Cardinal discussion

The reductio ad absurdum may help people understand the numbers here. Imagine a train that only served two cities, carrying 100,000 people between Decatur, IL and Maumee, OH. Yes, each station serves 100,000 people. And if you remove one of the two endpoints, the impact would not be a loss of 50% of ridership. You'd lose 100%.

With predictions one must be careful ... What would be the effect of a station stop in Oxford? Would every Oxford passenger be new or would many of those passengers choose to drive to Oxford instead of Cincinnati to catch the same train? I expect a slight increase in ridership when a new stop is added ... but close stops can simply move passengers instead of increase their counts. 33 miles could be far enough apart to avoid siphoning. (Oxford is only 25 miles from Connersville ... so there could be some siphoning in that direction.)

The same works for closing stations. Did Amtrak lose every Hamilton Ohio passenger when that station closed? Or did some/many/most just use the station with better facilities 22 miles south of Hamilton?

I consider 7/4 a bad formula for predicting ridership on a daily Cardinal. Some of the ridership will simply be a shift from other days. There will be some increase in ridership from those who want to travel on the currently unserved days or who would consider the Cardinal only if it was a daily train. But 7/4 is, in my opinion, optimistic. Even 7/4 of the passengers served at the 15 Cardinal only stations would be optimistic.

Arborway: You've convinced me the Cincinnati is responsible for 12 percent of the Cardinal's passenger activity because, as you suggest, even a one-way trip for one passenger needs two stations. One does need to be cognizant of the way these statistics are displayed. Much different for the train, a single station, a state, or whatever configuration of data is given. Thank you for the clarification.

--Mark Meyer

I don't think anyone in Amtrak wants to cancel the Cardinal.

The threat to it is external in regards to a line that might be downgraded.

The axe just moved a little closer. You most likely heard about the big CSX company email about reorganizing into four Regions
http://www.trainorders.com/discussion/r ... ?2,4200020" onclick="window.open(this.href);return false;

---------------------------------------------------------------------------------------
The operation divisions will consolidated to the following four. Northern Region, based in Indianapolis, stretches from Chicago to Albany, NY. The Eastern Region, based in Florence, will stretch from Albany south to Jacksonville and beyond. The Western Region, based in Atlanta or Nashville, will stretch from Chicago to New Orleans. The fourth and final region will be a "Shortline" Region based in Jacksonville and will encompass all subdivisions not on the "Core" routes. Examples are the the lines in the coal fields, and lines that don't have much through traffic.
----------------------------------------------------------------------------------------

Pretty much all of the CSX the Cardinal runs on east of Cincinatti were put in the "Shortline" division. The name shortline is telling. Many in the industry are joking this is the TO BE SHORTLINED division....

CSX is fighting for it's life here.... I fear that this summer will see the first Cardinal route downgrade.

I strongly agree with Arlington's assessment that the arrival of Southwest Airlines service in Cincinnati is a very significant event for Amtrak's Cardinal service. I do not think that this new airline service will be complimentary to the current trains. Southwest will compete against the Cardinal directly in terms of fares, convenience and timing with a major strength being their downtown station at Midway Airport.

This is a very smart move into a large but poorly served secondary market with minimal future prospect of competition from passenger rail.

east point wrote: Then the other way is compare revenue per passenger which has Capitol, Eagle, Star about same and LSL slightly above Cardinal.

Revenue/passenger and farebox recoveries for the LD trains with ranks (full-year FY2016 to smooth out differences in seasonality)
Star: $80 (#10), 46% (#4)
Cardinal: $73 (#14), 31% (#13)
Meteor: $108 (#7), 52% (#3)
Builder: $114 (#4), 45% (#5)
Cap Ltd: $83 (#9), 40% (#10)
Zephyr: $125 (#2), 45% (#5)
Chief: $118 (#3), 42% (#9)
CONO: $75 (#11), 44% (#7)
Eagle: $73 (#13), 39% (#11)
Sunset: $110 (#5), 23% (#14)
Starlight: $89 (#8), 44% (#7)
LSL: $74 (#12), 38% (#12)
Palmetto: $71 (#15), 84% (#2)
Crescent: $110 (#5), 42% (#9)
Auto Train: $316 (#1), 96% (#1)

