Since we last wrote about the Empire Builder’s woes, things have only gotten worse. When writers start doing novelty pieces about “riding the worst train in America”, you know things have gotten bad. In February 2014 the median arrival delay in Seattle was 4.7 hours, and the average delay was 5.6 hours (see chart below), excluding the 4 days the train didn’t make it to Seattle at all. After delays of up to 16 hours, even Amtrak spokesman Marc Magliari was saying the kinds of things you never want your PR person to have to say:

“You’ll get some people who’ll say ‘never again’ after going through that, and you and I can certainly understand why.”

Empire Builder 1
While previous issues still occasionally cause delays or cancellations, such as Pacific Northwest mudslides or high water at Devils Lake (an intractable problem in an endorheic basin), the blame this time around falls squarely on North Dakota’s oil boom and the constrained track capacity it has caused. While track expansion projects are underway, causing additional delay themselves, the problem will persist indefinitely.  Yet Amtrak has a mandate to run the train, and run it it will. To that end,  Amtrak has released a revised schedule that it hopes will at least be somewhat realistic. Beginning April 15, eastbound trains will now leave Seattle three hours earlier, at 1:40pm, and westbound trains will arrive in Seattle 90 minutes later, at 11:55 am.

I’m a strong supporter of intercity rail, but we do ourselves no favors to downplay this terrible performance or to sugarcoat the human misery these delays cause. As someone who particularly cares about eventually giving Spokane better than its current graveyard service, seeing our fate so miserably tied to North Dakota’s  is a shame.

Lost in this mess is the fact that the Empire Builder runs rather well here in Washington (see chart below and compare above), and we shouldn’t let miseries in the upper Midwest deter us from seeking better service across our state. This new schedule will have the (accidental but pleasant) effect of giving Spokane decent eastbound service, with a 9:00pm arrival time. (Westbound will get  worse, however, with a 3:45am departure.)

Empire Builder 2

Meanwhile, on this side of the Cascades, we have 7 trainsets making only 13 daily runs, an overcapitalized and underused fleet that awaits the completion of the Point Defiance Bypass and other projects. I would love to see WSDOT fund a temporary pilot project to give daily roundtrip service to Spokane through 2017 using one of those surplus sets, just to give Eastern Washington a taste of intercity travel that is better than either uncomfortable buses or unreliable trains. If the train performed poorly, we could cut it. If it performed well, we could fund it.

61 Replies to “New Empire Builder Schedule Begins April 15”

  1. Although the lower chart looks much better than the upper one, it still doesn’t look great. The top chart presumably is the result of delays that happen all the way across the country, whereas the lower one just captures delays between Seattle and Spokane. And for that comparatively short run, it’s not a good sign that there are two 5ish hour delays, two 2ish hour delays, and five other delays over an hour in a sample size of only 28.

    1. True, but note that many of those are related to late arrivals in Seattle not being able to turn on time the same day towards Spokane (scheduled layover is 6.5 hours). The delays in those cases are already baked in. Only rarely does the train get delayed because of problems in Washington.

      1. I thought the same thing, but I don’t see that. Taking the worst case on Feb 18 it was 5.4 hours late to Spokane, but only 2.4 hours late to Seattle. I actually don’t see any day where it was significantly late after arriving in Seattle very late. On Feb 7 after being 16.1hrs late to Seattle it was actually early to Spokane. Which actually doesn’t make much sense.

    2. There were mudslide embargoes on Feb 17 & 18 which could have caused service problems on 2/17-19 due to trainset(s) being captured in Seattle. Luckily, there’s also an extra trainset assigned to the current EB pool and the on-board staff is Seattle-based.

      1. During mudslide moratoriums, don’t trains run empty between SEA and EVR if needed? At least when the tracks are not blocked and freight trains are getting through.

  2. What is the schedule you see for the use of that one train set?
    Would it be a day trip or an overnight to use the train for a Spokane trip?
    What other Eastern WA cities would be served on this run?
    Sounds like a good experiment.
    How does one sign up?

