AND bring your train in to the station to allow the connection.

Newcastle is a very restrictive layout at times, especially when you start getting trains dumped there

 

[tehmarks]

The TPE departed from platform 7, and the Lumo arrived eventually into platform 4 which was seemingly available throughout while we were sat on King Edward VII Bridge for some fifteen minuted while we watched Northern's empties come past in the opposite direction.

I get that that's a gross simplification, but it feels a brave move to hold a 7 hour delayed train within walking distance of the station when it's full of passengers who are desperate for the toilet, desperate for some sort of sustenance, or just plain desperate to get off the train and get home for those alighting in Newcastle.

 

[cso]

Having been on the TPE service (as I got off 1S28) I suspect the amount of people off the Lumo service would not have been very comfortable on it.

 

[dmncf]

The following Realtime Trains page is for today's 0811 Sheffield to Cleethorpes. The train is cancelled and the page states "This service was cancelled due to a duplicate delay (PJ)." What does that mean?

Realtime Trains | 0811 Sheffield to Cleethorpes | 17/04/2026

Real-time train running information for 0811 departure from Sheffield to Cleethorpes on 17/04/2026. From Realtime Trains, an independent source of train running info for Great Britain.

www.realtimetrains.co.uk

(Not sure if this train is affected by today's Retford disruption or not; apologies if not and hence my post is off topic.)

 

[Jamesrob637]

This one doesn't appear to have passed Northallerton:

Realtime Trains | 1N86 1845 London Kings Cross to Middlesbrough | 16/04/2026

Real-time train running information for 1N86 1845 departure from London Kings Cross to Middlesbrough on 16/04/2026. From Realtime Trains, an independent source of train running info for Great Britain.

www.realtimetrains.co.uk

A friend was on it but only Peterborough to York.

 

[CW2]

1M16 Inverness sleeper spent a few hours at Newcastle (passing the northbound working 1S25 there) and has now reached Doncaster. We did the York Avoider, which was an unexpected bonus. The Doncaster stop seems like another crew change. ETA Euston some time early afternoon.

 

[yorkie]

The following Realtime Trains page is for today's 0811 Sheffield to Cleethorpes. The train is cancelled and the page states "This service was cancelled due to a duplicate delay (PJ)." What does that mean?

Realtime Trains | 0811 Sheffield to Cleethorpes | 17/04/2026

Real-time train running information for 0811 departure from Sheffield to Cleethorpes on 17/04/2026. From Realtime Trains, an independent source of train running info for Great Britain.

www.realtimetrains.co.uk

(Not sure if this train is affected by today's Retford disruption or not; apologies if not and hence my post is off topic.)

I advise against taking any notice of the Realtime Trains delay messages; they are not always accurate and frequently cause confusion by people who try to make sense of the amended wording on that site without being aware of the actual wording provided by the rail industry.

The actual text associated with code PJ reads as follows:

Delay Attribution Guide - Open Rail Data Wiki

PJ

Cancellation of a duplicate or erroneous schedule or an identified duplicate, false or erroneous delay (NOT berth off set related)

DUPLICATE

 

[Vexed]

The following Realtime Trains page is for today's 0811 Sheffield to Cleethorpes. The train is cancelled and the page states "This service was cancelled due to a duplicate delay (PJ)." What does that mean?

Realtime Trains | 0811 Sheffield to Cleethorpes | 17/04/2026

Real-time train running information for 0811 departure from Sheffield to Cleethorpes on 17/04/2026. From Realtime Trains, an independent source of train running info for Great Britain.

www.realtimetrains.co.uk

(Not sure if this train is affected by today's Retford disruption or not; apologies if not and hence my post is off topic.)

It's still running, just from Liverpool instead of Sheffield. The Sheffield to Cleethorpes was a VSTP (Very Short Term Plan) schedule which I assume was put in very late notice due to overhead line problems. The service I'm linking below was cancelled between Manchester Piccadilly and Cleethorpes at 06:55 this morning, then reinstated half an hour later.

I assume TPE planned to run the end part of it from the other side of the overhead line issues with a different unit (the service you linked), but this wasn't needed in the end as the original train was able to run through with a 30 minute delay.

