Numerology in Transportation

This post is a cautionary note for everyone who proposes, advocates for, or plans public transportation: please avoid numerology. What I mean by numerology is, it’s easy to target round numbers for trip time, ridership, capacity, or cost, but this may not be based on good design principles. Round numbers are memorable, which makes them attractive for marketing, but quite often the roundness percolates from public communications to system design, and then it tends to lead to bad results: excessive amounts of money spent on meeting a particular trip time, useful scope cut from a project to stay under a too tight budget, and general overpromising.

I’m tagging this incompetence because it is always bad, but even people who are generally good may unwittingly engage in numerology. I’m pretty confident I’ve done this in previous posts by accident. So I’m exhorting myself and good transit advocates and not just the usual politicians and power brokers.

10x and tech

The worst numerology that I’ve seen in technology is not specifically in transportation, but in the software industry of the American West Coast, which is obsessed with the concept of 10x, that is 10 times as good as normal. The most common variation of this is the 10x engineer, that is the programmer who gets 10 times the productivity of the average programmer, but (by implication) does not demand 10 times the average salary, or even 1.5 times the average salary.

Thanks to Elon Musk, the same concept of 10x has jumped into the transportation discourse – Musk promises a 10x reduction in construction costs for tunneling. It goes without saying he cannot deliver, but the telling thing here is the origin of the number. It does not come from some deep analysis finding that California’s tunneling costs are about 10 times as high as those of some target best practice, or even as high as those of a new method. (In fact, California is around 7 times as expensive to build in as Madrid or Seoul, the world’s cheapest cities to build in, so 10 is at the limit of plausibility.) Rather, the number came first: innovation in American tech is supposed to come in orders of magnitude, not continuous improvements, so the target was 10x, just as SpaceX’s target for space launch cost reduction is 10x even though so far the reality is maybe 1.5x or 2x.

The primary problem here is overpromising. Factor-of-10 improvements are almost nonexistent. The one example I am comfortable with in transportation is the tunneling costs in New York specifically, and even that is a problem that only emerged with the latest project, Second Avenue Subway Phase 2; Phase 1 and the 7 extension are off by a factor of 6 or 7 off the rest-of-world average (and about 15-20 off the very cheapest in the world), and East Side Access is a problem of overbuilding more than anything so I can’t even give it a specific factor. Many other things in New York are too expensive, but generally by a factor ranging from 1.5 to 3. Cutting operating costs in half, cutting rolling stock procurement costs by a third, and so on are both laudable goals, but 10x rhetoric skips them entirely. Thus comes the secondary problem with 10x-oriented numerology: just as it rounds up factor-of-7 improvements and overpromises a factor of 10, it completely ignores factor-of-2 improvements as they simply cannot plausibly be stretched to an order of magnitude.


It is common in marketing to promise round numbers for schedules: 2-hour trip times, 3-hour trip times, etc. This sometimes percolates into the planning world behind the scenes, leading to planning around discrete trip times in integer numbers of hours.

In France it’s a commonplace that high-speed rail is only competitive with air travel if the trains take 3 hours or less. The reality is very different on two levels: first, mode share is a continuous function of trip time, so the difference between (say) 2:55 and 3:05 cannot be very big. And second, in 2009, rail had a 54% mode share of all Paris-Toulon trips, on which the TGV takes 4:08-4:20, compared with 12% for air; the TGV held its own as far east as Cannes (34%), 5:26 away, and Nice (30%), 5:57 away. The 3-hour rule is alluring and may be true in one specific social class, namely airline and railway managers, but the numerology here makes it easy to stick to it even if the breakeven point keeps creeping up to 3:30, 4:00, 4:30, 5:00.

A more benign example of numerology is the 30-30-30 plan in Connecticut. Governor Lamont has proposed far-reaching investments to speed up trains to take half an hour on each of three segments: New York-Stamford, Stamford-New Haven, New Haven-Hartford. This is more or less feasible: a reasonable level of investment would reduce New York-New Haven to about 1:03 on express trains, with Stamford near the exact midpoint. However, the target trip times remain numerological: there is no obvious reason why 1:00 is so much better than 1:10. So far 30-30-30 has run into resistance from incompetent traditional railroaders, but it’s easy to imagine a future in which the governor approves the plan over their objections, and then has to decide how much money to spend on the final few minutes’ worth of speedup to meet the stated goals.

