Regular Thameslink disruptions over a week

[Nicholas Lewis]

Sometimes, but then they dont have 5MW trains on each line 16-20 times an hour.

Thats why Ludgate Cellars is often cited as the biggest DC traction substation in the world to support 24TPH in each direction and with plenty of redundancy. This contactor system has run very reliably for nearly a decade and follows on from two previous solutions installed along the Farringdon to Blackfriars route which had no reslience although an intensive maintenance regime.

 

[AM9]

I have seen the odd TL service run fast from St Pancras to Bedford. No real worry though, St Albans only has at least 9 other trains per hour to choose from…

... - as befits the considerable demand from fare paying passengers, that have no other TOC's services to rely on.

 

[Rail101]

The other issue with some of the stations where the 9J isn't initially booked to stop but then is added to the timetable as a result of disruption is you play a game of chicken as to whether or not they actually do stop or just go fast through the station. It's an absolute pain.

 

[Bald Rick]

I have seen the odd TL service run fast from St Pancras to Bedford. No real worry though, St Albans only has at least 9 other trains per hour to choose from…

Except when there isn’t. This year I have stood at St Pancras whe there has been no St Albans service of any type for an hour in the evening peak, through a combination of late running, cancellations and amended calling patterns. Now while it is a fair shout to say ‘only an hour?’ and deploy a eye roll, St Albans has peak hour flows to/from London of around 3000 an hour, and therefore any gap in the peak service of more than about 30 minutes automatically makes the first service crush loaded and unboardable at St Pancras northbound (or St Albans southbound). In the example above, it was 4 services, and almost another hour before any significant number of people were able to board at St Pancras.

 

[D7666]

IME Thameslink, on the Midland side at least, started seriously going downhill from about mid 2024 onwards.

So much so that from 01/01/25 I have been making a detailed time log of each trip.

I work commute - part time - TuWeThO from Luton to usually Blackfriars but can be to any core station (STP ZFD CTK BFR) as needed.

I also make the same trip into London on other days, mostly STP or ZFD or LBG, 1 to 3 times a week.

To date, 01/01/25 to 14/12/25 I have made 231 return journies, or 462 single trips.

I am using released doors / step on platform as my arrival time, not what time the train hit whatever track circuit berth or other criteria is used for reporting.

As at today, my AVERAGE delay for ALL 462 trips is 8.26 minutes. 3861 minutes total. 64 h 20 m of delay.

When you consider I am able to work flexibly, and can plan to avoid disruption, and do so when I can, and don't count that in my stats, i.e. if I sit at home or in the office for an extra half hour, I don't count that, you see how bad it is.

There have also been 3 different dates when TL advice was 'not to travel' - and these also are not counted in my stats.

Of the 231 return journies, ONLY 31 (13%) have been right time at destination both ways, and another 35 within 1 min of right time one way or both ways.

At no time have I had more than 4 consecutive return journies of either right time or within 1 minute - that to me is "regular" disruption.

Now you can whine response all you like about expecting trains to be within 1 min right time is too much - but while the TOC pares down Delay Repay to the exact minute, I will do the same.

On 65 occasions (14%) the planned train was cancelled.

On 23 occasions, 2 or more consecutive trains were cancelled.

On 3 occasions I had to abandon TL completely - and it happens to be more convenient for me to travel via Watford and StAlbans (and has the advantage of avoid the melee at STP, whether for TL trains or permitted EMR travel, that Baldrick alludes to when service restarts they things are so packed you can't get near one, or if you do, it is intolerably crushed).

One thing about these delays is over and over again, they cancel one, but the next one is right time, you are 15 min late, but they don't and won't pay out on that. They play to exactly 15 min, to the minute. Which is a pain when their knee jerk reaction on the Midland side is to can the Three Bridges / Bedfords (ad often remain canned hours after the original troubke has cleared).

On only TWO occasions have I arrived at destination earlier than planned - once Down from BFR where they preceding service I was not expecting to catch was late, another on an Up trip from Luton where by fluke I got there as they were non stopping a Rainham to STP leaving just ahead of my planned train.

Anyone can nit pick over use of the term "regular" or "frequent" but to me the two here have blurred and merged into one.

