Lies, Damned Lies and Statistics!

[S.A.C Martin]

Reading Hornby Magazine this month, and the "Is the Wishlist Wishing Well Running Dry?" I noted the following statistics:

 

Locomotive Classes available RTR for the Years 1958-1959

 

Western Region: 48.78% of classes or 73.33% of wheel types

 

Southern Region: 29.16% of classes or 53.33% of wheel types

 

London Midland Region: 40.62% of classes or 66.6% of wheel types

 

Eastern Region: 24% of classes or 56.5% of wheel types

 

British Railway Standards: 64.2% of classes or 100% of wheel types.

 

Early/Experimental Diesel Classes/Gas Turbines: 12 Classes and 58% available RTR.

 

 

The above is just the overall breakdown of the article's assertions regarding the locomotives available for each region and the diesel/gas turbine classes. There's a few things which bother me with this list and the breakdown of classes within it. A short disclaimer however - I am only mentioning the Standards and the Eastern Region as my knowledge of the other regions is not by any means comprehensive enough to warrant an opinion on those regions.

 

One thing is the assertion that 64.2% of the BR standard classes have been done. I note in the 4-6-2 category for British Railways that only 66.3% of the classes have been done. But when you consider this category includes the lone 8P Duke of Gloucester as the missing type, the statistic of 66.3% of the classes is mathematically correct, but the number of locomotives within a class isn't taken into account.

 

So in short, of the locomotives of the wheel arrangement 4-6-2, you can in theory model all of the British Railway locomotives bar the sole 8P from RTR sources. So the reality of what is model-able in the Pacific stakes is almost as makes no difference 99% of the Standard Pacifics that were in use with British Railways in that year (with the caveat that the air pump fitted Britannias are not available RTR?).

 

Then there's the Eastern Region's W1. Obviously being the only 4-6-4 tender in the whole of the British Isles, if it's not available RTR than the percentage of wheel types is 0%. But this contributes - a single locomotive - to an overall percentage of 56.5% of classes available RTR for the Eastern Region. Is this fair, as an overview of the locomotives of the Eastern Region, that one locomotive has more influence on the overall statistics than the many classes of 100s (looking at the J type locomotives particularly)?

 

I note that in the same section, the number of Pacific locomotives is numbered as seven ( 7 ). A3, A4, A2/1, A2/2, A2/3, A1/1 and the Peppercorn A1 and A2s makes eight ( 8 ) classes. All of the above classes were still in use in 1958/59. So the statistic of 57% of the Pacific types is further warped because one class has been left out entirely in the mathematics.

 

I think overall that this might have been a lot fairer if it was also noted how many were in which class - staying with the Pacific example the A1/1 is a single locomotive and the A2/1s numbered only four. That's before we get into detail differences between locomotives within the same class, which may mean that certain locomotives within certain classes, are not possible to model at this time without major surgery and chopping and changing of tenders/boilers/cabs and similar.

 

The idea behind the article was nice, and I feel it was a good idea in terms of highlighting the overall number of locomotives and class types British Railways used at that time, but I'm not convinced that the way in which it was conducted was entirely conducive to telling us anything about the availability of locomotives RTR for each region. The question that remains with me - what if the models available aren't practical as donor models for the engine you want to portray?

 

Does that still mean if you have an RTR model of an A3 Pacific, you can model any member - 100% of the class - as the article seems to suggest?

 

I'm not saying we want every RTR release to cater for all variations within a class, but it strikes me as somewhat missing the point to include classes with very few engines actually built within it in comparison to those whose numbers may be in the hundreds. If the point of the article is to highlight that we still only have about a quarter of locomotive classes that were in use in 1959 for the Eastern Region, then (taking into account the error in Pacific classes), it would be a fair statistic but it doesn't also then highlight how many locomotives this actually represents.

 

In short, is there a better way to calculate what percentage of locomotives and locomotive classes on [Region] are available RTR, taking into account the number of locomotives in a class and their model-able variations?

[Forester]

I notice the Scottish Region is not mentioned above! The statistics on that would be interesting and would challenge those who believe RTR has now covered most significant types.

 

But how many of us model a Region, anyway?

 

Most model local areas of particular interest to us. In the steam era local areas often relied on small older types, of which there are all too few available RTR across the board for a start.

 

 

I'm a Southern Region modeller.

Boo! Hiss! You lucky .....

........But wait.

I'm a South Eastern modeller. A very different story. South Western has a great variety of RTR steam available. Very few South Eastern types are available in RTR.

 

So really statistics by Regions mean very little. What is needed is analysis by area - and perhaps the best way would be to use pre-grouping territories. If that were done the statistics would show large variations.

[TheSignalEngineer]

Some areas are well served, in the West Midlands of my childhood and youth we had Coronations, Princesses, Scots, Jubilees, Patriots, Black 5s, Crabs, 8fs, Super D, Midland 3F & 4F 0-6-0, Jinty, Fowler, Stanier and Fairburn 2-6-4T, 2P and Compound 4-4-0, Ivatt 4MT and 2MT, B1s, Standard 4s & 5s Britannias and 9Fs on the LMS side. The main gaps are the Stanier 2-6-2T and 2-6-0 tender locos and Midland 2F 0-6-0, we will even have the Garratts soon.