In contrast with numerology based on round numbers, there is a much more solid planning paradigm based on trip times a few minutes short of a round number. In that case, the trip time is a round number including turnaround time, which makes it easy to run trains on a clockface schedule. Differences like 1:05 vs. 0:55 are not important enough to bother passengers about, but differences in frequency between hourly and every 1:10 are critical – passengers can remember 9:05, 10:05, 11:05, 12:05 much better than they can 9:05, 10:15, 11:25, 12:35. Therefore, the integrated timed transfer plan of Switzerland and the Netherlands aims at trip times that are not very memorable, but that together with connection or turnaround time enable memorable schedules.


In addition to the tech industry’s 10x concept, more traditional cost estimations can suffer from numerology as well. Here it is important to distinguish relative from absolute costs. Relative costs are relative to an already-decided budget; in that case, it is useful to force agencies to stay within their promised costs, to discourage lowballing costs in the future (“strategic misrepresentation” in Bent Flyvbjerg’s language). Absolute costs are about numbers that sound big or small, and in that case, there is no good reason to force costs to hew to a specific number.

In the case of absolute costs, politicians may fit the program to the cost in either direction. Reportedly, the size of the stimulus bill passed by the Obama administration at the beginning of 2019 was designed to be in the hundreds of billions and avoid the dreaded trillion number, even though some of the administration’s advisors argued for $1.2-1.8 trillion. In transportation, I do not know of specific examples, but there is so much political pressure among various people who think they’re fiscally conservative that there’s bound to be pressure to go underneath a round number, in other words a political equivalent of pricing a product at $99 instead of $100.

In the other direction, visionaries may think they’re being bold by making up a high number, usually a catch round  figure like $1 trillion for US-wide infrastructure. The numerology here operates on a different level from the relatively small band of just under a limit vs. just over a limit: here the main problem is that the cost figure is arbitrary, and then the list of projects to be funded is chosen to match it. If there aren’t enough good projects, agencies will either bloat the budgets of projects by lading them with semi-related spending, for example bundling a light rail line with  street reconstruction and tree planting, or go forward with weak proposals that would otherwise not be funded.


  1. Mikel

    This might be a somewhat tangential point, but when I read your (excellent) posts about the Swiss/Dutch takt and its German adaptation, I often think about the arbitrariness of the 24-hour day. If for some historical accident days were divided into 10 hours or 20 or 32, the travel times necessary for a clockface schedule would be different. So your recommendations wrt what infrastructure Germany should or should not build to achieve a takt might differ significantly!

    • Nilo

      Humans seem to have an innate tolerance for how much time they’ll spend getting to work each day. You can see this in shapes in sizes of cities like Ancient Rome or pre-omnibus Manhattan. Because the only real form of transportation for most is walking, you see consumption of very small amounts of floor area per person, and multistory building. I suspect though it would be denominated in different units the good takt times would be close to what we use today.

      • keaswaran

        With our current 24 hour day and 60 minutes to the hour, good takts are once an hour (every 60 minutes), twice an hour (every 30 minutes), three times an hour (every 20 minutes), four times an hour (every 15 minutes). With a 10 “unit” day and 100 “littles” to the unit (so that a little is 1.44 minutes), good takts would likely be twice a unit (every 72 minutes), four times a unit (every 36 minutes), five times a unit (every 28.8 minutes), ten times a unit (every 14.4 minutes). These are in the same order of magnitude as the ones we currently use, though there wouldn’t be one that is close to every 20 minutes, and they would align well with some slightly different trip distances or speeds.

        • Nilo

          A sixth of a unit is pretty close to 20 minutes. Could do 17 littles for the first five intervals and round down to 15 littles to finish up the final interval.

