Routeing query - Chippenham to Moreton in Marsh

[whole diary]

[Worcester_Passenger]

I can't get past the non-functional Captcha.

Ditto, using Firefox or Microsoft Edge.

[Robin Summerhill]

For Melksham to Evesham, you also get "no fares found" sometimes unless you put in "not via Reading". Simples.

The same problem been there for years, and it includes a wide range of other places north of Oxford on various lines. I remember helping a customer coming to us on a training course who had a similar problem looking to get a Coventry to Melksham journey.

I think that this relatively innocent query has thrown up some major shortcomings in the information being provided to the public, and is actually very similar to your Melksham queries/ glitches.

To recap, the story so far:

Given my location and the directions trains run in from it (and indeed my previous location of Bristol 40+ years ago), the North Cotswold is not one that I would normally use. Other than the occasional visit to Charlbury to see grandchildren in Ascot underWychwood, I have only used the line throughout once before, and that was in 1976! So my motivation in planning this trip was to travel the whole line – indeed, I might not even get off at Moreton when we get there!

So I start off by checking BR^▸ (British Rail(ways)) Fares to see what it will cost, and (secretly!) whether a circular round trip is possible. So I now find that Any Permitted or Not via Reading routeings are available, but I don’t know from this whether using the Midland line north of Cheltenham would be a permitted route. So I ask here.

BNM reminds me that I haven’t remembered the bleedin’ obvious, to put the journey into the NRE^‡(resolve) search engine and see what routes are offered. So I do, and it tells me that no return fares are available and suggests I buy first class advance singles via Reading for a total of £57.30. But I already know, because I’ve checked BR fares, that I can do the journey for £24.75 with a return ticket, or £25.50 for the via Reading option. So why the devil does it tell me that there are no return fares available? It is not because the system is picking out fastest trains only, because the £25.50 option is valid on those as well.

Then I amend the NRE entry to “avoid Reading” and the lower return fares appear, even though the engine has previously just told me that no return fares were available. But now all the options shown are via Didcot/ Oxford. So now I edit it again, this time specifying a route via Cheltenham. At this point I finally find documentary evidence that the route via Worcester is a valid one

Now imagine trying to get to the bottom of that lot on the TVM^▸ (Ticket Vending Machine) at Melksham!!

Now, I am fortunate in a way in that I know enough about railway routes and fares to be able to carry out this research. But what if Bert and Beryl down the road decide to have a day trip to Moreton with their railcards to visit the weekly craft market? If they relied solely on the NRE website they would pay £114.60 for the pair of them, when other tickets are available on exactly the same trains that would have cost them £51.00

It is very difficult to think of any justification for this at all

[stuving]

It is very difficult to think of any justification for this at all

I think I did find a possible (partly guessed) reason, though that obviously is not the same as a justification.

• There is no permitted route via Reading, so ordinarily the OJP^▸ (Online Journey Planner) would not find any.
• But until recently that route was permitted, and Advance fares are still offered using it. Why? Maybe they are generated in from a separate database, which hasn't had that easement updated? The route code numbers are different, but that may just reflect "GWR^▸ (Great Western Railway) only".
• If Advance fares exist, the OJP has to look for them. Note that they may "exist" even if there are none on offer that day, in the sense of being on a list of advance fares that may be offered or may not or may have run out.
• It is a limitation of the OJP that it finds journeys and then looks for valid fares, and leaves journeys with no fare on its listing. I suspect there are good reasons for this, and trying to eliminate it would introduce other problems just as bad.

You can still say that the routeing rules should not have such complications and contradictions. I'd say it's the result of trying to meet incompatible objectives, and the only solution is to Gordianly simplify them. But a lot of people will not like the result!

[stuving]

So why the devil does it tell me that there are no return fares available? It is not because the system is picking out fastest trains only, because the £25.50 option is valid on those as well.

I think that can be part of this problem too. If the OJP^▸ (Online Journey Planner) does step one of its job - find journeys for each way - by making a list of routeings, it has to include via Reading. At this stage it ignores what fares are offered, it just uses its list. It is set up to never offer journeys that are overtaken, as they are not the fastest. I've tried complaining this should not be an absolute rule, and at least you should be able to turn it off. But they seem convinced it's what we all want, so the more other people complain and say that isn't so the better.