    1. I’m not sure if you need to add any other Eastern Washington cities. Spokane is the second biggest city in Washington (although it is essentially tied with Tacoma). It is also the biggest city for many miles. Along that northern corridor, you won’t find anything bigger until you get to Minnesota (a long ways away). Boise is just a tad bigger, but still a long ways away (much further than Spokane to Seattle). This is why it is crazy that the Spokane to Seattle run is being held up by anything to the east. There is nothing substantial to the east unless you go so far (Minnesota) that a train ride becomes a huge, multi-day ordeal. Spokane to Seattle is a run that makes sense. It is far enough to gain a big speed advantage over a car if rail ever becomes high speed through there, but not so far that a plane will easily beat it. This run is just another example of our messed up “serve everyone, but serve everyone poorly” national rail system.

      1. @aw I am also in favor of this idea. Significant capital improvements will be needed for the Stampede Pass route though.

      2. FYI: Spokane the city is tied with Tacoma; Spokane the county is behind King-Pierce-Sno, and just ahead of Clark.

        If you included a quick stop in Cheney and Wenatchee…would be interesting if that proved to be a draw. Or for that matter, Leavenworth: hiking season & general summer fun starts along Icicle Creek on June 1st.

        I think Zach’s got a great idea in moving a Sounder trainset over there.

    2. One scenario could be an early morning departure from Seattle, probably around 6:30am for a 1:30pm arrival, a quick one-hour turn, and a ~2:30pm departure from Spokane to arrive back in Seattle around 9:30pm. With Talgo sets (or older surplus equipment) and good dispatching, I think the trip could be done reliably in 7 hours, which is 30 minutes faster than the Trailways bus and 90 minutes slower than Greyhound. And doing the trip that way would ensure that the train overnights in Seattle for maintenance.

      1. Not only that, but this would open up other things that have previously been impossible, such as visiting Leavenworth without having to stay two nights, or using the train to connect to flights either at SEA or GEG.

      2. That could and would work. BNSF is, of course, the final arbiter. A stop in Snohomish and/ or Monroe might be nice. The underutilization of our Talgo fleet is a travesty. Do WA and OR own all seven now?

      3. +1 LEAVENWORTH !!!

        and Wenatchee.

        Also, a stop at Monroe would benefit the whole Sky Valley and Duvall, Redmond, Bothell, Woodinville…

      4. Perhaps it could make the same stops as the Trailways bus. Stevens Pass could run a ski shuttle from Skykomish.

      5. Unfortunately, I believe BNSF operates the single track westbound in the morning and eastbound in the afternoon. Swapping directions for Amtrak would have ripple effect on every freight train that uses the line.

    3. Are you accounting for the fact that the Seattle – Spokane track is not as good as the Seattle – Portland track and thus Talgos couldn’t go as fast?

      The Empire Builder route serves only Everett, Leavenworth, Wenatchee, and Ephrata en route to Spokane. Everett is a city, Wenatchee might get some minimal ridership, and Leavenworth attracts tourists and skiers. A more productive route would be Stampede Pass, Ellensburg, Yakima, Pasco, and Spokane. That would serve all the cities in central Washington except Wenatchee. There’s freight tracks all the way through but I gather it would cost a significant amount of money to restore passenger-train quality to Stampede Pass. Still, Washington should add it to its long-range plan and have at least daily service on both routes.

      1. The north route to Spokane is more direct, so I suspect we will want both routes eventually. Both routes are in need of investment though.

        I could envision a scenario where we restored both sets of tracks and ran trains in both directions. This would give Spokane double the amount of potential service to Seattle, while also serving most of the major towns and cities on the East side of the Cascades.

      2. I know of two Talgo train sets up for sale, with some investment in the cascade tunnel, and signaling and siding improvements I’m sure BNSF would be happy to provide more train service. On the Seattle-Evt-SPK, I’d recommend adding stops in Monroe, and Skykomish (to serve the winter market). on Seattle-PAS-SPK, I’d add a stop at Auburn, Cle Elum, Ellensburg, Yakima, Pasco and Ritzville. Now to find the $$$. good luck with that :-(

      3. With I-90 so overbuilt and the railroad tracks so decrepit, is trains really the solution we should be shooting for for Seattle->Spokane travel, or should we be aiming for buses down I-90 that don’t have the Greyhound stigma associated with them.

