Cardinal discussion

The Cardinal's equipment going to NYP has the additional advantage of the equipment cross interchanging at different times to the LSL, Crescent, Silver Star & Meteor. Cardinal equipment often needs to go to MIA for some more than just turn servicing.
IMO the only need for turning sleepers at WASH would be the ability to get more longer distance passengers. Example Crescent , Star, Meteor each drop 1 or more sleepers at WASH and then south bounds of those trains pick up cars which ever needs to go to MIA. The Cardinal can go to SSY and interchange with at of the 4 single level routes out of NYP.

Alcochaser & East Point hit the nail on the head. Sunnyside is where the supply of spare single level LD equipment is kept. So is the supply of spare parts and mechanics trained on those specific car types. And, as well documented, robbing Peter to pay Paul is not unheard of when it comes to getting any of the LD trains 'out the door'. Cycling through SSY also allows frozen cars to head South every now and then to thaw out.

Also, other than Auto Train (which is separate unto itself, as I understand), there are NO long distance OBS crew based in Washington. If there were, could you imagine what would happen if two OBS staff called out sick and there wasn't anybody rested or even in town to take their place? At NYP, as well as CHI and other major cities where multiple LD trains originate/terminate, there's usually an LD OBS person or 3 that can take an extra run when needed. Not so in WAS.

re: adding or dropping cars at WAS...If the 48/49/448/449 'crew' at Albany/Rensaeller is any gauge, the trains would lose an extra 30 minutes over and above the engine change...and longer periods with no HEP, as well! Oh, and I just experienced a 16 minute engine change at NHV on train 148 a mere 20 minutes ago. They used to be faster when they split the train to go to Springfield...consistently 12-14 minutes as I recall.

Not good news for the Cardinal's economics (though WV fans will also see it as less of a reason to retime to chase CIN traffic, its going to be grim in what, today, is the Cardinal's #5 market by revenue, CIN-CHI ): Southwest airlines to fly from Cincinnati (CVG) to MDW (Chicago Midway) and BWI starting in June 2017:

New city! Cincinnati - Baltimore/Washington (BWI) 3 daily nonstops
New city! Cincinnati - Chicago (Midway) 5 Su-F, 4 Sa

I doubt additional air travel will take away passengers, especially if such routes already exist. I can't imagine Cincinnati doesn't already have direct flights to both the Chicago and DC areas (although might be to O'Hare and Dulles or National). The people who take Amtrak now probably wouldn't care how many flights there were and people who think of flights first are already lost to Amtrak. I doubt Amtrak can legitimately compete with planes for LD travel until it gets its speeds up.

Cincinnati long had high "fortress hub" prices on Delta. Service on Southwest means low fares and heavy advertising (and Delta will have to lower fares and increase capacity). It will hurt, the question is how badly. CIN is roughly 14% of the Cardinal's traffic (14k out of 100k+) noted I'd thought that with retiming we could have seen 60k or more.

Arlington wrote:CIN is roughly 14% of the Cardinal's traffic (14k out of 100k+) noted I'd thought that with retiming we could have seen 60k or more.

This is a common mistake. When Amtrak reports ridership at stations, it adds "boardings and alightings" meaning it counts both entraining and detraining passengers, so if you were to add up all such passengers you would reach a total of double the train's ridership because each boarding at Cincinnati is an alighting counted somewhere else, but it's only one passenger. Each alighting at Cincinnati is a boarding somewhere else, counted twice, but only one passenger. Since travel is not exactly the same in each direction, it's impossible to say (from this information) whether some, most, or all of Cincinnati's Amtrak patrons entrained or detrained there. But to get a more reasonable determination of the amount of usage at a station it's best to divide the figure of "boardings and alightings" in half at each station. That would yield a figure that should be equal to the entire ridership of the train, for which Amtrak counts a boarding at one station and an an alighting at another as one passenger.

Therefore, in 2016, the Cardinal carried 104,831 passengers. Cincinnati had 12,481 boardings AND alightings, or likely contributed about 6,241 passengers (half the 12,481) to ridership. So ridership at Cincinnati is only 6% of the total patronage on the Cardinal.

--Mark Meyer

12% of the Cardinal's passengers start or end their journey in Cincinnati (2015 numbers).

44% start or end their journey in Chicago. 18% start or end their journey in DC. 13% start or end their journey in NYC.
Sure, the numbers should add up to 200% ... but train journeys are double ended. A person no longer using the train at Cincinnati will also be no longer using the train at another station (probably Chicago). The effect is every passenger who uses the station.

(They could use a different station ... but that would be more common if we were talking about closing one station near another. The people allegedly siphoned off of rail at Cincinnati due to better air service are not going to show up at the next train station down the line.)

That being said ... I do not expect the new air service will have a large effect on train travel. The new flights do not directly compete with red-eye train travel.

Thanks, so we're looking at a 3% hit if half of CIN switches to CVG airport.

I think, though that the "uses per capita" math remains basically correct. CIN is low, and would grow tremendously if it was used at the same rate vs population at other stations with better service times.

justalurker66 wrote:12% of the Cardinal's passengers start or end their journey in Cincinnati (2015 numbers).

Nope. Only 6% do. It has 12% of "Activity" which is not ridership.

You're incorrectly dividing the FY2015 Cincinnati Ridership of 12,326 by the total ridership of 102,500 (a number rounded by Amtrak). That does yield 12% for Cincinnati, but that's not the ridership, and Amtrak says as much, because it says "Activity" by station, not ridership. The activity is "boardings and alightings" which doubles the ridership.

