|
|
Post by Baikal on Mar 31, 2024 13:23:38 GMT -8
Table showing loco models and their cumulative percent of each railroad's fleet, ordered by the number of units, most to fewest.
Selected railroads & years. July 1 for all except SAL just prior to merger and SCL just after merger. C&O shows all-time diesel roster.
Shows up to 50 percent of each RRs fleet in order to reduce the table size. I have an extended version going up to 90%. My work, free to share. Please let me know of errors.
Mods- ok to move this post elsewhere.
|
|
|
|
Post by Colin 't Hart on Apr 1, 2024 6:12:01 GMT -8
I'm curious what the BN percentage looks like in the SD40-2 era, especially if you lump together the big 6-axle EMD power: SD40, SD40-2, SDP40, SDP45, SD45, F45.
|
|
|
|
Post by packer on Apr 1, 2024 9:46:38 GMT -8
I'm wondering if that chart counts stuff that hasn't been patched for BN. Pretty sure they had close to 2x as many GP9 as GP7
|
|
|
|
Post by Baikal on Apr 1, 2024 9:54:26 GMT -8
I'm curious what the BN percentage looks like in the SD40-2 era, especially if you lump together the big 6-axle EMD power: SD40, SD40-2, SDP40, SDP45, SD45, F45.
I'd also like to see that. Same with UP after the deluge of SD40-2s.
My BN rosters are for 1970 & '71 only. But I do have it broken out by previous owner & BN purchased. Which is pretty interesting in itself.
I have about 20 other RR rosters broken out by year (usually 1954 thru sometime in the 70s) and by model. SP, UP, , SP&S, WP, and L&N are pretty complete. ATSF and ACL/SAL/SCL by year & model are in s l o w progress. Others are for a single year only- developed due to a particular interest I might of had. The data is out there, it's just time consuming to put it into a common format that's easy (for me) to use.
CNW and NdeM I'd like to have but I won't live long enough.
|
|
|
|
Post by Baikal on Apr 1, 2024 10:06:19 GMT -8
I'm wondering if that chart counts stuff that hasn't been patched for BN. Pretty sure they had close to 2x as many GP9 as GP7
That's what the table shows. The BN roster should be for July 1 1971, I put 1970. My error. But the numbers will be very close.
In 1971 BN had 281 GP9s (about 13.3% of the roster) and 144 GP7s (about 6.8% of the roster). Cumulative total about 20.2%.
The roster has nothing to do with paint patches, only ownership. If BN owned it, it was counted.
|
|
ed
Full Member
Posts: 132
|
Post by ed on Apr 4, 2024 8:17:04 GMT -8
Shouldn't the % amounts add up to 100%?
|
|
|
|
Post by gevohogger on Apr 4, 2024 8:39:28 GMT -8
Shouldn't the % amounts add up to 100%? He only did the numbers up to 50%.
But it raises the question: What did we (him?) learn from this exercise, if anything? I suppose it proves that any list of numbers can be turned into a statistic.
|
|
|
|
Post by riogrande on Apr 4, 2024 9:14:48 GMT -8
Nothing for D&RGW?
|
|
|
|
Post by prr 4467 on Apr 4, 2024 10:18:38 GMT -8
I'm sure it was a useful exercise to get the big picture view if one is trying to match the overall roster and assuming you have a big mainline operation on your HO pike.
However, a closer look, depending on the railroad and era, would reveal differences that might be important for the modeler:
I have not historically been a fan of PRR or their steam locos in general, but it is the area where I live, so I just can't escape it. A quick read of PRR steam loco books available from Morning Sun reveals, particularly after WWII, that certain locos were used mainly in certain regions only and not the railroad as a whole. For example:
The Class A-5 0-4-0 switchers were popular in the City of Philadelphia which had numerous tight radius industrial spurs, so most of them were actually assigned there. One would not have seen them elsewhere. If one is into switching heavy industries, one might need them.
The remaining Class E-6 Atlantics were mainly in service on the PRSL hauling passengers to the beach in New Jersey, so would have been unusual (but not impossible) to see elsewhere.