    • Alon Levy

      That is definitely true! And humans also have ten fingers, so counting is in factors of 10, which is why the metric system is useful and the 950272.5 stones to a furlong system isn’t.

      • adirondacker12800

        Because you were taught to count to ten on your fingers. Other people have come up with other methods that work equally well. Base 12 makes life a lot easier. For instance one third is nice and tidy. Quarters and sixths too. … A carton with half dozen eggs in it works a lot better than a carton with five…

          • Herbert

            There are systems to count with knuckles that are surprisingly common in different parts of the world…

          • adirondacker12800

            Because that’s the base the Persians picked. You’d be counting to twelve on one hand if they had picked base-12. Which has much tidier fractions, makes everybody’s life much easier.

  2. adirondacker12800

    The 3-hour rule is alluring and may be true

    Was true when flying was a glamorous experience and one could arrive at the airport 15 minutes before flight time. And before airports got so big they need their own railway systems. Four, four and half will get significant share.

    • Herbert

      There still is a type of pax that can arrive ridiculously close to departure time and make their flight. And they’re the kind of pax that doesn’t give a crap about price. On the other end, the average VFR pax will take an eight hour bus ride if it saves ten bucks over the faster modes…

      • adirondacker12800

        And some of them buy first class tickets on Acela. There would be more of them if the trip time was cut by an hour because the train ride is as long as the limo rides to and from the airport.

  3. Herbert

    I think a related concept is “ton ideology”. Back in Soviet days they said stuff like “we produced x tons of this y tons of that” and so on. Well in public transit there’s a danger to just count the kilometers of new line built and see them as the only measurement of quality

    • Alon Levy

      The American freight industry has something similar, leading it to pursue operations with lots of switching because tonnage switched is one of the key metrics to be tracked.

      • Brendan Dawe

        The American Freight industry rather seems to be more obsessed with unit trains than high switching statistics

  4. Herbert

    A common “argument” by austerity advocating penny pimchers is “euros per minute”. They take the schedule time savings (e.g. Stuttgart 21 promises to make train x a 3:21 ride instead of a 3:35 ride) and then divide the cost by the minutes to argue that “y euros are too much for x minutes of time saved”. Of course this figure ignores how many passengers get to benefit from that time saving and whether said time saving enables better equipment utilization. If anything the sensible measure if one is to use that figure at all is pax minutes per Euro or its inverse…

    • Alon Levy

      The main point of the integrated takt timetable is that it takes all of this into account, so speeding up a train from 1:05 to 0:55 has massive savings in equipment utilization whereas speeding one from 1:15 to 1:05 or from 0:55 to 0:45 doesn’t.

  5. michaelrjames

    Alas, you are spitting into the wind. Apparently numerology is profoundly innate to human brains, probably in some real senses, hardwired–or it becomes so as we become numerate when young. Possibly similar to human appreciation/creation of music and our innate detection of tonality and whether ‘in tune’ etc.

    However, be that as it may be, the only partial solution to falling under its sway is via strong government by independent executive (ie. unelected public service) backed up by transparency and regulation (what John Keane has called “monitory democracy”. Yes, I’m going to bang on again about how this has been progressively perverted over the past 40 years, especially in the Anglosphere and is directly related to their huge difficulty in transport (or any large infrastructure) planning and implementation in any reasonable manner.

    I am going to shamelessly use that as a segue into a very tangential topic, but surely of interest to all readers of this blog. There is growing insurrection on the streets around the world. From the relatively mild ExtinctionRebellion, Gilets Jaunes and Brexit, to the more serious calls for true independence and self-governance in Hong Kong and Cataluyna, to sheer pent up frustration at misgovernment in Chile, Cairo and Beirut following decades of cronyism and worse. These invoke all kinds of emotions and perhaps the least important issue is mere property damage (as Mao said, a revolution is not a dinner party …), yet somehow the destruction inflicted on their Metro systems in both HK and last week in Chile is very depressing. In HK perhaps it is still just largely cosmetic but in Chile it was very destructive with entire trains and stations destroyed by fire which cannot be repaired by replacing some glass panels. The Chilean protests began with transit charge increases so that is partly the reason for anger directed to the Metro, but it seems so self-harming; on tv news they showed some Chileans in tears at seeing their Metro reduced to ashes, perhaps because it is one of the few things they have that functions quite well and serves all the people?

