Passenger Miles Per Train Mile Metric

I was looking at the 1979 Amtrak cuts which included routes such as the National Limited and the Floridian among others. The cuts were somewhat based on a report to Congress by Brock Adams, Secretary of Transportation under Carter (https://en.wikipedia.org/wiki/Brock_Adams). The report had suggested more cuts than were actually made.

https://www.fra.dot.gov/eLib/details/L04153

The recommendations were loosely based on the metric of passenger miles per train mile (PM/TM). Per the report:

"To use those criteria, the Department first determined the number of passenger-miles per train-mile each route under consideration has generated and estimated the ridership each would generate. As discussed in the Preliminary Report, the Department considers passenger-miles per train-mile to be a valid primary measure of market and population requirements. The number of passenger-miles per train-mile represents the number of passengers, on the average, aboard a train at any given moment during its journey."

I have usually pointed to R & R to measure trains but usually if trains are longer the costs increase. So the PM/TM takes that into account.


A previous report from 1978 listed PM/TM for all Amtrak routes in FY 1977 (Table 3-11):
https://www.fra.dot.gov/eLib/details/L04177

Here are the PM/TM's for the LD trains at the time (listed by endpoints although it is easy to deduce which train was which):
Los Angeles to Seattle 225
New York to Florida 215
Chicago to Los Angeles 200
Chicago to New York/Washington 172
Los Angeles to New Orleans 172
New York to Savannah 161
Chicago to New Orleans 142
Chicago to New York/Boston 142
Chicago to San Francisco 127
Chicago to Seattle (Havre) 119
Chicago to Houston 113
Chicago to Seattle (Billings) 99
Kansas City to New York/DC 89
Seattle to Salt Lake City 87
Chicago to Florida 74
Chicago to Washington (Cincinnati) 60
Chicago to Laredo 57

The Southern Crescent was not an Amtrak train in FY 1977. The 1979 report said service would begin February 1 although it did propose discontinuing the route, saying its PM/TM was 117.

All of the routes whose PM/TM was 117 or below were recommended to be discontinued. They also proposed a route from CHI to California which would have gone through Kansas City and Denver and split in Ogden, Utah, replacing what was then known as the Southwest Limited and the San Francisco Zephyr. The Ogden to LA train eventually became the Desert Wind but that was cancelled.

They meant for the Empire Builder to be the CHI to NW train. The Pioneer between SEA and SLC stayed until 1997.

Both Chicago to Texas trains would have been cut although they essentially wound up being merged into the Texas Eagle.

https://en.wikipedia.org/wiki/Texas_Eagle

They wanted to have the LSL go through Detroit and Canada. "To avoid delays caused by customs formalities the train should operate on a "closed door" basis through Canada. Then the BL would've gone through Cleveland and split in Pittsburgh to serve the old Shenandoah route "as far as Cumberland, Maryland". I don't believe the LSL ever went via Detroit. The Capitol Limited began Oct. 1, 1981. It originally shared the Ft. Wayne route with the BL but became its own train and eventually did get rerouted through CLE in 1990.

https://en.wikipedia.org/wiki/Capitol_Limited_(Amtrak_train)

Finally the Cardinal was discontinued on Sept. 30, 1981 (the day before the CL was born) but was resurrected on Jan. 8, 1982 "by Congressional mandate".

(https://en.wikipedia.org/wiki/Cardinal_(train)%C2

I also found PM/TM data for the LD trains from the various PRIIA reports from 2010 to 2012.

Auto Train (FY12 Q1) 364
Coast Starlight (FY12 Q1) 220
Lake Shore Limited (FY11 Q1) 215
Silver Meteor (FY11 Q1) 208
Texas Eagle (2010 report, FY 2009 Data) 204.9
Empire Builder (FY12 Q1) 201
Southwest Chief (FY12 Q1) 197
Capitol Limited (2010 report, FY 2010 Data) 193.0
Silver Star (FY11 Q1) 185
City of New Orleans (FY12 Q1) 164
California Zephyr (2010 report, "Last 12 months"): 163.0
Crescent (FY11 Q1) 159
Palmetto (FY11 Q1) 142
Cardinal (2010 report, FY 2010 estimate): 119.2
Sunset Limited (2010 report, FY 2009 Data) 107.9

Comparing from the 1977 data,

The Silver routes and Palmetto have lower PM/TM than back then (the New York to Florida did not specify a specific train) but that might be in part due to the Auto Train which cleans up in that stat.

The Empire Builder is roughly what the 1977 EB and North Coast Hiawatha trains added up to.

