Speculation on Daylight/Starlight Scheduling

[Anderson]

I dredged this up on the same blog that had a lot of linked info on the FEC operations (posted in the HSR forum). In short, it speculated on the possibility of the Daylight more or less taking the daytime slot between LAX and EMY/SFO, and the Starlight adopting an overnight schedule between the two cities.

The biggest downside is that it basically screws Portland in the scheduling, but on the other hand it could also allow for a re-extension to San Diego if all parties were interested.

Mind you, this is all speculation, and my guess is that the plan would just be to re-space the trains to get/keep the departures a few hours apart. and expect a large amount of bus traffic (plus a combination of existing LAX-SLO traffic, new traffic attracted by the presence of a one-seat ride, and some cannibalized short-hop traffic on the Starlight) to make the train "work". Still...I swear, SoCal-NorCal is begging for an overnight connection given the length of the run (both in hours and in miles).

 

[Trogdor]

The biggest downside is that it basically screws Portland in the scheduling, but on the other hand it could also allow for a re-extension to San Diego if all parties were interested.

The problem with San Diego is equipment maintenance. There is a sizeable facility in Los Angeles where equipment is serviced, maintained, inspected, etc. Seattle doesn't have that kind of a setup, so you'd need to rotate the equipment somehow.

 

[johnny.menhennet]

The current plan isto run the morning departure out of LA going north, the least popular trains in the Surfliner system, to the Bay Area. It would be an extension of that service that would become the Coast Daylight. This leaves almost a 3-hour difference between the departure times in LA. The Daylight would have a little less padding, but many more stops, as it would stop at all current Pacific Surfliner/Coast Starlight stops plus all of the stops that will be added as a part of the Slainas extension to the Capitol Corridor. The stops would be:

LA

Glendale

Burbank Airport

Van Nuys

Chatworth

Simi Valley

Moorpark (maybe)

Camarillo

Oxnard

Ventura

Carpinteria

Santa Barbara

Goleta

Lompoc-Surf

Guadelupe

Grover Beach

San Luis Obispo

Paso Robles

Soledad

Salinas

Castroville

Watsonville/Pajaro

Gilroy

San Jose

the busier Peninsula stops

San Francisco.

The schedule would require a lot less padding as it would be run as a corridor-type operation. I think the current schedules are good, and I would rather see a third-frequency overnight train before seeing either one of the trains screwed up. The Coast Starlight runs on the right schedule. Both the Southwest Chief and SL are timed to arrive in LA to make an easy connection to the CS. Makign it leave at 9 or 10 like the current SL departure would make for a LONG connection.

 

[Anderson]

The biggest downside is that it basically screws Portland in the scheduling, but on the other hand it could also allow for a re-extension to San Diego if all parties were interested.

The problem with San Diego is equipment maintenance. There is a sizeable facility in Los Angeles where equipment is serviced, maintained, inspected, etc. Seattle doesn't have that kind of a setup, so you'd need to rotate the equipment somehow.

Point...though didn't the Starlight run to San Diego at one point?

The answer that comes to mind is running through cars (for which I suspect there would probably be demand for a sleeper and a coach).

Johnny: I agree that such would be preferable. However, I'll admit that LA-SF has rather burned for an overnight run in my mind.

To the benefit of running both trains, there's a post nothing that the bus operations connecting SLO to the Bay Area generate something like 100,000 rides per year. That's a lot, though obviously there would still be some folks stuck with the bus because they'd be connecting to/from a different train.

The biggest problem with actually getting a separate train on the route is, as far as I can tell, equipment (namely sleepers, currently in highly irritatingly short supply)...though I say this at the same time as Amtrak may or may not have some spare stuff laying around. IIRC, there was supposedly enough equipment to add a sleeper to the Builder and a sleeper to the Chief (requiring five sets apiece, I think); abandoning one of those plans would give you four or five sleepers for an intrastate overnight train (which would allow two per set). And of course, there's also the talk of converting the CONO to single level, which would free up another couple of sleepers.