    • Alon Levy

      Thoughts: “insurrection on the streets” is 100% people making up narratives. For example, one might note that the largest single protest this year, Fridays for Future, was not on this list of “insurrection on the street,” even though so far among these movements it’s the most successful, in that European governments (Merkel’s Germany!) are becoming less denialist in response. But because Greta doesn’t throw Molotov cocktails for the sake of throwing Molotov cocktails, that somehow isn’t shoehorned into the grand narrative of protest movements.

      • michaelrjames

        Not quite sure what you mean (made up narratives?) but maybe it proves Mao’s point. A dinner party (or street party a la Fridays for Future) is not going to change anyone’s mind that needs changing. However Greta did actually get the attention of old white angry shock-jocks. Here they really went off the deep end in the personal invective. They revealed themselves to be the real snowflakes. So she did get thru to them, and a pretty wide public.

        Also, in my lazy way I was lumping Thunberg in with ER.

        • Alon Levy

          ER is a tiny group of people who dance on the streets and think that this is political activism; they are 100% irrelevant in any serious political discussion. They’re tiny even by GJ standards, and the GJs themselves at their peak mobilized around 0.15% of France (about 100,000 people), vs. around 3% of Berlin marching on 20.09, 20-25% of Hong Kong regularly protesting Carrie Lam, etc.

          • michaelrjames

            But Alon, that is what I was saying””relatively mild ExtinctionRebellion, Gilets Jaunes and Brexit”. Or intended to.
            Incidentally, while we’re bashing the Brits, there has been a meme about ‘despite all the disagreement’ they have managed to do Brexit in a non-violent way, a civilised British way. In today’s papers, the Grau’s tame conservative, Simon Jenkins was boasting about this: “… the genius of the British [unwritten] constitution. It has guided parliament through crisis after crisis, from the Glorious Revolution of 1688 to the Great Reform Act of 1832 and the Parliament Act of 1911. Britain’s parliament at no point lost control of events, whether to the crown, the judges or the streets.” IMO, that may be exactly what is wrong. Mao, again. Time to throw some bricks and pavé. Those Hong Kongers are an amazing example.

      • michaelrjames

        Why Chile’s Massive Protests Started With a Subway Fare Hike
        Juan Pablo Garnham, Nicolàs Alonso, 26 Oct 2019.
        That is why, she says, many of the subway stations in the most vulnerable areas of the city are now in ashes.
        “If they don’t listen to you, what’s next? To shit all over the place,” Rivera says. “People used anger against the subway because it was the only way of getting attention. They don’t listen only with words.”

        It’s not clear who burned the stations of Santiago’s subway—known as Metro—or in what context it happened. President Sebastián Piñera accused groups with logistical skills of “a criminal organization,” but public opinion has been skeptical. What is clear is that last Friday, a series of attacks burned down 19 stations, which moved Piñera to declare a state of emergency and a night curfew in the biggest city.

      • michaelrjames

        A few weeks late but:
        Why protests around the world often involve public transportation
        In Hong Kong, Santiago, and New York City, protesters have disrupted, delayed, or even boycotted subways.
        By Terry Nguyen, Nov 7, 2019

        …. “We’re each protesting out of our own conditions,” says Amin Hussain, an activist and organizer with Decolonize This Place, one of many New York advocacy groups that planned the protest.
        The social and political conditions that have bred protests in cities like Hong Kong and Santiago, Chile, are of course different from those in New York. But in each bout of unrest, public transit systems have taken on new meaning, manifesting into a locus for protest.
        In acts of civic resistance, protesters have taken up space in stations, sometimes disrupting or delaying services. And since transit systems are often seen as an extension of local government and of the officials that run them, they’ve become a ripe setting for civil disobedience.
        “Transit is a basic service that is central to citizens’ everyday life,” says Celina Su, a professor of political science at the City University of New York. “So it makes sense for it to exist as a locus of protest and mobilization, especially around fare hikes.”
        An increase in subway fares or police presence can easily strike a nerve, Su adds, especially in a city dissatisfied with its public officials and socioeconomic situation.
        …. Mobilization occurs at a massive scale when residents are deeply unsatisfied with their public institutions. Su says outsiders might not understand the frustration over a fare hike of a few cents — or pesos, in Chile’s case — but the political message it sends is significant: “A lot of protests are about whether people feel represented. The decision-making that goes into a hike doesn’t feel transparent without proper consultation, especially when people are barely making ends meet.”

