GWR Performance Figures

[whole diary]

[IndustryInsider]

Lots of data to chew over this month given the large number of cancellations we've been witnessing and discussing at some length on other threads.

Firstly, my usual HSS^▸ (High Speed Services) and LTV^▸ (London [and] Thames Valley) performance summary graph is attached and just judging by that you might think all is well.  HSS posted a respectable figure of 89.1% which lifted the MAA^▸ (Moving Annual Average) by 0.5% to 87.2%.  LTV's figure was 84.9% which again lifted the MAA, this time by 0.3% to 82.5%.  It goes without saying that the MAA rose so well as this time last year results were unseasonably poor.  Another poor month for the Bristol region saw figures of 84.9%, dropping the MAA 0.2% to 89.1%, which is now just 0.1% above the trigger level.

Turning to reliability, which is the current hot topic of course, and the following figures were posted by the various sectors:  HSS 98.9%, LTV 97.9%, Bristol 98.6%, Devon 99.3%, Plymouth & Cornwall 99.3%, and South Wales to South Coast 98.3%.  Whilst most of those are worse than normal it has yet to become bad enough to threaten the charter discount thresholds, except in the case of South Wales to South Coast which is now bang on the trigger mark of 98.5%.  LTV's figure of 97.9% is only 0.1% below the trigger level so it will have to get quite a bit worse to reach that level for the MAA which is currently 0.7% above that trigger level.  Recent removal of driver incentives for overtime (and other factors) might well make it get quite a lot worse though.  It will be very interesting to see next months figures.

[BandHcommuter]

It's worth reminding ourselves of the small print relating to punctuality and reliability statistics in GWR^▸ (Great Western Railway)'s passengers' charter:

All Monday to Saturday services are included in these calculations, except former First Great Western Link services. For those, we include Monday to Friday services. And we only include morning and evening peak services in the punctuality calculation. Peak services are the trains that arrive at London Paddington from 7am–10am and trains that leave London Paddington from 4pm–7pm. We don’t include Sundays and Bank Holidays in these calculations.

That might explain why those of us who have been heavily inconvenienced recently whilst trying to travel on local services in the Bristol/Westbury areas on Sundays might have difficulty recognising the reported performance levels as representative of our own experience. And as IndustryInsider suggests, we might expect next month's figures to deteriorate further.

[IndustryInsider]

Yes indeed, BandHcommuter.  I've also been logging data for 'all-day right time' and 'all-day PPM^‡(resolve)' for the last five months which also covers the weekend trains and intend to produce a graph when I hit six months worth of data next period when the data will look meaningful.

[Chris from Nailsea]

Thank you for all of your work in compiling, interpreting and publishing these figures here, IndustryInsider. 

[IndustryInsider]

The latest results are in...

Following a steady slump in punctuality over the last six months, the 'Bristol' sector has now breached its charter target for the Moving Annual Average (MAA^▸ (Moving Annual Average)) and so will trigger 5% season ticket renewal discounts.  Elsewhere, GWR^▸ (Great Western Railway) certainly won't be winning any awards as MAA's stayed pretty much level, although the South Wales to South Coast is 3% points lower than this time last year and only needs to drop 1% for discounts to apply there too.

Reliability wise, another month of high cancellation rates led to low figures of 98% and 98.7% being recorded for LTV^▸ (London [and] Thames Valley) and HSS^▸ (High Speed Services) respectively.  Though with trigger levels of 98% and 98.2% there will need to be several months of this sort of performance for the charter figure to be threatened.  South Wales to South Coast did drop below the magic trigger level having hovered close to it for a couple of months, so that's 5% of renewals until it improves!

From the situation a year ago, where only one trigger level was being breached (LTV performance), there are now three more - HSS performance, Bristol performance, and South Wales to South Coast reliability.

The usual graph for HSS and LTV is attached.  Following comments about the usefulness of the data on that graph, I've also included another graph which has the data for all-day figures, not just the peaks, and covers 'Right Time' and 'Public Performance Measure (PPM^‡(resolve))' data for GWR overall, and the three main sectors - HSS, LTV and West.  The GWR MAA figure is also represented.  Looks a bit cluttered, but the figures are there for all to see - for example in mid-summer, if you caught a HSS train, there was less than a third of a chance it would arrive on time!