  3. You could try a morning run eastbound, departing around 7am. It’s an 8 hour run on the EB, but I’m sure a Talgo could do it in 7 or 7.5 hours. Return could make a quick turn and leave at 3pm, arriving in Seattle just after 10pm.

    Oregon isn’t really taking advantage of the new trainsets. They are improving the schedule between Portland and Eugene, but there are no new trips. There has to be equipment sitting idle in Seattle.

    1. Attempting a single day round-trip with one equipment set would be courting disaster. Asking BNSF for another northbound slot during the AM commute between Seattle and Everett would be difficult. Also, much of the route between Everett and Wenatchee is single track (including the tunnel at Stevens Pass) with lots of slower freight traffic chugging up hills and the westbound Empire Builder to weave around. Getting to Spokane in 7-7.5 hours would be difficult and then attempting a same day return trip would inevitably lead to very late trains and very upset customers.

      If WSDOT wants to fund another Seattle to Spokane trip, it should be protected with 2 trainsets doing one trip per day. I’d suggest leaving Seattle after the morning Sounders have arrived and have the trip timed to offer connections at Everett from Vancouver BC, Bellingham and Mt. Vernon via the morning Cascades train. That means leaving Seattle about 930am and arriving in Spokane in the late afternoon–an excellent schedule. Westbound, the connection to the northbound Cascades train at Everett would require a longer wait, but it’s important to provide a schedule that can be trusted.

      1. Obviously, that schedule is better with lower risk, but gives you terrible equipment and crew utilization. We don’t have two extra trainsets to use. They need 5 for Vancouver to Eugene, and they want one for backup. What would you do with just one trainset?

      2. One crew can’t do a single day round trip SEA>SPK>SEA, there’s a limit of 12 hours of work in one day.

        A 7-7.5 hour daily utilization isn’t excellent, but it isn’t terrible. Imagine the year 2017 when it could be possible for a train to leave Portland at 0615, arrive Seattle 0930, depart for Spokane at 1000 and arrive Spokane at 1700. That’s better equipment utilization.

      3. Guy,
        Just because there is a 12 hour limit doesn’t mean you have to make them work a 12 hour day. I believe an eight hour day is better. That way, if there are delays, you don’t strand a train out in the middle of nowhere when the 12 hour limit is up.

      4. Seattle to Whitefish would be about 13 hours. Two sets could easily run this with morning departures from each city. And yes, we should be more assertive about picking up those two Talgos originally built for Chicago to Madison service once the litigation ends. And get ID and MT to buy in.

  4. Devils Lake is just the beginning. Lake Agassiz is beginning to reform. Things will get interesting.

  5. I would really be tempted to do two trainsets doing counter rotating loops.
    Starting in Seattle, the CW loop would follow the path of the Empire Builder to Spokane using similar stops. From Spokane it would turn south and hit Cheney, Ritzfille, Tri Cities, Sunnyside, Yakima Elensberg, Cle Elum, Auburn and back to Seattle. The CCW loop would obviously follow the reverse path.

    This would give most major EW cities two trains a day, each train going in opposite directions.

    Politically they would wee that they are getting something out of it ;)

    1. I also like this idea. Lots of investment needed to bring the tracks up to snuff though.

      1. The Stampede line would need a full CTC implementation, since BNSF is using it for directional running it would probably also need more and longer sidings. I wouldn’t be surprised if they asked for full double track from Spokane-Pasco, although they are making strides on their own.

        Stevens would be similar, although it already has CTC. Probably stretches of double track, or enhanced sidings, I’m sure they would ask for improvements to the cascade tunnel to allow more trains through it (upgrade the ventilation equipment to allow more than one or two trains per hour through there)

      2. on Stevens, add a third rail / Catenary system and all electric locomotives to provide the power through the tunnel. Have the diesel locos, idled all the way down while going through the tunnel.

        This would probably require sidings immediately after the tunnel on the east side so that the electric locomotives could be decoupled immediately for reuse.

        Could the Diesel Locos be modified to use third rail / catenary where available?