With today's clockface timetables, it's quite possible for all journeys via one routeing to be suppressed by this. That does seem to happen here, and for a lot of days it finds none via Didcot not Reading. But it then can't find a fare, as the only valid fares that way are Advances, and sometime there aren't any - or only first class ones.

So the OJP is a beast of moderate brain, not bad at its job but unable to cope with the flawed data it's fed on.

[ChrisB]

I suspect the journeys via Reading are faster than the ones via Didcot (only) - is there a long wait plus stopping to Oxford, where another change/wait is needed, therefore are supressed by the OJP^▸ (Online Journey Planner)

[Robin Summerhill]

I suspect the journeys via Reading are faster than the ones via Didcot (only) - is there a long wait plus stopping to Oxford, where another change/wait is needed, therefore are supressed by the OJP^▸ (Online Journey Planner)

Northboud via Didcit/ Oxford the connections are not good so the journey time is considerably longer. Southbound, with a 5-minute same-platform connection, either route will connect into the same train at Reading or Didcot

[Bmblbzzz]

If the OJP^▸ (Online Journey Planner) never shows services which are overtaken, what does the "fastest trains only" button do?

[stuving]

If the OJP^▸ (Online Journey Planner) never shows services which are overtaken, what does the "fastest trains only" button do?

Nothing, I guess. That was one of the things I asked them, but it went unanswered.Maybe it's worth asking as a single question on its own.

Firstly, I would like to inform you that the functionality of our Journey Planner is to display the fastest route for any journey planned.

[Robin Summerhill]

Just in case my earlier query returned freak results I tried it again in the NRE^‡(resolve) website this afternoon. The parameters I entered were leaving CPM^» (Chippenham - next trains) after 1000 and returning from MIM after 1600. I did not specify a route. Before I started I hovered my cursor over the “i” button next to fastest trains only. The message I got was “Excludes services which are overtaken by other trains. The fastest services are not necessarily the cheapest” That button remained unchecked

I am still being told that no return tickets are available and the best price I can get is with two standard class singles for £38.00. Various other ticket options are available but they are all singles, not returns

All of the outbound services show one change at Reading. Rather intriguingly, all of the return journeys show two changes at Oxford and Didcot.

I have taken screen dumps of all this in case I say something that upsets NRE and they want to sue me. I would rather welcome the publicity. They, however, would not...

The quality of this information is appalling. We know from the BR^▸ (British Rail(ways)) fares website that return fares are available, anytime, off peak and super off peak. We also know that the off peak and super off peak fares are cheaper than the “cheapest price” NRE are offering, and we also know that they are valid on all the trains identified on their list of services.

The NRE website warns its visitors that the fastest trains are not necessarily the cheapest, and then appears to ignore its own advice by not showing slower, cheaper services

Now I very much doubt, or at least I fervently hope, NRE have not set out with the deliberate intention of telling lies to its users. Nevertheless, it is clear from the results that the way the search engine is currently configured “behind the scenes” is having exactly that result. Others on this thread have put forward reasons why this may be happening, but have also said that explanation does not equate to justification. I very much agree with that sentiment

We have also been told on this thread that the system is set up to never offer journeys that are overtaken, as they are not the fastest, when the website appears to contradict this with its advice about selecting fastest journeys only. Even worse, we have been told that this has already been complained about and not acted upon on the grounds that “this is what the customer wants.” Perhaps if the “customer” was fully informed about the implications they might “want” something else...

We have been told by Graham that similar issues have arisen at Melksham with travel to or from north of Oxford so it is clearly not a one-off that just affects Chippenham and Moreton in the Marsh

NRE should rectify this as a matter of urgency

[stuving]

That's the same as I was describing, Robin. Having Advance tickets offered over what is not a permitted route is more than the OJP^▸ (Online Journey Planner) can cope with. Is it meant to be possible or not? I have no idea.

Have you tried the cheap fare finder? Using both can be helpful, especially with that useless "show only fastest trains" selector. Obviously it's the same engine, just given different metrics. The link down at the foot of the page preserves the main journey settings.

If you do try that, you will see that in this case the OJP's behaviour is even sillier than we thought. For the outward leg, rather than a single change at Reading, it says to change at Didcot and Oxford, as for the return. But the timings northbound are such that the stopper gets you to Oxford in time to get on the same train as you'd have got at Reading!

