Best measures for "good" value rail fares

[MaurieBunde]

Hello,

I hope this is ok to post here.

I am doing a data science project in which I must devise a method of finding "good" value rail fares. I was wondering if anyone in this forum had any interesting metrics for measuring how good value a rail fare is?

So far I have a few of my own: price per mile as the crow flies, price per mile of track, price per hour of journey and price per hour saved from travelling by car.

If anyone has more suggestions of metrics or criteria I should consider I would really appreciate it.

Thank you.

 

[miklcct]

I will say:

(Price + time * value of time) / sqrt(straight line distance + some measure of natural barriers such as mountains and sea)

 

[yorkie]

Hello,

I hope this is ok to post here.

I am doing a data science project in which I must devise a method of finding "good" value rail fares. I was wondering if anyone in this forum had any interesting metrics for measuring how good value a rail fare is?

So far I have a few of my own: price per mile as the crow flies, price per mile of track, price per hour of journey and price per hour saved from travelling by car.

If anyone has more suggestions of metrics or criteria I should consider I would really appreciate it.

Thank you.

Are you talking about walk up fares or advance fares?

It may be worth you joining us at a forum event (e.g. walk / meal / fares workshop) to talk about this in person.

 

[MaurieBunde]

Are you talking about walk up fares or advance fares?

It may be worth you joining us at a forum event (e.g. walk / meal / fares workshop) to talk about this in person.

Thank you for your reply.

For now, walk up fares. I am still at the planning stage of the project and sizing up the job. It's possible that including advance fares will be too much. I am thinking of writing a program where the user inputs their location and can find the best value fares available to them according to various metrics.

When and where are the next forum events?

 

[SickyNicky]

It's definitely worth going to an event. There are people here who can help in a number of ways, including my company, and I may well be there. I'm sure yorkie will give you a list of dates.

 

[alistairlees]

Are you aware of http://faremap.ml/?

 

[MaurieBunde]

Are you aware of http://faremap.ml/?

I was not! Thank you very much for this. It's very useful as I'm going to look into what people have done on this topic already.

 

[spyinthesky]

I would’ve thought that anyone that has time to look up value fares won’t be very interested in walk up ones.

 

[miklcct]

I would’ve thought that anyone that has time to look up value fares won’t be very interested in walk up ones.

Not really. If I ain't sure which train I need to take I'll need to use walk up fares.

 

[AlterEgo]

Hello,

I hope this is ok to post here.

I am doing a data science project in which I must devise a method of finding "good" value rail fares. I was wondering if anyone in this forum had any interesting metrics for measuring how good value a rail fare is?

So far I have a few of my own: price per mile as the crow flies, price per mile of track, price per hour of journey and price per hour saved from travelling by car.

If anyone has more suggestions of metrics or criteria I should consider I would really appreciate it.

Thank you.

Price per average speed is a good metric. Price per hour and per mile don't really reflect value in a meaningful way.

 

[mmh]

Price per average speed is a good metric. Price per hour and per mile don't really reflect value in a meaningful way.

Price per mile is a very good metric! Some routes are faster than others, of course they are. I really don't understand why speed should be the main consideration here.

 

[Ken H]

as you are doing walk up fares you have to consider time of day and day of the week 9weekend/not weekend) - the walkup fare can vary as walk up cheap day returns become available. They may be good value but of no use if you have to be somewhere by a certain time.

 

[Watershed]

The 'generalised journey time' (particularly in comparison with the car or plane) is one factor which certainly warrants consideration. It's no use having a really cheap walkup ticket if the frequency is so poor that it's essentially an Advance, and a ticket being cheap is of little solace if it requires an extremely slow and roundabout route.

 

[mmh]

The 'generalised journey time' (particularly in comparison with the car or plane) is one factor which certainly warrants consideration. It's no use having a really cheap walkup ticket if the frequency is so poor that it's essentially an Advance, and a ticket being cheap is of little solace if it requires an extremely slow and roundabout route.

That isn't what happens in practice though. Do you have any examples of such really cheap walk up fares on infrequent routes?

