Will Americans ever take sleepers again?

[cirdan]

Could it be that since Europe has gotten so much HSR its sleepers are no longer needed for many of the shorter routes.

Yeah. The two sleeper services in the UK are really almost too short in duration to run sleepers. High speed lines are causing the same situation with a bunch of Continental routes; the trips are just too fast for sleepers to make sense.
But also, the longer routes in Europe, which still should have sleepers, mostly cross national borders, sometimes several times. There is substantial bureaucracy involved in doing that. This has caused many of the rail operators to not want to deal with cross-border sleeping cars. (Russian Railways is the exception, and is still running sleepers all the way from Moscow to Paris.)

Another problem in Europe lies in the transition from locomotives + cars to fixed consists. Because of this there are fewer and fewer locomotives, fewer and fewer switching locomotives and switching crews and station track layouts are slowly being rationalized to reflect this. Thus many railroads are asking, can they justify keeping the extra facilities for a once a day night train when the rest of the infrastructure is attuned to an hourly or more fixed-consist service. Many night trains split or join en route and this calls for middle of the night switching. Many sections also start their first leg attached to some day train, and if that day train goes from locmotive and cars to a fixed consist, you lose that option and either face the higher costs of that leg becoming a standalone train, or more likely, you dump it completely.

About 20 years ago, the European Union launched a blueprint for a trans-European high-speed network, with glossy brochures talking of Madrid to Moscow being the rail market of the future. Some of the ongoing investments such as Stuttgart 21 still claim to be serving this goal. But what has happened in reality has been the opposite. We have seen the emergence of lots of standalone high speed services, often with incompatible trains and lots of city pairs that previously had direct connections now require mutiple changes, sometimes even overnight stays in hotels, and despite the high speed, now take longer than they did 20 years ago. The beneficiaries of this are the low cost airlines.

20 years ago, most LD car were internationally normed and with the exception of the UK, Ireland, Spain and Poprtugal, which for various reasons were incompatible, these cars could go anywhere and could be mixed in the same consists. Even in the case of Spain there were some international trains using UIC cars. These had their trucks switched at the border. The same for trains from Germand and Poland to Ukraine and Russia. So with this high degree of standardization, international trains were not a big issue. Of course the locomotives were not as compatible and with a few exceptions (such as the French-Benelux multi-system locomotives, and also some German ones) these were changed at borders. But the passenger rarely noticed much of this as many border stations had slick and efficient techniques and the overall delay was not too bad. Today the opposite is true. Fixed consists are normally only suitable for the specific service for which they were designed. When a train is introduced that can run in different countries, this is trumpeted as a huge achievement while the PR folks hope we'll forget that 20 years ago that was the minimum you could expect.

It would be as if the Silvers were cut back at Washington DC and the Keystones to Philadelphia because the NEC was setup to accept only fixed consist Acela trains

 

[Anderson]

Cirdan:
What you described is my main fear at this point, actually: We get CAHSR, AAF, JR Central's project in Texas, NEC Future, and possibly something in the Midwest...and not a whole lot in between. Texas Central and AAF are going to have dubious connections with the broader system (and if that **** maglev project starts happening, add another built-in misconnect in the system if it doesn't manage to land within a few blocks of Union Station). It is entirely possible to envision a situation in the US where some major localized improvements start breaking apart the rest of the system.

To put this another way, if we start getting 220MPH+ trains running BOS-WAS, that could actually undermine SEHSR in a significant way due to folks suddenly wanting/needing to transfer at WAS to the faster service...but breaking the through operation of that service would have a chance of losing even more ridership.

 

[neroden]

Well, thankfully the Illinois hub projects and the Northeastern regional projects are pretty much running to the same standards, with the unfortunate exceptions of boarding height & overhead clearance, where there's an "eastern" standard and a "western" standard. Even there, Chicago is seriously proposing to restore a high platform at Union Station, supposely for "HSR" trains, which would completely match Northeastern standards, and would probably serve the LSL on day one.

CAHSR is going to run directly into LA Union Station, where nearly everyone has to change trains already anyway, and LAUS is actually being improved with run-through tracks for those who aren't changing trains.

The Texas Central project is unlikely to happen. Lone Star Rail would be on the same route as the Texas Eagle. Amtrak is still supposed to move over to the TRE route between Dallas and Fort Worth.

Florida... eh, Florida's gonna sink under the waves and the problem will be figuring out where to resettle the refugees. I'm not going to worry about a standards mismatch in Florida.

