Why do TOCs lie about delay times?

[MKB]

I've had a few delay repay compensation payments over the past year. One thing I've noticed, particularly with Cross Country trains, is that the letter they send with the compensation nearly always understates the actual delay by a minute or two.

So far, this has not made any difference to my compensation, but it came close today. A train I was on pulled to a stop in Leeds 32 minutes and 10 seconds late. The doors opened 4-5 seconds later. XC did pay up, but felt the need to point out that the train was in fact 31 minutes late. This is untrue. I know for sure my timings are accurate to the second. A secondary timing source -- the station clocks -- disagreed with me by no more than 1 second.

If this was a 30.5-minute delay that the TOC claimed to be 29.5 minutes, this would be significant. It seems clear to me that someone is massaging the figures by the odd minute or two in the belief that they can get away with it. Who is vetting the TOC's timings?

 

[Anvil1984]

The train running systems don't show when the train doors open / close, it works off a timing point which is usually just outside the station, if you go somewhere like York, you will notice the screen changes to "Arrived" well before it actually has, so its not down to the TOCs lying per se, just the information they have

 

[Tomnick]

Presumably it's all driven off the train describers - and the train description will step into the TD berth in the platform as soon as it passes the last signal in rear. You'd have to go to great lengths to do otherwise though...!

 

[The Planner]

You usually have a berth off-set though, so it hits the last signal section and a pre-determined time is added or taken off to allow for the train to stop or to pass a junction etc....

 

[radamfi]

The train running systems don't show when the train doors open / close, it works off a timing point which is usually just outside the station, if you go somewhere like York, you will notice the screen changes to "Arrived" well before it actually has, so its not down to the TOCs lying per se, just the information they have

But they are effectively lying if the TOCs are aware of this loophole.

 

[Tom B]

But they are effectively lying if the TOCs are aware of this loophole.

Quite; the train hasn't arrived (from the point of view of a passenger) until it is in the platform and the doors have been unlocked.

 

[moonrakerz]

Why use such an emotive term as "lying" ?

There could be a multitude of perfectly good reasons why the timings used by the TOC would vary from those used by a passenger with a stopwatch.
I've often heard lots of people muttering about an aircraft flight being "late" - these timings are always from "doors shut" to "doors open", not from when you leave the ground or arrive at the airport or..................

If people feel that strongly perhaps they should write to the TOCs concerned and accuse them of "lying" in writing - oh ! perhaps not

 

[radamfi]

Why use such an emotive term as "lying" ?

It wouldn't matter if TOCs weren't boasting about how punctual they are, or if compensation didn't depend on how late trains were. It is clear that the TOCs have a vested interest in exploiting such loopholes as leaving early, timetable padding and pretending you have arrived even though you are outside the station.

 

[Tomnick]

You usually have a berth off-set though, so it hits the last signal section and a pre-determined time is added or taken off to allow for the train to stop or to pass a junction etc....

Thanks for that, another little bit of education which really proves that there's no element of 'fiddling' the figures - evidently delays are being reported as accurately as could be expected.

Just another thought - are delays (in the context being discussed here) rounded down to the full minute? That'd explain how a 32'10" delay might be recorded as 31' - rather than a 'lie' in excess of one full minute, it'd be a matter of 10 seconds or so bringing the recorded figure under 32'.

 

[jon0844]

If the TOC is using the time measured at a track circuit outside the station, they should surely be adding on a set, agreed, time to allow for it to actually arrive at the platform and open the doors. Sure, it might still not be accurate - but it will be more so.

Say, 30 seconds or 45 seconds, which sounds reasonable. No train is going to be able to go so fast that it arrives earlier, causing a later time to be recorded than it actually is.

But, imagine how many borderline cases there must be at the moment where such an add-on would tip the balance and see many more trains registering as late.

 

[embers25]

Rounding down if it does happen is wrong as well as thats fixing the figures to make them less late. How woud you feel if you went to the shop for a 1p sweet and gave the cashier a £1 coin but got no change as the change was rounded down to the nearest pound!!! These performance figures are blatantly fixed in the interests purely of the TOCs and not in the interests of passengers. Why can't the have sensors on the platforms the pick up arrivals rather than outside stations. It's not rocket science but then again that would not be in the TIC's interest as they'd be late more often. I for one do not think Lying is too harsh a term as clearly all these performance figures are massaged and small difference do have big effects just ask the poor suffering Southeastern customers who missed out on a season ticket discout by 0.1% which given the cheating TOC's are using clearly shows that in actual fact Southeastern's unmassaged figures were less than the discount threshold and so discounts should have been given.
In summary:

Trains leave up to 2 minutes early due to early door closing
Trains have up to 10 minutes recovery time added at their destination station
Trains are considered arrived despite not even being in the station

In addition a train can be 9.5 (I think) minutes late before its even classed as delayed and thats based on the trains working timetable which could have yet another 2 or 3 mniutes worth of recovery time in it.