It is true, as you say, that there are many journeys where an "Anytime" ticket is effectively no different to an advance.

 

[AlterEgo]

Price per mile is a very good metric! Some routes are faster than others, of course they are. I really don't understand why speed should be the main consideration here.

Which is a better value rail fare? £50 single London to Manchester with Avanti in 2h12m, or £40 single with, say, LNR and CrossCountry, taking nearly four hours?

Which is better value? £46.60 Folkestone to London with plus High Speed, or £39.80 avoiding the HS lines? 54 min an 94 min respectively.

Etc, etc.

Clearly somewhere with a rapid intercity service is going to provide better value for the same fare than a slower, regional route.

 

[alistairlees]

Shrewsbury to Swansea (via the Heart of Wales line) has many Advances for £20.50, while the walk up fare is £30.80.

 

[mmh]

Shrewsbury to Swansea (via the Heart of Wales line) has many Advances for £20.50, while the walk up fare is £30.80.

TFW have many routes where the advance is pointless, given that most people want to return.

They also have journeys where a shorter journey has advances, but a longer one doesn't, and others where day returns are available on longer journeys than those that aren't.

Their ticketing is, like most things about them other than the excellent staff who have to put up with their incompetence, a shambles.

 

[alistairlees]

Sure.

But my statement:

Shrewsbury to Swansea (via the Heart of Wales line) has many Advances for £20.50, while the walk up fare is £30.80.

was a response to your question:

That isn't what happens in practice though. Do you have any examples of such really cheap walk up fares on infrequent routes?

 

[mmh]

Sure.

But my statement:


was a response to your question:

Thank you for proving my point. Advances which are more expensive than half the walk up return fare are pointless and shouldn't exist. Sadly I imagine lots of people buy them, and that's why the exist

 

[Watershed]

Ha, that's the example I was about to give!

The heavily reduced ScotRail timetable currently in operation gives rise to several more examples - such as Glasgow to Mallaig, for which there is an Off-Peak Day Return (naturally, valid at any time!).

Even with the normal timetable, there are only two trains in each direction which you could use to make a day trip. In the current timetable that's reduced to two westbound trains and one eastbound train (with a change onto the Sleeper at Fort William, seat reservations permitting).

Thank you for proving my point. Advances which are more expensive than half the walk up return fare are pointless and shouldn't exist. Sadly I imagine lots of people buy them, and that's why the exist

The fare quoted is a day return, so if you are making a period return - as is likely to be the case on a journey of this length and with such a low frequency - two Advances can still be the cheapest option. I note that the equivalent Off-Peak Return costs nearly twice as much as an Anytime Day Return.

Another example would be the XC only fare between Glasgow and Edinburgh. At £13.70 SDR, it is less than half the price of the interavailable SDR. However, it limits you to just three trains per day in each direction. That said, they are reasonably well timed for a commute or day out from Glasgow to Edinburgh and indeed there's also a season ticket available, costing 40% less than the interavailable equivalent.

 

[miklcct]

Advances which are more expensive than half the walk up return fare are pointless and shouldn't exist.

But they are still cheaper than walk up single.

I sometimes travel on such Advance tickets on one-way journey, especially on a longer journey where the return is by plane / on a different route.

Advances which are more expensive than the walk up single fare are pointless are shouldn't exist.

 

[30907]

What is the purpose? Comparing different rail routes, or rail vs road?

If you use mileage, "as the crow flies" isn't relevant, and I would tend to favour using road mileage as motoring costs are more obviously mileage based.
(The question is then, which route? - a wellknown map program will offer me three routes from Shipley to York, two reasonably direct but one via motorway which is at least as fast! So perhaps using rail mileage is easier after all...)

 

[miklcct]

If you use mileage, "as the crow flies" isn't relevant, and I would tend to favour using road mileage as motoring costs are more obviously mileage based.

I normally use "as the crow flies" as it's a single, well-defined measure which can't be manipulated to anyone's advantage.

