With the refined options for connecting Caltrain (and High Speed Rail) to San Francisco’s new Transbay Transit Center, redeveloping the 20-acre Caltrain Depot at 4th and King Streets and tearing down the northern terminus of I-280 in San Francisco expected to be ready by the end of this year, the odds of the existing Caltrain Station at 22nd Street surviving appear to be on the decline.

While a minor relocation or redevelopment of the 22nd Street station is a possibility, as recently noted by the Citizens Working Group for the Railyard Alternatives project, which was formed by the City to help identify community priorities and concerns and provide context for Planning’s recommendations, the 22nd Street Station, which primarily serves Dogpatch and Potrero Hill, “could also be eliminated, which would likely result in significant cost savings.”

At the same time, the Working Group notes a strong preference for tunneling versus a street-level, or even trenched, approach to any realignment of rail tracks into the city, the cost of which could be a difficult sell.

The selection of a Preferred Alternative from the vetted options will take another 12-18 months to complete, after which it’s likely to take another two to five years to secure the necessary environmental clearances to proceed.  And of course, there’s the issue of funding.

105 thoughts on “Could Potrero and Dogpatch Lose Their Coveted Caltrain Stop?”
  1. Never understood why our only commuter rail system from South Bay has two stops so close together in SF. Also don’t get why Caltrain hasn’t merged with the BART system.

    1. merged in what way? Just fare structure? They came to be in very different ways and if nothing else the Bay Area does not like to consolidate anything related to government

      1. Merged in oversight/management, making a more robust system that coordinates things like timetables, fares, expansions etc.

    2. Boston has Back Bay and South Station which are close together and serve two separate commercial corridors.

      1. are they the only 2 stops for a commuter rail within boston proper? if we had several stops serving downtown that would make sense, but potrero/dogpatch is isolated and seems like a waste of time to stop there since most commuters are heading to central SF.

        1. I forgot to add that I was talking about the two stations at 4th/Townsend and TTC, not 22nd St.

          One of the major problems with Caltrain is that its location is so far removed from the rest of the city. They can spend billions on undergrounding it from Dogpatch to the TTC, but folks taking Muni from other parts of the city still have to spend eternity getting to the station. I live in Parkside. If I want to travel down the peninsula I drive versus taking the L all the way downtown, transferring to the N or K to Caltrain and then hoping I don’t miss my Caltrain connection. That alone is an hour. If Muni is delayed, like it normally is, then it’s even longer.

          1. TTC will help some people. Not you, but it will help East Bay people get to the peninsula.

          2. Yeah I think 4th / TTC will serve more than the 22nd street station (which I was referring to). If they close the potrero location, moving it south closer to Bayview (where lots of housing is being built) could be ideal for access to Bayview / Potrero / Mission / Bernal / etc.

        2. No, the city limits of Boston extend pretty far south-west, so there are a number of other stations in Boston. The next one south of Back Bay is Ruggles, next to Northeastern University

          There is also North Station, which serves a different set of commuter lines.

        3. You neglect the commuters heading *out* of SF – the 22nd Street station is in a residential area more dense than 4th and King and which has attracted residents specifically for the train station.

    3. The 22nd street station is VERY busy, so it seems to me to be quite necessary. Getting to that 4th and King location efficiently is impossible.

      1. Is 22nd station busy with park and riders going South? Seems to be not very accessible, maybe they can all just drive to Bayshore.

        1. 22nd station has about one eighth the boardings of King St. There is only one station in the entire BART system that handles fewer riders than the 22nd Caltrain station. One lane of any “busy” freeway handles more passengers in an hour than the 22nd station handles in a day. Doubt there is such a thing as a “busy” train station anywhere in Caltrain service. Phenomenal under utilization of a dedicated public ROW in a well populated area.

          1. You are the fact king but then you always throw in a dubious or misleading assertion at the end

            “Phenomenal under utilization of a dedicated public ROW in a well populated area.”

            Compared to other commuter rail? Or BART? Why do you always compare Caltrain to a freeway lane? It seems to not help

          2. zig, both Caltrain and BART are commuter rail systems. Caltrain ave weekday ridership (AWR) is ~64k (2016), on 77 miles of track and 32 stations; BART AWR is ~432k (2015), on 107 miles of track and 45 stations. BART’s two busiest stations handle 150% as many passengers as the entire Caltrain system. In Millbrae, BART’s station has ~6.4k AWR, while Caltrain’s station has ~3.6k.

            Caltrain ROW occupies at least as much ROW as an extra single lane of freeway in each direction. In most places it occupies a wide enough ROW for a 2-lane eachway non-freeway. Of course, Caltrain’s track nearly parallels Hwy 101. The 4th/King 280 freeway ramp handles at least seven times as many people as the adjacent train station, which is Caltrain’s busiest. Oh, and the 6th St 280 ramp handles another four times as many people.

            Caltrain is a heavyrail commuter rail system with ridership stats worthy of a lightrail system. Santa Clara County has the most riders at ~27k AWR. SF MUNI has four different lightrail lines with higher ridership than that as well as several bus routes with higher ridership. Within SF, the cable cars carry more riders than Caltrain. Between SF and SV, the corp bus fleets carry more people than Caltrain.