The Texas Eagle is a bit higher than the combined totals of the 1977 trains to Houston and Laredo.

Back in 1977 there was a Shenandoah between WAS and CIN with a 1977 PM/TM of 33. The current Cardinal's PM/TM is higher than the old Cardinal plus Shenandoah.

The CL is slightly higher than the old 1977 BL.

The big surprise is the Sunset Limited which was way higher in 1977 (and according to an old timetables.org schedule was also three times a week back then).

http://www.timetables.org/full.php?group=19770622&item=0036.

The route back then served Phoenix while the current does not. Back then they ran through cars between NYP and LAX via ATL and Birmingham allowing for a transcontinental route between NYP and LAX although only 3 times per week (in addition a similar combo existed between NYP and LAX using the then National Limited and Southwest Limited.

Do these two changes account for a much lower PM/TM in 2009 than 1977 or are there other factors that led to worse performance?



Three of the recommended changes in the PRIAA:

The CL/Pennsylvanian connection was projected to increase the PM/TM from 193.0 to 226.8.
http://www.amtrak.com/ccurl/152/943/PRIIA-210-CapitolLimited-PIP.pdf

The Sunset Limited/Texas Eagle combination was projected to have a PM/TM of 132.0.
http://www.amtrak.com/ccurl/970/304/PRIIA-210-SunsetLtd-TexasEagle-PIP,0.pdf

A daily Cardinal was projected to increase the PM/TM up to 125.5. if Option #7 had been chosen, the PRIAA was projected to go up to 142.
http://www.amtrak.com/ccurl/536/878/PRIIA-210-Cardinal-PIP.pdf

I don't know if PM/TM is any more accurate than R & R but it is the statistic primarily quoted in the Department of Transportation's reports back in 1978/1979.

Does anyone have more up to date PM/TM? How do you calculate it?

Paulus said:

You can find the most up to date info here

Click to expand...

Thank you so much Paulus.

Auto Train was not an Amtrak operation in the 70s. I believe it was Graham Claytor's time when it was resurrected from the dead by Amtrak after the original operators went bankrupt and shut down operations.

Anyway since PM/TM can be doctored by offering more or less accommodation on a train to create whatever number you want, I don't trust that as a good measure for making any decision.

PM/TM is a reasonable measure for routes which never sell out (which always have enough capacity), which have relatively fixed prices and service levels, and which haven't been subject to recent disasters. So it's pretty decent for most commuter-rail systems. Jis is right that it can be doctored given Amtrak's car shortage and yield management system, though it's still a useful measure *in conjunction with* measures of ticket yields and identification of sell-out conditions.

It's also less meaningful with multiple trains per day unless you compute it for the whole collection of trains as a group: so a Piedmont PM/TM by itself makes relatively little sense without including the Carolinian running along the same route segment. It's also less meaningful with multiple trains per day, period, as a lower PM/TM can be achieved by simply running trains more frequently -- which is very popular -- but a lot of the operating costs are fixed whether you run two trains a day on a line or a dozen.

Of course, a low PM/TM may simply mean "this train hasn't been running on time recently and passengers are temporarily avoiding it for that reason", which was actually the situation with most of the trains cut in the Carter Cuts. So it's not a rational basis for getting rid of routes: if anything, a low number may mean that a route needs capital improvements.

The most important thing to note from these numbers is that PM/TM is higher than it was in 1977 for the majority of trains, and much higher for New York - Chicago. It's in the same ballpark for most of the rest, including the Coast Starlight (down very very slightly).

The significant fall for the Sunset Limited is mostly due to losing Phoenix service. It's the third largest metro area on the route, after LA and Houston; losing service was disastrous for the route.

Looking at the more recent numbers, it is notable that the PM/TM numbers do roughly correspond with a reasonable gut sense of how full the trains are. The Hoosier State has the worst ratio (63), followed by the Piedmont (72), Capitol Corridor (85), River Runner (87), Heartland Flyer (89), and Downeaster (90). (The Capitol Corridor is simply distributing its passengers over a lot of trains.)

Meanwhile, the best ratios are on the Auto Train (383) and Lynchburg (340), both of which are packed full and profitable.

Paulus: thanks for the FRA links. I wish they'd fill in Table 1A and 1B, "percent of short term avoidable operating costs covered by passenger-related revenue, including/excluding state revenue". My belief is that the long-distance network looks excellent on that basis, but it's all "not available".

Withdrawn! Facts are facts, I didn't do my research!