Realtime Trains | 1B66 0614 Liverpool Lime Street to Cleethorpes | 17/04/2026

Real-time train running information for 1B66 0614 departure from Liverpool Lime Street to Cleethorpes on 17/04/2026. From Realtime Trains, an independent source of train running info for Great Britain.

www.realtimetrains.co.uk

 

[55002]

This one doesn't appear to have passed Northallerton:

Realtime Trains | 1N86 1845 London Kings Cross to Middlesbrough | 16/04/2026

Real-time train running information for 1N86 1845 departure from London Kings Cross to Middlesbrough on 16/04/2026. From Realtime Trains, an independent source of train running info for Great Britain.

www.realtimetrains.co.uk

A friend was on it but only Peterborough to York.

maybe went to Newcastle?

 

[Segedunum]

There was a pretty colossal displacement of crew and stock last night. Its not the case that all services have been cancelled, but the service has been thinned out to make things more manageable. Namely all fast Edinburgh to London services (that only stop at Newcastle and York) and the Harrogate/Bradford/Lincolns have been as well. This still gives 1tph to Edinburgh, 1tph to Newcastle and 1tph to Leeds.

Trains were arriving at Newcastle this morning after they should have been emerging from the depot to commence service.

 

[185143]

maybe went to Newcastle?

Almost certainly. Middlesbrough station likely locked up. Suspect any passengers will have been taxied from Darlington

 

[D1537]

maybe went to Newcastle?

Yes, it went to Newcastle to help out with the large numbers of people unable to travel north from York.

 

[Banham7]

5D25, which was dumped at Arlesey last night following its failure at Stevenage is not expected to be moved until at least midday today.

An LNER train developed a fault yesterday and teams are working hard to get this train moved but we expect this may not be until towards midday. The train is blocking the northbound line after Arlesey which means trains cannot call here or they would be stuck behind it.

https://www.greatnorthernrail.com/s...487010000#INC861C913FE8E74076A583A5F7773259A6

 

[Tom]

I advise against taking any notice of the Realtime Trains delay messages; they are not always accurate and frequently cause confusion by people who try to make sense of the amended wording on that site without being aware of the actual wording provided by the rail industry.

The actual text associated with code PJ reads as follows:

Delay Attribution Guide - Open Rail Data Wiki

You are more than welcome to propose changes at https://github.com/realtimetrains/nr-reference-data (although PJ is pretty damn close to what it is, considering it needs *a* level of simplicity)

 

[Horizon22]

The following Realtime Trains page is for today's 0811 Sheffield to Cleethorpes. The train is cancelled and the page states "This service was cancelled due to a duplicate delay (PJ)." What does that mean?

Realtime Trains | 0811 Sheffield to Cleethorpes | 17/04/2026

Real-time train running information for 0811 departure from Sheffield to Cleethorpes on 17/04/2026. From Realtime Trains, an independent source of train running info for Great Britain.

www.realtimetrains.co.uk

(Not sure if this train is affected by today's Retford disruption or not; apologies if not and hence my post is off topic.)

It is regularly mentioned that RTT doesn’t pick up the passenger-facing reasons. There will be a different actual reason (probably trespassers).

 

[Johnnie2Sheds]

Controversial opinion: Keep running. This is madness. 10mph through the affected area. Police out in force to remove the idiot(s).

 

[397301]

Was on one of the northbound hull trains affected last night. Were sat for 3 hours outside Grantham with limited info (just knew there was a tresspasser on the track causing a delay). Was chaos at Doncaster, Hull Trains effectively abandoned us all at Doncaster without any info on how to get home (beyond blind faith a bus would eventually turn up). Very stressful

 

[Trainguy34]

Controversial opinion: Keep running. This is madness. 10mph through the affected area. Police out in force to remove the idiot(s).

I think they had to isolate the OHLE, so running some trains (cl.801/803) would simply not be possible (without the use of the backup diesel generator).

 

[185143]

I think they had to isolate the OHLE, so running some trains (cl.801/803) would simply not be possible (without the use of the backup diesel generator).

I don't think 803s have one at all, but surely the whole point of having it in the generator in the first place is to allow the units to "self rescue" if there's no OLE power. Obviously doesn't work if the wires are physically down, but just switched off it will.

May well be slow. But I'm sure if you'd done a poll on any of those trains last night and said we can move at 5mph for the next 5 miles. Or stay here UFN. I know what the unanimous response would have been!