            I strongly suspect that if the Caltrain ROW was converted into a highspeed BRT it would carry far more passengers at far lower cost, especially if corp/private bus fleets were allowed access for fees. BART is creaky but vital because it spans the Bay and the Bay Bridge has been at capacity for decades. Caltrain is antiquated and never recovered from the post-WWII non-self-driving automobility, and the construction of 280.

            I’ve lived near the proposed/approved/revised/reconsidered/etc path of the DTX for over 20 years and they still don’t have funding. Studies like these are all fun and games and full employment for planners with a peach crayon short of a full crayola pack. Just don’t expect to see any of this in the next 20 years, cause ain’t no money and don’t make much cents.

          3. Jake, you forgot to mention that if it was electrified, Caltrain could carry many more people.

            Electrification is the easy way for Caltrain to increase capacity.

          4. How many more people and for what cost? The official Caltrain Modernization Project forecasts they will less than double ridership by 2040 to 111k AWR at a cost of about $2 billion. That would bring Caltrain of 2040 up to about the ridership of BART in the 1970s.

            Now, I’m sure there are many fascinating trainy details about Caltrain’s plans to modernize that I will never bother or much care to learn. And there must be sections or examples of Caltrain that are horribly capacity constrained that ditching diesel for electric would aid, and better signalling, etc, but y’all can’t be serious that any of this is gonna make much difference or is economical.

            Seriously, Caltrain has many stations almost no one uses because they can more easily drive to wherever they need to go. Almost the entire geographical area served in San Mateo and Santa Clara counties has abundant free parking at home and work and toll-free roads connecting them and serves some of the wealthiest communities in the USA that have an average of two cars per household and few households don’t have at least one car (~2% in San Mateo County, ~5% in Santa Clara County). Far more people in those communities work from home than ride Caltrain. Ya wanna reduce carbon and ease road congestion, like the justification for Caltrain, then subsidize telecommuting and raise the price of parking at work. There is more land in the silicon valley covered by free surface parking lots than by office buildings.

            Caltrain is an expensive, ineffective relic. And I wish it well, and godspeed.

          5. Do you take Caltrain? The trains are jammed. Doubling capacity would be excellent for the peninsula and for SF.

            (It’s also faster than driving, regardless of how much free parking there is on the peninsula)

          6. Did you actually think ranting against (better) rail service was going to go well for you on this site?

          7. It is disingenuous to compare BART and Caltrain because they are both “commuter rail”

            You well know that BART is a very bloated and wildly expensive commuter rail system to expand and maintain and has the distinct advantage of downtown SF stations and servicing the East Bay while Caltrain is a legacy ROW that does not have access to downtown SF and serves an area without any bridges

            How does it not make sense to link the our TransBay transportation terminal and downtown to conventional rail? That seems to make more sense than fraudulent BART to San Jose

          8. Ah, “the trains are jammed” and yet the ridership numbers are pathetic. I’ll repeat a hint: Phenomenal under utilization of a dedicated public ROW in a well populated area.

            Caltrain publishes their ridership by train and station (namelink). From their own stats, Caltrain achieves fewer than 300 riders/train on average when their trains transit a station on a weekday. Even at the busiest station, 4th St in SF, most of the trains arrive with fewer than 300 passengers. The “jamming” is all jammed into a few trains. Only 6 of the many weekday trains that arrive at the SF terminal average 700 or more passengers, per official 2016 Caltrain stats.

          9. Of course I’ve Caltrained. I’ve lived within walking distance of the 4th St station since the good old days when there was an RV park across the street. In the time it takes me to walk there, onboard, and then wait until the train accelerates out of the turn under 280; I could self-drive my over-privileged self to the San Mateo County line on 101, to SFO on a speedy day. No way is Caltrain “faster” door-to-door, unless your trip endpoints are both very close to their stations and/or the roads are bollixed, which is why better than 9 out of 10 SFers pick automotive over rail transport along the narrow corridor Caltrain serves. And in the wilds of SM and SC counties it must be more like 99 out of 100. Antiquated, nearly obsolete relic. If only someone would fund/create a “better mass transit service” that actually achieves at least BARTian effectiveness.

            By contrast, a single bus lane can deliver 10,000 passengers per hour. That’s how the Bay Bridgers justify the contraflow lane proposed for bringing in more of those Oakland gentrifiers to slave in the office sweatshops of SF. People on SS complain that SF is building a billion dollar or whatever bus station, well on a price performance basis that’s cheaper than Caltrain.

          10. Jake, you can drive from your house to downtown Palo Alto in 45 minutes during rush hour? To downtown Mountain View in 50 minutes?

            I sure couldn’t. What route do you use?

            Caltrain is in fact jammed in rush hour and electrification would allow more trains then.

          11. You’re complaining that it’s obsolete and simultaneously complaining about the solution that will make it no longer obsolete.

          12. SFRealist, sure I’ve driven many times to the googleplex/nasaAmes/pagemillrd area in 45-50 min. That quick would usually require traveling at the beginning of the many-hour long AM commute fest, before the expected 92 and 84 clottings on 101; 280 has less variation.

            People that want to minimize congestion risk adjust their work hours. For me that would mean leaving SF southbeach before 7AM or after 9AM. 20 years ago that would have been leaving as late as 7:30 AM. That’s why the Bay Area “rush hour” is about an hour wider than 20 years ago. Progress, of a kind.