Only 13% right time means 87% are not. That is regular. Over a year, not a week.

I reckon it all started going downhill from when the date of termination of the current incumbent "operator" was announced. They have given up.

 

[LMS 4F]

I have decided the next time I do my infrequent trip to London I will stay close to the West Coast line and go in to Euston and avoid Thameslink altogether. At least I will have a choice.

 

[Bald Rick]

IME Thameslink, on the Midland side at least, started seriously going downhill from about mid 2024 onwards.

So much so that from 01/01/25 I have been making a detailed time log of each trip.

I work commute - part time - TuWeThO from Luton to usually Blackfriars but can be to any core station (STP ZFD CTK BFR) as needed.

I also make the same trip into London on other days, mostly STP or ZFD or LBG, 1 to 3 times a week.

To date, 01/01/25 to 14/12/25 I have made 231 return journies, or 462 single trips.

I am using released doors / step on platform as my arrival time, not what time the train hit whatever track circuit berth or other criteria is used for reporting.

As at today, my AVERAGE delay for ALL 462 trips is 8.26 minutes. 3861 minutes total. 64 h 20 m of delay.

When you consider I am able to work flexibly, and can plan to avoid disruption, and do so when I can, and don't count that in my stats, i.e. if I sit at home or in the office for an extra half hour, I don't count that, you see how bad it is.

There have also been 3 different dates when TL advice was 'not to travel' - and these also are not counted in my stats.

Of the 231 return journies, ONLY 31 (13%) have been right time at destination both ways, and another 35 within 1 min of right time one way or both ways.

At no time have I had more than 4 consecutive return journies of either right time or within 1 minute - that to me is "regular" disruption.

Now you can whine response all you like about expecting trains to be within 1 min right time is too much - but while the TOC pares down Delay Repay to the exact minute, I will do the same.

On 65 occasions (14%) the planned train was cancelled.

On 23 occasions, 2 or more consecutive trains were cancelled.

On 3 occasions I had to abandon TL completely - and it happens to be more convenient for me to travel via Watford and StAlbans (and has the advantage of avoid the melee at STP, whether for TL trains or permitted EMR travel, that Baldrick alludes to when service restarts they things are so packed you can't get near one, or if you do, it is intolerably crushed).

One thing about these delays is over and over again, they cancel one, but the next one is right time, you are 15 min late, but they don't and won't pay out on that. They play to exactly 15 min, to the minute. Which is a pain when their knee jerk reaction on the Midland side is to can the Three Bridges / Bedfords (ad often remain canned hours after the original troubke has cleared).

On only TWO occasions have I arrived at destination earlier than planned - once Down from BFR where they preceding service I was not expecting to catch was late, another on an Up trip from Luton where by fluke I got there as they were non stopping a Rainham to STP leaving just ahead of my planned train.

Anyone can nit pick over use of the term "regular" or "frequent" but to me the two here have blurred and merged into one.

Only 13% right time means 87% are not. That is regular. Over a year, not a week.

I reckon it all started going downhill from when the date of termination of the current incumbent "operator" was announced. They have given up.

Thank you for taking the time to post that. Hard data is good evidence.

You have mentioned your mean delay, but one thing that would help is also your median and mode. As Thameslink trains can’t arrive at most stations more than a minute or two early, a couple of big delays will skew the mean.

I will say that your data doesn’t tally with my experiences, but then I haven’t kept detailed info like this. I travel on average 4 times a week, all days of the week. Tuesdays, Wednesdays and Thursdays are busier and that in itself casuses more delay and also makes things more difficult when things go wrong. Monday and Friday mornings appear (to me) to be more reliable.

I, too, keep an eye on the trains before setting out to use them. In the past year I reckon I have had my intended train cancelled perhaps twice, in both cases because I had forgotten to check and didn’t amend my plans. There are certain southbound morning trains which are the first to be cancelled if there is a crew shortage (0722, 0737, 0752 from Luton for info) so if those are your regular trains you will experience a higher rate of cancellations than the norm.

Also, I routinely experience early arrivals into both St Pancras and Blackfriars by 1 occasionally 2 mintues - probably half my trips arrive early southbound. Also early into St Albans , much more rarely, perhaps 1 in 10. Typically arrival into St Albans is 1-2 mins late.