 

[johnny.menhennet]

The biggest downside is that it basically screws Portland in the scheduling, but on the other hand it could also allow for a re-extension to San Diego if all parties were interested.

The problem with San Diego is equipment maintenance. There is a sizeable facility in Los Angeles where equipment is serviced, maintained, inspected, etc. Seattle doesn't have that kind of a setup, so you'd need to rotate the equipment somehow.

Point...though didn't the Starlight run to San Diego at one point?

The answer that comes to mind is running through cars (for which I suspect there would probably be demand for a sleeper and a coach).

Johnny: I agree that such would be preferable. However, I'll admit that LA-SF has rather burned for an overnight run in my mind.

To the benefit of running both trains, there's a post nothing that the bus operations connecting SLO to the Bay Area generate something like 100,000 rides per year. That's a lot, though obviously there would still be some folks stuck with the bus because they'd be connecting to/from a different train.

The biggest problem with actually getting a separate train on the route is, as far as I can tell, equipment (namely sleepers, currently in highly irritatingly short supply)...though I say this at the same time as Amtrak may or may not have some spare stuff laying around. IIRC, there was supposedly enough equipment to add a sleeper to the Builder and a sleeper to the Chief (requiring five sets apiece, I think); abandoning one of those plans would give you four or five sleepers for an intrastate overnight train (which would allow two per set). And of course, there's also the talk of converting the CONO to single level, which would free up another couple of sleepers.

What is your stance on the breaking of all connections at LA and in the NW? Persoanally, I like the idea of through cars on a overnight train-Zephyr to serve Reno, Dalt Lake, Denver, and Omaha from SoCal. For a few years in the 1970's, you're right that San Diego was the terminus of the Coast Starlight. In the 90's, the tried adding two through-coaches, but the timekeeping-unreliability of the train led that to be cancelled. I know that this was back when the Tehama Cutoff wwas used, but back in the ealry 70's as well, the Coast Starlight schedule had the train leave Seattle at almost noon and arrive in LA at 7pm, which would allow both a Vancouver and San Diego connection on both ends. I can't imagine that the Tehama Cutoff would have saved 5 hours. I routinely see on the Status Maps that the train has at least an hour of padding alone between Chico and Sacramento. In the article that was linked at the beginning, they stress so much that the frequencies are so close and that a major problem would be loss of ridership. I think that a 3-hour gap plus many more stops but on a similar schedule will be a sure winner. Iguess what my point was was that I don't think the hassle of completely changing the CS schedule would be worth it is there were two trains per day, because I believe that you could sell both out as is if the planned schedule of the Coast Daylight went into service. It's just not worth it.

 

[Anderson]

I'm actually not horrified at some broken connections, though I'm wondering exactly how many Builder-Starlight connections there really are either way...any trip from California east doesn't require the Builder, and any trip from Chicago to the Northwest doesn't require the Starlight. Additionally, NB/EB, the Starlight-Builder connection is only marginally worse than the Cap-Zephyr/Chief connections in Chicago, and it's entirely possible that an hour or so might end up getting stuck onto the Starlight's schedule.

Likewise, I'll go ahead and point out that moving the EB Builder earlier (and possibly moving the WB Builder later) by an hour or two might actually be a good thing, system-wise, since it would actually allow more connections in Chicago (most notably with the Eagle; right now, there is no legal connection from Minneapolis to either Dallas or Kansas City, and the only options for some places involve rather obscene arrival times (Detroit can be connected to with a 12:30 AM arrival, for example). Mind you, either of these would likely have knock-on effects at MSP...but at least EB, that is probably not saying that much, since the EB Builder is not the most reliable of creatures by the time it gets that far east.