In general, we have the advantage over Europe that we are *not* a bunch of separate squabbling countries who refuse to cooperate.

 

[Barciur]

I wonder about Russia, as that seems to be the country with the most successful overnight sleeping service currently? Is it because the towns it serves have no alternative transportation, or is it still cheaper to travel by rail than to travel by air between those points?

Worth remembering that there is no coach seating on the eastern European LD's - Ukraine, Belarus, Russia - all of these have sleeping cars only.

 

[afigg]

CAHSR is going to run directly into LA Union Station, where nearly everyone has to change trains already anyway, and LAUS is actually being improved with run-through tracks for those who aren't changing trains.

The Texas Central project is unlikely to happen. Lone Star Rail would be on the same route as the Texas Eagle. Amtrak is still supposed to move over to the TRE route between Dallas and Fort Worth.

Florida... eh, Florida's gonna sink under the waves and the problem will be figuring out where to resettle the refugees. I'm not going to worry about a standards mismatch in Florida.

In general, we have the advantage over Europe that we are *not* a bunch of separate squabbling countries who refuse to cooperate.

I disagree on the Texas Central project. The actions in Dallas and Houston of the past several months have convinced me that it is a real project. It is by no means a sure thing, but I give it good odds of actually getting built in the next 10 years.

As for LA Union Station, it could rival Chicago or DC Union Station in daily total passenger numbers in 20 years between

CA HSR, Surfliner, other corridor services, heavy and light rail transit lines, Metrolink.

Instead of separate squabbling countries, we have 50 separate sometimes squabbling states, a dysfunctional Congress, and a deeply divided political system. Not clear that is much of an advantage at the present.

 

[JayMadison]

I can only speak for me, I've flown for nearly all of my 35 years, and only really experienced trains the last 2. At this point I'll always take a sleeper over a plane. I think the average traveler may just want to get there, but I do think getting there on a train in half the fun. The reason I keep riding the train is basically the level of service and all the interesting people I meet. I don't know what will happen, but I certainly hope long distance trains are here to stay.

 

[Anderson]

I'm with afigg on Texas Central for a few reasons, but the big one probably centers around the death of the Wright Amendment. A large part of what killed the Texas TGV was Southwest fighting like hell against it. This, in turn, was because Texas TGV under the Wright Amendment would have killed Dallas Love Field for Southwest (since flights out of Love Field couldn't go further than LA, AR, OK, or NM and to book beyond you had to actually get a separate ticket...the base of Texas TGV would have knocked out several of those destinations, and there would be a non-trivial chance that further expansion of the line could start running down hard on the remainder).

20 years later, the Wright Amendment is dead, but it took out 12 gates at DAL. The airlines as a whole seem happy enough to dump short-haul business to conserve slots for longer-haul flights...I suspect that Southwest wouldn't mind cutting a bunch of Dallas-Houston flights and swapping them for Dallas-Los Angeles, Dallas-Chicago, or Dallas-New York flights. They're not likely to fight too terribly hard. At the other end of the spectrum, you have Virgin America expressing interest in operating trains (although in California, the point still stands that they're up for operating them).

My understanding is that this is also the case in the instance of Los Angeles-San Francisco: You have a truly massive airline market, but it's a market that doesn't make much (if any money) but that can't be backed out of for fear of losing connecting business and the like. Again, every round-trip flight LAX-SFO is two round-trip flights that could be reallocated to another route (one at each end). I believe the same principle applies to LA-Las Vegas as well, and that's the second most massive airline market in the country if I'm not mistaken.

In several cases, the alternative is building a train route or massively expanding airport capacity...and the latter is not cheap: There's only so big you can make a narrow-body plane (and only so short of a route a widebody makes sense to run on) and only so much room you can squeeze out of an existing airport before you start having to fight over land again (see also: London Heathrow's third runway fight). There are exceptions where either a huge surplus of land was acquired or an airport is likely to remain in a semi-rural area for a while, but those tend to be the exceptions and I can't see SFO or LAX getting expanded without a hellacious fight.

 

[Daylight]

I have to travel quite a bit between Los Angeles and the east coast, mostly for business, and that usually means taking the bus with wings. I have also taken long distance trains, which from Los Angeles means a trip longer than 24 hours. However, if I have the time, I prefer to take the train. Flying today is so physically uncomfortable to me that it now takes me an extra day just to work out the kinks. I find that when I fly, I can't wait to get to my destination, but when I take the train I'm sad when the trip ends. When I board the train and it starts to rock-and-roll, a wave of relaxation comes over me (probably the infant inside of me).