Never mind!

neroden said:

Paulus: thanks for the FRA links. I wish they'd fill in Table 1A and 1B, "percent of short term avoidable operating costs covered by passenger-related revenue, including/excluding state revenue". My belief is that the long-distance network looks excellent on that basis, but it's all "not available".

Click to expand...

Yeah, I've been looking forward to those tables, but they've stopped publishing those reports at all.

Here is PM/TM for each individual Surfliner incidentally

Phoenix MSA - 4.489 million

San Antonio MSA - 2.329 million

Austin MSA - 1.943 million

Phoenix city - 1.537 million

San Antonio city - 1.437 million

Austin city - 0.913 million

Passenger metrics are very complicated and need more thought. Passenger miles per train mile ( PM/TM ) will only be reliably measured for those trips that have one or more legs not sold out. The more legs sold out the more constrained the PM/TM will be. So to measure real PMs/TM every train ( or at least 90% ) would have unsold seats. But fare levels ( cents per mile ) affect the number of passengers wanting to travel on any train. For multiple trains on a route time of day affects PM/TMs. ( ex. CHI - MKE ).

Another measure will be load factor. Average load factor on multi stop trains is very hard to get over 50 - 55% due to certain legs ( especially near end of train run ) running lower. ANY limiting load factor is the max load factor leg. Great example is the Crescent. At one time more days than not it would sell out Charlottesville - WASH limiting any sales thru that segment. Once the Lynchburg train was initiated almost no times is the Crescent sold out there enabling more longer PM/TM.

a. changes depending on which day of week.

b. changes for time of year

c. direction of travel

d. Holidays

e. Auto Train is probably the purest train that can be measured but even that is affected due to having to balance equipment such as 350 passengers south and 150 north ( some seasons ) and vice verses at other times.

f. Further complications for the Crescent example is CVS - WAS very occasionally selling out in coach but ATL/CLT - WAS often selling out for sleepers now. So we have one metric PM/TM for coach and One for sleeper on any train / route.

Then rate of return comes into play.

If 100 passengers are paying $0.20 per mile compared to NEC rates of $1.00 per mile then for same capacity train then which do you choose to operate ?

Operating costs divided by mile ( LD car mile costs about $4.00 per mile )

Crew costs assuming same capacity

a. number of sleeper crew

b. Diner and lounge crew

c. number of T&E crew

d. Slower routes requiring more OBS crew hours. ( delays factor in also ) ( Crescent longer than silvers ) ( Cal Z longer the SW Chief ) NEC T&E can do round trips.

It does appear that as new equipment is added ( big IF ) saving the older equipment as a surge fleet has big dividends

IF

a. storage costs less than additional revenue

b. part time employees can be had for peak period runs

c. Some one in congress doesn't push to scrap equipment not be used all the time. ( The scrapping of Heritage coaches and sleepers back when ) .

d. Granted now all remaining Amtrak Heritage cars will be scrapped except maybe diners the Amfleet-2s and Superliner-1s will remain useful for many years if they enter a surge fleet.

The Surge fleet can be used for charter trips ( added to present trains ) or exclusive trains such as the Safety patrol Florida trips to Washington. Note before Amtrak many Washington train schools trips originated all over the east. Now mostly bus.

neroden said:

Of course, a low PM/TM may simply mean "this train hasn't been running on time recently and passengers are temporarily avoiding it for that reason", which was actually the situation with most of the trains cut in the Carter Cuts. So it's not a rational basis for getting rid of routes: if anything, a low number may mean that a route needs capital improvements.

Click to expand...

Historically, the only use of PM/TM that I have seen is to give a "scientific veneer" to a decision to cut a bunch of trains, which were woefully poorly managed to start with. This practice spans both private operations and Amtrak. I agree that it is a useful measure if used with other relevant metrics. But that is not typically how it has been used in the past by Amtrak or its predecessors. The usual technique is to first mismanage a train, and then say its PM/TM is too low to continue operating it. So I have habitually come to not trust some one who spews PM/TM in the absence of any other measures. But then again after you mismanage a train, the revenue numbers ain't gonna be that great either. That was the method used to deep six the Broadway and Three Rivers. So be very afraid. The fact they managed to pull it off on a potentially lucrative New York - Chicago Corridor speaks volumes of what can be achieved with appropriate obfuscation while sounding scientific. Bottom line is I don't really trust Amtrak as an able guardian of the LD network. but that is all we've got for now.

jis said:

neroden said:

Of course, a low PM/TM may simply mean "this train hasn't been running on time recently and passengers are temporarily avoiding it for that reason", which was actually the situation with most of the trains cut in the Carter Cuts. So it's not a rational basis for getting rid of routes: if anything, a low number may mean that a route needs capital improvements.