    2. Alaska Air does, or has done, something similar to this in Montana between Missoula, Billings and Seattle. Enables daily RT’s between small towns that would otherwise require two days to accomplish.

  6. It’s a complete abomination that just when the people of this country have started to come around to using trains, many precisely for their beneficial effect on the environment, we have to have one of our main lines put permanently behind schedule over cargoes that are nothing but sticky airborne filth.

    This country can also afford to find new industries to put Dakotans to work on projects that will benefit the planet, the passenger trains, and them- especially when, like any other depletable resouce, the oil runs out. Best we don’t wait for that to start shifting.

    Meantime- what’s the farthest east from Seattle, and the farthest west from Minneapolis that good passenger rail service can be maintained? It’s a good bet that, however much the State of North Dakota values, or doesn’t, Amtrak service, their elected representatives in DC routinely demand it. And they need to understand that as taxpayers concerned about budgets, we’re not going to pay for service we’re promised won’t word.

    And meantime, let’s enlist these senators and congressmen to help in passing a Federal mandate that all interstate railroads side-track freight trains whatever the cargo in favor of passenger service- which I believe used to be standard practice.

    As citizens of the State of Washington, let’s us start doing everything we can to keep both oil and coal out of the way of passenger trains in our own state. And also start making whatever adjustments in our own way of life are needed to be sure the requisite industries understand we’re willing to put money and effort where our mouths are.

    Which, like with oil workers when the wells run dry, and coal miners when the coal’s all gone, we might as well do sooner than the inevitable later.

    Mark Dublin

    1. And that’s “don’t work”- increasingly like my editing capacity. Eye doctor’s appointment next Thursday.


    2. North Dakota citizens love Amtrak because there’s no east-west bus service up there, you have to take a north-south bus down to I-94 and another one back up. They may be time-coordinated but not necessarily for all trip combinations, and each of them runs once a day. Before the oil boom, North Dakotans were like half the passengers on the Empire Builder. I don’t know what the state government thinks of Amtrak.

      1. Thanks for the word on North Dakota citizen’s view of AMTRAK. Wouldn’t hurt to get some more information, especially North Dakotan’s affection for passenger rail service as opposed to their affection for everything the “Oil Boom” is bringing them. Including what the Boom leaves behind when it “Busts” -like they all do.

        I’d like to see the population there surprise me and win the vote for AMTRAK- if there ever is a vote. However, I suspect that whatever the population thinks, their legislators, being the ones who each get a lot more oil company money than their constituents, Liberty itself means freedom to put a mile of tank cars in front of every passenger train between them and every coast.

        Did I read online the other day that the Wyoming state legislator just voted down “Common Core” standards for their schools, giving the reason that an oil state shouldn’t be teaching science that disses liquid fossil fuel? If so, likely the oil industry will put Wyoming grads in their lobbying offices- but none of their engineering offices and certainly not on any oil rig.

        Having talked with working people from Washington State who support coal transport through our territory, and having lived in the upper Skagit Valley for a couple of years, though, my word to environmentalists is that we have to start lobbying real fast and very hard to get decently-paid work to people who haven’t had any for 40 years.

        Do that and balance of votes between passenger trains and oil ones could shift faster than oil out of a derailed tanker car can spill and catch fire.

        Mark Dublin

    3. You might be surprised how many Empire Builder riders are working in the oil industry.

  7. Under current conditions, Stampede Pass is probably the most likely route for additional trains between the two. Last I knew, the route over Stevens Pass is at capacity due to a mandate that the Cascade Tunnel be limited to one train every half hour due to diesel exhaust in the tunnel, and giving it time to clear out.

    The bad news with Stampede Pass is that it is a very long route to get to Spokane due to going all the way to Pasco, and then back north to Spokane. Furthermore, the ex-Northern Pacific route between Spokane and Pasco is slow and curvy, and filled with long heavy trains headed to the Columbia River Gorge route to Seattle, Portland, and everywhere else.

    Maybe the UP route between Pasco and Spokane would be available? It is a slow route to get from Pasco to the UP line, but once up there the route is shorter than the old NP route, and it is much straighter. I could see that route being capable of some decent speeds.