Moving south, the point was made (and I believe remains valid) that the Daylight would partly fill in the gap...and indeed, it might end up being worth everyone's while to improve the Daylight as much as possible and cater to through passengers who would otherwise be kicking in for a sleeper with some sort of hot meal service (even if it's Acela style) and/or FC of some sort (even if it's just an "especially nice" BC, particularly considering the sheer length of the trip, which is only rivaled by the NPN-BOS Regionals...and those don't do a whole lot of endpoint/near endpoint business like I suspect the Daylight will). As to connectivity issues here...well, I'm wondering just how much connectivity is being lost on this end given both the massive corridor system in the LA area and the note of the Zephyr connection. Likewise, the connections in LA aren't broken...even if they're made less convenient, they're still quite valid.

And, of course, it would in theory be possible to operate the Daylight a bit later in the day NB to allow a Chief connection (even if a tight one). SB, you've got a long connection no matter what (and that really just can't be helped), but it's not like it would be the only lousy connection in the system. I'd put this up there with the situation surrounding almost anything off the Empire Corridor connecting through NYP between the north and anywhere south of WAS (the Maple Leaf, Lake Shore Limited, Adirondack, and so on all misconnect). However, at least here your wait is during the day (and what I wouldn't give for that at NYP or WAS...I'd sure rather wait from 8 AM-3 PM than from 8 PM-3 AM like I would coming from the SB Adirondack, and I'd rather wait from 1 PM-8 PM for the NB Adirondack).

I would rather see three/day between LA and the Bay Area, and I think there may be the market for them. However, considering the number of buses going to/from places at 3 AM-ish hours in the Amtrak California system (there's a 3:50 AM bus from SLO, a bunch of midnight-or-later buses, as well as a 10 PM bus from Oakland that connects with the 6:30 AM Surfliner at Santa Barbara and includes a 2:15 AM meal stop).

And a final thought: Could a 66/67-esque operation (i.e. a sleeper [or two] but not a full diner) be added in the slots of the truncated 785/768?

 

[jphjaxfl]

Has UP agreed to an additional train along the Coast Line from Los Angeles to San Francisco?

 

[bretton88]

Has UP agreed to an additional train along the Coast Line from Los Angeles to San Francisco?

Yes. Mostly through agreements with their predecessor, the SP. California has 3 additional track slots that they own (the UP already considers the existing daylight surfliners to be in the daylights track spot, just "annulled" at SLO). So in theory California could run 4 LA-SF trains on the coast route. Funding and equipment realities probably mean we'll only see one daylight though.

 

[Ispolkom]

Mind you, either of these would likely have knock-on effects at MSP...but at least EB, that is probably not saying that much, since the EB Builder is not the most reliable of creatures by the time it gets that far east.

But MSP-CHI is a big city pair for the Empire Builder, big enough to have a dedicated coach, even with coach tickets approaching the cost of flying, and considerable competition from Megabus at the other end of the cost spectrum. Changing the scheduled arrival time might have serious effects on this traffic, given that casual travelers focus on the schedule, not the actual arrival and departure times.

 

[afigg]

I dredged this up on the same blog that had a lot of linked info on the FEC operations (posted in the HSR forum). In short, it speculated on the possibility of the Daylight more or less taking the daytime slot between LAX and EMY/SFO, and the Starlight adopting an overnight schedule between the two cities.

I reject the primary premise of that blog that a Coast Daylight will cannibalize ridership from the Starlight. The addition of a second frequency grows overall ridership as people take advantage of the flexibility in the schedule. Besides, the Daylight will go to San Francisco which will attract significant ridership that would not consider taking the long distance Starlight. The southbound LD train is often late, which cuts into the corridor ridership potential.

We have the example of the Lynchburger, which while not that close a model to the Daylight vs Starlight because of the greater SF-LA trip time, has attracted a lot of ridership that rarely if ever took the Crescent because of the limited number of CVS-WAS seats. The Crescent is doing just fine because it is able to keep those seats for longer distance passengers. The Crescent is also likely getting passengers who discovered that there are trains in CVS or LYH and end up taking the Crescent to NC or Atlanta.