I've done overnights in coach and in sleepers, and I agree the sleeper prices have been rather steep of late, but I've rarely seen an empty room throught a LD trip. Also, on the train I get to see more than the freeway (when driving), and the people are generally in a good mood, unlike when flying.

I usually tell people that ask me about taking the train that if you have to be at your destination by a particular time, then fly. If you include the train ride as part of your trip, then you'll either love it or hate it; some people can't sit still for more than a few hours.

 

[neroden]

I wish Texas Central good luck, but they haven't even started dealing with land acquisition. Of which they will require a great deal.

I just don't think there's the political will for it in Texas at this point.

California had the political will for it.

AAF already owned most of the right-of-way, and most of the the rest was owned by friendly airport and road authorities... and they still had tremendous trouble getting the *few remaining parcels* needed for the the curve.

In Texas, the route acquisition will require lots and lots of eminent domain; I think it's going to be tied up for years, and since it's being done by a private operator, I'm not even sure the state government will cooperate. Isn't Texas one of the states which passed an anti-Kelo law to make it hard to use eminent domain for private projects?

 

[timetableflagman]

The traveling public would have to return to the rails, en masse, if the airlines and highways WERE'NT subsidized (which they are). But, given that the Big Oil interests depend on the public's mass waste of Big Oil's product, fuel, wasted by driving and flying (and, little doubt, Big Oil owners have massive stock in those modes for that reason), then, no, we won't see the public being able to, much less having any incentive to return to overnight rail passenger service--no matter how much more sense it makes.

Arguably, that's the real reason we don't see AMTRAK immediately re-scheduling every one of their trains to originate around 6p and terminate at around 6a, with that general schedule recurring along the lines at major cities every 500 miles or so. That one act of re-scheduling of all of AMTRAK's trains to serve every major city on all of their lines would offer that overnight sleeping car service between major cities within 300-500 miles apart and, rush-hour coach service, at least within about 50 miles of those same cities--on the same daily trains!

How would overnight and rush-hour service be utilized by, and the huge increase of fares of which not help AMTRAK's financial situation: between Chicago and Cleveland, St. Paul/Minneapolis, Omaha, Kansas City, St. Louis, Memphis, and (with daily service) Cincinnati; between Greensboro and Atlanta; between Fayetteville and Jacksonville; between Los Angeles and San Francisco, (with restored DESERT WIND service) Salt Lake City, Phoenix and Tucson; between Seattle, Eugene, Spokane and, (with restored PIONEER service) Boise; between Little Rock and Dallas; between New Orleans and Houston; between El Paso, Del Rio, and (with new service between El Paso and Shelby, MT) Albuquerque; between Albuquerque and Dodge City; between Shelby and Minot; and, between Denver, Salt Lake City, and (with new El Paso-Shelby service) Casper? Combining trains at junctions to make for seamless, no re-boarding transfers between trains whose consists would continue on as the next, connecting train would better utilize equipment.

The sane, accomplishable schedules involved in meeting these times at these points by current trains would better accommodate host railroad freight service. Paying the host railroads ("profit-sharing") the reasonable rate for running passenger trains on their tracks, from the influx of more reliable, though reasonable and uniform fares ($7 coach seat, $14 sleeping berth, $28 roomette, $56 bedroom for every station departed (left or passed) along a line, with no gouging for "peak" times) from the resulting high volume of reliant, daily coach commuter and overnight business and travel sleeping car passengers, should result in on-time performance and no problems with host railroads accommodating AMTRAK trains (IF privately-owned host railroads care about profits!).

 

[Paulus]

The traveling public would have to return to the rails, en masse, if the airlines and highways WERE'NT subsidized (which they are). But, given that the Big Oil interests depend on the public's mass waste of Big Oil's product, fuel, wasted by driving and flying (and, little doubt, Big Oil owners have massive stock in those modes for that reason), then, no, we won't see the public being able to, much less having any incentive to return to overnight rail passenger service--no matter how much more sense it makes.

No, you'd see a combination of higher prices, higher taxes, and an overall reduction in trips. Flying and fast driving significantly increased the number of trips that Americans were willing to make, if they can't afford them an overnight train simply isn't an option for most trips. Deregulation of the airlines, and the resulting crash in airfares, created far more trips than they could have ever conceivably stolen in marketshare from Amtrak.