        Also, I think the militarisation of city police everywhere is an awful trend. Just viewing their absurd armaments and protection gear and then their hyper-aggressive response, safely in my armchair on the other side of the planet, makes me squirm with sympathy for the protestors and suppressed feelings of violence towards those police who have the appearance straight out of some Orwellian dystopia.

        • Mikel

          Disruption of mass transit is a common protest tactic these days but it can also go the other way around. In HK, pro-CCP thugs attacked commuters in MTR stations. In Catalonia, pro-independence demonstrators sabotaged Rodalies lines to prevent protesters from working-class suburbs from attending an anti-independence march in Barcelona.

        • michaelrjames

          Pretty remarkable events in Hong Kong. Sad too, to see the Metro in flames (picture link below) but OTOH it is just stuff, though public and a great feature of the city and its people. No matter, it can and will be replaced. There are more important things to fight for.
          The remarkable thing is that the unrest continues and escalates, the Lam puppets continue to be as tin-eared as ever, and Beijing may have left it too late to stop this thing. One still retains a tiny scintilla of hope, that what some thought or wished for at the time of the handover, that it was possibly a covert reverse-takeover, ie. of China by HK. Naturally this makes it even more imperative for Xi Jinping to stop it, but seriously, what would it take? The Lam proxy for Beijing is intensely irritating to most Hong Kongers (she is a very effective recruiting tool for this movement!), and if mainland police (already there judged by the use of Mandarin instead of Cantonese) are brought in, or the PLA mobilised, just doesn’t seem likely to work, and could generate an urban guerrilla war instead. And actually, is Xi’s position now at risk? (I certainly hope so. One can only hope there are top-level insiders advising him that he’s a goner if he has to order an invasion of HK.)

  6. po8crg

    The original design of the UK’s HS2 was done to keep the planned budget to £50 billion. There was an initial proposal (the “inverted A”) that would have included extending the eastern branch to Newcastle (it currently stops just outside of York) and the western branch to Glasgow and Edinburgh, and to build a connection running Liverpool-Manchester-Leeds-York joining the two branches together, but those elements were dropped to hold the design under the budget.

    It has (as these things tend to do) since much exceeded that original budget since – I think it’s £78 billion as of the last figure I saw – but the initial design was constrained by an arbitrary number, rather than being defined by the quality of the project.

    • Alon Levy

      The fact that this amount of HSR costs so much is itself a travesty. Germany is a larger and less dense country than Britain and what I think it needs to spend on completing its HSR network is around €60 billion.

      • michaelrjames

        You know I am hardly about to defend the Brits (on rail or anything much) but the fact that the UK is so densely inhabited is part of the problem, especially for HSR because of the lower flexibility in route planning. And they have a finely developed NIMBY skill set, not to mention the actual real landowners (the 1% or 0.1%) who have political power as well. You can hardly shoot a rifle in any direction at ground level in England without risk of killing someone–in a nearby village etc.
        It’s why pre-2007 Eurostar into Waterloo was so slow, threading its way through villages and suburban London on bits and pieces of lines; and why HS1 had to use so much tunnel and take such an easterly route, and of course cost so much. If they’d stuck with Waterloo (to achieve true HSR) it probably would have required tunnel all the way from Ashford!
        It’s why some rail enthusiasts and actual experts suggested maglev, ie. more flexibility on route, tighter turning radii etc. And as it happens, who knows, maybe no more expensive. Distance London to Birmingham is only 6.3x the Shanghai maglev.
        Same factors apply to Japan and why they are putting the whole thing in a tunnel thru mountains. And to stretch a point, the NEC.

