Cotswold Line redoubling: 2008 - 2011

[whole diary]

[Burty76]

Dont think we're ever going to agree on this willc but I do take your points, and at least a reliable hourly service stopping everywhere bar the halts is an improvement.  I really wish FGW^▸ (First Great Western) had kept the Adelantes as with their automatic doors and faster acceleration they could have done Worcester to London in just over 2 hours with the same stopping pattern.  This could have been supplemented by two peak Hereford-London HST^▸ (High Speed Train)'s that missed out Pershore, Honeybourne and Hanborough.  As it is, with HST's having them stop everywhere will mean 2h 15 timings at the best I'd imagine.

You asked about the timings for the direct Worcester-Cheltenham-London trains....I was lookign at the wrong timetable, it was last years December to May TT.  The timings there are :

dep Shrub Hill  17:25
arr Cheltenham 17:42
Depart Cheltenham 17:44
arr Swindon  19:25

So as you can see it would be possible to run Worcester to London in less than 2 hours this way, bearing in mind you can do Swindon-London in under an hour(current timetable shows trains taking 55 minutes with a stop at Reading)

[Btline]

Unless I am missing something, it would be 3 hours. (5.25 - 8.25)

That's longer than the slowest Cotswold service!

[willc]

with HST^▸ (High Speed Train)'s having them stop everywhere will mean 2h 15 timings at the best I'd imagine.

The current morning Cathedrals Express reaches London in 2hrs 13mins from Shrub Hill, inclusive of stops at Pershore and Honeybourne - and the five minutes' extra padding everything between Reading and Paddington has at that time of day.

The latter won't change soon, due to Reading rebuilding, but at present the train has to hit a meet at Evesham with the 5.42 from London and can still be delayed at Ascott by the 6.48 if that is late out of Oxford. The redoubling will make the whole timetabling situation far more flexible between Worcester and Oxford, so I think near two hours is feasible, all the more so if the extra train paths allowed another London through train in the peak (which simply isn't possible now) which could cover, for example, the Honeybourne call.

It may be harder to hit this time in the afternoon/evening, but the issue there is that return travel from London tends to spread out over a longer period, so stopping most trains at most places is a must and what people really value is being able to know there's something going their way leaving Paddington every 30 minutes.

But there's still plenty of scope for reductions on evening journeys too, again by removing pathing issues - eg at present, the 17.51 loses six minutes sitting at Oxford, four at Moreton and five at Evesham - all due to the single line sections - and is allowed 15 minutes Oxford-Charlbury, which should take 12mins, and then gets 18mins over a normally 12-minute run from Pershore to Shrub Hill. Lose all that and you're on 2hrs 8mins, missing Hanborough and Honeybourne, but calling at Shipton.

PS: I'm assuming you meant 18.25 for Swindon - but did it make any intermediate calls between Cheltenham and Swindon? Bypassing Gloucester would account for 10 minutes or so, but I can't see where the rest of the reduction comes from, compared with a typical run between Cheltenham and Swindon of just over an hour.

[Burty76]

Yes I meant 6.25 at Swindon, sorry. 
That train is non-stop between Cheltenham and Swindon, which is what id propose if this was a regular service.  Gloucester and the other stations on that line are currently served hourly by a London train or a Cheltenham - Swindon stopper.

In theory HST^▸ (High Speed Train) timings should or could be something like this following redoubling :

Paddington  12:51
Slough  13:06
Reading   13:21
Oxford arr 13:45
          dep 13:47
Hanborough  13:56
Charlbury  14:04
Kingham  14:13
Moreton  14:21
Honeybourne  14:32
Evesham  14:39
Pershore  14:47
Shrub Hill  14:58

and going the other way....

Shrub Hill  14:23
Pershore  14:32
Evesham  14:40
Honeybourne  14:47
Moreton  14:58
Kingham  15:06
Charlbury  15:16
Hanborough  15:24
Oxford  arr  15:34
           dep  15:37
Reading  16:00
Slough  16:14
London  16:32

So theyd pass at Evesham and between Kingham and Charlbury, and fit into existing Oxford-London paths.  Is that do-able?

[IndustryInsider]

Appreciating (even if I don't personally agree with all of them) the points made by Andy, Burty, Btline, and others. I usually find myself agreeing with WillC's standpoint in most regards.  What might appear nice in principle just doesn't make sense commercially for FGW^▸ (First Great Western). And if it doesn't then it simply won't happen.
 