[Chris from Nailsea]

Thanks, yet again, for that very informed detail, IndustryInsider. 

[initiation]

I get notifications via text of short formations or cancellations for key times in the morning and evening on my route (still quite a short period of time) and it is more unusual to get a day that I don't get a warning vs everything working fine with no text!

With the new stock finally arriving perhaps things will slowly improve. I bet just as my renewal in April comes up we will sneak back over the trigger %.

ChrisB helpfully previously confirmed you don't get anything in this situation.

Correct, coz, on average over the full length of your ticket, they weren't below target.

Which is fine when viewed in isolation but this is a renewal for a ticket I had purchased for several years. So if someone else happened to have their renewal this November they would get the discount but come April, if the 12 month rolling average improves then I won't get anything. Despite having exactly the same delays, just spread over 2 tickets rather than one. Them are the rules they set though....

Hoping for 15 min delay repay in the next franchise.

[IndustryInsider]

Them are the rules they set though....

Hoping for 15 min delay repay in the next franchise.

They are indeed the current rules, and they are unfair in certain instances.  I would be very surprised if delay repay 15 doesn't feature in the next franchise.

[Sixty3Closure]

Can't help but feel the discounts should be based on the "all day" numbers. This year in a completely non scientific way seems to have been one of the poorer one in my 20 years of commuting form Twyford.

[ChrisB]

But they would dilute (big time) the stats if based as is at the moment, possibly that regular late running in the peaks of say, 20mins, combined with few off-peak delays, wouldn't drop the stats enough to ever get your 5%.

Bear in mind that the calculation is for the benefit of seasons, generally used in the peaks

[IndustryInsider]

Latest graphs attached, and a pretty unremarkable month worth of data showing on them as similar stats were achieved to last year.  Certain specific sectors continue to struggle though with regard to Passengers Charter performance.  Two more very bad months for the Bristol and South Wales/South Coast sectors saw their charter performance MAA^▸ (Moving Annual Average) drop to 88.1% and 89.4% respectively.  Bristol had already dropped into discount territory last month, and South Wales/South Coast now only needs to drop a further half percent to join it.

Reliabilty wise, Bristol Suburban's figure of 97% has now dropped its MAA to just 98.8%, which is just 0.3% above the trigger level.  Another bad month and that could mean a 10% renewal discount for those renewing over the New Year, which of course is quite a lot of people.

Usual graphs are attached...

[harrybruno]

Hi. Thanks for posting these charts - GWR^▸ (Great Western Railway) don't seem too willing to post the history of the HSS^▸ (High Speed Services) season ticket PPMs^‡(resolve) (which seem to be higher than the HSS PPM data - this i understand is a function of season ticket renewals being based off peak arrivals and departures to Paddington only?)

Do you have the underlying data/numbers for the monthly HSS PPMs (with the most recently figure being 83.2 for the 4 weeks from October 15th 2017)?

I'm trying to work out what figure GWR need to hit/magic up in December to take the HSS PPM above the 88.0 target! It must have to be near or higher than the highest 4-week PPM figure they've managed through the whole of 2017 (and the past couple of weeks have been pretty disrupted)

Thank you

[IndustryInsider]

Hi, harrybruno and welcome to the forum.

The grey line on the HSS^▸ (High Speed Services) Punctuality graph is the monthly Peak PPM^‡(resolve) that season tickets are based on.  As you say, the most recent figure is 83.2%.  Regarding next month, the equivalent month last last year was 81.2%, which although poor, I reckon this month would have to be pretty flawless in order for the 88% target to be met.  And of course it won't be, so I will eat my hat if that target is reached next month.

I've attached the actual figures for each month from the start of 2016.

[harrybruno]

Great, thanks very much. Much appreciated.
Well if we assume a simple average and weight each period equally (though, for instance, were there a lower number of HSS^▸ (High Speed Services) running during the Swindon-Parkway electrification closure, which was a >88 period, in which case equally weighting may not be accurate), though 81.2 period drops out, i calculate that next month needs to be in excess of 91.5 to push the MMA to >87.95

I'll help you eat that hat! Thanks again

[t0m]

For the first time ever I might get a discount on my season ticket renewal... Compensation at last!

Now bring on DelayRepay...