The heavy mainline freight trains were still in the hands of more than 500 I1sa or I1s 2-10-0's. The fast mainline freight trains (reefers etc.) were in the hands of M1a/M1b 4-8-2's.
The really big engines ran from Altoona westward and were apparently always concentrated on the "Lines West". Especially westward from Crestline, depending upon the locos.
The J-1 2-10-4's were built to replace the I1sa's, but both classes ended up serving till the end of steam--however, Ian Fisher noted that they were primarily used west of Altoona, so would not have been very common east of Altoona though clearly did get to Harrisburg on occasion.
The T-1's and other duplexes were on the west end of the railroad west of Crestline where their speed could be used (except that the 50-mph freight speed limit of the PRR was counterproductive to maximizing the efficiency of all these locos--their peak horsepower occurred WELL above the PRR maximum speed limit). One could legitimately argue that they were utterly wasted on the PRR but might have been better suited on a road like Atlantic Coast Line which allowed 90 mph running for some freight trains.
For those fans of the 2-8-8-2 articulateds, they generally were limited to only two areas of the PRR: transfer between Enola Yard and Hagerstown, MD (occasionally, idk how often but it actually happened, the big N&W power like the Y6b made it as far as Harrisburg) and heavy coal and ore service out of Columbus, OH and/or to Mingo Junction. They most likely wouldn't have gotten elsewhere on the PRR.
It's worth noting that PRR hated articulateds, built one or two really good ones that were ahead of their time at 147,000 pounds starting tracitive effort in 1919 and actually broke drawbars (Class HC-1), but deemed them too slow, and swore off any further articulateds despite having a railroad seemingly tailor-made for them west of Altoona.
So if I were modeling PRR, even in the transition era with whichever early diesel I like (they bought some of everything!) my motive power choices could be dictated by what region of the railroad I was modeling rather than by percentage of the whole. They actually were the "far from standard railroad of the World", lol.
|
|
|
|
Post by Baikal on Apr 4, 2024 10:45:38 GMT -8
Here's the entire DRGW roster for 1 Jul 1968:
Model #units cumulative %
GP40 30 13.0
GP30 28 25.2
SD45 26 36.5
GP9 24 47.0
F7B 23 57.0
GP35 22 66.5
F7A 21 75.7
GP7 14 81.7
SW1200 10 86.1
SW1000 10 90.4
SD9 10 94.8
SD7 5 97.0
F9B 3 98.3
F9A 2 99.1
30-ton Dav 1 99.6
Sw slug 1 100.0
Four EMD models make up almost half the roster. The seven most-numerous models are all in the same range- no single model dominates. A 1962 or 63 roster would have looked much different- PAs, FMs, FTs...
|
|
|
|
Post by Baikal on Apr 4, 2024 10:59:32 GMT -8
I'm sure it was a useful exercise to get the big picture view if one is trying to match the overall roster and assuming you have a big mainline operation on your HO pike. However, a closer look, depending on the railroad and era, would reveal differences that might be important for the modeler... (localization info)
True. How to model specific models is great, but we can also model rosters by time & place. Unless someone has an unlimited budget...
If you're a SP modeler and you're only able to model, say, 15 locomotives, due to whatever reasons, how do you pick the locos that best represent what was present on the prototype, then and there?
I have an SP roster spreadsheet that partially breaks-out units by operating area and year. Portland, Bay Area, Central Valley, Sierra, LA Basin, Sunset Route, generalized T&NO/Texas, Cotton Belt. A work in progress.
In the late 60s, the Bay Area was Alco switcher territory. LA had a lot of Baldwins. Portland was almost exclusively Baldwin switchers and SD9s & Baldwin roadswitchers.
|
|
|
|
Post by jonklein611 on Apr 4, 2024 10:59:37 GMT -8
Here's the entire DRGW roster for 1 Jul 1968:
Model #units cumulative %
GP40 30 13.0
GP30 28 25.2
SD45 26 36.5
GP9 24 47.0
F7B 23 57.0
GP35 22 66.5
F7A 21 75.7
GP7 14 81.7
SW1200 10 86.1
SW1000 10 90.4
SD9 10 94.8
SD7 5 97.0
F9B 3 98.3
F9A 2 99.1
30-ton Dav 1 99.6
Sw slug 1 100.0
Four EMD models make up almost half the roster. The seven most-numerous models are all in the same range- no single model dominates. A 1962 or 63 roster would have looked much different- PAs, FMs, FTs...