Arguably, that's the real reason we don't see AMTRAK immediately re-scheduling every one of their trains to originate around 6p and terminate at around 6a, with that general schedule recurring along the lines at major cities every 500 miles or so. That one act of re-scheduling of all of AMTRAK's trains to serve every major city on all of their lines would offer that overnight sleeping car service between major cities within 300-500 miles apart and, rush-hour coach service, at least within about 50 miles of those same cities--on the same daily trains!

No, that has somewhat more to do with other scheduling reasons and the fact that the commuter railroads aren't exactly going to be thrilled with LD trains messing up their lines at peak hours.

How would overnight and rush-hour service be utilized by, and the huge increase of fares of which not help AMTRAK's financial situation:

Adjusted for inflation, fares have significantly dropped over the years.

The sane, accomplishable schedules involved in meeting these times at these points by current trains would better accommodate host railroad freight service. Paying the host railroads ("profit-sharing") the reasonable rate for running passenger trains on their tracks

The reasonable market rate for a track slot is something like $150 per mile; more really if you account for the fact that on a primarily freight railroad a single passenger slot displaces multiple freight slots.

from the influx of more reliable, though reasonable and uniform fares ($7 coach seat, $14 sleeping berth, $28 roomette, $56 bedroom for every station departed (left or passed) along a line, with no gouging for "peak" times)

So you want to significantly cut coach revenue while massively increasing sleeper fares?

from the resulting high volume of reliant, daily coach commuter and overnight business and travel sleeping car passengers, should result in on-time performance and no problems with host railroads accommodating AMTRAK trains (IF privately-owned host railroads care about profits!).

You know what gets high volumes of reliant commuters? Frequency. Guess what doesn't and won't have the requisite number of frequencies? Slow speed overnight trains traveling on freight tracks.

 

[Lawdude]

I think Paulus nails most of it.

One of the big things that I think a lot of people don't realize (not just railfans, but also people who talk about air and bus travel as well) is what might be called the "democratization of travel". A lot of trips simply didn't happen in the past. Most middle and working class families, if they took vacations at all, took them very close to home. A vacation might have meant a trip to an amusement park, or the beach (if they lived in a coastal or river city), or camping in the local wilderness.

I was talking to a college student recently who was complaining about not being able to afford airfares to fly home at Christmastime. And I was thinking about the fact that I didn't go to school THAT long ago (in relative terms) yet large numbers of my classmates simply did not fly home at Christmastime. That wasn't seen as a universal right. Lots of people couldn't afford it. People think about the $390 fare they paid to take a train OR a plane across the country in 1983, and they don't think about the fact that the average middle class income was something like a third of what it is now and that $390 took a much bigger bite. Most people just didn't pay it. Especially students and middle and working class people.

So what has happened is that a lot more people travel now than who used to travel. But most of those people are middle and working class people who have very short vacation schedules. If you get 5 vacation days a year, an Amtrak long distance train is a very difficult option. The people who took the Super Chief, or even the El Capitan, way back when were not simply people traveling in an era when driving and flying were less attractive and less available (though that is true); they were also, generally, people who had more time to travel. They didn't need to be back at work and on the clock in a few days. If they had been, they could have never taken the train anywhere. And they were also, generally, people who had more money to travel, either because they were generally upper class or because they saved money up over a long period of time to take a trip. Nowadays, people can pay airfare for a trip out of their disposable income.

And I don't see how long distance trains with sleepers captures more than a small sliver of that market. As I said, people are very time-sensitive. They are also money-sensitive-- and if sleepers are priced closer to cost (like they are in Canada) and without free dining service and with a more limited Guest Rewards program (as they would need to be if Amtrak wanted to make a marginal profit on sleeper service and expand it), it's very hard to see how they could compete with a plane ticket in most instances.

Now, there are some routes where you could conceive of a point-to-point sleeper service that leaves at 9 pm and gets in at 7 the next morning or something, and such a thing is theoretically convenient compared to all the headaches of air travel. But then you run into the problem that the tracks are shared with freight and commuter railroads, and there's no guarantee they will be cooperative in allowing Amtrak to run such a schedule.

I really think in the end, all (rail)roads lead towards the sort of frequent, corridor service between cities a few hundred miles apart that doesn't excite some railfans (because it doesn't involve the sleeping and dining and lounge cars) but which provides something that is of real use to a lot of Americans, along with some long distance trains run along the lines of the Essential Air Service program (i.e., focused on providing transportation options for people who live in small towns along the route, rather than on people who want to take 2 day land cruises with free meals). But the air travel market has evolved the way it has because it provides something of real use to working and middle class Americans with limited budgets of time and money, and I don't see how trains can take more than a small bite out of that.