This is farcical and blatantly wrong and rips off passengers as the train may be 30 minutes late but in reality due to all this recovery time it may only be 15 minutes late meaning no compensation. No other country operates this level of fiddling and so yes the figure are lies and they are costing all passengers money in lost delay repay and season ticket discounts.

 

[barrykas]

This is farcical and blatantly wrong and rips off passengers as the train may be 30 minutes late but in reality due to all this recovery time it may only be 15 minutes late meaning no compensation. No other country operates this level of fiddling and so yes the figure are lies and they are costing all passengers money in lost delay repay and season ticket discounts.

You're right. No other country operates this level of "fiddling", they build the recovery time into each part of the journey instead.

Which is how I managed to do Frankfurt - Nuremberg by ICE a couple of years back at an average speed of 75mph and still be early at every intermediate station...

Cheers,

Barry

 

[VideozVideoz]

I've had a few delay repay compensation payments over the past year. One thing I've noticed, particularly with Cross Country trains, is that the letter they send with the compensation nearly always understates the actual delay by a minute or two.

So far, this has not made any difference to my compensation, but it came close today. A train I was on pulled to a stop in Leeds 32 minutes and 10 seconds late. The doors opened 4-5 seconds later. XC did pay up, but felt the need to point out that the train was in fact 31 minutes late. This is untrue. I know for sure my timings are accurate to the second. A secondary timing source -- the station clocks -- disagreed with me by no more than 1 second.

If this was a 30.5-minute delay that the TOC claimed to be 29.5 minutes, this would be significant. It seems clear to me that someone is massaging the figures by the odd minute or two in the belief that they can get away with it. Who is vetting the TOC's timings?

My recent experience with XC compensation... I claimed a 55min delay (unfortunately I couldn't find a way of making it over 1hour for 100% refund). After submitting my claim form, they wrote back to me advising I had incurred a delay of 90 mins??? I'm not complaining but if I told them I was delayed by 55mins that's amazing they changed it to 90mins and paid out more...

 

[The Planner]

TRUST is part of the issue here, it will always round down even at xx:59 seconds. So if a WTT schedule is timed on a half minute, and gets timed at xx:00 TRUST will see it as 1 late even though it is only 30 seconds.

 

[313103]

I dont know why you are so shocked about this, it is common knowledge this goes on even the BBC and Channel 4 have picked up on this. It is all about Boxes being ticked. Two groups of boxes apply in my company. PPM (Public Performance Measure) and KPI (Key Performance Indicators). As a member of staff i dont agree with it. It looks poor when after every five to six stations you have to wait in order to keep time. The jouney time on my route is now slower then when i started working on it in 1993. End to end was 57 minutes in 1993 in 2011 it is an average 61.

 

[WelshBluebird]

you will notice the screen changes to "Arrived" well before it actually has

That is very annoying.

There have been quite a few times where the screen has said "arrived" but the train has still taken a further 10 to 20 minutes to turn up.
Even worse is when the train is clearly delayed, yet the screens still say "on time".
Or what about when trains which are yet to arrive just disappear off the screens (although I am guessing that is down to the system assuming the service left on time).

 

[MKB]

Why use such an emotive term as "lying" ?

You're right - I was in a bad mood last night. Not helped by opening the letter from XC telling me (incorrectly) that my claimed timings were wrong. Given it made no difference in this instance, it would have been good grace simply to apologise and pay up.

That said, sometimes being deliberately contentious is a better catalyst for sparking a debate.

 

[Tomnick]

Why can't the have sensors on the platforms the pick up arrivals rather than outside stations. It's not rocket science but then again that would not be in the TIC's interest as they'd be late more often.

It might not be rocket science but it's certainly not easy either. The reporting systems currently in use, as described earlier, rely on the train describers - part of the signalling system - to determine when a train is approaching the platform. Again, as described earlier, the reporting system then compensates, by adding a predetermined value, for the fact that the train's not yet actually in the platform. Given that that should be pretty accurate for the most part (give or take a few seconds - the train will have passed the last signal at this point so won't be delayed further awaiting a platform or for a conflicting move), that would appear to be sufficient. In this case, a possibly small discrepancy may have been exaggerated by the unfortunate (but not deliberate, I'm sure!) rounding-down feature of TRUST.

