Midwest High Speed Rail

[CG9602]

The Revenue Passenger Mile is an accurate unit of measurement for passenger volume.
Note that it is *_not_* the same as "numbers per passenger" basis. If you focus upon the. Examining the passenger trains on a per passenger basis ignores the figures, from Amtrak, that demonstrate the load factors in excess of 60 percent for the LD trains, versus the 40 - 45 percent for short distance trains (Source: Amtrak). As for the corridors such as the NEC being profitable, it would be interesting to separate the NEC from the rest of the system. Then all could really see the costs of the stations, the caternary, the maintenance, et cetera. Perhpas the claims that one form of train subsidizes the other would finally be put to rest, when people see just how very expensive the NEC really is. See link: http://archive.unitedrail.org/documents/numbers.htm , http://archive.unitedrail.org/documents/highcost/index.htm , http://archive.unitedrail.org/documents/longdistancedemand.htm , http://www.midwesthsr.org/news_library.htm, http://www.nationalcorridors.org/PressRel10022006.html.

You will also find political support for train shrink when constituents tell their representatives that if they don't get their train, then they don't want their dollars going to support someone else's train.

Also, the numbers that have been listed are in contradiction to my personal experiences aboard the LD trains, in which I have found to be operating close to capacity.

[wjstix]

I don't think anyone's advocating ending long-distance train service, it's just that on many lines, that's all that's available. If you live in eastern North Dakota and want to visit the Twin Cities, your only rail choice is the Empire Builder, which picks you up at 3 a.m. on it's way east, and brings you back home at like 4 a.m. It would be nice to have a second train running the same route but on a different schedule, and maybe some local trains say Fargo-M/SP.

Also we need shorter distance connecting trains. To get from here (St.Paul) to Denver, I would have to take the train to Chicago, stay overnight, then go from Chicago to Denver. If I could go from here to Kansas City and transfer to the Denver train there, it would cut a day off my travel time or more.  Having a great LD train doesn't do any good if you can't get to it!!

[Datafever]

Also, the numbers that have been listed are in contradiction to my personal experiences aboard the LD trains, in which I have found to be operating close to capacity.

Which numbers do you feel are contradictory?  Amtrak's numbers regarding ridership levels?

Which LD trains do you take, and for which segments?  If your experience is not an adequate sample of the entire LD network, then there is the possibility for bias. 

[Suburban Station]

The Revenue Passenger Mile is an accurate unit of measurement for passenger volume.
Note that it is *_not_* the same as "numbers per passenger" basis. If you focus upon the. Examining the passenger trains on a per passenger basis ignores the figures, from Amtrak, that demonstrate the load factors in excess of 60 percent for the LD trains, versus the 40 - 45 percent for short distance trains (Source: Amtrak). As for the corridors such as the NEC being profitable, it would be interesting to separate the NEC from the rest of the system. Then all could really see the costs of the stations, the caternary, the maintenance, et cetera. Perhpas the claims that one form of train subsidizes the other would finally be put to rest, when people see just how very expensive the NEC really is.

or they'll see how passenger rail can be run without political constraints from uncertain budgets to misinformed citizens in other areas of the country. perhaps freed from the burden of providing commuter access at cost (rather than market rate)and having politicians set your pricing policy (see law that ended the large discounts).

 

See link:

 http://archive.unitedrail.org/documents/numbers.htm (1984), http://archive.unitedrail.org/documents/highcost/index.htm (1984), http://archive.unitedrail.org/documents/longdistancedemand.htm (2001-2), http://www.midwesthsr.org/news_library.htm, http://www.nationalcorridors.org/PressRel10022006.html.