There is no doubt that the last couple of weeks on Thameslink have been awful, and please be assurred that no one is at all happy with it, and it is being monitored at the very highest levels. Very little this week has been down to the train operator, it has been Network Rail or ‘externals’ (two trespassers in the core on successive evenings, for example). I can assure you that Thameslink haven’t ‘given up’ since the announcement on transfer to DfTO; quite the opposite, as those further up the organisation want to keep ther jobs.


The biggest single issue for cancellations is traincrew. Two things have happened there. Firstly the rate of absence shot up during Covid and has never really recovered; to be fair to Thameslink this affected many other operators too. Unfortunately for Thameslink that landed concurrent with the training for ETCS and ATO in the core which required more drivers off the footplate too. To counter that, last year Thameslink revised how they calculate the number of drivers they need to cover the service, upwards, and by a big number. Those have all now been recruited and in training, many are now fully productive. Also the ETCS / ATO training should (at last) be complete next year which will free up more drivers. The number of cancellations due to crew is falling now (on an annulaised basis), and the full benefit will be felt next year.

 

[OscarH]

The biggest single issue for cancellations is traincrew. Two things have happened there. Firstly the rate of absence shot up during Covid and has never really recovered; to be fair to Thameslink this affected many other operators too. Unfortunately for Thameslink that landed concurrent with the training for ETCS and ATO in the core which required more drivers off the footplate too. To counter that, last year Thameslink revised how they calculate the number of drivers they need to cover the service, upwards, and by a big number. Those have all now been recruited and in training, many are now fully productive. Also the ETCS / ATO training should (at last) be complete next year which will free up more drivers. The number of cancellations due to crew is falling now (on an annulaised basis), and the full benefit will be felt next year.

This is good news! There was one point where it was pretty common at 2am to get through total cancellations of every other train for the whole day (on 9Js) because of train crew, so more stability there is welcome

 

[Minstral25]

Thank you for taking the time to post that. Hard data is good evidence.

You have mentioned your mean delay, but one thing that would help is also your median and mode. As Thameslink trains can’t arrive at most stations more than a minute or two early, a couple of big delays will skew the mean.

I will say that your data doesn’t tally with my experiences, but then I haven’t kept detailed info like this. I travel on average 4 times a week, all days of the week. Tuesdays, Wednesdays and Thursdays are busier and that in itself casuses more delay and also makes things more difficult when things go wrong. Monday and Friday mornings appear (to me) to be more reliable.

I, too, keep an eye on the trains before setting out to use them. In the past year I reckon I have had my intended train cancelled perhaps twice, in both cases because I had forgotten to check and didn’t amend my plans. There are certain southbound morning trains which are the first to be cancelled if there is a crew shortage (0722, 0737, 0752 from Luton for info) so if those are your regular trains you will experience a higher rate of cancellations than the norm.

Also, I routinely experience early arrivals into both St Pancras and Blackfriars by 1 occasionally 2 mintues - probably half my trips arrive early southbound. Also early into St Albans , much more rarely, perhaps 1 in 10. Typically arrival into St Albans is 1-2 mins late.

There is no doubt that the last couple of weeks on Thameslink have been awful, and please be assurred that no one is at all happy with it, and it is being monitored at the very highest levels. Very little this week has been down to the train operator, it has been Network Rail or ‘externals’ (two trespassers in the core on successive evenings, for example). I can assure you that Thameslink haven’t ‘given up’ since the announcement on transfer to DfTO; quite the opposite, as those further up the organisation want to keep ther jobs.


The biggest single issue for cancellations is traincrew. Two things have happened there. Firstly the rate of absence shot up during Covid and has never really recovered; to be fair to Thameslink this affected many other operators too. Unfortunately for Thameslink that landed concurrent with the training for ETCS and ATO in the core which required more drivers off the footplate too. To counter that, last year Thameslink revised how they calculate the number of drivers they need to cover the service, upwards, and by a big number. Those have all now been recruited and in training, many are now fully productive. Also the ETCS / ATO training should (at last) be complete next year which will free up more drivers. The number of cancellations due to crew is falling now (on an annulaised basis), and the full benefit will be felt next year.