        • Alon Levy

          Okay, so first of all, the Netherlands is denser than England and managed to build HSL Zuid at sub-HS2 costs even with a tunnel under prime Holland farmland. And second, it’s not hard to find a route for HS2 that avoids villages and built-up urban areas, the problem is that it might pass through a meadow that some toff never goes to but likes pointing to and saying “this is mine.”

          • michaelrjames

            I don’t disagree with you. Quite the contrary. Indeed it is the combination in the UK that is so toxic, plus, compared to the Dutch, the unwillingness to spend money at the right time & place and by government (who the hell else builds railways today and how many PPP/PFI clusterfks does it take for the Brit neocons to concede it doesn’t work). It’s why I have given up on the Brits. Can’t talk rationally with so many of them on so many things: there is a rabid, feral idiocy about HS2, especially the nonsensical pseudo-eco arguments (FFS, England which is an eco-desert anyway and compared to roads or airports?). Pile on the upper-class twits and entitled types (J Rees-Mogg is a truly wonderful model for this, straight out of Monty Python Ministry of Silly Walks sketch).

            Don’t really want to pick at this but are you really sure the Netherlands is denser? The Randstad is 8,287km2 with 8.2m people (50% of nation), very roughly comparable to London though greater London has closer to 11m; but SE England has a much higher population. If you drive to the south coast (Brighton-Hove), to the north-west (Oxford) or north (Cambridge) or north-east (Norwich), all commuter towns for London, there’s not a whole lot of open land. Some 84% of UK lives in England and Wiki says: “With a density of 424 people per square kilometre, it would be the second most densely populated country in the European Union after Malta.[171][172]” South-East England, a kind of greater-greater-London is even denser.
            Just saying your statement is too black and white.

          • Yom Sen

            Most of dutch HSL is within Randsdad. The southern part, from Rotterdam to the belgian border is only 35km and includes the city of Breda so we’re still clearly in high density area
            Density for the whole country is 418/km2, almost the same than England (424/km2).

          • michaelrjames

            Well, Randstat includes 50% of national population (17m, which is one third of England’s population, approx. 53m).
            And as I pointed out, SE England is denser than the rest; I can’t quite find the popn of SE England* but it’s probably between 25m to 30+m, and assuredly is a lot denser than the average density for all of England which means it has to be denser than Randstat.

            *It would include the Home Counties plus Bedfordshire, Cambridgeshire, Hampshire and Oxfordshire. Couldn’t find it in a few searches and gave up.

          • Alon Levy

            The combination of London and South East England is around 840 people per km^2, and this goes down if you include London commuter belt counties in the East of England, i.e. Essex and Bedfordshire. The combination of North and South Holland is 1,190. North West England is actually denser than South East England, because it includes Manchester and Liverpool whereas South East England is drawn to exclude the region’s big city.

            Not that any of this matters, though. The worst place I’ve looked at for high-speed rail impact on residential areas is the American South, which has low background density, but more importantly has extreme urban sprawl and mostly unusable legacy approaches, so any line that enters a city has to carve right-of-way through endless suburbia. It’s somehow even worse than Fairfield County, Connecticut, which sprawls endlessly but has semi-reasonable rights-of-way, both the Shore Line and I-95.

            In England there’s less sprawl – towns are mostly dots on a map, and HS2 can be drawn to avoid them – and the legacy rail approaches to the major cities are good for 160+ km/h. Moreover, the legacy approaches for HS2 have spare capacity, because outside London there’s no real commuter rail capacity crunch, and within London the WCML has six tracks thanks to the Watford DC Line so it’s okay to give two of them to HS2. Finally, England is not a major agricultural producer, and nothing on the way is prime farmland, whereas Dutch farm productivity is very high, forcing a short tunnel in Holland.

          • Matthew Hutton

            For England you’ve got to take off the area south of Exeter and north of Leeds/Manchester other than Newcastle where the population density is pretty low. That loses you probably 1/5-1/4 of the area of England.