I didn't read the midwest HSR one, the only recent one was a bit editorial for my liking, esp since it was written by the president of the national corridors initiative. I think the most telling is how much the numbers have changed over the years. clearly revenue growth in the nec has far outsripped the long distance routes. there are few measures that show long distance to be more profitable. please check prices from phl-to nyp for friday, they are as high as $101 for a regional train (not an acela). it's 90 miles. you're telling me that ridership isn't capped? chicago-milwaukee is $21 and it's about the same distance. Load factors are low because Amtrak wants it that way. If they lowered the prices, load factors woudl be way up and the train would be bursting at the seems. And please email them and ask them how they account for monthly riders (which exist on the corridor services). there is no such discrepancy in the long distance data. From a more commonsense standpoint, if it takes 18 hours for a long distance train, how many crew members is that? engineers? dining car and cafe car?

You will also find political support for train shrink when constituents tell their representatives that if they don't get their train, then they don't want their dollars going to support someone else's train.

bingo. of course, it's okay that their state gets billions in federal farm aid while others don't. politics, not a way to run a business.

Also, the numbers that have been listed are in contradiction to my personal experiences aboard the LD trains, in which I have found to be operating close to capacity.

that's probably true, since I imagine amtrak doesn't want to operate empty cars...esp long distance. It's my understanding that load factors on the acelas are frequently 80-90%.

[Datafever]

Recognizing that there are corridors embedded within the current LD routes doesn't mean the end of through traffic.  It might just mean that some small number of through coaches and sleepers may connect between trains on the route.

Not even worrying about LD passengers, just those who must make a connection between corridors -

Let's say that there are two trains a day between Denver and Omaha, and two trains a day between Omaha and Chicago.  That's two round trip trains.   The passenger arriving in Omaha that is continuing on must now make a connection.  If the train is late, do you 1) delay the connecting train?, or 2) make all connecting passengers wait for the next train (which may not be for another day)?

The way I see it, corridor trains are really only useful for the people who are travelling within that corridor, unless the number of trains in the connecting corridor is rather large (probably four or more a day). 

[nanaimo73]

Since Cleveland "service" consists of the Lake Shore and the Capitol in the middle of the night, would it be feasible to add a daytime (no sleepers) train from Chicago to Cleveland, which would split there for Buffalo and Pittsburgh. Those trains could be serviced in those cities overnight and continue to NYC in the morning. NYC to Buffalo, and the Pennsylvanian, could do the same westbound, arriving in Chicago the next day. 

[daveklepper]

Cleveland, Erie, Buffalo, and Toledo all deserve better service.   The first step would be for the NY State track and signal improvements to continue through Buffalo and then join  with Pennsylvania and Ohio rail improvements to continue to Cleveland, with a dawn to dusk NY-Cleveland schedule, replacing one of the existing NY-Buffalo round trips.   Then Cleveland would be part of the Midwest high-speed network, with corridor service of some type to Chicago and to Cincinnati and St. Louis.  Possibly also to Pittsburgh as you suggest.   And the Lake Shore on a somewhat faster schedule, would continue to be the through train, much as the Florida trains and the Crescent provide through service on NE Corridor.   And obviously good connections at Albany are required, better than the one train a day, to Boston.   But Massachusetts thinking is still car oriented, possibly because of rather than despite MIT.   The absense of North Station - South Station rail connector is the best evidence.

 

Suppose Federal and State spending had been limited to getting the farmers out of the mud and insuring decent paving and road connections to all communities.   And then all expressways and parkways and interstates were constructed by private enterprise and required tolls to earn a profit, pay real-estate taxes, and pay off the debts incurred in construction.   I would bet than an impartial economic evaluation of what would have happened would turn up the conclusion that much of the passenger network that survived to just before Amtrak would be profitable today, assuming the same kind of cost reduction and labor saving and added efficiencies that have been applied to freight railroading had had their analogies in the passenger service, as they have been on some of the more forward looking commuter operations.   (mechanised coach cleaning, power operated doors, etc.)

[CG9602]

I didn't read the midwest HSR one, the only recent one was a bit editorial for my liking, esp since it was written by the president of the national corridors initiative. I think the most telling is how much the numbers have changed over the years. clearly revenue growth in the nec has far outsripped the long distance routes.