The cumulative % I think is throwing people. May be better to show % of each model individually. | Model | # of Units | % of Roster | | GP40 | 30 | 13.0% | | GP30 | 28 | 12.2% | | SD45 | 26 | 11.3% | | GP9 | 24 | 10.4% | | F7B | 23 | 10.0% | | GP35 | 22 | 9.6% | | F7A | 21 | 9.1% | | GP7 | 14 | 6.1% | | SW1200 | 10 | 4.3% | | SW1000 | 10 | 4.3% | | SD9 | 10 | 4.3% | | SD7 | 5 | 2.2% | F9B
| 3 | 1.3% | F9A
| 2 | 0.9% | 30-ton
| 1 | 0.4% | SW Slug
| 1 | 0.4% |
Total: 230 100% |
|
|
|
Post by Baikal on Apr 4, 2024 11:10:34 GMT -8
The cumulative % I think is throwing people. May be better to show % of each model individually.
But it's not what I'm interested in looking at/sharing. I have the individial percents, obviously. I figured most people could understand a different way of looking at rosters. (It would be nice if other people posted info)
|
|
|
|
Post by gevohogger on Apr 4, 2024 11:39:48 GMT -8
The cumulative % I think is throwing people. May be better to show % of each model individually.
But it's not what I'm interested in looking at/sharing. I have the individial percents, obviously. I figured most people could understand a different way of looking at rosters. (It would be nice if other people posted info)
I'm not sure the cumulative total is very useful, though, if at all. Take PC for example.... The SW1 bumps the cumulative total from 31.3% up to 35.2%. But it still doesn't tell us how many SW1s they had.... A lot, probably, to make a difference of several percentage points on such a large roster. On a smaller roster like WP or SP&S it would take just a handful to move the total up that far. We still don't know how many SW1s they had or how many locomotives there are on the roster.
I just don't see the end goal. Unless there is some sort of divine modeling wisdom I am missing here.
|
|
|
|
Post by riogrande on Apr 4, 2024 11:46:57 GMT -8
Here's the entire DRGW roster for 1 Jul 1968: Model #units cumulative %
GP40 30 13.0
GP30 28 25.2
SD45 26 36.5
GP9 24 47.0
F7B 23 57.0
GP35 22 66.5
F7A 21 75.7
GP7 14 81.7
SW1200 10 86.1
SW1000 10 90.4
SD9 10 94.8
SD7 5 97.0
F9B 3 98.3
F9A 2 99.1
30-ton Dav 1 99.6
Sw slug 1 100.0 Four EMD models make up almost half the roster. The seven most-numerous models are all in the same range- no single model dominates. A 1962 or 63 roster would have looked much different- PAs, FMs, FTs...
I was being lazy. I have some books at home and UtahRails.net also had D&RGW Diesel data. By the early 1980's the mix had changed quite a bit: GP7's - 1 GP9's - 24 GP30's - 28 GP35's - 22 GP40-2's - 37 GP40's - 66 SD7's - 5 SD9's - 10 SD40T-2's - 73 SD45's - 26 SD50's - 17 (1984) SW1000's - 10 SW1200's - 10 |
|
|
|
Post by Baikal on Apr 4, 2024 12:02:06 GMT -8
But it's not what I'm interested in looking at/sharing. I have the individial percents, obviously. I figured most people could understand a different way of looking at rosters. (It would be nice if other people posted info)
I'm not sure the cumulative total is very useful, though, if at all. Take PC for example.... The SW1 bumps the cumulative total from 31.3% up to 35.2%. But it still doesn't tell us how many SW1s they had.... A lot, probably, to make a difference of several percentage points on such a large roster. On a smaller roster like WP or SP&S it would take just a handful to move the total up that far. We still don't know how many SW1s they had or how many locomotives there are on the roster.