 

[chakk]

Lots of people traveling in the sleepers on the CL this time of year. For the past week, this train has had 3 sleepers plus the transition dorm sleeper, plus 3 (sometimes 4) coaches.

 

[DC Traveler]

Consider me a disinterested party, but whenever I see the justification for supporting long-term sleepers, the reasons seem to boil down to:

1- Looking out the windows

2- Talking with people

3- Eating food

Basically, these are things you do on a vacation like on a cruise line. Maybe I'm biased with my east coast routes, but based on these discussions and also the trip reports, it seems that many of the supporters of long-distance trains play the "comfort and luxury" angle, rather than the "utility and efficiency" angle. Which is fine, but I don't know if that's the best use of scarce resources, such as capital and political influence.

The truth is that the western US (excluding the coast) isn't very dense, and the cities themselves aren't very dense. This isn't an ideal situation for rail service at all. I appreciate that many of these small towns don't have airports, but at those levels of densities, I have to assume that it would be cheaper for the state-level DOT agencies to simply contract/subsidize Greyhound for intercity/city-to-airport service.

Even in Europe, sleepers are somewhat on the wane. The healthy sleeper trains are the one-overnight, PM departure/AM arrival which works for businessmen and tourists. Even with that, cheap airfares are certainly an important part of the market share. However, if you want the "luxury train trip" experience, you can do that, but it's not done under the auspices of national rail companies, but private operators.

People will take sleepers, but only when they're cost and time competitive with other modes. I can fly from DC to LAX in 5 hours, and even after you add in marginal time at both end, it doesn't come close to transcon sleepers, completely overlooking delays. And nevermind the fact that you also have to get early to the RR station for your trip.

TLDR: The numbers seem to imply that corridor services are the best bang for your buck. Tie it in with hubs at airports, and you'll get a good regional transportation network. LDs don't seem to cut it for utility or price.

 

[Anderson]

*sighs*
Again I feel compelled to differentiate between a one-night trip (ranging somewhere in the 7-16 hour range at the outsides...six hours and change being about the best you can manage after work and still get in by midnight or so, and sixteen being the most you can theoretically do outside of working hours) or things a hair longer and multi-night trips. The former have a viable, scalable non-tourist market; the latter generally do not (with some exceptions, generally centering on a lack of other alternatives...you've got various rural air markets that are basically stuck with pre-deregulation fares for coach and thin service that qualify here).

Honestly, beyond valuing being able to sleep with a bit of privacy (and having access to decent meals...a microwaved burger does not qualify here, thank you very much) I don't mind being a bit light on the frills. Something akin to the old Slumbercoaches would probably do me fine, even if I had to check a bag instead of having all my stuff in the room with me. I don't think I'm the only one in that boat.

Bear in mind that such efforts are not incompatible with corridor development; if anything, they mix nicely with it. IMHO the biggest issue would be encouraging the sorts of equipment shuffling that was done back in the 50s and 60s. You'd need to do a lot of "plug and play" at NYP, but if you could do that there are a number of destinations in the 4-10 hour range (or potentially within it) from NYP that could be served: BOS, MTR, TWO, BUF, PGH, WAS, ROA/LYH/CVS, and RVR/NFK/NPN all jump to mind depending on what you're willing/able to do as far as pairing them off with one another (i.e. MTR-LYH may not be doable, but BOS-LYH would be). If you were willing to set up 2-3 "pulses" at NYP (one hitting around 2200-2400 and the other at 0200-0400) you could shuffle quite a lot effectively, and in a lot of cases most of the needed trains already exist (66/67 and its counterparts are a key part here) or have been mooted (the evening train out of ROA/LYH comes to mind here).

You'd fundamentally only have 2-3 trains each way on the trunk of the NEC (i.e. PHL-NYP), with more branching from there...but a lot of those "back ends" have their own markets to serve (such as LYH-WAS or NPN-WAS). You're just running a sleeper and/or coach through to aid in connecting the far ends of the line and generally using an already-scheduled train to do it. Moreover, in some cases you're already looking at cases where second/third frequencies are verging into odd hours for one end or another...you'd arguably be adding value if there was a through connection of some kind available.