            Very small percentage of my san mateo and santa clara county destinations have been within a short walk of a Caltrain station.

          13. That’s not the question. He question is how long it takes to drive from your house to downtown Mountain View or Palo Alto during rush hour. It’s great that you don’t have to make the trip during rush hour, but many people don’t have that freedom.

            It’s not just me who thinks so. The rush hour trains are jammed. It is entirely feasible to create more rush hour trains, which makes sense for everyone.

          14. Caltrain ridership is obviously unimpressive by world standards, but a commuter train line can carry huge numbers of people– the busiest in the world have a capacity of 100k/hour.

            You can say that the ROW is phenomenally underutilized, but my response would be that there is a huge amount of land in the Bay Area that’s phenomenally underutilized– as you say, “there is more land in the silicon valley covered by free surface parking lots than by office buildings”. In both cases, we have a lot of land, devoted to transportation, being used in a fairly inefficient way. But doubling down on cars and buses to preserve the current land use is hardly going to help.

            In addition to that, one of the main ways in which free parking and roads are maintained is by putting strict limits on development, and limiting the number of people in a given area, so that there’s room for all their cars. This also leads to underutilization, and is one of the main causes of the spiraling cost of living we have.

            If you actually want to increase the utilization of the corridor, of public land, of investments in infrastructure, etc., the obvious answer is to upgrade Caltrain to carry many more people, while reforming the land use laws to allow dense development around its stations. It makes no sense to marginally increase the throughput of the corridor without considering what happens to the users once they leave the corridor.

          15. SFRealist, you will get a more accurate answer if you ask google maps. You can use South Park as a surrogate for my sheltertent and anywhere you want as the destination, then set “Depart at” to whatever is your “rush hour” or “Arrive by” as your time of “don’t have freedom.” Oh, I personally usually do 5-10% better than google’s estimate, but they have the best god’s view.

            I think the policy question you raise is how much is it worth to the overall community to subsidize people that choose to live and work ~40-50 miles apart in a notoriously congested region. And for a very long time the answer has been a concerned, sympathetic shrug and another plan/meeting/crayola diagram/etc, but not much money. Which is why the few trains of the relatively few people that made those live and work choices which left them so little time freedom are jammed during the few hours of their rushing. I’m not unsympathetic, just a realist.

          16. While I agree that it’s questionable how useful it is for the public to subsidize transportation for people living 40 miles away from their jobs, subsidized transportation is a fact of life in America.

            For what it’s worth, Caltrain has a 60% farebox recovery ratio, which is quite good by American standards. Generally speaking, bus systems are worse at that– for one, they tend to need a lot higher driver/passenger ratio. Converting the Caltrain line to buses or cars on the grounds that it’ll be more cost-effective doesn’t seem like it’s likely to succeed.

            In any case, subsidized transportation is a huge component of American welfare– our de-facto affordable housing policy, after all, is “drive till you qualify”. Ending these policies would have some far-reaching consequences.

          17. Factually speaking, you’re not going to get to downtown Palo Alto or Mountain View in 50 minutes at rush hour.

            It’s a fair question about what should be subsidized. The problem is asking where the subsidies end. 101 is free to drive, but cost money to build and has costs to maintain. Farther afield the low cost of gas is subsidized by our military, and our children are subsidizing the cost of the higher carbon emissions from cars. It’s a fairly pointless exercise.

            In the real world, our population is increasing in the Bay Area and people need to get around. What is the best way to get more people around? One easy way using well proven technology is to electrify Caltrain. What is your alternative?

          18. There are about 2 million jobs in the three counties served by Caltrain, of which Caltrain only transports about 30k. If all goes according to their $2 billion electro-modernizing plan, that will get all the way to around 55k in a mere 20+ years. Better than spending the money on the USS Cheney Enhanced Interrogation Mindsweeper, but factually something that is much less than the rounding error in commute projections.

            If I was one of the very tiny minority of SV workers that longterm needed to be at my job in downtown PA or Mt. View in the rushing hours, then I would probably move to somewhere south of 92 and north of 17. They do have housing down there, and I don’t like commuting.

            BTW, most of the jobs in Mt View are not within walking distance of Caltrain. They are lined up along 101, a mile or so north. That’s generally the case in the Silicon Valley: most of the vast and enormously wealthy fields of dream jobs are not near Caltrain, they line the major freeways 101, 280, 237, 880, 84, 92, and the fwy feeder roads like Tasman, Trimble, Shoreline, Marine,…

            The Caltrain ROW is a relic of the 19th century, pre-auto and pre-Silicon era. Electrifying it ain’t gonna move it to where the jobs are or expand it into the neighborhoods where 2 million people live. Caltrain has very little network effects, which makes it perpetually an ineffective transportation system. Sheesh, it has a station within a mile of Oracle HQ, yet only 300-400 people use the Belmont station. Oh, and Oracle built lovely multi-story parking garages, which still have a bigger footprint than their office buildings. And yes, I have parked in them, in the real world.

          19. It is true that most jobs in Palo Alto and Mountain View are not near the train stations. But many of them are. That’s why the trains and jammed. It’s faster to get there on the train. (Nice to see you give up on those last two arguments, by the way!)