 

[enginedin]

Reading some of the above I feel lucky that I managed to get home by a LNER-provided taxi, only 5 hours late (at 3:30).

It does seem a bit of a shame that there is such different behaviour from different Train Managers (even on trains from the same TOC on adjacent platforms at the same station e.g. in this post https://www.railforums.co.uk/threads/major-disruption-through-retford-16-04-2026.301187/post-7754338 - I'm sure he has enough on this plate but I wonder if @David Horne has ever been told about these kind of diverse experiences). The LNER staff at Newcastle (when we finally got there) were good, taxis for everyone being organised based on final destination (there were some Glasgow and Falkirks, didn't hear of anyone going further!) - and most people were being patient with them. The one thing from my point of view going northbound is wondering why the LNER unit stranded at York couldn't have just reversed to go back to Edinburgh. I'm sure there are good reasons, but even the LNER staff said they thought this should be the sensible thing to do

 

[H F Stephens]

The TPE departed from platform 7, and the Lumo arrived eventually into platform 4 which was seemingly available throughout while we were sat on King Edward VII Bridge for some fifteen minuted while we watched Northern's empties come past in the opposite direction.

I get that that's a gross simplification, but it feels a brave move to hold a 7 hour delayed train within walking distance of the station when it's full of passengers who are desperate for the toilet, desperate for some sort of sustenance, or just plain desperate to get off the train and get home for those alighting in Newcastle.

By any objective measure, from the outside, last night's performance was a total shambles from a passenger point of view, and it was obvious that it was rolling out of control from at least 21:00h.

I expect that the efforts of many if not all front line staff were probably borderline heroic, and their commitment unquestionable, but at a network/institutional level, it was a farce.

This has probably been the first real test of the resilience of ECML operations post December 2025 timetable change, and the designed-in susceptibility of a vast geographical area to critical disruption was clearly demonstrated - and still is this morning.

What was also notable last night was just how poor information dissemination was, which is completely unrelated to the timetable intensity, but suggests a major failure in data systems.

As an example, the LNER website was, at a point last night, advising that an Edinburgh departure was 'currently' ten minutes late from Waverley, when it should have been at King's Cross for some time, and was actually at Newcastle.

In 2026, that's indefensible.

 

[Tetchytyke]

I'm sure he has enough on this plate but I wonder if @David Horne has ever been told about these kind of diverse experiences)

It's nothing new with LNER, nor VTEC before them. Some of the staff are very good, most of them are in the middle, and some are terrible. If Mr Horne cared it wouldn't keep happening.

In my experience, the customer-facing staff at Newcastle station are excellent and genuinely can't do enough for people. 100 miles away in York, though, and they all do a magic disappearing act the second anything goes wrong. Leeds is somewhere in the middle. I've never been stuck at Doncaster so can't comment about there.

Nothing about the stories from last night surprises me. Including the gridlocking of Newcastle station by abandoning trains in the platforms there rather than getting someone to at least move them to Heaton. On more than one occasion I have spent over an hour sat on the King Edward Bridge because the station is full and it is beyond the wit of anyone to get someone to move the empty trains.

 

[MontyP]

By any objective measure, from the outside, last night's performance was a total shambles from a passenger point of view, and it was obvious that it was rolling out of control from at least 21:00h.

I expect that the efforts of many if not all front line staff were probably borderline heroic, and their commitment unquestionable, but at a network/institutional level, it was a farce.

This has probably been the first real test of the resilience of ECML operations post December 2025 timetable change, and the designed-in susceptibility of a vast geographical area to critical disruption was clearly demonstrated - and still is this morning.

What was also notable last night was just how poor information dissemination was, which is completely unrelated to the timetable intensity, but suggests a major failure in data systems.

As an example, the LNER website was, at a point last night, advising that an Edinburgh departure was 'currently' ten minutes late from Waverley, when it should have been at King's Cross for some time, and was actually at Newcastle.

In 2026, that's indefensible.

The information point is a very valid one. This is indefensible in the 2020s. How many times have we commented on this forum that a rail enthusiast can piece together a better picture of what is going on, where the delays are, what the alternatives are, than the staff can? I don't know enough about the back end systems but as far as I can see, the raw data is all there (and we can see it via Traksy, RTT, etc) but there doesn't seem to be the ability to make this available to front line staff in a joined up way that enables them to relay it accurately to customers or to make sensible decisions on rerouting options etc.