If you don't believe me, add up all the "Activity" for Cardinal stations in FY 2015 from the same document, and you come up with 204,988, or exactly (given the Amtrak figure of 102.5 thousand, and nothing more specific) double the amount of ridership. Math being what it is, the number you should be dividing the 12,326 is 204,988 (not 102,500), and you get the same result: 6%

--Mark Meyer

Actually, justalurker is right. Adding a couple words for emphasis: "12% of the Cardinal's passengers EITHER start OR end their journey in Cincinnati." If someone who would have gotten on or off in Cincinnati flies instead (or stays home, or walks, or whatever), that will remove one whole passenger from the ridership number. It'll also remove a boarding or detraining from wherever that person would have gotten off or on at the other end of their trip. Total station activity will go down by 2; total ridership will go down by one. Either way, if 12% of the pax on the Cardinal are going to OR coming from Cin, and half of them do something else instead, there will be 6% fewer pax and 6% fewer boardings.

The discussion of CIN brings up a question not related to CIN. Why has Amtrak not posted the number of connection passengers at connecting stations ? And does a connecting passenger = a terminating and one originating passenger ? Total of 2 ? Also how does Thruway passengers fit for boarding and alighting. Train to train connecting stations are ----- CLT, RGH, Selma, Petersburg, Richmond, VA, WASH, PHL, NYP, New Haven, BOS*BON . Springfield, Albany, Pittsburg, CHI, STL, MKE, SEA, PDX, EMY, SJC, LAX, NOL, Sacramento, Spokane. Also some minor connecting stations between the above listed

Arborwayfan wrote:Actually, justalurker is right. Adding a couple words for emphasis: "12% of the Cardinal's passengers EITHER start OR end their journey in Cincinnati."

I suppose that's one way of looking at it, but it smacks of being disingenuous, suggesting that something is more than it is. When dealing with percentages, we expect the "whole" to be the sum of the parts and to equal 100%. I know there are instances when the summation can exceed or be less than 100%, but this is not one of them.

Taking this to its logical conclusion and using your phraseology here, if one started at Chicago and listed every station along the Cardinal route, it would go something like this: "44% of the Cardinal's passengers EITHER start OR end their journey in Chicago; 1.6% of the Cardinal's passengers EITHER start OR end their journey in Dyer", etc., etc., until the final station, " 13% percent of the Cardinal's passengers EITHER start OR end their journey in New York." And then logical summation of all the stations: "All told, 200% of the Cardinal's passengers EITHER start OR end their journey at all Cardinal stops combined." And of course, that can't be because there never can be an amount more than 100%!

The problem is mixing ridership on the one hand and combined boardings and alightings on the other. When you specifically state, "EITHER start OR end" you're talking about boardings and alightings. When you talk about "passengers" you're talking about one trip. So it's an apples-to-oranges comparison. And when you're talking about the worthiness of a stop with regard to percentages of the route as a whole, your numbers better add up to 100% lest the anti-passenger train people out there call you on it.

I don't see anything wrong with Cincinnati generating only 6% of the Cardinal's ridership. But instead of mixing and misleading percentages, what's wrong with simply stating that (in 2015) 12,000+ people got on an off Amtrak trains that run three days per week and in the middle of the night but still generate activity equal to about 40 people per train....and leave it at that? I think that alone would be enough to continue a conversation as to how ridership could grow if the train at a different time or every day.

One needs to state the impact of a station ... your method understates that impact. It is statistically true that 12% of the Cardinal's passengers used Cincinnati as an endpoint. Dividing the actual passenger count in half is disingenuous. And the same can be said for dividing other station's passenger count in half. Like it or not, 44% of Cardinal passengers in 2015 used Chicago as an endpoint. The station served 44% of the Cardinal's passengers.

One would be wrong if the station numbers were added together to create a statistic such as adding all the Virginia stations together and trying to claim "x% of passengers traveled to Virginia" because one would be double counting trips within Virginia. But for a single station the math is correct. Cincinnati served 12% of the Cardinal's passengers. (One could add together the stations north/east of DC since travel between those stations is not allowed on the Cardinal.)

What percentage of passengers used each station as an endpoint for their trip on the Cardinal?

How many Cardinal passengers are on the Cardinal because they want to get to or from Cincinnati? Around 12,000 a year. If everyone whose trip started or stopped in Cincinnati decided not to take the train, there would be around 12,000 fewer one-way journeys (what Amtrak counts as "passengers"; could also say "boardings") on the Cardinal each year. That would be around 12% fewer rides. If someone decides not to get on in Cincinnati, they don't take the train. If someone else decides not to get off in Cincinnati, they also don't take the train. If someone does decide to get on in Cincinnati, they do take the train. If someone does decide to get off in Cincinnati, they do take the train. If we were looking at the stats for the train, instead of the station, we would need to count only boardings or only detrainings, or else we'd get twice as many one-way trips as there really are. But if we're just counting the effect of the one station of Cincinnati on the ridership of the train, we need to count everyone who goes through the station.

I'm not sure why I'm arguing this point. You're right that it's impressive and promising that so many people to or from C on a lousy schedule. Better schedule would bring more riders.

This discussion on cutting back or cancelling the Cardinal may need serious rethinking. Some very interesting items that have been disclosed for the first time by the October and November Amtrak performance reports just out.
All these statements are predicated on counting the Cardinal as 7/3rds of its 3 times a week operation comparing it to daily trips except the Sunset. Now this poster cannot say 7/3rds will work for the Cardinal but let us take an assumption that it will. this is predicated on combining October and November.

A daily Cardinal would have carried more passengers than Capitol (33% lower ), Auto Train, CNO, Eagle (285 lower ) , Sunset ( 3 days a week ) 12% lower, Meteor, Crescent 10% lower.
Then the other way is compare revenue per passenger which has Capitol, Eagle, Star about same and LSL slightly above Cardinal.

So lets cancel the Capitol ? No way same answer as cancel Cardinal.