If we break both into better/worse (with an #8 being neither) on both, we have:

better rev/passenger, better farebox
Auto Train
Zephyr
Builder
Meteor

better rev/passenger, worse farebox
Crescent
Sunset
Chief

worse rev/passenger, better farebox
Palmetto
CONO
Star

worse rev/passenger, worse farebox
Cardinal
Cap Ltd
Eagle
LSL

Not much of a correlation between revenue/passenger and farebox. If we ignore the outlier that is the Auto Train, the Palmetto becomes the worst LD on average fare but the best on farebox recovery and we have 7 LDs where the correlation holds (plus the starlight which is midpack on both) and 6 where it doesn't.

That's a very interesting list. I had no idea the California Zephyr was that successful.

Thanks, Mark, and you're welcome.

On the Zephyr: It's a great, fast overnight train from Chicago and western Illinois to Denver (a thousand miles between 2 pm and early morning westbound, between 9 and midafternoon eastbound, faster than most people would like to drive that route), and then an incredibly scenic train from Denver to California, with several good tourist spots and a few big cities along the way. Times are lousy at Salt Lake, but even so I've always been one of quite a few pax on or off there (admittedly often around Christmas). I'm actually a little surprised it's farebox ratio isn't better than it is, because I would have thought they could charge more. Has anyone else noticed that sleepers are a better deal out west (or maybe on Superliners): lower fares per meal and per night? If I haven't imagined that, it could have something to do with it.

At CZ Illinois/Iowa/Nebraska speeds, the Cardinal would do Cincinnati to Chicago in not much more than the time it now does Indy to Chicago. I'm not sure how to compare the whole route to any segment or combination of segments of the CZ, at least not usefully. I live in Indiana and I want real corridor service with real connecting buses to places like Terre Haute (prob. from Crawfordsville to/from Chi and from Indy to/from eastward points) or with an east-west corridor maybe Columbus-St. Louis. The LD train is nice, but I've used it once to Washington compared to a dozen or a couple dozen trips to Chicago, mostly on the ex IC route.

I'm sort of surprised the LSL does so poorly.

I definitely think with some work (a bit faster and perhaps another sleeper and coach) it could do better.

I seem to recall previously the LSL was 1-2 hours faster. I think this is one where cutting down the time in each direction by 2 hours would definitely help make it more practical for overnight city to city travel.

I strongly believe the LSL would perform better if it ran at better hours during the day. One specific example, it leaves Chicago much too late to be of any value serving cities in Indiana and Ohio. But there is a reason why it leaves Chicago so late, it acts as the mop up (last) train for transferring passengers of other trains. I believe it would do better if Amtrak ran another train, like the old Three Rivers, to be the mop up train. Of course, that other train would have just as bad a performance. Some train has to take that hit.
Take another look at the trains performing well, excluding the one of the kind Auto Train. Those trains could be described as two day trains are either end connected by an overnight segment in between in mostly rural areas. There is no time during the year when one could describe the LSL leaving Chicago as a day train.

Greg Moore wrote:I'm sort of surprised the LSL does so poorly.

Me too but as with the Cardinal I think the critical weakness is the inability to sell capacity into the demand wave. Amtrak is simply unable to capitalize on the primary advantage that railroads have over airlines, flexible consist capacity.

In re: CSX Shortline Region (which sounds suspiciously like the old Soo Line Lake States Transportation Division) , what kind of clearances does the Dyer-IND-CIN-Huntington route have? Obviously enough for Superliners, but would double-stacks clear? The FRA under the next several administrations would do well to look at public-private clearance & grade/curve-easing projects, and to encourage eastern railroads to remake their low-traffic coal & secondary lines into faster intermodal routes. In this case, more railroad freight traffic at higher speeds benefits the public directly by allowing faster Amtrak trains, and indirectly by making intermodal train travel more attractive versus the fleets of trucks clogging interstates through the mountains.

leviramsey wrote:

east point wrote: Then the other way is compare revenue per passenger which has Capitol, Eagle, Star about same and LSL slightly above Cardinal.