I just don't see the end goal. Unless there is some sort of divine modeling wisdom I am missing here.
Sometimes looking at things differently is a goal unto itself. Made you think.
|
|
|
|
Post by Colin 't Hart on Apr 4, 2024 12:18:18 GMT -8
Shouldn't the % amounts add up to 100%? The numbers are cumulative, left-to-right. Eg BN, GP9+GP7+NW2 accounted for 26.8% |
|
|
|
Post by peoriaman on Apr 4, 2024 15:39:42 GMT -8
So this is just a big April Fool's prank?
LOL You got us good!
|
|
|
|
Post by kentuckysouthernrwy on Apr 4, 2024 16:27:34 GMT -8
Oh, ok….  numbers lock up my mind….that’s why they’re called NUMBers…  |
|
|
|
Post by prr 4467 on Apr 4, 2024 16:38:11 GMT -8
Well, my life is numbers...and drawing pictures for other people to build.
I got it. However Baikal wants to express the numbers is ok with me.
I think most of us are always looking for an excuse to have this or that particular favorite loco or rolling stock item on our layout, while perhaps a more mature or wholistic view of a railroad's roster IS useful.
|
|
|
|
Post by kentuckysouthernrwy on Apr 4, 2024 18:20:21 GMT -8
perhaps a more mature or wholistic view of a railroad's roster IS useful. But so uncharacteristic here about....  |
|
|
|
Post by lvrr325 on Apr 4, 2024 23:23:31 GMT -8
The Lehigh Valley is pretty easy, with small enough numbers of units it's not impossible to model all of them. By 1975 the percentages were fairly even for C420, U23B and GP38-2, with slightly more C628s (15) and slightly fewer RS11s (8 including the PRR leasers). It's more a question of what paint they wore in a given year than how many there were, although it varies some year to year from 1968-1976.
|
|
|
|
Post by peoriaman on Apr 5, 2024 2:55:48 GMT -8
So did Baikal delete his post about falling math scores in the U.S.?
Sorry, it isn't the math that we're having issues with. It's the usefulness of a table of numbers which doesn't tell us (any of us) how many of any particular locomotive a railroad has, nor how many they own in total. If we knew, for example, PC or BN or whomever owns a thousand locomotives, it is simple to reduce the numbers and discover they own however-many GP9s, SD40s, GG1s etc. But we don't know if they own a thousand, so we can't even do ANY of that math. It's totally useless. I suppose if I wanted a collection of PC or BN engines, I could use the table to get the proportions right, but that differs depending on era and locale I want to model. Northeast Corridor in 1968? Heavy on the GG1s. Indianapolis in 1974? Give me a lot of GP38s.
Why not just print the numbers right there on the table so we can see they have XXX GP9s, etc? No one cares what part of the cumulative total GP9s + SD40s + GG1s is.
I still think his table was a prank and he is challenging us to share it with others.
|
|
|
|
Post by gevohogger on Apr 5, 2024 4:08:50 GMT -8
This is probably a good time for me to present some of what I am working on. Allow me to present: The Average Area Code in Major Cities 1. Los Angeles - 458 2. Dallas - 552 3. Miami - 579
4. Denver - 581 5. Phoenix - 602
6. New York City - 628 7. Atlanta - 639 8. Chicago - 652 9. Boston - 737 Feel free to share with all your acquaintances! |
|
|
|
Post by Baikal on Apr 5, 2024 7:46:12 GMT -8
So did Baikal delete his post about falling math scores in the U.S.? Sorry, it isn't the math that we're having issues with. It's the usefulness of a table of numbers which doesn't tell us (any of us) how many of any particular locomotive a railroad has, nor how many they own in total. If we knew, for example, PC or BN or whomever owns a thousand locomotives, it is simple to reduce the numbers and discover they own however-many GP9s, SD40s, GG1s etc. But we don't know if they own a thousand, so we can't even do ANY of that math. It's totally useless. I suppose if I wanted a collection of PC or BN engines, I could use the table to get the proportions right, but that differs depending on era and locale I want to model. Northeast Corridor in 1968? Heavy on the GG1s. Indianapolis in 1974? Give me a lot of GP38s. Why not just print the numbers right there on the table so we can see they have XXX GP9s, etc? No one cares what part of the cumulative total GP9s + SD40s + GG1s is. I still think his table was a prank and he is challenging us to share it with others.