            You’re also right that the trains run on antiquated technology. They’re still packed with people. If service was improved by electrifying them, then they would be able to carry even more people. Improving the trains would also encourage more jobs to be located on the train line. This is not controversial.

          20. “…most of the jobs in Mt View are not within walking distance of Caltrain.”

            True, though Caltrain provides excellent support for riders who bike+ride, placing an enormous amount of jobs (and homes) within an easy 10 minute bike ride from Caltrain stations.

          21. Caltrain’s route has been antiquated/obsolete since 101 between SF and SJ was widened to 10 lanes in the 1950s. Connecting it to the MUNI rail at 4th St has been a key to increasing ridership and is one of the very few rail ribs connected to the spindly Caltrain spine. Once MUNI’s Central line is connected, much of the added benefit of DTX to the TTC will decline. (It is not faster for me to get to Sand Hill or Page Mill Roads via Caltrain in the rushing hours, thanks though for ignoring the real issue, guess this is all about your commute).

            However few jobs are actually within walking distance of the PA and MtV Caltrain stations is trivial in the countings of the herd flows of the valley. Not that the people or their work are trivial, just their count in the scale.

            I’m all in favor of whatever ligthrail, biketrail, shuttlebus lines, etc can allow people to travel the last mile or five to their actual places of work and abode. Not likely we will see Caltrain stations open at the GooglePlex, FacebookThumbsUpyours, Apple ring, etc, etc, etc.

            Meanwhile, lucky us that our forebears had the foresight to construct an actual transportation network that does connect where people live and work, usually to within a ~1 min onboarding. This wonder of the modern age carries millions daily in the tri-county Caltrain service area.

            The policy debate or controversy is about how much money to spend on our inherented antique vs the next best thing. I hear there is this information superhighway that allows folks to work together without having to hurl themselves long distances. Cheaper to move bits than butts. Maybe I’ll mention it at the next meeting of the Hyperloop Homebrew Club at SLAC.

            As to the Caltrain farebox take mentioned by Alai above, sure, Caltrain loses money on every fare, but makes it up with lack of volume.

          22. “it has a station within a mile of Oracle HQ, yet only 300-400 people use the Belmont station” — this is probably because Oracle runs shuttles to San Carlos and Hillsdale stations, not Belmont, presumably because few trains currently stop at Belmont.

            At peak hour, today, Caltrain (“obsolete relic” that it is) carries about twice as many passengers per track (~700 x 5 trains) as there are vehicles per lane on 101 (18400 peak per hour for all lanes combined). And 101, clearly, is running at maximum capacity. Caltrain, if it were modernized, could increase its capacity many times.

          23. Alai, unless you know people/vehicle you can’t compare the two transport streams. The Bay Bridge moves more than 5000 passengers per lane per hour during the rushing. They do it for ~4 hours AM and PM. They claim a dedicated bus lane would move 10,000 per hour. I don’t know the passenger/vehicle ratio on 101 in SV, but I also don’t know of any reason it could not approach the Bay Bridge ratio.

            Caltrain only achieves ~3500 passengers/hour from Hillsdale to SF in the AM northbound commute and only for about an hour at Hillsdale and two hours (~8-10 AM) in SF. The southbound AM out of SF barely reaches ~1500 passengers/hour for about 2 hours. The southbound PM only achieves ~3500 in SF for about an hour (~5-6 PM).

            By comparison, many long sections of Bay Area freeways and bridges run at peak for more like 3-4 hours in both AM and PM commute. Some do it in both directions. Also, an interconnected network is resilient. Caltrain has a track issue and has to run a bus ferry around it.

            You claim Caltrain can “increase…many times”, but provide no evidence at all. Unless you have something to back up your claim wrt to the realities of Caltrain and the paucity of the feeder network, all I got is Caltrain’s own unfunded modernization project that forecasts ~70% ridership increase by 2040.

          24. Caltrain’s “track issues” are pretty much “person on the tracks” or “car on the tracks”. That should be fixed by grade separation, and that’s part of the plan. For all the resilience of roads, accidents cause huge delays on freeways as well, and are harder to prevent.

            My claim that Caltrain can increase capacity many times is based on the existence of commuter train routes that handle many times more people. The East Rail Line in Hong Kong has a capacity of 100,000 passengers per hour. BART moves 25,000 (the same as the entire Bay Bridge). Increasing Caltrain above 3500 isn’t exactly a technical feat.

            Yes, the feeder network and the last-mile problem are real. Fifty years of land use that’s been planned entirely around automobiles will do that. Fortunately, a lot of the changes that would help (ie increasing density around stations) actually pay for themselves, so it’s a lot more likely to happen than yet another unfunded billion-dollar transportation investment.

            You seem to be promoting BRT, and suggesting replacing Caltrain with a road for BRT. And I’m sure it could carry 10,000 people per hour, in theory, just like a rail line can in theory carry many times that. I’m just not sure why you think that would be a worthwhile investment. Bus systems tend to have lower cost recovery, so you’d need to increase the operating subsidy to serve the same number of people. At least Caltrain’s unfunded modernization project should result in a more cost-effective system, not a less cost-effective one.

          25. BART and the freeways don’t have surface level crossings though.
            It would be interesting to have a 3-lane high speed autonomous vehicle track in the caltrain right of way, but the crossings would limit utilization.