I shall be interested to see what you all make of my little timetable experiment. What little hair I had left is in serious danger of being ripped out. Squeezing in any extra trains, whilst taking into account the seriously restrictive single line sections from Evesham-Norton and Worcester-Henwick really are making it difficult to improve things too much - even with all this nice shiny new double track further up the line.

Burty76's timings are fairly close to the mark - though I would say there needs to be the odd minute or two here and there for padding. Something like a bike waiting in the wrong place, or a door being left open at the other end of the platform need to be allowed for, and with those timings there is no slack at all. I've nearly finished mine now though...

[Burty76]

Interesting points Industry Insider.  And youre right I expect there will be more padding, which would take the timings up to the 2H15 mark as I suggested.  Thats too long IMO^▸ (in my opinion).
Looking forward to seeign your proposed timetable 

[James_H]

Is it seriously worth spending that amount of money on redoubling if it's not going to allow any extra trains... thought having an hourly service each way was the main reason for this? 

Also, what exactly is the 'Cathedrals Express'...?

[IndustryInsider]

Is it seriously worth spending that amount of money on redoubling if it's not going to allow any extra trains... thought having an hourly service each way was the main reason for this? 

My previous post wasn't too clear. I should clarify; You'll be able to get the hourly off-peak service with no major problems and most trains can be quicker - in some cases quite substantially. It's mainly the peak hour service that it's difficult to improve on - offering an improved local commuting service from/to Worcester for example.

Also, don't forget, the main reason for the redoubling (and the reason the ORR^» (Office of Rail and Road formerly Office of Rail Regulation - about) agreed to the funding for it) is that it should allow FGW^▸ (First Great Western)'s punctuality on the line to improve from the pretty dire state it's currently in to something percentage wise in the early 90's. Too many extra trains will simply negate any extra flexibility.

For example, you could, in theory, have an hourly stopping service from Oxford to Moreton and an hourly fast service allowing for through trains from London to Worcester to be speeded up. In practice that would mean the single line between Wolvercote and Charlbury being occupied by the four trains for nearly 50 minutes out of each hour, which gives no scope for service recovery when things go wrong. You can just about get away with that level of service in the peak hours - indeed you have to, but to operate it all day would be asking for trouble.

[Andy W]

.  What might appear nice in principle just doesn't make sense commercially for FGW^▸ (First Great Western).

Prove it. It's not passenger numbers that are important it's revenue.

Take Hanborough, what is the split of Paddington vs Oxford Park and Ride customers?

There are no figures I've seen to make any bold commercial statement - what I can tell you is the 'business' traveller on expenses is far more likely to travel 1st class - ^129 lost Wos - PAD^» (Paddington (London) - next trains) against ^3.70 (you surely don't travel 1st for Park & Ride) Hanborough - Oxford.

Lots of passengers does not necessarily mean lots of revenue.

[IndustryInsider]

.  What might appear nice in principle just doesn't make sense commercially for FGW^▸ (First Great Western).

Prove it. It's not passenger numbers that are important it's revenue.

Quite right. Though my current job within FGW has no responsibility for service development or revenue splits and so I am not able to prove anything. Many of the passengers at Hanborough joining the first services of the day are bound for London though - and importantly they're regular season ticket holders which pay their way far more than a business traveller from Worcester on an occasional basis would do, 1st class or not! If you have, say 20 people,  joining at Hanborough then that's not far short of ^100000 in the coffers a year for standard class. That equates to a lot of 1st class business travellers! And I would ask the question, is that revenue worth risking by taking out stops and banking on the few minutes saved bringing in lots of extra custom from Worcester? I would say not, but that of course is only my opinion.

FGW's money making machine is not stupid. If there's revenue to be had they'll sniff it out it they can. I'm sure if they had some spare HST^▸ (High Speed Train) sets, and up-to-date infrastructure then I'm sure they would introduce faster trains from Hereford & Worcester to London (although I think there would be juicier cherries to pick on other routes first). Until such a situation is reached then there will be little change in the stopping patterns of trains - that I would be prepared to bet on.