YOU don't find it useful. But other people did. Sound's like YOUR problem. For whatever reason, it sure seems to bug you.
You ask "Why not just print the numbers right there on the table so we can see they have XXX GP9s, etc? Ans: Because I didn't want to.
Why don't you post some roster info- in any format?
I didn't want to rub people's faces in it, but it's true, Americans don't do well in math (as you demonstate)
Here 'ya go:
Math scores for U.S. students hit all-time low on international exam
|
|
|
|
Post by Baikal on Apr 5, 2024 7:48:27 GMT -8
The Lehigh Valley is pretty easy, with small enough numbers of units it's not impossible to model all of them. By 1975 the percentages were fairly even for C420, U23B and GP38-2, with slightly more C628s (15) and slightly fewer RS11s (8 including the PRR leasers). It's more a question of what paint they wore in a given year than how many there were, although it varies some year to year from 1968-1976.
Now I need to do a LV roster... Great road to model, beautiful scenery too.
Add the proximity to D&H, PC, EL... nice.
|
|
|
|
Post by gevohogger on Apr 5, 2024 8:14:51 GMT -8
I don't think the math is the problem. It's the added step you're putting into the formula to glean any information from it.
EDIT: Never mind my point, you're just going to argue it until you're blue in the face. Have a good day.
|
|
|
|
Post by Baikal on Apr 5, 2024 10:46:00 GMT -8
Lehigh Valley locomotive models 1 July 1968. Cumulative percent of roster, ordered largest to smallest.
SW8 16.3
C-628 26.5
S-12 34.9
SW8M/900M 43.4
SW9 51.2
C-420 58.4
RS-11 64.5
DS-4-4-10 69.9
NW2 74.1
RS-2 78.3
F3B 81.9
F7A 84.9
SW1 88.0
SW7 91.0
GP18 93.4
VO-1000 95.2
GP9 96.4
RS-3 97.6
S-4 98.8
DRS-4-4-15 99.4
S-1 100.0
Total n = 166, about the size of WP's fleet. With that the n of each model can easily be deduced.
Interesting that four of the five most-numerous models were switchers, and that makes up over half the roster. Very un-like the WP. What does that say about their respective operations...? Note that 18 (2/3rds) of LVs SW8 fleet had dynamic brakes and MU.
|
|
|
|
Post by jonklein611 on Apr 5, 2024 11:12:15 GMT -8
Lehigh Valley locomotive models 1 July 1968. Percent of roster, ordered largest to smallest.
SW8 16.3
C-628 26.5
S-12 34.9
SW8M/900M 43.4
SW9 51.2
C-420 58.4
RS-11 64.5
DS-4-4-10 69.9
NW2 74.1
RS-2 78.3
F3B 81.9
F7A 84.9
SW1 88.0
SW7 91.0
GP18 93.4
VO-1000 95.2
GP9 96.4
RS-3 97.6
S-4 98.8
DRS-4-4-15 99.4
S-1 100.0
Total n = 166, about the size of WP's fleet. With that the n of each model can easily be deduced.
Interesting that four of the five most-numerous models were switchers, and that makes up over half the roster. Very un-like the WP. What does that say about their respective operations...? Note that 18 (2/3rds) of LVs SW8 fleet had dynamic brakes and MU.
Sounds like lots of locals, which makes sense given their route map:  |
|
|
|
Post by hudsonyard on Apr 5, 2024 13:51:53 GMT -8
By 1970 the Lehigh and Hudson River had 9 C420s and 3 RS3's supplemented by run through CNJ power, easy peasy!
|
|