        2. But if there are so few jobs in downtown PA, MV, and other locations, why is Caltrain so popular? It’s jammed every day.

          Then there’s the flip side, which is all the jobs in SF near 4th/King. Those trains are jammed too, from people coming up to SF.

          Thanks for informing me of the internet! I had no idea! One question: why is it that companies keep wanting people to come into the office every day? Are they all wrong?

          1. In the peak hour of the southward commute, about 1200 Caltrain passengers get off at the PA station and about 700 get off at the MV station. That is accomplished by 5 trains carrying an average of about 500 passengers on arrival at the PA station, and an average of about 300 at the MV station. Not so jammy, according to the official Caltrain 2016 stats. FWIW, that’s about 25% of the capacity of a single BRT bus lane. And a bus line would have far more than 5 departures an hour, which would save you wait time. Maybe that is why the corp bus fleets carry more passengers than Caltrain. Are they all wrong?

            At Caltrain’s busiest station, 4th St SF, in their busiest hour (8-9 AM), they deliver less than 3500 passengers on an average workday. That is so far below the capacity of a single lane of not jammed BRT that even if Caltrain does achieve their targeted growth by 2040, they will still be around 60% of the capacity of a single lane of BRT at their peak station in their peak hour.

            If the taxpayers can be con(vinced) to spend billions to subsidize you and others to live far from work and y’all willing jam into a measly 4-5 trains/hour to rush to work, then why shouldn’t a company require your friendly presence for the appointed hours of no freedom?

          2. Thanks for making my point. The physical trains are jammed. So the sensible thing to do is to run more trains, in order to carry more people. There’s only one way to run more trains–electrification. More trains delivering more people. It’s not controversial.

          3. ~500 passengers on a train with ~700 seats is jammed? cry me a train whistle. You like to repeat that things aren’t controversial, but yet our little controversy persists; and Caltrain’s yuge Federal funding controversy persists: “Feds delay Caltrain electrification money, potentially derailing project. California’s Republicans in Congress had asked the Trump administration to block the funds” — SJ Mercury News, February 17, 2017.

          4. So your evidence that the trains aren’t jammed is that they aren’t full when they get to Palo Alto? Why are they not full when the get to Palo Alto? Oh right, because people get off before. They are jammed when they leave San Francisco, then people get off them. That’s how trains work.

            They are also jammed when they get to San Francisco in the morning. Standing room only.

          5. SFR, my evidence is Caltrain’s own 2016 ridership stats, freely available on the info supertrainway. Elementary timeseries analysis of the data shows that your southbound AM Caltrain commute achieves a max of 2803.4 passengers/hour at the San Carlos station at 8-9 AM. That’s for all Caltrain stations for every sliding one hour timewindow for the entire AM. FTR, the max for SF is 2046.4 passengers/hour at 22nd St 7:30-8:30 AM.

            The peak times have 5 trains/hour, so your jammy AM commute on average does not reach 600 passengers/train, and leaving the SF 4th St station has 303 passengers/train on average during the peak hour. Train 220 is the most popular leaving 4th St in the AM, with an average of 425 passengers. Maybe you can start doing a daily headcount and fix Caltrain’s unjammy looking stats.

            Caltrain only publishes the measured average for each train at each station. If they gave me their actual measurement data, then I could fit a distribution and run an Erlang traffic model to compute the blocking, which is the odds that a passenger would not have a seat. BTW, that’s Erlang the math stuff, not Erlang the programming language, which is awesome but more suitable for building the train control and communication system than the numerica. This isn’t rocket science, ya know. Nope, it is network analysis, which at this very trivial level the telephonies mastered a hundred years ago or so.

            Now, it would be nice to over provision Caltrain to reduce the blocking (odds of standing room only). I know they turn away bikes due to blocking. But as any traffic engineer should know, less blocking costs more money. Of course there is another option, the last worst option: you and a few thousand of your travelmates catch an earlier or later train to spread the jam wider. And the usual way a network operator achieves that is with pricing.

          6. Thanks for filling me in on Erlang distributions. If you know about it you must be very, very smart. But your point, if you have one, is still mysterious.

            Yes, there might be the occasional empty seat on a rush hour train. They are still standing room only because every single person does not have visibility into every seat on every car. That’s how trains work.

            Electrification is an easy way to add capacity. It won’t work for everyone in the Bay Area. But it will help many people. More trains, more bullet trains, more bike cars.

          7. Jake – There’s a disconnect in the stats you’re looking at and the reality on the trains. During the rush the most popular express trains are SRO. Most people don’t like standing for long rides and this drives some people away from Caltrain and onto the roads.

            Caltrain added a sixth car on those most popular trains which provided some relief. But after a year they’re SRO again. This indicates that demand is flexible and depends on capacity. Caltrain is not able to keep up with demand and that limits its impact on traffic relief.

            Please try riding Caltrain baby bullets during rush hour and you will see for yourself.

          8. MoD, I doubt there is a stats problem. At the time of the 2016 stats, looks like the southbound (sb) bullet trains were only 650 seats, per Caltrain.

            The most popular AM sb bullet is the #322, which averages 663 passengers during it’s peak which is the 15 min segment from Millbrae to Redwood City. So, that train on average does achieve ‘jammed SRO’ state, per Caltrain stats. Not when it leaves SF, but for about a quarter of the one hour trip to San Jose.