Have to say I'd wish I could travel on the same services as you as my experience (without detailed records) feels more like the previous poster.

Whilst I'd agree that the biggest reason of problems on 9J (PBO to HOR) could be drivers, along the Redhill route the biggest reason is the constant removal of the 9R service (BED - GTW) by control. It is a continuous issue and means stations like Earlswood and Salfords regularly have no service for hours and as the 9J is so unreliable, constantly non-stop in disruption from East Croydon to Gatwick, missing all our local stations it often leaving us with no service. (And don't say we could use the 4-coach southern service from Victoria as it normally cut whenever the 9R is cut).

Redhill route should be at least 2/3rds of St Albans patronage, but the constant cancellations mean many locals have given up on using Rail services and it is shown as Rail usage has been recovering much more slowly than elsewhere, despite thousands of new homes being built near the rail line since 2020.

 

[SteveHFC]

IME Thameslink, on the Midland side at least, started seriously going downhill from about mid 2024 onwards.

So much so that from 01/01/25 I have been making a detailed time log of each trip.

I work commute - part time - TuWeThO from Luton to usually Blackfriars but can be to any core station (STP ZFD CTK BFR) as needed.

I also make the same trip into London on other days, mostly STP or ZFD or LBG, 1 to 3 times a week.

To date, 01/01/25 to 14/12/25 I have made 231 return journies, or 462 single trips.

I am using released doors / step on platform as my arrival time, not what time the train hit whatever track circuit berth or other criteria is used for reporting.

As at today, my AVERAGE delay for ALL 462 trips is 8.26 minutes. 3861 minutes total. 64 h 20 m of delay.

When you consider I am able to work flexibly, and can plan to avoid disruption, and do so when I can, and don't count that in my stats, i.e. if I sit at home or in the office for an extra half hour, I don't count that, you see how bad it is.

There have also been 3 different dates when TL advice was 'not to travel' - and these also are not counted in my stats.

Of the 231 return journies, ONLY 31 (13%) have been right time at destination both ways, and another 35 within 1 min of right time one way or both ways.

At no time have I had more than 4 consecutive return journies of either right time or within 1 minute - that to me is "regular" disruption.

Now you can whine response all you like about expecting trains to be within 1 min right time is too much - but while the TOC pares down Delay Repay to the exact minute, I will do the same.

On 65 occasions (14%) the planned train was cancelled.

On 23 occasions, 2 or more consecutive trains were cancelled.

On 3 occasions I had to abandon TL completely - and it happens to be more convenient for me to travel via Watford and StAlbans (and has the advantage of avoid the melee at STP, whether for TL trains or permitted EMR travel, that Baldrick alludes to when service restarts they things are so packed you can't get near one, or if you do, it is intolerably crushed).

One thing about these delays is over and over again, they cancel one, but the next one is right time, you are 15 min late, but they don't and won't pay out on that. They play to exactly 15 min, to the minute. Which is a pain when their knee jerk reaction on the Midland side is to can the Three Bridges / Bedfords (ad often remain canned hours after the original troubke has cleared).

On only TWO occasions have I arrived at destination earlier than planned - once Down from BFR where they preceding service I was not expecting to catch was late, another on an Up trip from Luton where by fluke I got there as they were non stopping a Rainham to STP leaving just ahead of my planned train.

Anyone can nit pick over use of the term "regular" or "frequent" but to me the two here have blurred and merged into one.

Only 13% right time means 87% are not. That is regular. Over a year, not a week.

I reckon it all started going downhill from when the date of termination of the current incumbent "operator" was announced. They have given up.

Interestingly, I make a very similar journey - but from Leagrave rather than Luton, so I have slightly less trains to pick from. I travel to/from Blackfriars 2-3 times a week, and very rarely have any issues in the morning, the train normally arrives at Blackfriars no more than 2 or 3 minutes late. In fact, I've not put a delay repay claim in for the morning commute since November 2024. There has been more disruption in the evenings this year, but the vast majority have been infrastructure related rather than because of driver shortages.

