GWR Performance Figures

[whole diary]

[IndustryInsider]

Latest graph attached with another month of figures.  Both LTV^▸ (London [and] Thames Valley) and HSS^▸ (High Speed Services) hardly setting the world alight, but an improvement on last month's very poor figures and the Moving Annual Average hasn't suffered as a result.  It will be interesting to see what effect the rush hour signalling problems we had recently has on the figures for the next period, as that very poor week of delays started the day after the attached graph is for.

[ChrisB]

Bear in mind any performance better than the equivalent period last year will see their figures improve, even if it's an appalling period (but still better than last year's equivalent. It's a 12 month rolling average

[IndustryInsider]

Yes indeed.  Hence the MAA^▸ (Moving Annual Average) for the LTV^▸ (London [and] Thames Valley) service slowly creeping up over the last year despite the odd very poor month.  Autumn and winter weather also play a crucial role as last years very mild winter (contrary to Daily Express predications) meant the railways had it pretty easy.  This Autumn has so far been very benign, but should windy and wet conditions hit us in November then that could also hit the figures hard.

[IndustryInsider]

Here's the last period of figures which covers the first of the two periods most prone to leaf-fall related delays.

The LTV^▸ (London [and] Thames Valley) sector recorded 80.3% which sounds pretty bad (well, it is!), but much better than the equivalent month last year as the Moving Annual Average rose 0.3% to 85.7%.  Not quite so good on HSS^▸ (High Speed Services) with an 81.8% dragging the MAA^▸ (Moving Annual Average) down to 88.9%.

[TaplowGreen]

Here's the last period of figures which covers the first of the two periods most prone to leaf-fall related delays.

The LTV^▸ (London [and] Thames Valley) sector recorded 80.3% which sounds pretty bad (well, it is!), but much better than the equivalent month last year as the Moving Annual Average rose 0.3% to 85.7%.  Not quite so good on HSS^▸ (High Speed Services) with an 81.8% dragging the MAA^▸ (Moving Annual Average) down to 88.9%.

To be honest, this is pretty meaningless from a customer experience perspective, unless there is another chart showing the true picture, incorporating all the cancellations and delays without all the permitted "exclusions".

[ChrisB]

You want the 'Right Time' stats.....some TOCs^▸ (Train Operating Company) do publish them. Includes all delays and shows stats for delays of 5mins or more

[IndustryInsider]

To be honest, this is pretty meaningless from a customer experience perspective, unless there is another chart showing the true picture, incorporating all the cancellations and delays without all the permitted "exclusions".

I do agree with you in part, though its fair to say that a good month for charter figures is more than likely a good month for all day figures including exclusions.  That's another reason why delay repay is a better system of compensation when we eventually get it on GWR^▸ (Great Western Railway) - especially the new 15 minute claim threshold.

There are more detailed provisional figures published on a daily basis including all services operated by GWR and they can be found at:  https://www.firstgroup.com/uktrain/kpi/gw/

[Sixty3Closure]

I don't know how you would do it but the stats apart from measuring a very narrow range don't reflect the customer experience.

My regular train home is normally a moderately crowded 8 coach 125 and on Tuesday was replaced by a 2 coach turbo. I'd guess the performance figures reflect this as running and on time even though most people couldn't board it?

A starting point though would be as Taplow Green says would be figures measuring the whole day.

Then there needs to be a way to measure over crowding, cleaning, toilets out of order, no seat reservations etc.

[IndustryInsider]

Aren't the all-day stats generally better than the peak time ones used for the charter figures, even if you were to include things like fatalities?

Regarding the short formed train, those do get recorded and have to be submitted to the DfT^» (Department for Transport - about) IIRC^▸ (if I recall/remember/read correctly)?

For those interested in all-day or specific train performance, there are a number of third-party sites using the freely available data feeds from Network Rail that can help, including:

http://www.recenttraintimes.co.uk

http://www.liveppm.co.uk

(resolve)/" target="_blank">Http://www.easignalmap.altervista.org/PPM/

[SandTEngineer]

I started a separate topic concerning the far Southwest service performance: http://www.firstgreatwestern.info/coffeeshop/index.php?topic=17665.0

Perhaps it might be better merged to this topic (mods?).

[broadgage]

IMHO^▸ (in my humble opinion), if a train is so badly overcrowded that significant numbers of intending passengers are unable to board, then it should be regarded as cancelled when collecting performance data.

The purpose of running the service is to convey passengers, if from a certain point the train is so crowded that no more passengers can board, then the service if in effect cancelled from that point and this should be reflected in performance data.

[IndustryInsider]

Latest charter figures attached and a very poor second leaf fall period which saw LTV^▸ (London [and] Thames Valley) drop to just 73.3% punctuality.  That was so low I had to reconfigure the axis of the graph!  The Moving Annual Average (MAA^▸ (Moving Annual Average)) dropped as a consequence to 85.0% from 85.7%.

HSS^▸ (High Speed Services) fared a little better at 81.2% but that was still the worst period of the year.  Not totally surprising as the leaf fall months are nearly always the worst unless heavy snow or exceptional heat come along.  The MAA dropped half-a-percent to 88.4% which is now just 0.4% above the season ticket 5% discount trigger.  GWR^▸ (Great Western Railway) must be very relieved virtually all the leaves have now fallen!

[ChrisB]

You might ask NR^» (Network Rail - home page) about their treatment train PPM^‡(resolve) too....TOCs^▸ (Train Operating Company) aren't very happy.....

[Sixty3Closure]

Rather than take other threads off topic this seems a better place to ask about trains that skip stops.

If a service drops some stations and runs non stop does this affect the targets? Are there penalties for doing this or is it only the arrival time at the final stop? When I've looked into this in the past it seems to be the arrival times of certain services between certain times e.g. for Thames Valley all Paddington services between 07.00 and 09.00 (to make up a random example) which would suggest that skipping stations isn't penalised but this was only a short summary.

I'd imagine though the actual performance measure are  not a small document. I think I read somewhere about several pages dedicated to bird strikes and depending on the size of bird where responsibility lay and how much.

[grahame]

If a service drops some stations and runs non stop does this affect the targets? Are there penalties for doing this or is it only the arrival time at the final stop? When I've looked into this in the past it seems to be the arrival times of certain services between certain times e.g. for Thames Valley all Paddington services between 07.00 and 09.00 (to make up a random example) which would suggest that skipping stations isn't penalised but this was only a short summary.

I don't know how it reflects into performance / penalties fares ... but I do know that an unplanned stop on the fast main lines at Slough, Maidenhead, and Twyford can have a knock-on effect for hours as there's simply no slack in there.   Stops at one of these stations are typically put into services leaving Paddington in the slot ahead of a Heathrow Express service, so that the train after it slowdown, station duties and restart ends up in the slot the HEX would have occupied had it not branched off to the airport.  Move that train a few minutes later and it ends up creating a ripple effect on following trains if it makes the stops ... and that makes the following trains delayed to knock on as they get further west over Cotswold single lines, or with failing connections at places like Swindon.

With a system running so close to capacity, the only remedies are to increase the capacity or improve the reliability / timekeeping;  moving the existing deck chairs around isn't going to provide a solution - just a marginally better set of winner and looser tradeoffs.

Here from Recent Train Times is the actual arrival time graph for trains at Reading from Paddington in the last 2 weeks, evening peak.   I am struck by a 'wobble' on 13th December at about 18:00 which then pushed everything back for quite a while thereafter; no sudden recovery which provides some evidence for what I've written above.