When you consider what has been happening to the US economy as a whole, numbers are up for all trains, not just the ones in the NEC. Loss per passenger does not reflect the high infrastructure and capital costs of the NEC, which, while providing a necessary service, are so high that they obscure or wipe out any profits that are coming from the LD trains. When examining FY 2000 data, the LD trains show high passenger miles, 284 million, while the Acela had 35 million passenger miles. The other NEC trains had 156 million passenger miles. The LD trains had higher passengers per train-mile that any of the other types of trains in the Amtrak system, at 181 passengers per mile. The Acela had 176 passengers per train-mile, while the remainder of the NEC had 163 passengers per train-mile.

There are few measures that show long distance to be more profitable.

Except for the ones that discuss train-miles and revenue passenger miles, which are the units of measurement that are the most accurate.

please check prices from phl-to nyp for friday, they are as high as $101 for a regional train (not an acela). it's 90 miles. you're telling me that ridership isn't capped? chicago-milwaukee is $21 and it's about the same distance.

When you consider how Amtrak has been under pressure from Our Elected Representatives to keep its costs in line, one cannot blame Amtrak for charging what the market will bear. This also ignores the fact that the average LD trip distance is over 500 miles (Source: Amtrak. For FY 2000, the average trip legnth was 719 miles). Route legnth is not a major factor in determining a train's financial performance, trip legnth is. The LD routes are actually numerous itineraries (trips) that happen to take place on the same route. The overall train load may be upwards of 66 percent - functionally sold out- while embarking or disembarking even a modest number of passengers at each stop. So, when discussing profitability or loss on a passenger train, one has to use the most accurate measurements, and also take into account how far each passenger is traveling, and the capital requirements.

Load factors are low because Amtrak wants it that way. If they lowered the prices, load factors woudl be way up and the train would be bursting at the seems. And please email them and ask them how they account for monthly riders (which exist on the corridor services). There is no such discrepancy in the long distance data. From a more commonsense standpoint, if it takes 18 hours for a long distance train, how many crew members is that? engineers? dining car and cafe car?

You also failed to answer how discontinuing the routes will prevent decreases in passenger volume, as when people cannot connect, they will be less likly to take a train. Cutting or reducing a network while expecting it to maintain viability is about as smart as reducng the number of paved roads, in that reducing the network will reduce the amount of overall traffic volume.

It's my understanding that load factors on the acelas are frequently 80-90%.

No. According to Amtrak's FY 2006 numbers, the Acela is around 45 percent.

[Suburban Station]

When you consider what has been happening to the US economy as a whole, numbers are up for all trains, not just the ones in the NEC. Loss per passenger does not reflect the high infrastructure and capital costs of the NEC, which, while providing a necessary service, are so high that they obscure or wipe out any profits that are coming from the LD trains.

put down the peace pipe, LD trains do not make money. we've covered this ad naseum. you've provided antiquated data prior to the southwest effect (1984) and prior to any reporting improvements made by David Gunn. We all know that pre-Gunn data is extremely unreliable. According to the new issue of trains, railroads never knew what their costs were.

When examining FY 2000 data, the LD trains show high passenger miles, 284 million, while the Acela had 35 million passenger miles. The other NEC trains had 156 million passenger miles. The LD trains had higher passengers per train-mile that any of the other types of trains in the Amtrak system, at 181 passengers per mile. The Acela had 176 passengers per train-mile, while the remainder of the NEC had 163 passengers per train-mile.

what's the point of talking about 2000?

When you consider how Amtrak has been under pressure from Our Elected Representatives to keep its costs in line, one cannot blame Amtrak for charging what the market will bear.

but aas long as you realize they are filling a given set of equipment, not allowing it to increase. and we've already shown that the nec runs a gross profit (please note that this is different from net income). also note that the load factor woudl indeed be much higher if they charged the same rate per mile as they did for long distance trains.