My other point is around some of the decision making - it was clear that this would be a major problem from around 1930, but it seems that trains continued to be despatched from Kings Cross and other stations rather than held in platforms there or at intermediate stations. With the result that there were something like 18 northbound services marooned at various points, mostly between stations.

 

[cats_five]

The information point is a very valid one. This is indefensible in the 2020s. How many times have we commented on this forum that a rail enthusiast can piece together a better picture of what is going on, where the delays are, what the alternatives are, than the staff can? I don't know enough about the back end systems but as far as I can see, the raw data is all there (and we can see it via Traksy, RTT, etc) but there doesn't seem to be the ability to make this available to front line staff in a joined up way that enables them to relay it accurately to customers or to make sensible decisions on rerouting options etc.

My other point is around some of the decision making - it was clear that this would be a major problem from around 1930, but it seems that trains continued to be despatched from Kings Cross and other stations rather than held in platforms there or at intermediate stations. With the result that there were something like 18 northbound services marooned at various points, mostly between stations.

The rail enthusiast can get on with doing it without having to deal with a train-load of passengers, and so on.

 

[Pakenhamtrain]

The information point is a very valid one. This is indefensible in the 2020s. How many times have we commented on this forum that a rail enthusiast can piece together a better picture of what is going on, where the delays are, what the alternatives are, than the staff can? I don't know enough about the back end systems but as far as I can see, the raw data is all there (and we can see it via Traksy, RTT, etc) but there doesn't seem to be the ability to make this available to front line staff in a joined up way that enables them to relay it accurately to customers or to make sensible decisions on rerouting options etc.

It seems to be a problem with information flow from control centres to the front line on railways in general.

Down here there have been times I've been sitting a home and my mate driving a train has less of an idea of why his train is delayed than I do.

 

[Horizon22]

Controversial opinion: Keep running. This is madness. 10mph through the affected area. Police out in force to remove the idiot(s).

Sometimes they cannot be removed because of where they have position themselves. And then if you approach, they jump and that’s going to be a whole lot worse…

== Doublepost prevention - post automatically merged: 17 Apr 2026 ==

It seems to be a problem with information flow from control centres to the front line on railways in general.

Down here there have been times I've been sitting a home and my mate driving a train has less of an idea of why his train is delayed than I do.

Sometimes there’s no information to share. Sometimes there’s no update, especially if emergency services have taken primacy of the scene.

It is also much easier to send a delay notification via Darwin with a few clicks that it might be to speak to a guard via phone (with poor signal), who then has to tell the driver if there are also 8 other actions to do. It is not a criticism of all crew and stations, but sometimes the information IS available, but don’t have the knowledge or confidence to know where to look for it.

A lot of it comes down to workload; to shift from very little to multiple phone calls and actions simply cannot be done in a timely manner for all stakeholders (passengers, train crew, stations, managers, updating systems). Ideally you’d double the amount of staff in control rooms, but frankly 95% of the time they wouldn’t be needed. This is a conundrum that rears its head every few months or so and has no easy solution.

I’d also say that this was probably the night shift team too, which will probably have some element of reduced staffing.

 

[MontyP]

The rail enthusiast can get on with doing it without having to deal with a train-load of passengers, and so on.

If they had the right information then perhaps it would make the job of dealing with a trainload of passengers easier?

 

[ModernRailways]

The information point is a very valid one. This is indefensible in the 2020s. How many times have we commented on this forum that a rail enthusiast can piece together a better picture of what is going on, where the delays are, what the alternatives are, than the staff can? I don't know enough about the back end systems but as far as I can see, the raw data is all there (and we can see it via Traksy, RTT, etc) but there doesn't seem to be the ability to make this available to front line staff in a joined up way that enables them to relay it accurately to customers or to make sensible decisions on rerouting options etc.

My other point is around some of the decision making - it was clear that this would be a major problem from around 1930, but it seems that trains continued to be despatched from Kings Cross and other stations rather than held in platforms there or at intermediate stations. With the result that there were something like 18 northbound services marooned at various points, mostly between stations.