So this, valid, route has exactly the same journey time as the invalid one via Reading. I can only guess that the "never knowingly overtaken" rule is backed up by a tiebreaker rule to choose fewer changes over more. If two routes have the same number of changes ... does it flip a coin? Perhaps, or quite possibly crash.

Anyway, I've banged in an "inquiry" or two and we'll see what they say. Do, please, do the same; they may weigh these inputs to see if they can be ignored.

[Trowres]

I'm sure I've read somewhere that the Great British Passenger is eagerly anticipating the arrival of tap-in, tap-out ticketing, where the system retrospectively calculates the best fare for the journey.

Good luck with that.

And train operators are under pressure to reduce manning at stations to cut costs. Who needs good advice from a booking office?

[grahame]

Who needs good advice from a booking office?

I suspect that's a question you are asking without expecting an answer.

There is (or was under franchises) a requirement on ticket office staff to offer the best value ticket for the journey requested - so that (for example) someone walking up to the ticket office at Chippenham and asking for a day return to London, on a Saturday, would be sold a super off peak return, rather than an off peak return or an anytime return. It is limited in that the ticket office staff are not required (and indeed not encouraged) to offer ticket combinations - so for example "return to London" in Monday morning peak will result in you being sold an anytime return, even if you're not coming back in a peak and two singles would be cheaper. If you're coming back later the same day on a long distance trip where no day return is offered, it could be cheaper to but two day returns for two "legs" of the journey - in our example the classic is Chippenham to Didcot and Didcot to London, but again the ticker office is not required to offer you these.

That requirement, partial though it is does not, as I understand it, extend to web sites or ticket vending machines. Not sure how it applies to touch in and out systems.  So no requirement to have a full range of tickets on a TVM^▸ (Ticket Vending Machine), nor for it to easily offer you the best value even to the ticket office staff standard.

Passenger need good advice from ticket staff - however, you could argue that as far as the rail industry is concerned, they are something of a an enemy within to the financial model, directing people who are willing to pay higher fares toward the lower priced tickets.  In my view, that is a very shortsighted argument ...

[Robin Summerhill]

There is (or was under franchises) a requirement on ticket office staff to offer the best value ticket for the journey requested

.......
 
That requirement, partial though it is does not, as I understand it, extend to web sites or ticket vending machines.

OK, but in this example we have the NRE^‡(resolve) website telling intending passengers that no return fares are available, whilst return fares are actually available and at a lower price than the one the NRE engine is offering.

Is that "within the rules?" Because if it is the rules need chaging as a matter of urgency

[IndustryInsider]

Agree 100%.  One for Transport Focus, Robin?  Barry Doe might be interested as well as he often criticizes the online journey planners.

[stuving]

I have a reply to my first question last night, which was:

You told me, when answering a previous inquiry, this:
"Firstly, I would like to inform you that the functionality of our Journey Planner is to display the fastest route for any journey planned."

So why is there a check box labelled "Show only fastest trains"? Does it do anything? Apart from misleading users, of course - it clearly does that!

This is the reply:

Thanks for taking the time to email us. I was sorry to learn that you find our journey planner to be misleading. I understand that you would like to know the purpose of the check box "Show only fastest trains" on our journey planner. I further understand that our journey planner already shows the fastest possible route.

I would like to confirm that our journey planner certainly displays the fastest possible route between two stations; however, at times there are longer journeys which do get displayed because we offer passengers choice of travel options that are possible. This in most cases would be a direct train which would take a longer duration than a train with one change at one of the intermediate stations. So by changing trains one can arrive at the destination station earlier. If you check the box labeled "Show only fastest trains" the slow direct train will be removed and the journey with one change will continue to be displayed, as it is a faster journey.

I trust I have answered your query, and I hope you have a lovely day.

Thanks again for contacting us. Please let me know if I can be of any further assistance.

So it is, as I suspected, something to do with always including direct, or in this case more direct (fewer changes) journeys. The "Show only fastest trains" options would, if that reply is correct, suppress these more direct options if they were not in fact quicker. That does not help for this aberrant CPM^» (Chippenham - next trains)-MIM case because the one-change option via Reading is not slower, it's exactly as fast.

Note the dubious claim that "we offer passengers choice of travel options that are possible". I think that works out to not happen where fast trains are frequent enough that any slower train is overtaken by a faster one.