I looked back at my delay repay claims since I started the commute in August 2023, and the number of claims I make has dropped significantly over the past 12 months - in fact, before the recent spate of issues in the core over the past couple of weeks, I hadn't put a claim in for 3 months. Also, I've yet to put in a claim for a morning trip in 2025.

My experience is that things have improved significantly on the MML side of Thameslink overall. Yes, when it goes wrong, it can go badly wrong, but either I've been very lucky, or things really aren't that bad. Yes, in an ideal world everything runs right time, but we all know things can happen that cause disruption. Maybe I'm more understanding because I worked at Network Rail for 9 years so have more understanding of what's going on behind the scenes when things go pear-shaped.

Yes, the past couple of weeks have been awful, and I've been lucky that I've not been caught up in the worst of it - Monday just gone was the worst for me when I was on a crush loaded train home and was 35 minutes late into Leagrave.

 

[SAPhil]

Except when there isn’t. This year I have stood at St Pancras whe there has been no St Albans service of any type for an hour in the evening peak, through a combination of late running, cancellations and amended calling patterns. Now while it is a fair shout to say ‘only an hour?’ and deploy a eye roll, St Albans has peak hour flows to/from London of around 3000 an hour, and therefore any gap in the peak service of more than about 30 minutes automatically makes the first service crush loaded and unboardable at St Pancras northbound (or St Albans southbound). In the example above, it was 4 services, and almost another hour before any significant number of people were able to board at St Pancras.

When that happens, I usually jump on a train going the other way and wait at Blackfriars!

 

[southern442]

there’s 11 Thameslink trains a day at Oxted, all in the peak (6 in the peak direction, 5 contra peak). They are the first to be removed in the contingency plans.

Whilst I can understand this from a strategic perspective, the idea that a set group of trains are immediately removed at the first sign of major disruption simply just isn't good enough. People cannot rely on them and therefore don't use them; they always seem more lightly loaded than the trains from Victoria.

 

[Bald Rick]

Whilst I can understand this from a strategic perspective, the idea that a set group of trains are immediately removed at the first sign of major disruption simply just isn't good enough. People cannot rely on them and therefore don't use them; they always seem more lightly loaded than the trains from Victoria.

I can assure you they are very well used on the north side!

What do you suggest as an alternative in a contingency plan?

 

[southern442]

What do you suggest as an alternative in a contingency plan?

Pie in the sky maybe, but not cancelling the same group of services every time there is disruption would be nice. Cancel some different ones from time to time as a first resort... it just doesn't seem fair that they and the three bridges services are always the first ones to go...

I know it isn't as easy as that in any way, but it's just no good having services which are the 'first to go' if they ALL end up going quite a substantial proportion of the time, it creates an extremely unreliable service. Through whatever means, I would argue that they are far too unreliable at the moment and something should be done so that it doesn't feel like once a week they just all don't run.

 

[jon0844]

When things go wrong, TOCs will need to respond quickly and so there are predetermined plans that can be implemented relatively quickly, and with people knowing and understanding that they're following 'plan xx' so as not to cause further confusion. It's important that everyone is on the same page. Choosing the appropriate plan may well be done automatically by computer, based on what and where the issue is.

These plans aren't always the best plans though, and as times change then the plans need to change too. I have no idea what the plans are in such cases, or when they're revised and reviewed, and some decisions during disruption might seem odd - but if well thought out then they should reduce the risk of further problems with trains ending up stranded, crew in the wrong places etc.

Does anyone know if these plans are created 'by hand' or computer modelled scenarios based on specific data?

For the core, there are very clear plans to thin out the service if they have to impose bi-directional working. But in the recent issues, they had even less capacity than that.

Obviously if a particular route suddenly gets all trains cancelled, it's equally important to make sure that there are plans on how to cover those stations with other services. All too often that doesn't seem to happen, and when the ECML struggles along with its latest incident of the week, it's disappointing that while some services can keep running, it's common to have trains not call at places like WGC to take people north, so leaving huge gaps in service.

 

[Bald Rick]

Does anyone know if these plans are created 'by hand' or computer modelled scenarios based on specific data?

By hand, but based on experience of varous incidents and the details of the timetable.

