I thought the 3rd and 4th paragraphs interesting. I'm not sure how he calculates origin/destination (O/D) opportunities. How does he get 1,640 O/D pairs between Chicago and Seattle? He makes an interesting observation regarding the Texas Eagle connecting with the Sunset Limited.
I don't know if I completely agree with the article, but he makes a point for additional roomettes, sleepers, and premium coaches on L/D trains.
Richard
Posted by Bob from MA (Member # 4686) on :
I belong to MinnARP and regularly see articles by Selden in its newsletter. He is a great advocate for the long-distance trains, claiming they are profitable, unlike the Northeast Corridor. He is very knowledgeable on rail matters. Some time ago he was a candidate for head of Amtrak.
Posted by TwinStarRocket (Member # 2142) on :
I even attended the same Sunday School as Andy Seldon in the '50s and he was one of the brightest bulbs in the room in an affluent suburban church. I have also followed his opinions for decades as a MinnARP member.
Unfortunately, all of his MinnARP newsletter articles are print only, so they do not easily lend themselves to be shared for discussion on this forum. Since his impressive career in Minnesota was in accounting and law, his opinions regarding how Amtrak allocates costs and assesses profitability certainly have credibility.
Of course as a fan of the LD's, I want to believe his opinions are correct.
Here is a short recent example I found recently of Seldon's analysis of how NEC vs LD costs are misrepresented.
[Long winded answer deleted]
Posted by RRRICH (Member # 1418) on :
Geoff and others -- it's simple math. There are 41 stations on the line. Whatever station one boards at, he has a choice of 40 other stations at which to disembark.
41 (possible origin stations) X 40 (possible destination stations) = 1640
Posted by RRRICH (Member # 1418) on :
Actually, if you consider say, Whitefish to Minot the same "city pair" as Minot to Whitefish, then there are only half as many "city pairs" available, or 820.
Posted by TwinStarRocket (Member # 2142) on :
I'm not much of a mathematician, but as a foamer I can propose the theory that there are really only 40 total stops on #7/8. East Glacier is replaced by Browning on Oct 11, and switches back in the spring. This is noted by footnote #40 on the timetable. Coincidence?
Posted by yukon11 (Member # 2997) on :
I'm not much of a mathematician, either.
I think Andrew Selden's reference to origin-destination pairs is significant and needs to be explored, further.
However, to my way of thinking, you also have to ask yourself, for each passenger, questions regarding "who, why, and where". Who is the passenger, why is he/she taking the train, and where is he/she going. Why the train instead of an automobile, plane, or bus? How does he/she get to a depot and how does he/she get from the depot to his/her final destination. For example, many might opt for the Empire Builder because of dangerous and icy roads during the winter months, or because there aren't many towns or cities, along the route, with commercial airports.
I think one of the problems, with Amtrak, is the constant trend to think small. To prove Mr. Selden's point, I think Amtrak would need a good prototype. Equip just one long distance train with the type of cars & services which might greatly enhance revenue. Many more sleepers, superb dining car meals, bistro cars, and upscale coach cars and lounge cars, with all the amenities. Then take a look at revenue. Weigh the revenue against the cost of the upgraded cars and services over time. If Mr. Selden is wrong, well, at least his concept is put to the test.
Richard
Posted by TwinStarRocket (Member # 2142) on :
Here is a simple math problem I have always wondered about. If a Superliner sleeper has 20 rooms at $300/night and runs full, the revenue is $6000/day. They have been in service now 35 years? $6000 x 35 years x 365 days = $76 million in total revenue for a rail car that cost $1-2 million? (I have no idea). In any case, the cost of building one should be recovered in a short time relative to the cars (currently indefinite?) lifespan.
So does an operating staff of one, daily meals for 30-40 people, furnishing and maintenance really come to more than the $2+ million in annual revenue? Or more simply put, does a long distance sleeper really lose money? And if not, why not build some?
The same might apply to coach, but I am not sure they always run as full as sleepers.
If Andrew Seldon's city pair theory is really a way to increase ridership, first you need the equipment.
Posted by PullmanCo (Member # 1138) on :
Remember, you have 4 trains running at any moment, and two more in the wings. So, for every day, you're paying 4-6 mandays on 3/4, 5/6, 7/8, 11/14...
Also, sleepers do not have 100% filling, including at least one space for the Porter him/herself.
Posted by PullmanCo (Member # 1138) on :
Finally, how many people run end-end, and how many are "shorts?"
Posted by yukon11 (Member # 2997) on :
quote:Originally posted by PullmanCo: Finally, how many people run end-end, and how many are "shorts?"
**************************** It could, Mr. Pullman, depend on which route. Selden, in the article, says no passengers went all the way from Chicago-LA or LA-Chicago in Feb. I'm sure some do take the SWC the complete route, but I have seen other stats which show the majority of passengers don't. Probably more ride the EB the complete route, CHI-SEA and SEA-CHI, but it might be interesting to look at the o/d stats for most major LD trains and see how many are short trippers.
Richard
Posted by TwinStarRocket (Member # 2142) on :
PullmanCo, I do most of my traveling in the off season. I find from personal experience there are very few unoccupied rooms that I see on the train. And I still always have to work around sold out dates. There are 14 roomettes, 5 bedrooms, a family room and the handicap room on a superliner so that leaves one for the attendant when I say 20.
When it is not off-season, finding a room on the day I want within a current week is very difficult. Many days are sold out of first class.
But most importantly, when I choose $300 as revenue for a one day (24 hours), the actual price paid for a room might be more like $500-$1200 for 24 hours. Hopefully this would compensate for unoccupied times.
If I look at Amsnag right now over the next 5 days, from MSP to Whitefish (24 hours): Roomettes run about $582, Bedrooms about $520-$1285. Family rooms are sold out all 5 days on #27. Bedrooms are sold out Thu on #27 and Sat on #7. This is not only off-season for Whitefish (Glacier and skiing are virtually shut down), but perhaps the most sparsely populated 24 hours of traveling you can do on Amtrak.
I was trying to create a simple example for one superliner sleeper. Most likely my guess of $6000 revenue per day is low. I would love to see some actual data to dispute that. If it cost more than $6000/day to run one, that I could believe. But I ave no idea what that cost might be.
Posted by PullmanCo (Member # 1138) on :
Looking at Mr TwinStarRocket's post...
Factor in as well...
- Pro rata share of the baggage car operations expense. - Pro rata share of the lounge cars expense. - Pro rata share of the locomotive's expense. - Pro rata share of the train operating crews expense. - Pro rata share of the cost for the train to ride someone else's rails. - Pro rata share of overhead (Beech Grove Shops, National offices, stations, ad infinitum)
All of those are costed into the ticket.
Posted by TwinStarRocket (Member # 2142) on :
Yes! And these pro rata shares PER CAR would go down when you add the first car. I believe also what is paid to the host railroad might be per train for passenger rail, unlike freight.
Posted by PullmanCo (Member # 1138) on :
GBN: You're the CPA, and were a railroad accountant. Professional thoughts on the factors going into a ticket rate?
Posted by Geoff Mayo (Member # 153) on :
Though this blog post is more about Metrolink, Amtrak is mentioned and similar principles should apply for many of the costs. Paul Druce, the author, is somebody I've followed (in the electronic sense!) for a while now and his analysis of figures is usually pretty good.