This also ignores the fact that the average LD trip distance is over 500 miles (Source: Amtrak. For FY 2000, the average trip legnth was 719 miles). Route legnth is not a major factor in determining a train's financial performance, trip legnth is.

yes and no. only if peopel are willing to pay more for a long route tha a short route. also note that sleeper revenues include meals whcih means the cost are much higher for a given sleeper customer in terms of labor. not to mention they've gotta change crews at least once during the tripe.

You also failed to answer how discontinuing the routes will prevent decreases in passenger volume, as when people cannot connect, they will be less likly to take a train. Cutting or reducing a network while expecting it to maintain viability is about as smart as reducng the number of paved roads, in that reducing the network will reduce the amount of overall traffic volume.

 

It's my understanding that load factors on the acelas are frequently 80-90%.

No. According to Amtrak's FY 2006 numbers, the Acela is around 45 percent.

guess I can't quibble with you there, although I heard it from a fairly reliable source. now that you've called me dumb, i have to tell you long distance riders do not pay the bills. overall rail ridership woudl be much higher if a given set of dollars went to servce corridor markets. It doesn't matter how much antiquated, probably inaccurate data, you pull up.Even with capital costs the nec loses less than LD trains.  

[Anonymous]

Recognizing that there are corridors embedded within the current LD routes doesn't mean the end of through traffic.  It might just mean that some small number of through coaches and sleepers may connect between trains on the route.

Not even worrying about LD passengers, just those who must make a connection between corridors -

Let's say that there are two trains a day between Denver and Omaha, and two trains a day between Omaha and Chicago.  That's two round trip trains.   The passenger arriving in Omaha that is continuing on must now make a connection.  If the train is late, do you 1) delay the connecting train?, or 2) make all connecting passengers wait for the next train (which may not be for another day)?

The way I see it, corridor trains are really only useful for the people who are travelling within that corridor, unless the number of trains in the connecting corridor is rather large (probably four or more a day). 

You're forgetting - most LD rail travellers are in it for the casual haul, not really worried about time constraints, e.g. in vacation mode.  Therefore, if an inbound train is late and the connection with the outbound is missed, more than likely they won't complain too much as they adjust to an unexpected period of local site-seeing et al.  As long as the rail company accomodates them and doesn't leave them to fend for themselves.........

[Anonymous]

You're forgetting - most LD rail travellers are in it for the casual haul, not really worried about time constraints, e.g. in vacation mode.  Therefore, if an inbound train is late and the connection with the outbound is missed, more than likely they won't complain too much as they adjust to an unexpected period of local site-seeing et al.  As long as the rail company accomodates them and doesn't leave them to fend for themselves.........

 

Yeah, "right"!  So I've got reservations at the South Seas Resort in Florida, for my one week vacation. The connection is missed in Atlanta so the RR puts me up for the night at the "Peachtree Dismal" for a fun night of frollicking through the cable channels while I use my $10 meal allowance on a $50 room service bill , and a box of Goody's.

And all it costs me is one less day at the beach.....ohhhhh JOYYY!  

[CG9602]

guess I can't quibble with you there, although I heard it from a fairly reliable source. now that you've called me dumb, i have to tell you long distance riders do not pay the bills. overall rail ridership woudl be much higher if a given set of dollars went to servce corridor markets. It doesn't matter how much antiquated, probably inaccurate data, you pull up.Even with capital costs the nec loses less than LD trains.  

It was not my intention to call anyone dumb here. I was playing "Devil's Advocate," or in this case, LD trains advocate. I was pointing out that for all of the information regarding the NEC and the corridors, there are some stats that are contrary to the conventional asumptions re: Long Distance trains stated hereabouts. The point in talking about 2000 was 1. to produce evidence to support my argument, 2. to show that these stats have indicating what they've indicated for several years. The NEC does not make money any more than the LD trains do, and the capital costs are such that they cancel out any income that either group of trains brings in. To sum things up, when one counts the books differently, and assigns costs differently, one produces different numbers.