(For the record, I'd have to drag a bunch of timetables out in front of me, but I know most of this at least works on paper...the question of a bunch of switching in NYP is going to make someone in operations kvetch up a storm, but as long as you're managing it during the deep off-hours from about 2100-0500 you should be fine)

 

[neroden]

Now, there are some routes where you could conceive of a point-to-point sleeper service that leaves at 9 pm and gets in at 7 the next morning or something, and such a thing is theoretically convenient compared to all the headaches of air travel. But then you run into the problem that the tracks are shared with freight and commuter railroads, and there's no guarantee they will be cooperative in allowing Amtrak to run such a schedule.

The so-called "commuter" railroads are pretty cooperative. The so-called "freight" railroads have mostly been openly hostile to passenger service of any sort for 50 years, for reasons which became irrelevant decades ago -- though their attitudes have slowly been changing.
Don't consider train service to be competing with air travel on time or price; it doesn't *need* to. Train service competes with air service on quality, because air travel is utterly miserable for a lot of people, for various reasons. On time and price, the key is to compete with driving.

*sighs*

Again I feel compelled to differentiate between a one-night trip (ranging somewhere in the 7-16 hour range at the outsides...six hours and change being about the best you can manage after work and still get in by midnight or so, and sixteen being the most you can theoretically do outside of working hours) or things a hair longer and multi-night trips. The former have a viable, scalable non-tourist market; the latter generally do not (with some exceptions, generally centering on a lack of other alternatives...you've got various rural air markets that are basically stuck with pre-deregulation fares for coach and thin service that qualify here).

What Anderson said.

The future of sleeper trains is to be running on "corridor" routes, but a bit further. This is as it should be. The sleepers originally developed on top of a dense network of corridor routes. Where corridor routes cannot be supported, it is pretty hard to support sleeper trains.

Corridor trains should, of course, exist in Ohio, Indiana, Georgia, Florida, Louisiana, Texas, Wisconsin, Iowa, Minnesota, etc. They have been sabotaged in many of these states by political actions taken by people who are openly in the pay of the fossil fuel industry. If-and-when these corridors are built and operated, it will become obvious that the east-of-Mississipi sleeper services are effective and popular.

The empty deserts are a a different problem. I actually suspect that if you can't support two trains a day each way, you can't really support one train a day each way.

The majority of the sleeping car services on Amtrak run mostly on routes which would be good corridor routes, and some of which already are corridor routes. (This is the LSL, CL, Cardinal, Crescent, Star, Meteor, Auto Train, CONO, Texas Eagle, 9 routes -- I could add the missing Broadway Limited.) About 5 of these already contribute to Amtrak's bottom line (make money before overhead is allocated). The Cardinal would if it were daily.

All of them need to be faster (the Texas Eagle particularly), but then so do the corridors which they overlay. And, important point, they appear to require *less subsidy* than day trains on the same routes.

The other 5 routes are the "problem routes". Most of them also run on routes which would be good corridor routes, but spend substantially more of their time in the no-mans-land in between (Sunset Limited, California Zephyr, Empire Builder, Coast Starlight, 4 routes). The Southwest Chief has the least "corridor" running (basically nothing west of Kansas City).

Anyone who talks about "LDs" without distinguishing between the eastern and western trains is not paying attention.

 

[Anderson]

Honestly, the way I see it you can pack a number of trains full with a few bad geographic exceptions. Reno-Salt Lake City is one such gap (WP had dumped this line while UP was down to tri-weekly here by A-Day). The SWC and Sunset have similar problems.

The Empire Builder and Coast Starlight have a more complicated situation on the ground...both have, historically, had massive ridership (into the early 90s, combined ridership was over a million per year once the routes were both daily Superliner trains) but a crash-off in connectivity (loss of the Desert Wind and Pioneer) plus several infamous bouts of OTP problems and mudslides did a number on that...it took the Builder until the late 2000s to rise to where it was in the early 1990s, while the Starlight got hit by something in the early 1990s and never bounced back.*

Honestly, as I see it you've basically got a massive system that can mostly take care of itself...but cutting out a couple of random money-losing segments (SLC-RNO, for example) would make a hash of the rest of the system. So you keep a few "losers" in place for a bunch of reasons (connectivity being a big one...IIRC, something like 2/3 of passengers from the LDs into Chicago connect to another train), but anchor them with higher-frequency corridors (VAC-SEA-PDX and California for the Starlight, MSP-CHI and, to a lesser extent, SPK-SEA for the Builder).