Road mileages can be exaggerated by using motorways which makes a long detour but nevertheless faster than the shortest route. Also, if there are natural barriers which can't be crossed by a car (such as a sea with ferry services where passengers use them to cross the sea), road mileages are exaggerated as well. Nonetheless road mileages are still useful to compare motoring costs, but I am more interested in the true efficiency to get from A to B as a measure of how good the transport network is, instead of comparing it to using a car / taxi.

For example, if A and B, 2 km apart as the crow flies are separated by a hill which there aren't any direct tunnels, but the road distance is 10 km and the rail distance is similar, the fastest way may be to hike across the hill rendering neither rail nor road travel good value.

 

[Ken H]

I normally use "as the crow flies" as it's a single, well-defined measure which can't be manipulated to anyone's advantage.

Road mileages can be exaggerated by using motorways which makes a long detour but nevertheless faster than the shortest route. Also, if there are natural barriers which can't be crossed by a car (such as a sea with ferry services where passengers use them to cross the sea), road mileages are exaggerated as well. Nonetheless road mileages are still useful to compare motoring costs, but I am more interested in the true efficiency to get from A to B as a measure of how good the transport network is, instead of comparing it to using a car / taxi.

For example, if A and B, 2 km apart as the crow flies are separated by a hill which there aren't any direct tunnels, but the road distance is 10 km and the rail distance is similar, the fastest way may be to hike across the hill rendering neither rail nor road travel good value.

Crow flies distances can be a bit rubbish in the UK. They bring up silly distances.
Barrow in Furness - Lancaster
Southport -Llandudno
Cardiff - Plymouth.
Southend - Ramsgate.
Stranraer - Workington

All are closeish but there are esturies in the way. So unless there is a ferry or air service you have to go rond.

 

[miklcct]

Crow flies distances can be a bit rubbish in the UK. They bring up silly distances.
Barrow in Furness - Lancaster
Southport -Llandudno
Cardiff - Plymouth.
Southend - Ramsgate.
Stranraer - Workington

All are closeish but there are esturies in the way. So unless there is a ferry or air service you have to go rond.

The point I'm making is that the lack of ferry service make the transport network there bad. If there is a ferry/ bridge / tunnel the total generalised cost can be reduced to be comparable levels to elsewhere similar in straight line distance.

The comparison is not made between different modes of transport at a location. It is made between different towns and cities to see where has better transport.

 

[Haywain]

Cardiff - Plymouth.

Or Barnstaple to Swansea!

 

[185143]

Crow flies distances can be a bit rubbish in the UK. They bring up silly distances.
Barrow in Furness - Lancaster
Southport -Llandudno
Cardiff - Plymouth.
Southend - Ramsgate.
Stranraer - Workington

All are closeish but there are esturies in the way. So unless there is a ferry or air service you have to go rond.

Had to laugh recently at a well known hotel booking site listing nearby attractions to my Hotel in Weston-Super-Mare in Penarth recently!

 

[Ken H]

Had to laugh recently at a well known hotel booking site listing nearby attractions to my Hotel in Weston-Super-Mare in Penarth recently!

There is a commonly used api that finds nearest by postcode. It ignores estuaries so gives stupid results.

 

[jupiter]

You are looking at station-station cost though? What’s often more relevant to an individual is the total cost of home-station-station-destination.

I don’t live within (sensible) walking distance of a station so need to factor in either car travel and parking, the rather sporadic public transport or taxis for four segments of the journey when comparing with the cost of simply jumping in the car.

 

[MaurieBunde]

Not really. If I ain't sure which train I need to take I'll need to use walk up fares.

I was thinking of developing something to be used in a scenario where a person is in a, possibly unfamiliar, town/city and can easily find out the best rail fares according to certain metrics. Would this be at all interesting or useful in your opinion?

The 'generalised journey time' (particularly in comparison with the car or plane) is one factor which certainly warrants consideration. It's no use having a really cheap walkup ticket if the frequency is so poor that it's essentially an Advance, and a ticket being cheap is of little solace if it requires an extremely slow and roundabout route.

Can you explain this a little more, perhaps with an example? Do you mean I should do a comparison of the cost of the total time of the journey? So if I buy a walkup ticket, but the train doesn't arrive for 3 hours, that 3 hours should be included?