[Andy W]

Many of the passengers at Hanborough joining the first services of the day are bound for London though - and importantly they're regular season ticket holders which pay their way far more than a business traveller from Worcester on an occasional basis would do, 1st class or not! If you have, say 20 people,  joining at Hanborough then that's not far short of ^100000 in the coffers a year for standard class. That equates to a lot of 1st class business travellers! And I would ask the question, is that revenue worth risking by taking out stops and banking on the few minutes saved bringing in lots of extra custom from Worcester? I would say not, but that of course is only my opinion.

FGW^▸ (First Great Western)'s money making machine is not stupid. If there's revenue to be had they'll sniff it out it they can. I'm sure if they had some spare HST^▸ (High Speed Train) sets, and up-to-date infrastructure then I'm sure they would introduce faster trains from Hereford & Worcester to London (although I think there would be juicier cherries to pick on other routes first). Until such a situation is reached then there will be little change in the stopping patterns of trains - that I would be prepared to bet on.

Hi II

Unfortunately there are too many assumptions:-
1. That there are 20 people with season tickets from Hanborough -
2. Even if there are and some trains skipped on to Oxford FGW would lose their custom as there are not really many options
3. That 'occasional' business travel is an unreliable income stream
4. ^100,000 is 'a lot of 1st class tickets.

Well on 1 there are no breakdowns so it may or may not be reasonable but on 2 it is very unlikely that the revenue to Paddington would be lost.

On 3&4 there is a significant number of people who travel from Worcester to London using Virgin / Chiltern - this is lost revenue. If you take an average fare of ^70 at peak hours you only have to lose six people per day to have lost ^100,000 p.a. I can assure you the number is far higher than 6.

It's nothing to do with spare HST sets - it's all to do with understanding your customers (or more importantly potential customers). It is understanding that there is nothing more frustrating than sitting at Hanborough for someone to get their bike on the train to see them get off again at Oxford.

[IndustryInsider]

Some interesting points there, but I wasn't using assumptions, Andy. I was using a figure (20) as an example of what revenue that would bring in. The real number of season ticket holders from Hanborough-London is (I assume  ) more than 20. Then there's occasional people heading to Reading/London/elsewhere on business (+ then you add all the extra revenue, albeit smaller, to  Oxford as well of course). The main business train from Hereford and Worcester already misses Hanborough station, others stop there and I think that with the available infrastructure that is the best compromise.

The other options for Hanborough passengers would be to either drive to Charlbury (where there is no spare parking), drive into Oxford (when at peak times it is an absolute nightmare traffic wise), or drive to Bicester and use Chiltern. I'd suspect they'd do the latter - so you're simply swapping gains for losses.

The number of people using Warwick Parkway from the Worcester area may well be more than 6 per day, but do you really think that missing a couple of stops and saving, at the most, 10 minutes will make them all swarm back to FGW^▸ (First Great Western)? I think a reliable service punctuality wise is what people are looking for and that (combined with journey time improvements brought about by the effects of the redoubling anyway) will do more to entice new/lost custom.

I agree that bicycles can be annoying, especially if they are on a short hop to Oxford, though Charlbury is the biggest source of bicycle related delays (that is actually based on delay data).

And again I must stress that if better infrastructure was in place then I would be fully supportive of a limited stops service from Hereford and Worcester to Oxford/London. However it's not, and it won't be anytime soon.

[Btline]

I shall be interested to see what you all make of my little timetable experiment. What little hair I had left is in serious danger of being ripped out. Squeezing in any extra trains, whilst taking into account the seriously restrictive single line sections from Evesham-Norton and Worcester-Henwick really are making it difficult to improve things too much - even with all this nice shiny new double track further up the line.

Exactly, they are doubling the wrong section - probably to cut costs.

You make a good point - it is not only the Norton - Evesham section that will delay trains, but the whole Worcester area. Seeing as the Worcester area will not be re-signalled for a while, I would have thought NR^» (Network Rail - home page) would double the Perhore to Honeybourne section. A small bit of single line between Morteon and Honeybourne would be fine, as train crosses will be nowhere near here and the Aston Magna curve can stay at its max speed. I would also benefit Pershore (having a second platform, footbrige etc).

However, trains will be crossing at Evesham and in the Worcester area, leading to delays.