As for inaccurate data, individuals such as the Midwest High Speed Rail Assocation, as well as some Rail Passenger Assocations have argued, that the data being produced by Amtrak's Route Profitability System is suspect. This same data is what is being used to support claims that the NEC and the short trains makes money, when data points to the opposite. It is upon that foundation that the claims that the NEC is the financial Albatross around Amtrak's neck are made. That claim was first made in the 1980s and I think that it could still be made at the present time. That's why I brought up those old articles from the United Rail Passenger Alliance - the claims applied then, and they will continue to apply.

I concede your point that there is no telling which section loses more when the system which measures profit or loss may be flawed itself. I was making efforts to point out that there are plausible, acceptable reasons for continuing the intercity trains, and pointing out how they differed from the shorter distance trains. I'm sorry if I hurt any feelings here - That was not my intention at all.

[Datafever]

Not even worrying about LD passengers, just those who must make a connection between corridors -

You're forgetting - most LD rail travellers are in it for the casual haul, not really worried about time constraints, e.g. in vacation mode.  Therefore, if an inbound train is late and the connection with the outbound is missed, more than likely they won't complain too much as they adjust to an unexpected period of local site-seeing et al.  As long as the rail company accomodates them and doesn't leave them to fend for themselves.........

Futuremodal, did I really do such a poor job of explaining myself.  I thought that I made it clear that I was not talking about LD passengers.  I was talking about a passenger that needs to transfer from one corridor train to another corridor train.  Total trip may be 30 or 100 miles.

[n012944]

Recognizing that there are corridors embedded within the current LD routes doesn't mean the end of through traffic.  It might just mean that some small number of through coaches and sleepers may connect between trains on the route.

Not even worrying about LD passengers, just those who must make a connection between corridors -

Let's say that there are two trains a day between Denver and Omaha, and two trains a day between Omaha and Chicago.  That's two round trip trains.   The passenger arriving in Omaha that is continuing on must now make a connection.  If the train is late, do you 1) delay the connecting train?, or 2) make all connecting passengers wait for the next train (which may not be for another day)?

The way I see it, corridor trains are really only useful for the people who are travelling within that corridor, unless the number of trains in the connecting corridor is rather large (probably four or more a day). 

You're forgetting - most LD rail travellers are in it for the casual haul, not really worried about time constraints, e.g. in vacation mode.  Therefore, if an inbound train is late and the connection with the outbound is missed, more than likely they won't complain too much as they adjust to an unexpected period of local site-seeing et al.  As long as the rail company accomodates them and doesn't leave them to fend for themselves.........

 

Obviously written by someone who has never worked in a transportation customer service job. 

 

Bert

[oltmannd]

Recognizing that there are corridors embedded within the current LD routes doesn't mean the end of through traffic.  It might just mean that some small number of through coaches and sleepers may connect between trains on the route.

Not even worrying about LD passengers, just those who must make a connection between corridors -

Let's say that there are two trains a day between Denver and Omaha, and two trains a day between Omaha and Chicago.  That's two round trip trains.   The passenger arriving in Omaha that is continuing on must now make a connection.  If the train is late, do you 1) delay the connecting train?, or 2) make all connecting passengers wait for the next train (which may not be for another day)?

The way I see it, corridor trains are really only useful for the people who are travelling within that corridor, unless the number of trains in the connecting corridor is rather large (probably four or more a day). 

You're forgetting - most LD rail travellers are in it for the casual haul, not really worried about time constraints, e.g. in vacation mode.  Therefore, if an inbound train is late and the connection with the outbound is missed, more than likely they won't complain too much as they adjust to an unexpected period of local site-seeing et al.  As long as the rail company accomodates them and doesn't leave them to fend for themselves.........