Click to expand...

Historically, the only use of PM/TM that I have seen is to give a "scientific veneer" to a decision to cut a bunch of trains, which were woefully poorly managed to start with. This practice spans both private operations and Amtrak. I agree that it is a useful measure if used with other relevant metrics. But that is not typically how it has been used in the past by Amtrak or its predecessors. The usual technique is to first mismanage a train, and then say its PM/TM is too low to continue operating it. So I have habitually come to not trust some one who spews PM/TM in the absence of any other measures. But then again after you mismanage a train, the revenue numbers ain't gonna be that great either. That was the method used to deep six the Broadway and Three Rivers. So be very afraid. The fact they managed to pull it off on a potentially lucrative New York - Chicago Corridor speaks volumes of what can be achieved with appropriate obfuscation while sounding scientific. Bottom line is I don't really trust Amtrak as an able guardian of the LD network. but that is all we've got for now.

Click to expand...

Maybe the 1990 reroutes did in the BL the first time. What if Amtrak had let the BL go through TOL and CLE and the CL through Akron? You could say the Superliners but they wouldn't have fit through the tunnels so they had to give them to the CL. The CL did get to WAS faster than the BL got to PHL and that helped with southern transfers.

As for the TR, I thought the name was poor. I don't know why they didn't just call it the BL again? They were clearly on the bottom of the three daily CHI to NEC trains and I don't believe ever got a dining car. And maybe we were punished because of the Pennsylvanian which is paid for by PA. Would Amtrak had taken the BL/TR away if there was no train between PGH and PHL?

The decision to cancel trains should be based on performance but we all know they're not always and there are exceptions. If it was truly based on performance, the BL would've never been canceled.

If the CL had been assigned to the B&O route and the BL to the NYC route, then my bet is that CL's performance would have been way worse than BL's on the B&O route. As it is CL's performance on the more favorable route isn't exactly stellar. So yeah, I agree. The 1990 reroute together with certain, shall we say, prejudices at Amtrak HQ, did the BL in. They tried mightily to kill the BA at that time, which remains unexplained to me. We fought hard. But Amtrak is Amtrak. they'll just do whatever random thing they and their paymasters come up with.

I wonder if the western reorganization helped lead to the eventual TR cancellation. In the LSL/CL/TR days, you had three routes but two shared the CHI to PGH route and two share the CHI to CLE route. Out west, you really could have kept just the California Zephyr, Desert Wind, and Pioneer so all three could share the CHI to DEN portion, the Pioneer could run only DEN to PDX/SEA and the Desert Wind could run only SLC to LAX. I probably would've kept the SWC over the DW if forced to make a choice as the train runs to LAX much faster and serves Albuquerque. But I'd probably have kept the Pioneer and truncated the EB in MSP. That way you can save all those extra miles between MSP and the west coast. Now you have three routes with the only overlap between the CZ and SWC in Illinois/Iowa and it would be next to impossible to cut one or truncate one in 2005 or today. So of course the CHI to NEC market (and the NYP to Florida markets) had to lose because of the overlap. In reality, Amtrak should've just reinstated the through cars after cutting the TR.

I wonder if Katrina hadn't damaged the SL East would it still exist today and if so would something else have been cut?

jis said:

If the CL had been assigned to the B&O route and the BL to the NYC route, then my bet is that CL's performance would have been way worse than BL's on the B&O route. As it is CL's performance on the more favorable route isn't exactly stellar. So yeah, I agree. The 1990 reroute together with certain, shall we say, prejudices at Amtrak HQ, did the BL in. They tried mightily to kill the BA at that time, which remains unexplained to me. We fought hard. But Amtrak is Amtrak. they'll just do whatever random thing they and their paymasters come up with.

Click to expand...

Or whatever Sen. Byrd comes up with.

Philly Amtrak Fan said:

jis said:

If the CL had been assigned to the B&O route and the BL to the NYC route, then my bet is that CL's performance would have been way worse than BL's on the B&O route. As it is CL's performance on the more favorable route isn't exactly stellar. So yeah, I agree. The 1990 reroute together with certain, shall we say, prejudices at Amtrak HQ, did the BL in. They tried mightily to kill the BA at that time, which remains unexplained to me. We fought hard. But Amtrak is Amtrak. they'll just do whatever random thing they and their paymasters come up with.