[G.Uard]

Fully expecting howls of execration, I think that a blanket ban on non folding bikes during peak hours seems sensible, but outside of the rush, the railway must also cater for leisure/off peak cyclists.  I am by no means anti bike, but I can't understand why dedicated cyclists don't either invest in a folding cycle or else fork out, (no pun intended), for 2 cheapo tredders,leaving one at either station as needed during the course of a commute.  BTW^▸ (by the way), I ride my trusty bike to the depot on an almost daily basis.

[willc]

Few people do park and ride to Oxford from Hanborough at present - the service is too sketchy for one thing, eg nothing into the city from 8.01 until 10.16 - and the car park is usually full by the end of the peak anyway. The current passenger mix is typically local walk-up or dropped-off custom from the village into Oxford, as well as locals and park-and-riders (from the likes of Witney) heading to Reading and London.

If there was something like a 30-minute interval service most of the day, that might change things, but unless someone can find a way to solve the parking question at Hanborough, it will be
Oxford custom from the village itself that gets a boost, by providing a regular, reliable service.

And I should think there are considerably more than 20 London season ticket holders there. I don't have to do it at present but in the past quite often travelled into Oxford on the first two trains of the day. The numbers boarding at Hanborough and Charlbury were always substantial.

The first train only started at Moreton then, so earned its corn entirely on the basis of custom at the eastern end of the route, but knowing that this revenue is there is what has allowed FGW^▸ (First Great Western) to experiment with starting the train from Worcester at 5am since December. Just like money earned here allowed the eventual provision of mid-evening trains from London all the way to Worcester.

Revenue is important, but guaranteed, steady revenue that season tickets and regular commuters bring is even more so. Business travel is nice money, but it's harder to predict and can't be relied on, especially in the current climate. Your first class business traveller from Worcester is unlikely to use the train every day, whereas every morning peak train calling at Hanborough will bring in that ^129 - and a deal more with London and Reading fares in the mix - five days a week.

If all these business travellers are so set against FGW as you suggest and have convinced themselves that driving all the way to Warwick or Birmingham International is a better option, even if some of us can't see any time advantage, I can't see how anything less than about 30 minutes off the Worcester timing would bring them back - and that can't happen without Reading rebuilding, Thames Valley resignalling, complete redoubling of the Cotswold Line, Worcester resignalling and electrification. Even then, the stations nearer to Oxford would still fill more seats and generate more revenue, day in, day out.

Btline, more single-line sections, whatever the length and wherever they are on the line, equal more potential for delay. It doesn't matter where the timetable says trains will pass - it's what happens when they hit problems that counts and the more bits of single line, the worse the problems get. And if Industry Insider had to factor another section into his calculations, then he certainly wouldn't have any hair left!

The example of an ailing HST^▸ (High Speed Train) v Campden bank that I cited earlier is pertinent in the context of your suggestion - it does happen from time to time and it does destroy the morning peak timetable. It would still do so, though perhaps not so completely, if the line remained single-track here.

The fundamental premise of the NR^» (Network Rail - home page) plan was not to allow lots of extra trains, or substantially faster trains - it was to deliver reliability and punctuality across the whole FGW network - that was the basis on which it was flagged up in NR's submissions to the ORR^» (Office of Rail and Road formerly Office of Rail Regulation - about).

Are you seriously suggesting that you know better than people at Network Rail, who spent almost a year running through every conceivable option, including the kind of thing you favour?

PS for James: the Cathedrals Expresses - marked CE in the FGW timetable - are the 6.43 from Hereford and the 18.22 from London. The name is from the train serving the cathedral cities of Oxford, Worcester  and Hereford.

PPS: In theory, if one believes the black blobs on the timetable, then bikes should not be loaded at any Cotswold Line stop on the 5.17 from Malvern and both the Herefords. The rule has never been enforced as far as I can tell and is practically unenforceable, since there are no platform staff and the guards are always in the middle of the train operating the doors, nowhere near the TGS, so anyone with a bike has loaded it before the crew can get anywhere near. It's not worth the trouble stopping them, but nor is it the crippling problem some make it out to be here.

In my experience, bikes cause far more delays with Adelantes than HSTs, because the crew have to open the bike stowage area doors, while the automated doors on the coaches mean the guards are ready to despatch the train quicker. And if an Adelante turns up instead of an HST, the cyclists are usually waiting in the wrong place on the platform as well.