            The earliest sb bullet, #312, is much less crowded on average, peaking at 482 passengers on the same 15 segment the #322 was SRO. The #312 leaves one hour earlier than the #322.

            The #216 only takes 8 min longer than the uber jammiest #322 to arrive in Palo Alto, leaves 37 min earlier; and has an average peak of only 432 passengers. Your daily extra 8 minutes of sitting unjammed on a train could save the taxpayers billions. Given the routine congestion delays on nearby 101, that would seem a bargain for everyone.

            To summarize, looks like the AM sb jammies are due to underpricing the preferred departure times and routes (bullet) so much that not enough people are willing to switch to another train that would either add an extra 8 minutes to their most preferred 42 minute trip or make them start an hour earlier. Cry me a train whistle. And yes, of course there is elasticity of demand, and yes it is more dynamic and easier to shape than supply for high CAPEX infrastructure like heavy rail, let alone for heavy rail severely handicapped by an ROW with an obsolete route encumbered by surface level crossings.

    4. Caltrain didn’t merge with BART because (a) it was still privately run when BART opened, (b) caltrain uses ‘normal’ train tracks with grade crossings while BART uses custom tracks on dedicated trackway, but most importantly (c) caltrain primarily runs through counties that didn’t want to pay for BART and thought caltrain (actually Southern Pacific at the time) and ‘expressways’ were good enough.

      1. Elaborating on the physical incompatibilities:

        – Diesel power (Caltrain) vs. third-rail electric (BART). If Caltrain electrifies it will use overhead wires.

        – Caltrain uses a 1435 mm gauge (distance between rails) which is standard across the US and much of the world for both at-grade railroads and dedicated right of ways. BART uses a completely bizarre 1676 mm gauge that’s not seen much outside of India, along with a weird rail profile and other non-standard elements which add to maintenance and equipment costs. Why? Nobody seems to know for sure. Wikipedia says “The original engineers for the system had background in aerospace (rather than railroads) and intended to make a state-of-the-art system for other municipalities to emulate.” Simple incompetence, or a vendor lock-in scheme?

        – BART trains most likely aren’t FRA approved for use on mainline railroads so the Caltrain line would have to *completely* separate from the national rail network (it’s still used for freight at the moment, and there’s HSR in the works), and/or seek an exemption from the FRA.

        – Various other incompatibilities such as the height and length of platforms and the shape of the cars themselves. It’s already enough hassle to get Caltrain and CAHSR compatible on these issues.

        But you’re right that political and organizational issues are probably the bigger obstacle. Even when systems aren’t physically compatible they can have integrated timetables and fare systems, which is done in many parts of the world but not so much in America.

        1. many systems have different types of vehicles/transit under one agency—I was referring to the organizational structure. Really it should all be under the BART umbrella (including muni) so we can unify our overly complicated region

          1. I would rather BART be disbanded and it all be run under a new organization as BART is dysfunctional

            See even on Socketsite we can’t agree!

          2. Having been a regular commuter on both systems, they both suck. But while BART feels severely outdated, Caltrain feels outdated AND third-world.

    5. Lots of commuter rail systems have frequent stop spacing downtown, allowing them to offer more one-seat rides and also double as city rapid transit. Caltrain isn’t a good example, but BART is — it’s both a city subway and a commuter rail system.

      1. But BART can’t run express trains so that it sucks as commuter rail and also is way too expensive for lower density stops.

    6. They can’t physically merge because Caltrain runs on standard gauge rails but BART trains use a very non-standard wide gauge.

    7. The answer is simple: DENSITY, if it was not needed 10 years ago, it looks like perfect 20/20 vision today!

  2. How do we get the MUNI T-Line that is currently on the surface to also go underground? The only thing reliable about this line is that it’s completely unreliable — largely because of obscene traffic when it leaves 3rd st –> channel –> 4th. And of course, the 4th/King debacle, but that’s where it’s going to go underground anyway. IMHO, it should be underground the whole way…or at least from Potrero Hill and North.

      1. It’s closer to $800 million/mile but I won’t pick at details. The point is: how much does it cost us to not have fully underground Muni lines? I mean this in terms of time lost, by passengers and by the agency itself in terms of operating costs and revenue. Consider how quickly Muni trains move between Embarcadero and West Portal or Duboce and Church, and their complete inefficiency once they hit the streets afterward (the exception being the Sunset Portal). The benefits over a hundred years, as noted by a SocketSite comment below, are what we should measure the worth of such permanent infrastructure by.

        For the record I agree with Bayview_Rising. I think the T-third works perfectly well as a (separated!) street line until the crossover to 4th. It should be undergrounded going through 4th and King.

        1. I don’t think we mostly need underground lines as they are too expensive. In the city we just need to bite the bullet and give LRT/BRT more priority and dedicated lanes like they do it pretty much in every other other mid-sized city in the civilized world. Cars need to be redirected. There could be lost parking. Life will go on.

          1. Certain sections of the city should have underground lines, but the existing surface lines west of twin peaks need to act more like light rail and less like streetcars. By that, I mean dedicated ROWs, boarding platforms and stations, not stops, every 1/3 mile min. SF needs to look at how other cities are building and operating light rail systems outside of their city cores. Otherwise, you’re stuck with trains stopping every couple blocks and running in mixed traffic.