Click to expand...

Or whatever Sen. Byrd comes up with.

Click to expand...

He being one of the "paymasters in chief" back then

Paulus said:

neroden said:

Paulus: thanks for the FRA links. I wish they'd fill in Table 1A and 1B, "percent of short term avoidable operating costs covered by passenger-related revenue, including/excluding state revenue". My belief is that the long-distance network looks excellent on that basis, but it's all "not available".

Click to expand...

Yeah, I've been looking forward to those tables, but they've stopped publishing those reports at all.

Click to expand...

Interesting and bizarre given that the reports are apparently a *legal requirement of PRIIA* (according to the reports themselves). I don't see any indication that the legal requirement has been removed.

jis said:

neroden said:

Of course, a low PM/TM may simply mean "this train hasn't been running on time recently and passengers are temporarily avoiding it for that reason", which was actually the situation with most of the trains cut in the Carter Cuts. So it's not a rational basis for getting rid of routes: if anything, a low number may mean that a route needs capital improvements.

Click to expand...

Historically, the only use of PM/TM that I have seen is to give a "scientific veneer" to a decision to cut a bunch of trains, which were woefully poorly managed to start with. This practice spans both private operations and Amtrak. I agree that it is a useful measure if used with other relevant metrics. But that is not typically how it has been used in the past by Amtrak or its predecessors.

Click to expand...

Brings to mind the old saying:

"Figures don't lie, but liars do figure"

Ever read Darrell Huff's classic book, _How to Lie With Statistics_? When my mom taught statistics, she often made it recommended reading.

If the people in charge are determined to abuse their numbers to serve a predetermined conclusion, they will. Perhaps the extreme example, in railroading, is the Milwaukee Road management, which managed to convince themselves that the profitable half of their business (the Pacific Extension) was unprofitable, and that the unprofitable half (the eastern granger lines) was profitable. The country is still hurting from their colossal mistake.

I've seen PM/TM used well in reports by and for commuter railroads -- with management which *wasn't* trying to find excuses to cut service, but was actually trying to improve service -- which is probably why I'm not immediately hostile to it as a measurement.

neroden said:

jis said:

neroden said:

Of course, a low PM/TM may simply mean "this train hasn't been running on time recently and passengers are temporarily avoiding it for that reason", which was actually the situation with most of the trains cut in the Carter Cuts. So it's not a rational basis for getting rid of routes: if anything, a low number may mean that a route needs capital improvements.

Click to expand...

Historically, the only use of PM/TM that I have seen is to give a "scientific veneer" to a decision to cut a bunch of trains, which were woefully poorly managed to start with. This practice spans both private operations and Amtrak. I agree that it is a useful measure if used with other relevant metrics. But that is not typically how it has been used in the past by Amtrak or its predecessors.

Click to expand...

Brings to mind the old saying:

"Figures don't lie, but liars do figure"

Ever read Darrell Huff's classic book, _How to Lie With Statistics_? When my mom taught statistics, she often made it recommended reading.

If the people in charge are determined to abuse their numbers to serve a predetermined conclusion, they will. Perhaps the extreme example, in railroading, is the Milwaukee Road management, which managed to convince themselves that the profitable half of their business (the Pacific Extension) was unprofitable, and that the unprofitable half (the eastern granger lines) was profitable. The country is still hurting from their colossal mistake.

I've seen PM/TM used well in reports by and for commuter railroads -- with management which *wasn't* trying to find excuses to cut service, but was actually trying to improve service -- which is probably why I'm not immediately hostile to it as a measurement.

Click to expand...

I have no reason to believe the numbers aren't genuine. If you believe that cuts had to be made, there had to be some justification as to which route(s) were cut. You may not agree with the statistic or metric but it seemed like a reasonable criteria to me ... when it was followed. Of course some people want to believe the rules and criteria (which were Congress's, not mine) don't apply to them and demand their route get saved at the expense of other routes which had higher PM/TM numbers. The report to Congress suggested the route be cut (and it's the second worst LD train), Amtrak actually does cut it, and somehow there's a Congressional mandate that restarts it. And ironically I didn't see any numbers in 1995 when the BL was cut (if you can find them, by all means post them or share a link). I find it hard to believe that one train has almost 3 times as high a TM/PM and suddenly almost 20 years later the other train is higher. Or was it? Until shown otherwise, I have no reason to believe the BL or the TR failed. I'm sure the BL's number was lower than 172 with competition from the CL but you're not going to convince me it was lower than you know what.