Revenue/passenger and farebox recoveries for the LD trains with ranks (full-year FY2016 to smooth out differences in seasonality)
Star: $80 (#10), 46% (#4)
Cardinal: $73 (#14), 31% (#13)
Meteor: $108 (#7), 52% (#3)
Builder: $114 (#4), 45% (#5)
Cap Ltd: $83 (#9), 40% (#10)
Zephyr: $125 (#2), 45% (#5)
Chief: $118 (#3), 42% (#9)
CONO: $75 (#11), 44% (#7)
Eagle: $73 (#13), 39% (#11)
Sunset: $110 (#5), 23% (#14)
Starlight: $89 (#8), 44% (#7)
LSL: $74 (#12), 38% (#12)
Palmetto: $71 (#15), 84% (#2)
Crescent: $110 (#5), 42% (#9)
Auto Train: $316 (#1), 96% (#1)

The revenue/passenger ranks seem to favor the western trains over the eastern trains with the obvious exception of the Auto Train. A passenger going from CHI to LAX or CHI to SAC is automatically going to pay a higher fare than CHI to NYP or CHI to WAS. So there is no way the eastern trains can compete with the western trains because while maybe only 10% of SWC passengers travel 2000 miles, 0% of SM passengers do (at least exclusively on one train). This is under the assumption that revenue is entirely based on mileage. So overall revenue all things being equal will also skew towards the longer trains as well. Additionally, the western trains have Superliner coaches which I believe have more seats/car than the Amfleet coaches. But costs of the trains also go up with more miles and since LD's generally lose money it's unfair to the west coast trains to blame them for bigger losses.

The PM/TM seems to be IMHO a fair way to compare trains of different lengths. It was used to determine the hit list for trains in the late 70's (assuming no government interference).

The most recent PM/TM's on the FRA website: http://www.fra.dot.gov/eLib/Details/L18308" onclick="window.open(this.href);return false;

Highest PM/TM among LD trains: AT (374!), CS (219), SM (214), LSL (206), SWC (191), CL (186)
Lowest: Cardinal (121), SL (133), Palm (149), Crescent (154), CONO (161)

Excluding the Auto Train (special category), the top five serve different geographical regions although most are "shorter" distance with the SWC being the exception.

While the two non daily trains are on the bottom of the list, remember that if the trains run daily and passenger miles go up as expected, so do the train miles. And in the 2010 PRIIA for the Cardinal (https://www.amtrak.com/ccurl/536/878/PR ... al-PIP.pdf" onclick="window.open(this.href);return false;), PM/TM for the 2010 FY Cardinal: 119.2, PM/TM projected for a daily Cardinal: 125.5 (still lower than the 3x/week SL!)

gokeefe wrote:

Greg Moore wrote:I'm sort of surprised the LSL does so poorly.

Me too but as with the Cardinal I think the critical weakness is the inability to sell capacity into the demand wave. Amtrak is simply unable to capitalize on the primary advantage that railroads have over airlines, flexible consist capacity.

Yeah, hence my thoughts about needing more coaches/sleepers for this particular route.

electricron wrote:

Greg Moore wrote:I'm sort of surprised the LSL does so poorly.

I definitely think with some work (a bit faster and perhaps another sleeper and coach) it could do better.

I seem to recall previously the LSL was 1-2 hours faster. I think this is one where cutting down the time in each direction by 2 hours would definitely help make it more practical for overnight city to city travel.

I strongly believe the LSL would perform better if it ran at better hours during the day. One specific example, it leaves Chicago much too late to be of any value serving cities in Indiana and Ohio. But there is a reason why it leaves Chicago so late, it acts as the mop up (last) train for transferring passengers of other trains. I believe it would do better if Amtrak ran another train, like the old Three Rivers, to be the mop up train. Of course, that other train would have just as bad a performance. Some train has to take that hit.

You have to have a cleanup train. The SWC leaves LAX after the evening rush and it would hurt the train to leave before it. The CZ leaves Emeryville around 9am and you probably don't want to go earlier than that. The Cardinal can't be it because if it left CHI at 9:30pm it wouldn't make it to NYP by midnight. The CL probably should be but a lot of CHI-Florida/Southeast passengers need it to transfer. So the LSL lost by default. My memory of the BL/TR was it was the cleanup train and I remember once the TR took passengers who missed the LSL. But having both CHI-NYP trains allowed you to get back to NYP before the evening rush hour which you can't do now.