          2. I believe the main issue with street cars in SF are the traffic laws and signaling systems. Muni trains often share their tracks with cars and traffic signals are not adapted to give preference to Muni vehicles.

            For example, Zurich, Switzerland has an old town and crooked street, yet it has a huge and extremely punctual system of Muni trains. Street lights turn green as street cars approach and cars are almost never allowed to use the Muni tracks.

            SF’s traffic systems are not THAT difficult to solve when compared to other cities.

          3. Unfortunately we can’t even competently build modest BRT systems here. Geary BRT has been in the works for 10 years and construction has yet to start. Once it does it will cost $300 million to move some sidewalks, reprogram signals, and paint a red stripe on the road.

          4. I actually think an el over Geary would be pretty good. The street is already nasty enough for pedestrians and it’s very wide.

          1. You’re correct. I also agree with your statements about dedicated ROWs, at least in the western portions of the city.

        2. T-third last year had its service interrupted due to a collision with a truck making an illegal left turn across the light rail lines… so I’m not sure that qualifies as working perfectly well.

          There might be some issues with grade separation since the line has to go over (or I suppose under) the creek. It would also have to go under the future currently planned below grade Caltrain tunnel at Townsend (part of the route changes for the upcoming central subway). So that is a lot of grade changes.

        3. Measuring infrastructure benefits over 100 years? That’s a rediculously long period of time – at a moment in our history when technological change is advancing exponentially.

          Will people even be commuting in 100 years? Will most human jobs be replaced by automation/AI and most of the planet is given some kind of guaranteed income? Who knows.

      2. Keep in mind that getting rid of the 4th Street & King railway station and rail terminal would free up 3-4 blocks of prime SoMa real estate for office or residential development. Money from the sale of this land should offset some of the costs for the tunnel.

  3. It sounds like the glass is either 3/4 full or 3/4 empty, depending, of course, on whether you add the “meh”s to the “Not” or the “Very”. (This article might have been helped by an explanation of how many people were polled and who they were)

  4. Another poll question would be: “If you were told the tunnel option is 20 years away, vs 10 for the trench option, does that change your opinion?”

    1. To quote a member of the CWG: “This is about the next century, not just the next 10 or 20 years. We cannot just talk about money (construction costs), we need to talk about the long-term benefits. We have a responsibility, and the Planning Department has a responsibility, to look out 100 years.”

      To which John Rahaim, San Francisco’s Director of Planning, noted that the closest thing to a 100-year decision that he could think of was the Golden Gate and Bay Bridges, which were completed about 80 years ago. And he urged the CWG members to “imagine how different San Francisco would be if either of those decisions were made differently.”

      1. The unnamed member speaks as only one spending “someone else’s money” can. S/he has a point, of course, but somehow people often forget to add up all the costs to offset all those nifty alleged benefits.

        And Mr. Rahaim gets a jump start on that overkill comparing this project to the Bay Bridge: THAT project allowed ‘Friscans to motor directly into New York …the World’s two most important cities brought closer together!

  5. The 22nd station is obviously important to Potrero and Dogpatch residents but surprisingly not that useful for anybody else. I commuted from Noe via Caltrain for a while and initially boarded at 22nd because it looked like the closest station. Turns out biking to 4th + King took almost the same amount of time and didn’t require climbing two stories of stairs either.

    To be sure, once DTX is built, we don’t need both 22nd and 4th+King, that’s overkill.

    1. I’ve taken this train from 4th and King every weekday and sometimes the weekend for over a decade. If I had to guess, 22nd st is less important in the commute direction, but I think there are more people getting on and off at 22nd st in the reverse commute direction (i.e., leaving sf in the morning) than 4th and king. The density of residents here is high and will only be getting higher. There are many south bay stations they should get rid of first before 22nd st.

      I think probably the worst possible idea is having Caltrain, Muni and HSR share the same set of tracks on 3rd st. Muni can’t even figure out how to run remotely on time (in fact they don’t even have a real schedule), I don’t want them interfering with any other transit systems.

      I’m not sure why people prefer tunneling over trenching except that it sounds nicer. I’m not sure how a tunnel would be that much nicer following its current alignment. At least with a trench they could dig it so it could be covered at a later date if necessary, deferring potentially useless expenditures into the future.

        1. I am not sure if I should respond to this specific bit of hyperbole, but it’s possible that a lot of “at a later date” things cost a lot more money but were not really economically viable. Things that are actually useful end of getting built, like the various BART extensions to Millbrae and San Jose.

          1. The BART extensions are hardly examples of money well spent. The Millbrae extension cost billions of dollars and ridership still is below initial expectations. Huge car-dependent commuter rail stations continue to be built by BART while it ignores the denser, transit-dependent inner cores.

  6. The Dogpatch Caltrain station is crucial! The main arguments are:

    1) If you think the 4th Street and King intersection is busy right now, imagine how it would be if 20-40% more people used it to board Caltrain. They would crowd busses, Ubers, sidewalk, bike lanes etc. A dogpatch Caltrain stop spreads out the traffic, in particular at peak hours.

    2) Major residential real estate projects are being built/planned in proximity to the stop (Pier 70, Potrero Projects Redevelopment, Indiana Street developments, even Hunter’s Point and India Basin). There could be 20-40k new residents in the South Eastern corner of the city in 10-15 years. Imagine if the only public transit option to the South Bay starts at 4th and King? Should they all cram into the T-line, which is already quite unreliable?