If you look at the numbers, it's surprising the Pioneer lasted until 1997.

I gotta stop looking at old stats because the more I see the more furious I get. Congress should decide which routes stay and which go? Last time I checked, Brock Adams was a member of Congress.

A measure of an agency's efficiency would be Passenger miles per Car mile. PM/CM running empty cars on a trip will lower that metric.

Another thing I found interesting is the end result of Texas service. There were two routes from Chicago to Texas. One was called the Inter-American which went via St. Louis and served Austin and San Antonio. The other was called the Lone Star which went via Kansas City, Wichita, and Oklahoma City and served Houston. Both trains stopped at both Dallas and Ft. Worth.

October 1978 Amtrak timetable: http://www.timetables.org/full.php?group=19781029&item=0042

According to the PM/TM scores, the Lone Star had a 113 while the Inter-American had a 57. In the 1979 report, both the Lone Star and the Inter-American were recommended to be discontinued. This would have meant there would be no service to Texas from Chicago, Dallas would have no service at all, and Houston and San Antonio would only be served by the Sunset Limited. I think most if not all of us are glad that recommendation didn't happen.

The Lone Star ended in 1979 (https://en.wikipedia.org/wiki/Lone_Star_(Amtrak_train)) while the Inter-American eventually became the Texas Eagle we know today (https://en.wikipedia.org/wiki/Inter-American_(train)). Originally Amtrak ran the Inter-American with through cars to Houston but once the TE was born (1981) they were removed. Essentially the Texas Eagle was essentially choosing one of two routes to Texas like choosing between the Empire Builder and North Coast Hiawatha for a route to Seattle.

For a short time there was a Houston branch from Dallas (https://en.wikipedia.org/wiki/Texas_Eagle). Now there is no Dallas-Houston service at all which I think would be a good route. Now Amtrak is at the mercy of Texas because of PRIIA if they ever want to start the branch again (or run a 750 mile train that happens to pass both cities).

If you had just gone by PM/TM, the Lone Star should have been kept while the Inter-American should've been the train cut. A compromise could have been to have the Texas train go CHI-STL-KCY-Oklahoma City-Dallas/Ft. Worth with a split with one branch to Houston and one to San Antonio/Austin. The STL/KCY branch would also serve as a route lost when they cut the National Limited. If I had to choose, I'd probably have gone with Houston over San Antonio/Austin. Little Rock would've lost service but when the Lone Star was discontinued Wichita and Oklahoma City both lost service. The whole state of Oklahoma didn't get back service until 1999. So which state do you cut off (even though I know most of you would say neither)?

One outside the box solution could've been to keep the Inter-American and merge the Lone Star with the National Limited and run that train New York-Dallas/Houston via STL and KCY. That would've given direct NY to Texas service. Or you could reroute the Lone Star via STL as I suggested and then merge the Inter-American with the National Limited (New York-Dallas/San Antonio via STL and Little Rock). Either way, you have one train from CHI to Texas and one train from New York to Texas. Perhaps that would've given them a reason to keep the National Limited although that would have admittedly been a long shot. The Floridian was probably out as well. At that time, the only remaining service to Florida was via the NEC as the Sunset Limited terminated at NOL back then. Of course, that's the case now.

The 1979 report has also suggested merging the Southwest Limited (now SWC) and San Francisco Zephyr (now CZ) into one train going CHI-KCY-DEN and then splitting at Ogden, Utah. They had also recommended cutting the Pioneer which then was Salt Lake City to Seattle via Portland although that train wasn't cut until 1997. I probably run separate SWC and CZ and run the Pioneer off the CZ at SLC, cancelling both the Empire Builder and the North Coast Hiawatha and running a new train only between CHI and MSP.

It's pretty clear that the Lone Star route (Kansas City & Houston) was better than the Inter-American route, by both speed and population numbers (St. Louis aside). The current route through Arkansas is slow, and impressive in its lack of online population.

I really don't know why the Lone Star got cut despite having much better PM/TM than the Inter-American. It would be interesting to see the political history there.

Philly Amtrak Fan presents what would have been the most sensible route: "CHI-STL-KCY-Oklahoma City-Dallas/Ft. Worth with a split with one branch to Houston and one to San Antonio/Austin".

I wonder what they were thinking in 1979. It still seems like they made a deliberate effort in 1979 to kill Amtrak by slashing *stronger* routes and keeping *weaker* routes

neroden said:

It's pretty clear that the Lone Star route (Kansas City & Houston) was better than the Inter-American route, by both speed and population numbers (St. Louis aside). The current route through Arkansas is slow, and impressive in its lack of online population.