The truth of the matter is a lot of rail staff aren’t enthusiasts and know what they’re training teaches them and that’s it. I have colleagues who use Trainline to see their trains, and find platforms. An enthusiast tends to have a better understanding and can use a few different ways to get a better picture, similar to how the control room will. Unfortunately, control rooms (at least for my TOC) have a handful of people on so when we get major disruption you can’t get through to them, and you have to check industry systems and hope that the messages they put out are good enough and frequent enough. My TOC is pretty poor at providing additional internal information, thankfully on routes that tend to receive the worst disruption it is shared with other TOCs who are very good and as we can see the information they put out it lets us get a better understanding - the majority of my colleagues will not check the other TOCs Arrakis page so won’t see the additional info that our own TOC doesn’t publish. I would also put a part of this down to being a bit more tech savvy.
Ultimately, staff on the frontlines can be fantastic but if the people sat in the office with a better overview are putting out poor/no information then they’re stuck in a horrid position.

Whilst I think my TOC is very poor with disruption messaging, they are very good with service recovery and management, we also have fairly good diversionary knowledge in my patch. A lot of people will moan that during disruption we are quick to cancel trains, but ultimately cancelling trains and moving people onto services that will move via a diversion is significantly better than having people sat on trains not moving and with little knowledge of what will happen next. Yes, you may be delayed by the diversion, but you’re not going to get to an area of disruption away from a station and be sat there for 3 hours. I appreciate LNER is a bit more advanced than we are, using electricity, but they were diverting some trains via Lincoln so getting as many passengers onto these services would have surely made more sense?

Continuing to send trains full of passengers into the affected area is a very strange decision. I’m not aware of the track layout but once it was realised this was a significant issue, trains already past Peterboro shouldn’t have gone past Grantham or Newark, with trains either tipping out in platforms, or turning around and going back to London where there are a good number of hotels and staff.

 

[Horizon22]

If they had the right information then perhaps it would make the job of dealing with a trainload of passengers easier?

The information is often there (in some form), but it needs to be actually accessed, interpreted and then know what the next plan is (just because you can see something on Traksy or equivalent, do you know what’s going to happen next)?

It is much easier for an enthusiast with the know how to just look on a map with an overview of everything, as opposed to a driver in the cab focussed on their train, or station staff trying to deal with lots of different queries about multiple trains; if trains aren’t moving as well sometimes there’s nothing to say. And if they try and call control, there’s another call to add to the list as they’re the 5th station to call.

There’s definitely some solutions to resolve this - use of radios at stations, notification based apps, greater staff training - but simply put there will always be more people that want to know in serious incidents than there is capacity to disseminate that information from a central source. Sometimes it’s a case of “joining together the dots” which isn’t ideal but does require some initiative.

Not to mention sometimes the control team is reliant on information from Network Rail, or teams on the ground, or the emergency services, or the signaller so even they might not know, or it will be delayed or contradictory to other sources.

This is all very generic, so it’s hard to comment on the exact circumstances of this incident - lots on social media say the Train Manager announcements were fantastic and informative, even if there wasn’t much progress - but provides some very basic detail on what might be going on.

== Doublepost prevention - post automatically merged: 17 Apr 2026 ==

Continuing to send trains full of passengers into the affected area is a very strange decision. I’m not aware of the track layout but once it was realised this was a significant issue, trains already past Peterboro shouldn’t have gone past Grantham or Newark, with trains either tipping out in platforms, or turning around and going back to London where there are a good number of hotels and staff.

At the very least it’s incredibly poor stranded trains management, which is something Network Rail are reportedly getting hot on and better at. Yet I keep seeing real-life examples to the contrary.

Sure it might be okay on a train with lots of staff, toilets, power still on and food facilities which larger gaps between stations, but it still isn’t ideal and gives no options for people to abandon journeys.

 

[The North Briton]

As an example, the LNER website was, at a point last night, advising that an Edinburgh departure was 'currently' ten minutes late from Waverley, when it should have been at King's Cross for some time, and was actually at Newcastle.

In 2026, that's indefensible.

Even outside of major disruption, LNER’s approach seems to be to announce the initial delay on JourneyCheck once, and then not update it until it ‘falls off’ the screen some time after it arrives at its final destination.