...doubtful.  If the delay was a couple of hours, maybe.  If it was 24 hrs, there'd be much wailing and gnashing of teeth.

You' connect the ends of the spokes with thru cars (maybe a single connecting train, if the spoke ends don't touch) with guaranteed connections.  If you went to a hub/spoke/corridor model for operations, the whole nature of Amtrak changes.  You wouldn't be talking about corridor trains generally running 2 or more hours late.  The plan would necessarily include proper capacity and operating incentives as part of the deal.

We seem to be always stuck on Amtrak LD vs corridor when the problem isn't chosing between them, but how best to deploy the assets overall.

[Datafever]

You' connect the ends of the spokes with thru cars (maybe a single connecting train, if the spoke ends don't touch) with guaranteed connections.  If you went to a hub/spoke/corridor model for operations, the whole nature of Amtrak changes.  You wouldn't be talking about corridor trains generally running 2 or more hours late.  The plan would necessarily include proper capacity and operating incentives as part of the deal.

We seem to be always stuck on Amtrak LD vs corridor when the problem isn't chosing between them, but how best to deploy the assets overall.

And this is the part that confuses me.  How do you get guaranteed connections?  If you make the outbound train wait for the inbound train, then you basically have an LD train, with LD train performance.

Example.  SF to Reno is 30 minutes late, so Reno to SLC leaves 30 minutes late.  It get delayed by two hours enroute, so SLC to Denver then leaves 2 1/2 hours late.  And so on.

Unless you take the trains off the freight lines, or give them absolute priority.  Occassional breakdowns in connections can be worked with.  If connections are consistently missed... 

[Suburban Station]

. The point in talking about 2000 was 1. to produce evidence to support my argument, 2. to show that these stats have indicating what they've indicated for several years.

However, since Amtrak apparently has improved its reporting, the numbers no longer indicate that. what does that tell you?

.The NEC does not make money any more than the LD trains do, and the capital costs are such that they cancel out any income that either group of trains brings in. To sum things up, when one counts the books differently, and assigns costs differently, one produces different numbers.

true, partially. The point isn't that, as of now, they make money, but lose less. From a capital standpoint, it makes good common sense.

.
As for inaccurate data, individuals such as the Midwest High Speed Rail Assocation, as well as some Rail Passenger Assocations have argued, that the data being produced by Amtrak's Route Profitability System is suspect. This same data is what is being used to support claims that the NEC and the short trains makes money, when data points to the opposite.

Let's assume it is suspect, there's no proof that it doesn't understate the cost of long distance trains. I have seen no data pointing that it's opposite. what you've given me is even more suspect than what Amtrak itself produces...not to mention dated.

.It is upon that foundation that the claims that the NEC is the financial Albatross around Amtrak's neck are made.

suspect claims on suspect data. Of course, if we really want to get into it, not all long distance routes are created equal either.

.That claim was first made in the 1980s and I think that it could still be made at the present time. That's why I brought up those old articles from the United Rail Passenger Alliance - the claims applied then, and they will continue to apply.

for those looking for evidence that their hunches are right but these numbers are not only dated, but they are from an era in Amtrak where no number should be used.

.
I concede your point that there is no telling which section loses more when the system which measures profit or loss may be flawed itself. I was making efforts to point out that there are plausible, acceptable reasons for continuing the intercity trains, and pointing out how they differed from the shorter distance trains. I'm sorry if I hurt any feelings here - That was not my intention at all.

My feelings aren't hurt, just letting you know that's how it came off. Intercity trains, IMO, are the corridor trains. While this thread has rambled, it's comeout that if the bill passes for Amtrak, Wisconsin will move to extend the Hiawathas to Madison. Hopefully improve travel speeds as well. Long distance trains are nice, as Kummant himself says, in the national park service kind of way. However, corridors are what most people seem to use trains for. I have a reserved optimism in that Amtrak's new boss is their first private sector CEO and hopefully will bring private sector quality to its financial reporting (which, of course, is usually a reflection of investment in IT from top to bottom...meaning, from the point of collection to reports)

[oltmannd]

You' connect the ends of the spokes with thru cars (maybe a single connecting train, if the spoke ends don't touch) with guaranteed connections.  If you went to a hub/spoke/corridor model for operations, the whole nature of Amtrak changes.  You wouldn't be talking about corridor trains generally running 2 or more hours late.  The plan would necessarily include proper capacity and operating incentives as part of the deal.

