SkyTrain runs very short trains, maxing out at 262 feet (equivalent to a three-car Link train), which means that it has to run those short headways to cope with demand. Automation has allowed it to do so without incurring a ton of cost, but it's still not ideal for a real urban rail system.

The point of The Times report was to highlight how awful our zoning rules are. Look at Lynnwood and Federal Way, where the city is rolling out the red carpet for 35-story buildings BEFORE light rail has even started construction. If Northgate ends up shorter than them, Seattle should be embarrassed.

How does SkyTrain do this? The same way as Amazon Go: No humans required. Because SkyTrain is automated, frequency is no longer linked to number of service hours you have to pay people and it doesn't really cost you much more in operations to have trains as frequent as you want.

This is exactly why we should have only had grade-separated transit for Sound Transit. Ok, we failed. But it was a first try. With our second line in design, there's still time to save it. Grade separated everywhere = unlimited frequency for little extra cost.

"SkyTrain ... services a region with a population of 2.5 million [...] Link services a region of 3.1 million"

Neither link nor SkyTrain serve their greater metro areas, or anywhere near it. To serve 3.1 million people, Link would have run from Arlington (north of Everett) all the way to Eatonville (well south of Tacoma).

If we (rather charitably) say that each system serves the entire populations of all the cities it runs through, then SkyTrain serves around 1.2 million people (can't count Surrey yet, really) while Link serves only around 0.7 million.

And SkyTrain has a 20-year head start; its service in 1989 was remarkably comparable to what Link has today. We should be comparing SkyTrain today to Link in 2039.

Once they pull the buses out of the tunnel, the length of trains can be increased. And if I'm understanding things correctly, once East Link comes on line in 6 years, the headways will improve between ID and Northgate, because east link trains will continue north after stopping at the ID.

Oddly enough, I lived in Vancouver when SkyTrain was being built back in the '80s. I vaguely recall a lot of provincial NDP supporters considering it a bit of a SoCred boondoggle (Bill Bennett, while a Milton Friedman style conservative in many senses, was hugely enamored of massive infrastructure projects).

Excellent, Charles.
Keep 'em comin.'

"city-changing success of the Link when its stations on Capitol Hill and University of Washington"- ST will NEVER have another increase like this one. You can't use the success of the capitol hill to u-district line as an indicator of subsequent openings because the stupid fucking thing was finally useful. U to Ballard, meh. Ballard to North gate, meh. Not gonna see the increased ridership that the u-line saw.

@8 The Ave and Roosevelt stations will be at least big of an event as Capitol Hill and Husky Stadium.

Husky Stadium is a *trek* for the tens of thousands of UW students whose entire lives exists between NE 40th St to NE 50th and I-5 to 17th Ave NE. They aren't using Link to go downtown or to Capitol Hill yet. The Metro 43, 49, and 71/72/73 are still absolutely crushed all the time.

I wouldn't be surprised at all if the Ave station opening bump totally smashes the Capitol Hill/Husky Stadium opening.

Opening Capitol Hill and Husky Stadium without the Ave station was and is *extremely* questionable, but of course the whole problem is that the region isn't spending the fortune it should be spending to build all this at the necessary pace.

@4:
Actually, both Link and SkyTrain frequently have more impressive headways than L.A. County Metro does, while serving a smaller regional population with far fewer miles of track. I've read and re-read your comment hoping to find and understand why you posted it. But, there doesn't seem to be any logical reason other than wanting everyone to know that you must've ridden a CTA or METRA train at some point in your life. Congratulations.

Unlike Seattle, Vancouver built a direct route from Downtown to the Airport. Our Politicians sold us the Mt Baker line as "developing the South end" for the "poor people." When it was really about protecting parking lot profits at the airport and ensuring tourists and business travelers never use Light Rail.
If we cared about removing cars from the road, this was the worst possible decision to make.

Leave it to Seattle to ignore a model system to copy just north of the border.

I've lived in WA for almost 7 years now and I've gotta say, the politics here are some of the strangest I've seen anywhere. Here they build apartments with no parking and light rail that was once used so infrequently they darkened the windows so the public couldn't see how empty the trains were.

And coming from someone who was born and raised in Connecticut, where private interests once forced I-84 to curve wildly west of Hartford so it wouldn't go through their land, that's really saying something.

Here's a simple formula for great urban planning, find a system that works and copy it, making improvements if you can and tweaking it to handle local needs. Copying success is one of the oldest tricks in the book because it actually works.

@2- grade separation is exactly right. having trains cross surface streets makes them about as useful as streetcars. The new lines need to be 100% separated and underground as much as possible. And let's start planning to move the Rainier Valley line to elevated tracks.

I'm just going to step in here and say that the SKYtrain is vastly superior to Seattles lightrail service, but ONLY because it exists and is more extensive. Portlands MAXX system is far superior to the skytrain.

The skytain suffers significant accessibility problems. Anyone with a mobility impaired relative who has ever visiting Vancouver will attest to this. So will most Yelp and Google canada reviews. Escalators and elevators are usually down in the most popular areas (Metrotown, Yaletown, King Station Surry, etc), there are few facilities for bathroom usage or change stations, and the routes are very limited. They dont have a grid like portland, they have a triple criss cross.

Seattle may want to look at Portland (who was able to complete 60% of the current MAXX grid in less than 15 years, compared to Sound Transit who are aiming at 25-30 years to connect the full metro with a single north-south link) as a better example than Vancouver. Vancouver barely works for Vancouver, but it couldnt work for Seattle/Bellevue/Tacoma/Everett.