I have no idea what the plans are in such cases, or when they're revised and reviewed,

Reviewed at each timetable change, rewritten when there is a significant timetable change.

For the core, there are very clear plans to thin out the service if they have to impose bi-directional working.

There are several core contingency plans; depending on the nature of the incident and what the impact on throughput is, and whether it is one line or both.

 

[Mattplans]

Wow. Another signal failure. Thameslink commuters have been very unlucky recently. Let’s hope there are less issues in the upcoming week

 

[D7666]

Re. my journeys, my Up commutes are on the post peak shoulder - either 08:59 or 09:59 from Luton depending on if I have a 'nine o clock' teams meeting (that I do from home); in both cases I start work from home c.08:00, as I can work flexibly.

The 08:59 Up seems quite prone to be cancelled for me - either a one off or they have canned all the ThreeBridges.

Cancelled might mean it runs but skips Luton etc - to me that is a cancel.

Down I am either clear of BFR before 16:30 or after 19:00 - only rarely in the high peak.

These are not typical 9-5 office commutes I know - which might be why they don't align with others travels; if there is a focus on getting the high peaks right, there may be a knock on to peak shoulders and off peaks.

Other days my travel can be any times.

6 monthly rolling averages are coming down -

6 mo Jan - Jun = 9.4 min
6 mo Apr - Sep = 8.6 min
6 mo Jul to date = 7.1 min

- but the rate of slowing is now declining if I look at each mo by mo rolling value - and even 7 min x 400 is a lot of delay over a year.


Just as an observation, more subjective than anything, it seems to me that since about May this year, on time Up trains doing the US > UF switch at Harpenden are held more often at Harpenden for longer than they used to be, often now 2-3 min, long enough to be late into the core, even with all that recovery before STP, and late enough to miss their core path, and likewise on time (at STP) Down trains doing the DS > DF switch at Carlton Road similarly delayed - they used to totter out of the tunnel on yellows, now they all seem to stop on reds, with the knock on they can then lose path at Harpenden to go back DF to DS so halt on the DF. I've not checked either WTT or RTT for EM timing changes or delays. I have not kept seperate records of this, so, as I said, it is subjective, being more likely to notice a prolonged dead stop than extended unfast running.


Other longer term statistics are that in the past 2 years, I have had to use my alternate route Luton / StAlbans / Watford four times. In my entire travel from BedPan since 1986, through all of the shadow TOU, GO the first time, and FCC, I only ever did that one time. Likewise I never ever had to use a Hitchin emergency bus until 2022, and again in 2023. ((Not counted here, obviously I have used Hitchin engineering works buses.)). And on one occasion this year, my plan was into London and out from Euston to Crewe - after 90 min of no trains anywhere, north or south, nor any buses, I gave up and took a local service bus to MKC. Never ever done that before either.


IMHO the issues are not the failures themselves, but GTRs seeming inability to recover and/or alternative plans.

Not going into actual or speculative reasons why - I am stating the fact that it is like that.


I have the impression the only contingency plans are to bulk cancel certain train patterns - of which the Bedford / Three Bridges seem to be first on the north side.

Since GTR took over, I am not sure I have ever known them to split the route either side of the core, that FCC used to do a lot of. None of it is ideal, but it did work. Nor when there has been a complete core / KNT / WHP issue blocking the lot (but not the OLE) I don't recall them putting in Hendon / Luton / Bedford shuttles, that again FCC used to do.

Also, there are never emergency replacement buses these days, well not at Luton, until several hours have passed . FCC used to seem to get them up and running within 1-2 hours. The trotted out excuse there are no buses or drivers available can't be right - FCC managed to do this somehow, so it must be down to contracts (lack of).

One last thing. I do actually work in railway control. My last 20 years in industry have been in different controls. Admitedly I am an engineer not an operator, but am fully immersed in control room culture, and completely aware of what Controllers can do and do when things go wrong, train crew issues notwithstanding, and we have daily shortages to . Knowing what can be done and is done elsewhere does permit me to be critical of GTR. GTR can only have inherited it's controllers etc from the previous TOCs - so then issue lies with upper management policy and interference and limiting what could be done at Ground Zero.