In the case of the Starlight, maybe Amtrak will shift the schedule a bit to move it to 4 hours apart from the Daylight, which makes for a second daily train type service. Don't have to make radical changes to schedule.

This is a long term debate, because CA really can't start the Daylight service until they get enough new bi-level corridor cars to cover the growing demand on the three current corridor services. Figuring CA will use state money to buy additional bi-levels beyond the 42 cars they are slated to get out of the 130 car order, it could be 2017 or later before they get enough bi-levels to support a Coast Daylight service.

 

[zephyr17]

Point...though didn't the Starlight run to San Diego at one point?

In Amtrak's VERY early days (1971), when the "Coast Daylight/Starlight" only ran through to the Northwest 3 days a week, on the days it ran to Seattle, there were Starlight Pullmans added to a San Diegan. It only lasted a few months. Don't know where those cars were serviced.

There were never any plans to change the Starlight to a night/day/night schedule like the SP's West Coast (pre cutback to Sacramento) or Owl/Shasta Daylight (the Lark via the Coast did not connect. Owl connected at Martinez) with the addition of the Surfliner Daylight. The plans were always to run the Surfliner Daylight earlier and run it into San Francisco rather than Oakland/Emeryville, and run it on an approximation of the Coast Daylight's schedule with earlier departure and arrival.

 

[Anderson]

Has UP agreed to an additional train along the Coast Line from Los Angeles to San Francisco?

Yes. Mostly through agreements with their predecessor, the SP. California has 3 additional track slots that they own (the UP already considers the existing daylight surfliners to be in the daylights track spot, just "annulled" at SLO). So in theory California could run 4 LA-SF trains on the coast route. Funding and equipment realities probably mean we'll only see one daylight though.

Huh...I actually didn't know that CA owned that many slots along the Coast Line.

And...I actually thought that the Daylight was supposed to get cars out of the 42-car order.

 

[Ozark Southern]

Likewise, I'll go ahead and point out that moving the EB Builder earlier (and possibly moving the WB Builder later) by an hour or two might actually be a good thing, system-wise, since it would actually allow more connections in Chicago (most notably with the Eagle; right now, there is no legal connection from Minneapolis to either Dallas or Kansas City, and the only options for some places involve rather obscene arrival times (Detroit can be connected to with a 12:30 AM arrival, for example). Mind you, either of these would likely have knock-on effects at MSP...but at least EB, that is probably not saying that much, since the EB Builder is not the most reliable of creatures by the time it gets that far east.

That sounds great. The only options from MO to MN right now is the 6:40 Lincoln from STL or 6:12 Illinois Zephyr from QCY. That schedule is bad enough for those who live in St. Louis or Hannibal, but it's practically impossible for those who live anywhere else in the state. If this can be corrected by a mere schedule change, so that KCY and LAP become options, I say go for it.

 

[Trogdor]

Keep in mind, moving the Builder later westbound and earlier eastbound would basically eliminate the same-day turn, and require a sixth set of equipment.

 

[leemell]

Has UP agreed to an additional train along the Coast Line from Los Angeles to San Francisco?

Yes. Mostly through agreements with their predecessor, the SP. California has 3 additional track slots that they own (the UP already considers the existing daylight surfliners to be in the daylights track spot, just "annulled" at SLO). So in theory California could run 4 LA-SF trains on the coast route. Funding and equipment realities probably mean we'll only see one daylight though.

Huh...I actually didn't know that CA owned that many slots along the Coast Line.

And...I actually thought that the Daylight was supposed to get cars out of the 42-car order.

I too, read somewhere the Daylight was supposed to get cars on the 42 car order.

 

[johnny.menhennet]

Apparently, the CD cars are not necessarily coming fromt he 42 car order. The 21 allocated to the PS are all being used for capacity expansion on existing trains, AFAIK.