There are a couple of very big differences between the systems. First of all, Vancouver's is urban in nature. As the article pointed out, the lines do serve suburban places, but those suburban places are still fairly close to the city, and more importantly, you serve many places along the way. Burnaby is about as far as Rainier Beach. Richmond is much closer to downtown than Tukwila. They have multiple lines all covering multiples stops not that far from each other, while we manage to only put in one station between the UW and downtown (skipping First Hill was a big mistake).

The other big difference is that the buses complement the trains really well. Look at a frequent transit map of Vancouver and it is a nice grid. The buses more or less go across, while the trains head toward downtown. As you can imagine the worst traffic is headed downtown, which means the buses go fairly fast, while the trains (like our trains) go fast as well.

We really don't have anything like that, nor are we building it. There are very few urban places being served by Link, and not that many new ones as it expands. Just look at some of the stops being added inside the city. Delridge, Avalon, Smith Cove. These are stations that at best serve as bus intercepts (but not in a good way) and at worst are simply stops that you might as well add. There are miles of track with no station (in both of the new city lines) while the Central Area basically gets nothing. But that isn't even most of ST3. Most of ST3 is to build a subway that extends way farther than what Vancouver's does -- thirty miles north and south. The result will mean hour and 15 minute rides from the Tacoma Dome (not even downtown Tacoma) into Seattle. It is a model used often by cities -- and it has failed every time. BART was the first one, and at least they have the excuse of being first. Dallas (with DART) failed miserably as well. The problem is that long distance lines fail -- and it becomes really expensive to run the trains frequently. The simple truth is that density plus proximity equals high ridership. Vancouver has that, we don't.

Bus integration is pretty bad with our system as well. We tend to try and replace bus service that is fairly fast with trains, instead of trying to complement the train service with the bus service. Consider a train from Ballard to the UW (which would continue to downtown). It would only be two minutes slower than what they are building for a trip from Ballard to downtown. But it would be much, much faster as a way to get to the UW. Faster, in fact, than driving, in the middle of the day. It would integrate really well with the bus service, and the combination would be competitive with driving (even if you include the transfer). You can't say that with much of our system.

We've failed to learn from successful systems (like Vancouver's) but managed to mimic failed systems (like DART). Oh well, it is only money.

@9 The Ave and Roosevelt stations will be at least big of an event as Capitol Hill and Husky Stadium...

Don't forget Northgate. The three stations will result in a significant increase in ridership. I agree that the U-District station will be a huge improvement over how things stand now, while the other two stations make for a real network. There will be trips that take a very long time right now that will be trivial. Northgate to Capitol Hill, for example. That is what a real subway does -- enable stops between urban neighborhoods (it isn't all about getting a few suburban commuters to ditch their cars). Suburban riders benefit from such a system as well, as not all jobs are downtown.

Opening Capitol Hill and Husky Stadium without the Ave station was and is *extremely* questionable, but of course the whole problem is that the region isn't spending the fortune it should be spending to build all this at the necessary pace.

Ha, well, no, not really. The train was supposed to go from the U-District to the airport -- that is what voters approved with the first ballot initiative (ST1). But then they ran out of money (it was a lot more expensive than they thought). One executive -- the mayor of Edmonds of all places -- suggested we start with the section between downtown and the U-District (including a stop at First Hill). But he was overruled (probably by suburban interests who don't understand transit) and we built downtown to SeaTac first. We built things out of order. Oops.

We continue to build things out of order. As the Times article said, we are spending more than anyone in the country right now on transit (per capita). The problem isn't that we aren't spending the money, it is that we are building the wrong thing. We are fixated on quantity, not quality. When all is done we will pass Chicago for longest transit system in the U. S. (behind only New York). We will have stations in Fife, Mariner, and the industrial land owned by Boeing in South Everett. We will have lines connecting one part of Issaquah with downtown Bellevue. But the Central Area and Belltown will have no train stops, and a trip from Ballard to UW will be by bus (despite the train stops nearby). Oops.

Vancouver has about 50 miles of track, and 475,000 riders a day. When they finally get around to building the line out to UBC, they will go well over half a million. Seattle will have close to double that in terms of miles of track, and will be lucky to have half that in terms of ridership.

Ridership per mile is a very important metric. The longer you run, the more expensive it is (trains are heavy, you need security, people to check fares, and in our case, drivers). The higher the ridership, the more you get back from the fares. Systems (like Vancouver's) that have high ridership per mile can afford to run quite often. Systems that don't (like ours), can't, and inevitably see loss of frequency (e. g. https://www.denverpost.com/2017/10/24/rt…).

@12 -- Welcome to the great Northwest! Yes, you are right, we fail to learn from other cities. We are one of the most provincial cities in the U. S. My guess is it is because we are relatively far away from everyone else. The closest city is Vancouver, but somehow people ignore that city, even though we could learn a lot from it (I guess they assume Canadians are very different, a "we are special, no one is like an American" attitude not unique to this city). So that leaves Portland before we have to travel quite a ways to another big city.

That's just a theory. There is no reason why we can't look around at cities like D. C. (which built a very expensive, but very effective subway system) or cities like Vancouver (which built a much cheaper, but also effective system). There is no reason why couldn't look at systems like BART and DART or Sacramento rail with a critical eye, and see what works and what doesn't (hint: urban transit works, suburban transit is just fine with buses and commuter rail). But no, that isn't the way we do it. Each and every ST board member has their own idea about transit, and it has nothing to do with research -- just gut feeling. Just about every board member also has a more important job (mayor, city council) and only one of them is a transportation expert (and that person is head of WSDOT, not exactly a transit organization). The result are folks who mean well, but don't know shit, and aren't willing to hire folks who will run studies to see what is most cost effective. They focus on pleasing the cities along the way, and winning the next election, instead of actually building a cost effective system.