 

The GW side had Kings, Castles, Halls, Manors, Granges, plus 28xx, 43xx, 57xx, 56xx, Large Prairies, WD 2-8-0, 9Fs, and the occasional County, 72xx, and 2251. Main shortage is the 64xx/74xx tanks.

 

We even have many of the "Rare Birds" seen only occasionally or on specials, such as LNER A1, A3, A4, V2, K3, O1, J39, and L1 or Southern MN, WC, BOB, Q1 and T9, plus BR Clan.

 

As for diesels we saw the 03, 08, 20, 24/25/26/27 variants, 40s, 44/45/46, 47, Lion, occasional 31s from the Eastern, Westerns, B-B Warships and Hymeks, LMS Twins and SR 1Co-Co1, Cromptons on test from BRCW........

 

I've probably missed a few, but it shows that with a few exceptions the mid 1950s to mid 1960s period in that particular area was very well served.

[JOHNMCDRAGON]

Hi S.A.C., yes I fully agree your point and I think that to talk about percentages of locos in this case is meaningless, the classes represented and how plentiful they were are much more relevant. By the way, I'm sitting here writing this with more than the average number of legs that people have. I 've got 2 perfectly good legs, but if you add up all the number of legs in the world, while most people have 2, some have one leg and some have no legs, but nobody has three, so the average number of legs is 1.9999999999999999999999999 . So much for percentages, not always the most appropriate wat to express things!

[34theletterbetweenB&D]

...statistics by Regions mean very little. What is needed is analysis by area - and perhaps the best way would be to use pre-grouping territories. If that were done the statistics would show large variations.

That's the one that makes sense for big four and steam era BR in particular. I would like the RTR companies to 'get logical' and round out what they have done so far to make the better covered pre-group areas essentially complete.

 

Agian I know ER/LNER best, so for example, the ex GN mainline is nearly there. A J6, J50, K1, N7 are the vital additions, the common, everyday small black stuff to supplement the super green stuff. We don't need any more ER greenstuff, really. Did someone say Howlden coaches?

 

The GER lines need some of their small black stuff, a D16, F5, J15 or 17, J67/68/69, N7 and you are in business. Look, the N7 turned up twice, and the ex-GER J's (tender and tank) were frequent GN area visitors, 2 for the price of 1...

 

The opportunity going begging is for a RTR manufacturer to stake out a near virgin territory. The NER, LNWR, L&Y, CR, NBR, look like open goals to me...

[Ozexpatriate]

What's a "wheel type", according to this measurement?

[Kelly]

The mundane are the real missing items. the fancy express and famous have been done for the most part, but the 0-6-0 tender/tanks (and 4-4-0, etc) don't have the glamour to market them, with a few exceptions of course.

[Catkins]

The mundane are the real missing items. the fancy express and famous have been done for the most part, but the 0-6-0 tender/tanks (and 4-4-0, etc) don't have the glamour to market them, with a few exceptions of course.

 

It seems to be the same in the Diesel and Electric era's, but the mundane there are the multiple units.

[Guest dilbert]

... the fancy express and famous have been done for the most part, but the 0-6-0 tender/tanks (and 4-4-0, etc) don't have the glamour to market them, with a few exceptions of course.

 

And not only once - in certain cases, twice & thrice ...dilbert

[S.A.C Martin]

And not only once - in certain cases, twice & thrice ...dilbert

 

Yes indeed - there are four different models of the A4 Pacific to different levels of detail over two manufacturers in OO, there are six (the super detail one with variants within counting as one) of the A1/A3 Pacifics over one manufactuer, and if we go to Diesels, the Class 47 was made by four manufacturers, is now made by three to differing levels of detail.

 

But is the mundane, as it was put earlier, becoming more desirable? I think so. The Bachmann O4 and forthcoming Hornby O1 seem to point to this.

[34theletterbetweenB&D]

And the 0-6-0s such as the 3F and SECR O1. If the demand holds up, at their usual steam loco release rate Bachmann have a twenty year pipeline of potential small black models based on classes with at least the thick end of 100 members in class in BR ownership. That sort of number is a good filter: getting on for a hundred in class means pretty wide dispersion within their operating territory.

 

B16, D16, G5, J6, J10, J11, J15, J17, J21, J25, J26, J27, J35, J36, J37, J50, J67/68/69, K1, K2, N5, N7, N15, O2, Q5, Q6.

 

A similar length list can be compiled for the ex-LMS lines, Stanier's scrap and build barely touched the small stuff...

 

There's a half dozen more from the minor groups too.

[Horsetan]

No Maunsell "U", "U1" or "N1" so far. No "K" or "W" class tanks either.

 

No Maunsell "D1" or "E1".

 

There are still plainly gaps that could be filled.

[Downer]

When talking about gaps needing to be filled, shouldn't some note be taken of the relative difficulties involved in filling them oneself? There are obviously some modellers skilled enough to attempt any kit, and others who'd shy away from the simplest, but many, I guess, are something like me - willing to take on an 0-6-0 tank or tender engine, but reluctant to risk our money on anything involving valve gear. The 4 notable RTR gaps for anyone modelling cross-London traffic IMHO are the ex-SR W 2-6-4, the ex-LMS Fowler 2-6-2T, ex-GER N7 0-6-2T and ex-LNER J50 0-6-0, and given that the latter pair lack valve gear and are available in relatively straightforward kits, I would far prefer RTR models of the first two.