 

[AlterEgo]

It's hardly a fiddle.....we all know TRUST isn't gospel, so if you train was 58 minutes late, we round it to an hour for Delay Repay purposes.

 

[ChrisCooper]

Surely the main thing is that they payup.

The only time I can see this being a problem is with connections if TOCs were strict about the times. I doubt they are though because they could not get away with it. Imagine if someone missed their connection because either their inbound service was late or the outbound left early, or both. They go to the ticket office to get their AP ticket changed for a later train and are told that it can't be done as the records show that the inbound train was ontime, and/or the outbound train was ontime and therefore they had more than enough time to make the connection and it's their fault they did not, therefore payup to get the ticket changed to the cheapest available on the later train plus an admin fee. The TOC then would be implying that the passenger way lying about the times the trains arrived and departed as it contradicts what is recorded, so the customer would have perfect right to say the TOC was lying about the times. If the customer took it to court the TOC would be obliged to come out with the truth that times were rounded and not representative of what really matters to the passenger (door open on arrival to door close on departure) and no judge would then take their evidance as anything more than it was, a load of rubbish (in context). The TOC would be paying the customer their money back (for the upgrade at least), plus costs.

 

[Minilad]

I am finding myself wondering more and more about how important all these little minutes really are.
Are we that much in a hurry everywhere that one minute makes that much difference.
Me personally, and I think the vast majority of the travelling public would rather have a reliable service than one operating on the edge just to save a few minutes.

OK here's a question
Lets say a journey from A to B done at full line speed everywhere and operating with no slack or recovery time at all would take 60 mins
Would you rather have the train timetabled to take 60 mins and be late 3 times a week
Or would you rather it be timetabled for 65 mins and be on time every time.

To me some of the comments made on this thread, and the thread about not running at 125mph are made by people who want the railway to run in some sort of fantasy world where nothing goes wrong and everything is able to run smoothly all the time.
For example the people who say the journey now take longer than it did in 1992 or whenever. Well yes, there are more trains running on the same infastructure so of course times will need to reflect that as bottlenecks are developing.

Anyway I can't go on for ever I got something important to do. Like counting up all the little minutes and collating them in a big spreadsheet, then cross checking them against the times I recorded on my stopwatch to make sure no one tries to do me out of any of those oh so important minutes

 

[WelshBluebird]

I am finding myself wondering more and more about how important all these little minutes really are

As I said in the other thread:
If you are trying to make a connection (be it train or bus or whatever), then those little minutes are very important.
It may only be a minute, but that minute can easily mean an extra hour or two.

 

[Tomnick]

Which surely comes back to XCDriver's point above - is it better to shave a few of those little minutes off journey times at the cost of a delayed arrival (and missed connection) 4 days out of 5, or to allow a sensible amount of recovery time to provide reliable connections?

 

[embers25]

Most journeys are now slower than they were years ago due to this extra slack in the timetable and yet they still want that extra 30 seconds for the doors. Also if the trains were on time nearly all the time (which isn't fantasy as it happens in Japan and most of the time in Switzerland) then maybe we could tolerate the extra slack as we'd make all connections. Unfortunately as XC illustrate so well no matter how much excessive slack you add they'll still run late (not always their fault I hasten to add) and miss connections and those connections won't wait. Also, as was illustrated on Sunday, people got very grouchy at waiting around for time and if you are getting off at the next stop you don't want to wait 2-10 minutes at the previous station as you want to get home or make that tight connection that you should make but you don't because after waiting 10 minutes you then get delayed on the approach to that next station and have the pleasure of watching your next train sail out!

Clearly we are all going to disagree on this issue although this thread has highlighted clearly "some" of the negative attitudes held by "some" rail staff...definitely not customer focussed and more driven by the desire to meet targets.

The more these issues are explored the more its plainly clear that ATOC and its members will stop at nothing to rip off, confuse and deceive their customers:

1. Simple fares are anything but simple.
2. The ATOC routing guide is a national joke that even ATOC refuse to adhere to without a fight.
3. Performance stats don't measure the actual performance and are heavily weighted in the TOC's favour.
4. The public timetable isn't the timetable used to measure delays.
5. The departure time can be up to 2 minutes before the stated departure time.
6. Many have stated that the departure time should be when the train starts moving and yet the arrival time is a random time before the train stops moving and certainly doesn't properly take into account the 30+ seconds to open the doors that is allowed for on departure.