We seem to be always stuck on Amtrak LD vs corridor when the problem isn't chosing between them, but how best to deploy the assets overall.

And this is the part that confuses me.  How do you get guaranteed connections?  If you make the outbound train wait for the inbound train, then you basically have an LD train, with LD train performance.

Example.  SF to Reno is 30 minutes late, so Reno to SLC leaves 30 minutes late.  It get delayed by two hours enroute, so SLC to Denver then leaves 2 1/2 hours late.  And so on.

Unless you take the trains off the freight lines, or give them absolute priority.  Occassional breakdowns in connections can be worked with.  If connections are consistently missed... 

Two things.   First the corridor part of the trip would have capacity for pretty good time keeping.  If it doesn't, then it's not really a corridor, it's just a joy ride route.  Second, the connections would have enough padding in them to make them reliable. 

Let's say, Phila to Pittsburgh is a corridor and Chicago to Cleveland is another, each with 10 RT a day.  Your NY to Chicago service would ride a NY-Phila-Pittsburgh schedule to Pittsburgh, and then depart as it's own train to Cleveland.  At Cleveland, there would be a gap of a couple hours before the through cars connected to the Cleveland-Chicago "corridor" train.

The train could be overnight, riding an evening train out of NY to Pittsburgh, with a morning arrival in Cleveland.  Then, the connection would be to a mid-morning Cleveland -Chicago train.

In this scenario, Amtrak would have to "invest" in NS's routes to support the Cleveland-Chicago and Harrisburg-Pittsburg portions of the route to support the new service level.  Perhaps a thrid main track or at least some healthly portions of triple track.  The "new" track could even be marginal high-speed (100 mph or so) and more-or-less passenger only.  In return NS would have to support a decent run from Pittsburg to Cleveland.  That wouldn't be too hard because only 150 miles or so is now in jeopary of being trapped on a tenant RR.  So, that Cleveland connection would be highly reliable.

[Anonymous]

You' connect the ends of the spokes with thru cars (maybe a single connecting train, if the spoke ends don't touch) with guaranteed connections.  If you went to a hub/spoke/corridor model for operations, the whole nature of Amtrak changes.  You wouldn't be talking about corridor trains generally running 2 or more hours late.  The plan would necessarily include proper capacity and operating incentives as part of the deal.

We seem to be always stuck on Amtrak LD vs corridor when the problem isn't chosing between them, but how best to deploy the assets overall.

And this is the part that confuses me.  How do you get guaranteed connections?  If you make the outbound train wait for the inbound train, then you basically have an LD train, with LD train performance.

Example.  SF to Reno is 30 minutes late, so Reno to SLC leaves 30 minutes late.  It get delayed by two hours enroute, so SLC to Denver then leaves 2 1/2 hours late.  And so on.

Unless you take the trains off the freight lines, or give them absolute priority.  Occassional breakdowns in connections can be worked with.  If connections are consistently missed... 

I also go by the "overnight rule".  If most of these HSR passenger trains are overnighters, then being a little late or early to the destination won't matter for the LD passengers, since the next connection isn't scheduled to depart until the evening hours.

Also, the overnight rule would allow HSR passenger trains to be complemented by conventional passenger trains.  Although an HSR overnighter would cover roughly 1000 miles give or take, the conventional overnight would cover 300 or so miles.  So, the HSR overnighter between say SF and Salt Lake is complemented by the conventional Boise-Salt Lake overnighter.