With this in mind, it is worth noting that on the basis of demand the Zephyr could run a heck of a lot bigger CHI-DEN, DEN-GSC/GJC, and RNO-EMY. You get a lot of deboards at GSC and GJC out of Denver, Denver-Chicago is a massive market by Amtrak LD standards (hovering at close to 1% of the LD system[!]), and so on. Amtrak has looked into tinkering with equipment utilization as a result (witness some booking antics one can pull in the face of a "sold out" train on these segments).

Of course, once you get out of no-man's-land the picture changes drastically, and most of the rest of the network has room to scale up substantially within reason.

It is worth noting that you don't compete directly with air service for the most part (exceptions do exist)...but in some cases, like New York City, competing with airplane-plus-hotel is entirely doable alongside the competition with driving. Let's face it, a hotel in Manhattan will easily run you $200/night after taxes (I felt like I got a good deal when I was able to combine a friend's government discount with a last-minute rate drop and only shell out $150). Montreal can be a bit better or a bit worse, depending on exchange rates.

Still, the point stands that the situation in the East, generally speaking, is different from the situation in the West. CA/WA/OR falls somewhere in the middle, as does a good part of the western Midwest and some parts of the South.


*Worth looking into is a shakeup on the Starlight in the early 80s in conjunction with the Spirit of California (and comparing notes with overall ridership trends on Amtrak in this time...IIRC Amtrak took a bit of a lump overall then).
Builder/Starlight rideship data: http://www.wsdot.wa.gov/NR/rdonlyres/F67D73E5-2F2D-40F2-9795-736131D98106/0/StateRailPlanFinal201403.pdf

 

[Paulus]

Wasn't the 1990s issue for the Starlight the introduction/expansion of cheap California<->PNW air travel?

 

[twa904]

If Amtrak is like the airline industry, you have a lot of cost in starting and stopping the train every 25, 50 or 100 miles. If the train could run 400 miles nonstop it will cost less on a cost per seat mile basis than to operate it on 50 mile segments.

I remember back when Ron Allen was CEO of Delta he said there was no way Delta could make money flying a DC-9 between Atlanta and Augusta, GA even if he could tie people to the wings.

In another article I saw, shortly after Piedmont Airline started flying from Charlotte to London, it

cost them less on a per seat mile basis to fly the plane to London than it did to fly a B737 from

Charlotte to Birmingham.

 

[jis]

Amtrak is not like airlines, and should not strive to be like airlines. It should take advantage of its unique capability to effectively serve many communities on the way without adding unreasonable cost to operations.

 

[jis]

Well, thankfully the Illinois hub projects and the Northeastern regional projects are pretty much running to the same standards, with the unfortunate exceptions of boarding height & overhead clearance, where there's an "eastern" standard and a "western" standard. Even there, Chicago is seriously proposing to restore a high platform at Union Station, supposely for "HSR" trains, which would completely match Northeastern standards, and would probably serve the LSL on day one.

CAHSR is going to run directly into LA Union Station, where nearly everyone has to change trains already anyway, and LAUS is actually being improved with run-through tracks for those who aren't changing trains.

The Texas Central project is unlikely to happen. Lone Star Rail would be on the same route as the Texas Eagle. Amtrak is still supposed to move over to the TRE route between Dallas and Fort Worth.

My bet is, if Texas Central happens it will be high level platform.

Florida... eh, Florida's gonna sink under the waves and the problem will be figuring out where to resettle the refugees. I'm not going to worry about a standards mismatch in Florida.

Meanwhile before it sinks 80 or 100 years from now HSR in Florida will be high level as well as of course LD service on the Atlantic Corridor. All commuter rail will likely be low platform. Any restoration of service through the panhandle will most likely be low platform.

In a well capitalized system this is typically not a problem. Denver and indeed all other cities with a ,mix of rail transit modes potentially have some semblance of mixed platform heights between light and heavy rail, specially with the advent of tram-trains where light rail vehicles operate intermingled with heavy rail vehicles on the same infrastructure. They just use different platforms.

In general, we have the advantage over Europe that we are *not* a bunch of separate squabbling countries who refuse to cooperate.

Do we really? If you lived in New York City you would not know, given the endless disagreements, squabbling and non cooperation among The Port Authority, MTA, NJ Transit, and between LIRR and MNRR within MTA. It is probably the largest city in the world currently without a single integrated coordinated fare system for its transit.