I really don't know why the Lone Star got cut despite having much better PM/TM than the Inter-American. It would be interesting to see the political history there.

Philly Amtrak Fan presents what would have been the most sensible route: "CHI-STL-KCY-Oklahoma City-Dallas/Ft. Worth with a split with one branch to Houston and one to San Antonio/Austin".

I wonder what they were thinking in 1979. It still seems like they made a deliberate effort in 1979 to kill Amtrak by slashing *stronger* routes and keeping *weaker* routes

Click to expand...

Ironically in the case of the Pacific NW service, Amtrak kept the train with the higher PM/TM (EB) over the one with the weaker one (North Coast Hiawatha). In the report to Congress, the EB was the train with the lowest PM/TM among LD trains that was recommended to be kept. The report discusses specific circumstances as to why to keep them while recommending merging the Southwest Limited and San Francisco Zephyr (if they went by the report, the entire portion of the Southwest Limited west of KCY including Albuquerque would've been cut) and recommending cutting both the Inter-American and the Lone Star.

Among the hit list, the LD routes that were spared were the Pioneer (until 1997), Cardinal, and Inter-American. Also, the SW Limited/SF Zephyr merge never happened and the suggested route from Utah to LAX via Vegas became the Desert Wind (although that died in 1997 as well).

Until the BL was cancelled in 1995, no train that was actually cancelled after 1979 was not on Brock's list of being discontinued. Of the trains that did get cancelled, only the Lone Star had a PM/TM over 100 (and it was lower than the EB). I would say the Lone Star was probably the only "stronger" train that got canceled (the National Limited and Floridian were really low at the time). If we take the Texas trains out of the picture, you could say it was a decision to keep two of four trains between the National Limited (89), Pioneer (87), Floridian (74), and Cardinal (60). Everyone knows which two I would've kept. Certainly I don't see why you would need both the Pioneer and the Empire Builder. Of course, I would've kept the Pioneer instead of the EB (or truncated the EB west of MSP). Maybe I would've rerouted the National from IND to CIN to COL then to PGH to serve CIN and avoid the IND to COL branch which I believe was the biggest problem of the National Limited route.

Looking at those 1977 reports, I'm realizing that the current tactic of the haters is to hide the avoidable cost numbers, which were readily available in the 1977 reports. Perhaps because they show so many routes generating a surplus. I don't see why Amtrak should stand for this; the avoidable cost numbers should be published as well as the ludicrous "fully allocated" numbers.

There's something else going on: Chicago-Laredo had a lower avoidable yearly cost than Chicago-Houston (probably it wasn't daily). North Coast Hiawatha had a lower avoidable yearly cost than Empire Builder (again it probably wasn't daily). I haven't checked the frequencies.

The report has a fabulous section near the end giving the true underlying passenger potential on all corridors, based on the sqrt( (city A pop) * (city B pop) ) / (distance)^1.33 model.The result (Figure A-1) shows the massive potential in Chicago-East Coast service -- as big as the potential in the NEC -- a potential *still* hasn't been realized due to stupid politicians. (Figure A-2, by contrast, is no good because it makes the fatal assumption that railroad infrastructure won't be improved.)

Armed with this analysis, I feel quite comfortable saying that speed and frequency improvements to NY-(intermediate points)-Chicago service are *more* important than *any* improvements on the NEC.

It would be very enlightening to do this again with current population rather than 1980 population projections. NY-Florida would probably look even better, and I'm not sure what would happen in other regions of the country. NY-Chicago would still look very good.

My proposals for trying to save both the Inter-American, Lone Star, and National Limited (or as much as possible) routes:

Proposal 1:

Reroute the National Limited east of PGH to serve WAS on the current CL route (was proposed by Brock Adams to spin off BL)

Reroute the National Limited between IND and COL via CIN (avoid IND-COL trouble and make sure CIN is taken care of)

Extend the National Limited from KCY to HOU along the Lone Star route

Keep Inter-American as is

So the new National Limited/Lone Star would be WAS-PGH-COL-DAY-CIN-IND-STL-KCY-Wichita-Oklahoma City-DAL & Ft. Worth/HOU

The National Limited would no longer serve PGH-PHL-NYP while the Lone Star would no longer serve CHI-KCY

The BL would remain and the through cars to BAL/WAS would remain.