    3) The Chase Center will bring major events to the Dogpatch. A smart public transportation system should try to steer the masses away from daily commuters and towards secondary public transportation hubs, where extra trains/busses can be used. This Caltrain stop can play a central role in this strategy.

    Having said that, I DO believe that the current 22nd street location is somewhat unpractical to commuters from the Mission or Noe. My suggestion would be to move the station further south towards Cesar Chavez, where busses can connect commuters from Bernal Heights, the Mission and Noe. It would also move the stop closer to the new residential developments in the Bayview.

      1. Yes, I heard of the Oakdale option but Oakdale has no existing public transportation links to the Mission, Bernal or the Bayview. It is also difficult to use it as a hub for crowds going to the Chase Center or AT&T park. Keeping a Dogpatch stop and improving the T-Lines connections to the Brisbane Caltrain stop will do the job IMO.

        1. Oakdale would have direct transportation connections into the Bayview. It’s walking distance to the T-Line and Bus Hub at Palou. And it’d serve the massive residential developments happening in the area, thousands of new units. Here’s Caltrains’ [Oakdale Station] study, which outlines service areas, demographics, and transportation connections.

          1. Also, Caltrain has nearly completed construction of the Quint street berm, which is designed to be able to accommodate a future Oakdale Station. Does anyone know the status of this?

  7. I take Caltrain south from 22nd Street 4 days per week. A fair number of people get on there and in the evening, a lot get off there.

    I personally prefer the tunnel under Pennsylvania St option. The tunnel immediately south of the 22nd St station that emerges near Cesar Chavez has a second bore on the west side that isn’t used. I am not sure about the particulars of the engineering involved, but it seems to me that that bore could be used as a pilot hole to bore under Pennsylvania, so the line would be underground from Cesar Chavez all the way to the Transbay.

  8. the 22nd street station serves a wide range of mission / bernal / noe commuters who get there by bike. It is quite busy. I used it for the last 7 years, commuting from the mission.

  9. I’ve used 22nd & 22 Fillmore on occasion to avoid Downtown. As the crow flies, halfway between 4th & King and Bayshore stations is about Jerrold. Could Muni adjust to a relocated station, would riders?

    1. Jerrold is deep in an industrial area, and near the work being done for the improved Southeast Treatment Plant. I don’t know if there’s room for a station, and it wouldn’t really have the connections to transit and walkability that a location like Oakdale would (which is about 1,500 ft south of Jerrold).

  10. Is this a misleading headline? The diagrams seem to show possible new subway stations either at Pennsylvania near 22nd, or slightly further south. Maybe station “moves” rather than “loss”?

    1. Or as we wrote in our second paragraph/sentence above: “While a minor relocation or redevelopment of the 22nd Street station is a possibility…”

      1. It remains a misleading HEADLINE, regardless of a second or third paragraph because they will NOT lose their coveted stop.

  11. The study is looking at two stations south of ttc: depending on the alignment one in mission bay or 4th/Townsend AND one near 22nd street although the 22nd street station may be relocated. There will be TWO stations plus the ttc in any configuration. A quick call to planning and their project manager would have cleared this up.

    1. The accessibility of 22nd street is abysmal. Relocating it could still serve the meighnorhoods while also providing a thoughtful design.

  12. There are plenty of places in the world that use Dual Guage tracks. Hundreds. Some of the most modern high speed rail systems have dual guage. Some of the busiest urban transit systems are dual guage. Not rocket science. Tunnel height is a constraint. The distance between the rails is not.

    1. Very true. Also, once electrified, an urban rail system could run more frequently and be extended points west, like under Geary, which would provide much-needed relief and access to a big chunk of SF. Instead of the trains ending at the TTC, the station could be under 2nd St. between Market and Mission providing direct access to both the TTC and the Market St. subway before turning west with stations at Union Square, Van Ness, Fillmore, Masonic, Arguello, etc.

  13. Question is who will have the development rights to the areas torn down and what mass transit systems will be in place for new communities and density proposed…? Also how does this next mega-phase shift affect the DTX completion and other Westside projects and major links not yet addressed such as Geneva harney to balboa park station via LRV?

  14. All of the links in this story are to previous Socketsite stories from months ago. Why is this news now? What information has just become available? What is the source for the images?

    1. The links are for background. The images are from a recent presentation to the Citizens Working Group (as is the 22nd street station elimination quote). And the news is the status of the study and insight into the group’s leanings.

  15. Caltrain was looking to add an Oakdale stop. I think that actually could be a better hub than 22nd street for all the bernal/noe/mission folks. Accessing by transit or bike would be a lot more straightforward.

  16. I live on Pot Hill and worked in SJ for a couple of years. I’d take Caltrain on average only once a week (usually Fridays) because it was both slower and less reliable than driving.

    An electrified Caltrain with more frequent trains, quieter trains, and grade separation (fewer accidents and trespassers) would actually be a superior alternative to driving. The current Caltrain is inferior.

    Still, looking at the options here I’d say that solving the quagmire at the 7th St/16th St intersection is more important than keeping the station. Option D, please.

Leave a Reply

Your email address will not be published. Required fields are marked *