I do not like the idea of switching the CS northbound until even 1 hour later unless many track improvements/less padding take effect south of Portland because there is no point in not switching the train to overnight but still harming connections. FWIU that connection is very popular. I think that afigg is right and I agree with him that a second daytime frequency will improve pax numbers on both becasue of the connectivity and more variation in the choices. I think that this is even a more pressing nedd thatn an overnight service.

Actually, if anything is being done to initiate an overnight train, I have a bold suggestion. What if the California Zephyr would leave Chicago a little later, and arrive in Emeryville closer to six. With a half hour break there, in Oakland, and in San Jose to take up some time, you could then send it to Los Angeles arriving early morning. I think this makes a lot of sense.

 

[jis]

CZ is already too long a run for it to be extended further. The consist is pretty ripe by the time it gets to OKJ, and in dire need for a little TLC.

 

[johnny.menhennet]

CZ is already too long a run for it to be extended further. The consist is pretty ripe by the time it gets to OKJ, and in dire need for a little TLC.

Still a good thought!!! The SL went nonstop form Orlando to LA, with a similar run time as well.

On a slightly different point, the early morning train that would result in the Bay Area northbound (the CZ) would connect with San Jose. I think it would be important for San Jose to have the nonstop service to Reno, Chicago, etc. because currently on the weekends there is no legal connection. The earliest northbound train is scheduled to arrive only 7 minutes before the CZ departs.

 

[Trogdor]

CZ is already too long a run for it to be extended further. The consist is pretty ripe by the time it gets to OKJ, and in dire need for a little TLC.

Still a good thought!!! The SL went nonstop form Orlando to LA, with a similar run time as well.

On a slightly different point, the early morning train that would result in the Bay Area northbound (the CZ) would connect with San Jose. I think it would be important for San Jose to have the nonstop service to Reno, Chicago, etc. because currently on the weekends there is no legal connection. The earliest northbound train is scheduled to arrive only 7 minutes before the CZ departs.

I think you mean direct service. No train has ever run across the country nonstop.

 

[johnny.menhennet]

CZ is already too long a run for it to be extended further. The consist is pretty ripe by the time it gets to OKJ, and in dire need for a little TLC.

Still a good thought!!! The SL went nonstop form Orlando to LA, with a similar run time as well.

On a slightly different point, the early morning train that would result in the Bay Area northbound (the CZ) would connect with San Jose. I think it would be important for San Jose to have the nonstop service to Reno, Chicago, etc. because currently on the weekends there is no legal connection. The earliest northbound train is scheduled to arrive only 7 minutes before the CZ departs.

I think you mean direct service. No train has ever run across the country nonstop.

Yes. Thank you for the clarification. In a way, it could almost help having the CZ in LA since the maintenance facility is bigger and (presumably) better. If the SL went daily, that would then give LAX three trains arriving morning leaving evening/night, plus the CS.

 

[George Harris]

I think an overnight between Los Angeles and the Bay Area sounds like a good idea. Logically it seems that it should be San Diego - Los Angeles - Oakland/Emeryville - Sacramento. It would seem that it should arrive at Los Angeles southbound and Emeryville northbound at about 8:00 am. This should be its own trainset, not an extension of the CZ.

 

[The Davy Crockett]

Leave 11 & 14 alone. The schedule works well the way it is now, providing good scenery and a civilized schedule for most riders. Adding an overnight train from LAX to OAK, with cross platform connections to Cap. Corridor services makes the most sense to me, as it would tap into a different market and its terminal points already have maintenance facilities. I know the problem is a lack of sleepers, but I speculate that CA should shoulder the cost. Or, since we are speculating, how about an All-Pullman train?

 

[Anderson]

Here's what I'm thinking:

-In general, I agree: San Diego-to-Sacramento is a good run for the train. With the right timing, you could probably pick up a moderate amount of "upsold" passengers leaving San Diego for Los Angeles, change them out for the batch of traffic that you're bound to pick up at LAX (since it's...well, LAX), and run overnight from there.