    What the Cascade Tunnel really needs is re-electrification, maybe using low voltage 3rd rail, so that it would be easy to run the 600 volt DC section of the locomotives off of that while in the tunnel. You would need special dual mode locomotives to do this, but otherwise a very good route between the east and Puget Sound is being wasted due to diesel exhaust.

    Another thing to consider would be getting some Spokane – Portland traffic into the mix.

    1. The stampede pass route isn’t just longer, it also needs more work. Last I heard it isn’t even set up for two way traffic.

      The stampede route makes more sense for serving Seattle to Pasco and Pasco to Spokane than as an actual Spokane to Seattle route, but having an alternate slow route could help relieve some of the congestion on the northern route, creating other opportunities…

      1. I see this as a quick off-the-shelf service addition, so it’d be the current route and stations. In the near-term, Stampede is a no-go because it’s dark territory, it’s jointed rail, and it’s really slow in lots of places (such as the Yakima River canyon).

        Stevens is doable, and sources at BNSF have told me that in the near future carbon monoxide detectors will be installed in the Cascade tunnel, potentially allowing the flush time to be reduced from 30-60 minutes to just 5-10 minutes. That by itself frees up significant capacity.

      2. @Zach Shaner it would be nice if it were actually off the shelf, but I am concerned that there is more than just the tunnel here that is a bottle neck. Aren’t there a number of single tracked portions that limit the capacity of this line?

      3. stampede is capable of bi-directional service, before BNSF started the iron triangle directional running scheme, they commonly sent the Pasco-Everett freight on there. Stampede’s problem is that is does not have full CTC signaling, it has CTC islands around the sidings so they have power switches. I’m sure this would have to be changed along with other upgrades.

  8. To be fair, the idea of a single 2000 mile train trip is completely ridiculous. (And in the distant past, I took the EB all the way from Seattle-Chicago, then got on another 1000 mi trip to NYC, an experience I would never repeat without a sleeping bunk.)

    In Europe even though you can cross the country, you often switch trains, so no one line would be responsible for such a long distance. Then there is the matter of sharing the train with freight.

    To me, a better evaluation would be how often the Empire Builder, Coast Starlight and Cascades, are on time for the intra-state routes. SEA-GEG. SEA-PDX. SEA-YVR.

  9. One potential use for the new schedule for Seattleites is a pub crawl. Take the train on Saturday to Spokane, arriving at 9 PM, just in time for the fun. Then you have an hour from closing time to when the train back to Seattle leaves. Sleep it off Sunday morning in your luxurious Amtrak seat ;-) and get home with a few waking hours left before turning in Sunday night….

    As to the maximum extent of “feasible” service from Seattle and Minneapolis, I would guess that Spokane is about as far east From Seattle and Fargo or Grand Forks is about as far west from Minneapolis as efficient “local” service could be provided. Reality is, however, that stops all along the route are well-used. In fact, I believe two trains a day along the entire route would be possible and would have enough ridership to justify them. Perhaps an approximate 9 AM and 9 PM departure from both Chicago and Seattle?

    1. In the 1970s, I think Amtrak did offer both a North Coast (Northern Pacific) route and the Empire Builder route, giving the bigger cities 2x per day and the smaller ones on each individual line 1x per day. However, I think that went away once gas rationing stopped.

      1. Yes, thanks to POTUS JECarter who offed the North Coast Hiawatha and a number of other trains in his administration as did WJClinton in his.

    2. The Chicago to Minneapolis section of the Empire Builder could use 6 trains a day if the states would pay up. I researched it in grad school and the train was routinely sold out, overpriced to discourage short trips, and had terrible stopping times at Wisconsin Dells which is a fairly big tourist draw. The times were such that you had to stay overnight and you lost half of each day there, making a weekend trip pointless without taking a 3rd day and staying 2 nights. There are tens (hundreds?) of thousands of 0 car households in Chicago, Milwaukee, and Minneapolis that would use additional trips if they made it possible to depart and arrive at logical times and were priced to attract riders rather than repel them. Also, given the bicycle network Wisconsin and other places have along the route, it is completely ridiculous to have to box up a bike to take it with.
      On the western side of the route things are more spread out, but I could see an extra trip each day with good arrival and departure times at Leavenworth and Wenatchee being used.