Potential transfers:

BL and NL/LS at PGH (NY/PHL to southern Ohio/IND/Missouri/Texas or CHI/northern Ohio to WAS, which might be faster than the CHI-PHL-WAS leg of the BL)

NL/LS and SWL at KCY (for travel between WAS/PGH/southern Ohio/IND/STL and Albuquerque/LAX or CHI to Wichita/Oklahoma/HOU)

NL/LS and Inter-American at STL (for travel between HOU/Oklahoma/Wichita and CHI)

Proposal 2:

Reroute the National Limited east of PGH to serve WAS on the current CL route

Reroute the National Limited between IND and COL via CIN

Reroute the National Limited instead of going to KCY to go to HOU along the Inter-American route (merge the National Limited and Inter-American)

Reroute the Lone Star CHI-STL-KCY-then continue along the 1979 Lone Star route

So the new NL/I-A would be WAS-PGH-COL-CIN-IND-STL-Little Rock-DAL & Ft. Worth/Austin/SAS

The National Limited would no longer serve PGH-PHL-NYP while the Inter-American would no longer serve CHI-STL

The PGH transfer would still be in play

You could transfer from the NL-I/A at STL to go to KCY/Wichita/Oklahoma/HOU (the NL-I/A would serve DAL/Ft. Worth, Austin, and SAS directly).

Since this route would no longer serve KCY, there would no longer be a connection between the NL and SWL. The only way to get between the East and West Coasts would be via CHI.

If I couldn't do either, I'd probably keep the NL (with CIN and PGH-WAS), reroute the Lone Star via STL and cancel the Inter-American.

I certainly feel the NL via CIN would be important if the Cardinal was dropped. I could run the Cardinal CHI-IND-CIN only or reroute the Floridian to serve IND (it served Bloomington, IN in 1979) to allow a transfer at IND for CIN-CHI traffic. I'm not completely sold on the PGH/WAS branch and might want to keep the NL going to PHL/NYP although that is a duplicate branch with the BL and it would give some passengers a quicker trip to WAS.

So my 1979 LD system:

NEC-Florida (2): Silver Star (then 81/82) and Silver Meteor (then 83/84)

NEC-CHI (2): Broadway Limited (40/41, 440/441 to WAS) and Lake Shore Limited (48/49, 448/449)

CHI-Florida: Floridian (56/57)

NEC-NOL: Crescent (19/20)

CHI-NOL: Panama Limited (58/59)

Texas/Midwest/WAS (2): Some combo of National Limited (30/31), Lone Star (15/16), Inter-American (21/22) based upon above proposals

CHI-LAX: Southwest Limited (3/4)

CHI-SAC/Bay Area: San Francisco Zephyr (5/6)

SLC-PDX/SEA: Pioneer (25/26)

SLC-Vegas-LAX: Desert Wind (35/36, introduced October 1979)

NOL-LAX: Sunset Limited (1/2)

LAX-SEA: Coast Starlight (11/14)

Cardinal (50/51) only CHI-CIN (reroute to IND)

Empire Builder (7/8) only CHI-MSP

Remember this was before the 750 mile rule.

If you no longer count the Cardinal and EB as LD trains, I would have had 15 LD trains. Assuming we'd have an Auto Train today it would be one more than number of LD trains as we have today. Probably by 1995 I would've had to cut 1-2 as was the case in real life but it probably would've been the NL and/or Floridian. Maybe I could've cut the Desert Wind and introduced LAX-Vegas service instead.

Before the 1979 cuts, there were 19 (counting three New York-Florida trains).

After the dust settled in 1982, they had 16 LD trains (National Limited, Floridian, Lone Star, and North Coast Hiawatha were cut and Desert Wind was added) and lost many markets I would've kept (Columbus, Dayton, Louisville, Nashville, Wichita, and Oklahoma City, among others). It was 15 when they canceled the Cardinal before Congress made them bring it back. The Auto Train came (current version) came in 1983 and the CL split from the BL in 1986.

The 1979 report recommended keeping only 9 (NINE!) LD trains and the Inter-American only CHI-STL. There would be no service between CHI and Texas, only one train from CHI to California, and only one train from NYP to Florida (they would've kept only the Silver Star and Palmetto to Savannah) which would mean that the Silver Star would've had all of the NYP-Florida traffic! In addition to losing the markets that did lose service they also would've eliminated service in Cincinnati, Indianapolis, Atlanta, and Dallas. Even though the BL would've been kept, even I would've thought that was extreme. I think the 1979 report was probably a drastic report and I'm sure Amtrak and Congress were negotiating through 1979 as to which of the trains on Brock's hit list would actually have been saved.