Scheduling-wise:

-The Los Angeles-Oakland run is 12:10 SB and 11:07 NB. Let's call it 11:20 (assuming a more Surfliner-esque stopping pattern/adding 2-3 stops along the way vis-a-vis the Starlight), since some of that extra time SB is pretty clearly pad. Of this, 1:00 is San Jose-Oakland.

-San Diego-Los Angeles is about 2:40. Here, I'm assuming some skipped stops (and you can move time between the two segments if need be).

-Oakland-Sacramento is 2:00.

11:20 plus 2:40 plus 2:00 is 16 hours. NB, assuming 5 PM out of San Diego, that would hit LAX at 7:40 PM, Oakland at 7:00 AM, and Sacramento at 9:00 AM. SB, assuming 5 PM out of Sacramento, you'd hit Oakland at 7:00 PM, San Jose at 8 PM, LAX at 6:20 AM, and San Diego at 9:00 AM. You could easily add an hour along the way if you want to nudge the Oakland/LAX arrivals later, and you can slide the schedule by a bit as well.

If you want to (for maintenance purposes) cut cars at LAX, it might be possible, or indeed desirable, to set out a set of cars at LAX primarily to facilitate exchanging cars for maintenance. Though you'd want to do some cleaning after each trip, the fact that the trip is a one-night/no days trip each way has some implications on this front (considering that the Zephyr, Builder, Chief, and Sunset all do two-night trips, you could probably do most work at one end or the other. On the really bright side, this would be doable with only two equipment sets, a real rarity in the system. If you needed to cut one end or the other to enable maintenance...well, I'm not sure, but I think cutting Sacramento might make more sense than cutting San Diego.

OBS-wise, it might be possible to run the train with something less than a "full" diner; as long as you have some sort of competent hot meal service (I'm thinking either something along the lines of the diner-club idea for the LSL or some sort of "lunch counter" service [i.e. walk up, order, and have your number called out when the food is ready]), you should be fine. You should aim for something better than 66/67 has tended to have (i.e. just the standard cafe fare), but the LA/Oakland times might make for an odd meal situation.

As to the issue of the Zephyr, don't run the train through as a whole. Maybe exchange cars (even this seems a hair pointless), but don't run it through. The Zephyr is, in many ways, too long of a run as it is.

Another somewhat related thought: An overnight Coast Starlight would force the NB/WB Chief connection to be all day, but SB/EB it would actually shorten the connection considering the scheduled 9 PM arrival of the Starlight now. So running the Starlight on a schedule like this (i.e. LAX departure in the evening and arrival in the morning) is probably ridership neutral, and might actually be a plus by removing an overnight wait.

 

[jis]

Yes. Thank you for the clarification. In a way, it could almost help having the CZ in LA since the maintenance facility is bigger and (presumably) better. If the SL went daily, that would then give LAX three trains arriving morning leaving evening/night, plus the CS.

That will also require a significant build out of the maintenance facility in LA to be able to handle 3 LD trains on a daily basis.

 

[George Harris]

Anderson: Of course this is all speculation, and given the California financial situation not too likely, but I think an arrival in Los Angeles much before 7:00 would be too early. Likewise, an arrival in Oakland at 7:00 would mean an arrival in San Jose about 6:00. I would think there would be a goodly amount of traffic to/from points on the Penninsula that would transfer between this train and Caltrain at San Jose, and for that a northbound arrival at San Jose at 6:00 would be a little early to roll out of your bunk.

I would think making this train an extension of the CZ would not be a good idea, as it could and almost certainly would hurt the timekeeping of the southbound which would destroy it appeal as an overnight for a reliable business day in Los Angeles. The main problem with making this train through to San Diego is that LA is a stub end station, meaning that the train would have to reverse in one direction, likely adding about 15 minutes to the schedule of whatever portion has the backup move.

(There has been a through track scheme for Los Angeles in the works for a long time, but it kicks several sacred cows, so may not ever be built.)