Re: SP Overnight service

Malcolm Laughlin <mlaughlinnyc@...>

Sorry for taking so lonfg to post this, but it was a very busy holiday weekend and this topic deserves some really thoughtful input.

I think that a lot of those comments on averages need a response, but first I’d like to provide some background information that is important for understanding what those statistics mean. These are based on my experience in car distribution management in various roles in the period 1963 to 1978. My first job related to car distribution was in 1963. At that time, except for substitution of computers for clerks and and small changes in average speed, utilization, etc. little had changed in 20 years. So, what I’m about to say is reasonably representative of the era of the STMFC.

There has been much commentary on average train speeds, car-miles, etc. From the perspective of model railroaders trying to have a reasonable replication of prototype operation of model cars, the average statistics are misleading to extent of being mostly useless.You can use averages to show that one doesn’t have to worry about drowning in a body of water with and average depth of two inches or that everyone in a room of less than 47 people is a billionaire if Bill Gates is among them

Now to address some specific comments.

Tim said
> “Most railroads were happy if the average LOCOMOTIVE managed to travel 200 miles every day.”

I assume this is about the 50’s, maybe 40’s It tells us nothing about any specific operation. Do you suppose NKP would have been happy to get that daily mileage with its Berkshires, or PRR with the T-1’s, NYC with the Niagaras and Hudsons, SP with the GS-4’s or UP with its Challengers and Northerns. The other side of the coin was the locals on short branch lines, the division way freights with their 100 mile runs and all those switch engines, transfers, etc. With all of those short runs, there must have been a lot of power that did 300-400 miles to get that 200 mile average. Then there were the locomotives in storage awaiting seasonal peaks, for example power for DM&IR’s ore trains and FEC’s passenger engines waiting for the four month winter season. Also if you are modeling the 50’s, you’re more than half way into the diesel era. That wouldn’t have happened at 200 mpd utilization. Which makes any conclusions drawn from average miles per day that isn’t separated
by steam and diesel really silly.

Tim also said
> “But 219 miles [per car day] sounds very high, especially given average freight train speeds of around 20 mph”.

Here we are seeing skepticism about a car mileage number that deviates from the average based on a train speed figure totally unrelated to car-miles and also not representative of any specific type of train operation. I’ll , first address the average train speed. In the 50’s, many railroads had a freight train speed limit of 50 on heavily traveled main lines, and on a few it was 60. If you saw a limit of 40 or 45, that told you that it was a drag freight railroad or mountain territory.

Then there were those way freights that spent most of their time doing useful work that didn’t generate train –miles, i.e., switching cars in and out of customer sidings. A way freight might have covered 80 miles in a ten hour day and handled 40 cars for customers. Its average speed was 8 mph. Does it make any sense to average that with the train that went 435 ,miles from New York to Buffalo in ten hours / 43.5 mph. It’s not just an apples and oranges thing. More like apples, bananas, peanuts and chicken nuggets.

So there was a time when if you took all of those freight train hours and divided them into the total train miles the answer was 20. But that number has no useful application, unless you consider it useful to have one more way to say that a railroad did better or worse than last year.

Another Tim comment
> “. So even though it only takes 15 hours (average) to travel 300 miles, those other activities can easily eat up a week or more.”

It rarely took so much time to travel 300 miles. That would happen only if there were a big delay in an intermediate yard. A through train at 50 mph would have done it in six hours, a drag at 40 with an hour at the division point, in eight to ten hours. In two division locals from a local station on one division to a station on the next division, 30 hours. So that average is meaningless.

Now to go back to that miles per car day number , which I will not dignify with any particular number. I’m looking at a car days per load report for the MILW for a week in October of 1974. I know it’s not the era for the list, but times hadn’t changed by 10 percent since then so it’s representative of our era. In four weeks, the weekly average varied from 24.0 to 32.3 days for ,wide door box cars, 11.7 to 13.6 for narrow door cars (this was the height of the grain season) and 11.9 to 20.5 for gons. Translate that to car-miles with your favorite average number.

Now consider all of those inactive cars. Ore jennies sitting in Duluth for the winter. Those obsolete PFE cars that Tony mentioned bringing the average down to three loads per year while waiting for the peak season. Cars on the cleaning track and coopering track. Bad orders. The multitude of cars out of service awaiting heavy shopping. Cars in the shop having racks for auto parts installed. Cars that will never see another load, but no one has taken the time to tell a carman to go out and paint a white line through the number. Automobile box cars sitting still during the several weeks of annual model changeover. Etc., etc., etc.

Now tell me the meaning of 15 days per load, or any other number.

> Another interesting comment from Tim that is quite out of context.
“”…cycle times are actually longer…more like 20-30 days. … someone who leases tank cars, and he has to lease one car per load, per month. (John Kneiling was right).”

There is big difference today because tank cars and covered hoppers are being used for storage of inventory to an extent that is tremendously greater than in the 50’s, or even the 70’s. John Kneiling was dead wrong about anything other than unit trains. That advocacy was good, but he was totally ignorant of the usefulness of the single carload freight business, which despite predictions in the 80’s of its demise, is thriving today. I could write several paragraphs on this subject, but it really is out of era for the STMFC and would be more appropriately discussed on the RwyOpSig list.

Another totally out of context comment from Tim.
> “Or looked at another way, 15 days * 50 miles/day would be 750 miles/cycle. If Malcolm were correct, 15*16 = 240 miles
per cycle. But that is far too low.”

I do make errors and I don’t mind corrections of them ,but this one is way out in left field. I mentioned specific cases, not averages. If you think my statements are questionable Tim, address the case that I actually mentioned.


In conclusion, it’s really dumb to say that any specific statistic is “astounding” or less than credible because it deviates from the average by 50 % or 200 % or more.

Malcolm Laughlin, Editor 617-489-4383
New England Rail Shipper Directories
19 Holden Road, Belmont, MA 02478

[Non-text portions of this message have been removed]

Join to automatically receive all group messages.