I realise these companies have to make money but is their a more corrupt non-passenger orientated rail system anywhere else.

The scale of the overall effect of all these "fixes" to the figures in terms of money saved by the TOC's in compensation payments is not known but it certainly isn't small as Southeastern carefully proved with their seemingly massaged performance figure. Transparency of any aspect of the railway is non-existant and passengers deserve better.

1 minute is not a lot but it can make a huge difference and in extreme cases can be the difference between life and death so it is by no means insignificant. Also if you think 30seconds matters that little then wait that extra unimportant 30 seconds and shut the doors at the scheduled departure time, as someone that does value that 30 seconds and couldn't get there any earlier for whatever reason might just be trying to catch your train.

 

[Tomnick]

As stated numerous times now, the recorded arrival time isn't random: it's estimated by taking the time at which the train passes the last signal in rear of the station and adding the time normally taken for the train to come to a stand in the platform. That's as good as you're reasonably going to get, and that won't ever account for the time taken to reach different stopping positions or a slower approach to an occupied platform.

It's important that the train starts to move at the booked time as it's not the only train on the railway! That extra half-minute in the platform could well put an extra half-minute into a conflicting move over the station throat - or later in the journey - and cause passengers on that train to miss connections. Likewise, the train waiting time at its penultimate stop then standing outside the final station probably wouldn't arrive any earlier if it didn't wait time there, if it's waiting for its booked platform to become clear. Whether the public departure time should be a minute or two earlier than WTT at certain major stations is a separate debate (it's certainly sensible, and I'd argue that passengers should be allowed to board right up to advertised departure time in that case, with platform duties completed to allow right time departure to WTT times).

 

[radamfi]

It's important that the train starts to move at the booked time as it's not the only train on the railway! That extra half-minute in the platform could well put an extra half-minute into a conflicting move over the station throat - or later in the journey - and cause passengers on that train to miss connections. Likewise, the train waiting time at its penultimate stop then standing outside the final station probably wouldn't arrive any earlier if it didn't wait time there, if it's waiting for its booked platform to become clear. Whether the public departure time should be a minute or two earlier than WTT at certain major stations is a separate debate (it's certainly sensible, and I'd argue that passengers should be allowed to board right up to advertised departure time in that case, with platform duties completed to allow right time departure to WTT times).

I think there is a strong case for the WTT to be 30-60 seconds later than the public departure time. Then trains leave on time according to the WTT and no one misses the train. Everyone is happy!

I'm not convinced that 30 seconds is that critical in most cases. The punctuality conscious Swiss are prepared to hold connections for a few minutes if necessary so if they are willing to do that then it can't be that important.

 

[WelshBluebird]

I'm not convinced that 30 seconds is that critical in most cases. The punctuality conscious Swiss are prepared to hold connections for a few minutes if necessary so if they are willing to do that then it can't be that important.

And there have been many times I've been on services where connections have been held for people in this country. I suppose it depends on the line and time. The last train from Exeter to Barnstaple for example could be held by a few minutes if needed because it wouldn't really affect any other service.

 

[Tomnick]

It can be important - a TPE fast train through the Hope Valley, to take a recently-discussed example, will end up stuck behind the stopper if it's held for more than a couple of minutes. If the stopper's held, then it'll delay the folllowing EMT fast which will catch it by New Mills South. There's perhaps not so much of an issue on a branch line with a decent turnaround at the far end and no booked crossings of anything else on a single line, but even then, in a world where connecting buses won't wait, there's an issue there potentially. I do think there's too much of a reluctance to hold connections where it might be appropriate to do so though.

On topic though, I'd always understood that this was the main reason for an earlier PTT departure than WTT, so it would seem sensible to apply that to remove the potential for confusion and missed trains.

 

[DavidBusDriver]

""On topic though, I'd always understood that this was the main reason for an earlier PTT departure than WTT, so it would seem sensible to apply that to remove the potential for confusion and missed trains.""

Oh but then, cue the threads of 'My train was booked to leave at 11:04, but it actually left at 11:05..."!

 

[radamfi]

To create a proper interconnecting network there needs to be adequate provision for stopping trains to be held to allow fast trains to pass. Situations like the Hope Valley example above need passing places to be constructed, ideally at stations. It is highly undesirable to have such tiny margins for error.