 

[neutralist]

In another article I saw, shortly after Piedmont Airline started flying from Charlotte to London, it

cost them less on a per seat mile basis to fly the plane to London than it did to fly a B737 from

Charlotte to Birmingham.

well because Birmingham is 125 miles northwest of London so it is actually further away from Charlotte.
killing off LD trains could be construed as a form of discrimination. there are people who can't fly because of health reasons, or end up in illegitimate no-fly list because of their ethnicity.

 

[neroden]

I was gonna write a lot, but instead, uh, "What Anderson Said".

in some cases, like New York City, competing with airplane-plus-hotel is entirely doable alongside the competition with driving. Let's face it, a hotel in Manhattan will easily run you $200/night after taxes (I felt like I got a good deal when I was able to combine a friend's government discount with a last-minute rate drop and only shell out $150).

Yeah. Amtrak has unfortunately not managed to arrange its New York City connections to avoid overnight stays (a particular irritation for those of us from upstate NY). It would be well worth tweaking the schedule to reliably allow same-day connections from the upstate NY, Toronto, Montreal, and Vermont trains, to, well, everything whatsoever heading south. Right now there's a lot of misconnects. This would be made a lot easier by upgrades on the Empire Corridor.

The future of long-distance trains lies in upgrades on the same tracks for "corridor service". This should be possible on most of every existing route except the 5 westernmost routes, and also should be possible on several Eastern routes which used to exist and don't any more (Broadway Limited, Silver Palm, eastern Sunset Limited); it is merely a question of electing state governments supportive of passenger train service.

As for the 5 "problem routes", the Empire Builder and Coast Starlight consistently outperform the other 3, and will continue to do OK even with the "empty gaps". For the Empire Builder, there's no parallel superhighway and plane flights are very expensive. For the Coast Starlight, I'm not sure why it remains so popular -- Redding to Eugene is simply *way too slow* (9.5 hours vs. 5 hours driving). There may be some cultural effect, there being more of a "train habit" in California, Washington and Oregon.

The California Zephyr has massive potential from Chicago to Denver (build that Iowa corridor route now!), and from Denver to the ski areas (revive Ski Train now!), and from Reno to the Bay (extend the Capitol Corridor now!); with the revival of train service in Salt Lake City, there may be potential for improved Salt Lake-Denver ridership, but not on the mountain route (which takes too long), only on the Wyoming route. For some reason nobody goes from Salt Lake to the mountains to ski by train, only from Denver. Salt Lake - Reno is always going to be a hole in ridership which will cost a lot to run.

The Southwest Chief, sadly, looks like it's going to stay on the unpopulated Raton Pass route rather than scooping up the population center of Amarillo and people driving in from Lubbock. Y'all know what I think should be done.

The Sunset Limited is a curious case. There is a giant empty section in west Texas. But west of there, the cities along the route have exploded in population since the 1960s. Daily service would probably be well patronized. The route is relatively speed-competitive with driving from LA to El Paso, given appropriate scheduling (which it has). I really suspect that daily service would make this route look a lot better, despite the expensive-to-run vacant section from El Paso to San Antonio. Even better would be to avoid the Mexican border and reroute via Odessa & Abilene directly to Ft. Worth, but just try to get UP to agree to that (sigh).

 

[Paulus]

As for the 5 "problem routes", the Empire Builder and Coast Starlight consistently outperform the other 3, and will continue to do OK even with the "empty gaps". For the Empire Builder, there's no parallel superhighway and plane flights are very expensive. For the Coast Starlight, I'm not sure why it remains so popular -- Redding to Eugene is simply *way too slow* (9.5 hours vs. 5 hours driving). There may be some cultural effect, there being more of a "train habit" in California, Washington and Oregon.

Starlight runs over three different rail corridors (Surfliner, Capitol Corridor, and Cascades) which will make it look better (since it can function as an extra frequency and/or final train) and probably has a plurality of travel within California (56.5% of its boardings/alightings are within CA). There's also a fourth pseudo corridor given the Thruway bus frequencies between Sacramento and Redding.

 

[twa904]

In another article I saw, shortly after Piedmont Airline started flying from Charlotte to London, it

cost them less on a per seat mile basis to fly the plane to London than it did to fly a B737 from

Charlotte to Birmingham.

well because Birmingham is 125 miles northwest of London so it is actually further away from Charlotte.
killing off LD trains could be construed as a form of discrimination. there are people who can't fly because of health reasons, or end up in illegitimate no-fly list because of their ethnicity.

I had meant Birmingham Alabama not England