      1. Yeah, implementing Seattle-Spokane service with newer equipment would also solve my biggest pet peeve (though no one else seems bothered by it): the inability to get a bike to Eastern Washington/North Idaho without either (a. driving, (b. boxing it for bus/train, or (c. saying your prayers and riding across the I-90 bridge at Vantage.

        I did a big bike trip out there two summers ago with a combination of a car and Greyhound, but it was a hassle and a half.

      2. The Eastern routes need more service even more badly than Minneapolis. And aren’t getting it. :-( The biggest villain here is the Ohio Governor, Kasich.

      3. My understanding was that the baggage cars were to replace all the Heritage baggage cars.

        The Viewliner II diners, sleepers and bag/dorms are intended for eastern single-level LD routes. Note that they ordered many more of the baggage cars than the other car types.

      4. Looking at Amtrak’s fleet plan, I see that there would still be 25 Heritage baggage cars in active service after the Viewliner II order is delivered.

  10. This is pure insanity. It was already hard to take the Amtrak, now leaving at 3:45am is basically impossible. At least earlier you could close down the bars, stubmle to the train, and wake up around snohomish for some lovely views and a trip to the diner car.

    I made the mistake of trying to take the train for a business presentation. I got the message at 1:45AM that the train was to be delayed a few hours, making my morning presentation in Seattle impossible to arrive at by train. I couldn’t get a refund with Amtrak but had to rebook on Alaska at crazy last-minute rates.

    One of my life goals is to help push forward high-speed rail between Spokane and Seattle. Ideally it would be a 1 hour commute from King St. Station to Downtown Spokane. It would change our region forever, and in a very good way. It would make Spokane a viable place to live as a Seattle commuter. It would unclog I-90 from people headed over for Mariners and Seahawks games (Spokane is a lot more progressive than it used to be, but there are not too many Sounders fans yet…). It would make boost tourism and commerce on both sides immensely.

    I want a new high-speed rail, but it we can’t do that we can at least run the trains we have on the tracks we have with sane hours. The Empire Builder tops out at 85mph, but the train is able to travel at over 125mph safely. Amtrak leases the tracks from Burlington Northern and they essentially refuse to maintain the tracks to a standard that would allow for safe travel at 125. Imagine 2-4 commuter trains running daily ala the sounder between Spokane and Seattle. I can tell you know that there is IMMENSE support from Eastern WA including the business community, GSI, the home builders, etc. Even the trucking people would like it because it reduces congestion on the interstate. There isn’t real opposition. What would it take to move it forward???

    1. “There isn’t real opposition”.

      Only true if you have a ferry godmother willing to pay for it. In the real world, the tax increases to pay for it would generate big opposition. See California for an example.

    2. First of all, you may not be able to get refunds on your Amtrak tickets (it depends on the situation) but it is very rare that they will not at least grant travel vouchers of some sort.

      Now, about those rediculous station times at Spokane. Those are not exceptionally easy to solve directly because with a long distance route like this, someone will always be inconvenienced. However, the way the Empire Builder was set up after the 1950 reconfiguration, this problem was partly solved. The way things worked then, several local Portland-Spokane and Seattle-Spokane cars were left in Spokane once the train continued east. Local passengers could then arise at their convenience. There must be something like 30 Superliner compatible cars of various types stored at Gateway Rail Services in the Saint Louis area. Rebuild and modernize a few of those and re-create the original Spokane service plan for the Empire Builder.

      However, any way you slice it, an off the shelf solution is still going to be a slow train. Rather than an early morning departure for the proposed new train, what about something later, like 9 pm? This then becomes an overnight trip, and the slow speeds on the line and frequent stops for freight interference aren’t as much of an issue. You would need a different mix of room types than a standard Superliner sleeper (ie, more rooms with showers). It might even allow for some crew efficiency, as it might be possible for a crew to do part of the trip on the Builder and part on the overnight train, so that more of them would be able to stay home rather than overnight on the road.

Comments are closed.