          • fjod

            I suppose the difference between the UK and Netherlands is that HS2 doesn’t use the legacy approaches while HSL-Zuid essentially gives up at Schiphol and spends 20 minutes trundling into a mainline platform at Amsterdam CS. In the UK, years of underinvestment and pushing existing infrastructure to capacity means there is little spare approach or platform capacity left to use. This is especially the case in Birmingham, where New Street is at capacity due to national rather than commuter usage. I imagine the new tunnels and stations of HS2 make up a large amount, if not the majority, of its construction cost – a cost which HSL-Zuid essentially avoided (HSL-Zuid has <10km of new tunnel).

            Really my point is that HS2 is no longer so outrageous in pricetag when you factor in the preceding decades of underinvestment and capacity-fitting service that characterise the UK's railways. Maybe most of the remaining excess cost is due to overbuilding caused by the desire to make HS2 a 'landmark'/'world-leading' project.

          • Mikel

            I imagine the new tunnels and stations of HS2 make up a large amount, if not the majority, of its construction cost – a cost which HSL-Zuid essentially avoided (HSL-Zuid has <10km of new tunnel).

            Even if you take that into account, the current estimate of £78B is completely outrageous. In Barcelona (which is just as old, and denser than London) the legacy approaches were also at capacity and suffering from underinvestment, so they built an urban tunel for €32M/km, according to this list. At that price, you could just tunnel all the way from London to Manchester for €10B and still have plenty of money left for a huge station on each end.

          • michaelrjames

            At that price, you could just tunnel all the way from London to Manchester for €10B and still have plenty of money left for a huge station on each end.

            Precisely (and they already have those big stations). I made the same argument about the outrageous estimate of HSR between Sydney and Canberra: 285 km which is only 1.87x the Gotthard Base Tunnel (57 km route length but a total of 152 km of tunnels). So the Swiss could tunnel all the way to Canberra for only US$19.3 bn.

            I suppose they have good reasons for using Euston rather than run it into/next to HS1 at St Pancras. After all Gare du Nord handles international HSR, not only to London, but Brussels, Amsterdam and Cologne, without any real fuss. Instead it was proposed to create a tunnel link just north of the two stations (maybe 1km long) so that some trains from the north-HS2 could progress directly onto HS1 and head for the continent. It was to cost $700m but it has been dropped, one wonders at what future cost to the Treasury and travellers, not to mention the “powerhouse” economies of the north? HS2 planners now say that pax will either have to catch a LU train across to St Pancras or walk the approx. 400m. Must be the same planner who designed Heathrow T1-T3.

          • michaelrjames


            The combination of North and South Holland is 1,190. (per km2?)

            Don’t know where that comes from as it contradicts anything I have seen. Unless it is using only “urbanised” area instead of raw population/land area (as all the other figures here are). Holland is dense but so is SE England (and clearly more than NE England unless you are doing contortions with “urbanised area”.) As Wiki says: “The population of Greater London and those counties adjacent to the green belt was 18,868,800 in 2011.” But SE England would include those additional counties of Bedfordshire, Cambridgeshire, Hampshire and Oxfordshire, which brings the total population to 22.7m and a crude density of about 800/km2.

            HS2 uses 28.8km of tunnel just to get out of central London and across the M25 before continuing in a ditch (cut & cover tunnel across farmland) for another 13km etc. and so it goes.

            But ok, we actually agree that they could build HS2 without too much drama, but they don’t/can’t because of NIMBYism which is a combination of those little-Englanders (like inhabitants of the fictional Midsomer village in Midsomer Murders which is roughly where HS2 passes–both fictional but hey …) and the landed gentry who have owned most of the land since Magna Carta … (or since the East India Company created most of the landed gentry …).

          • Alon Levy

            I’m counting all land area. Wikipedia has density figures for every Dutch province here; for England, add up South East England and London.

            Also, the 152 km figure for the Gotthard Base Tunnel is single-track, which is why it’s more than twice the route-length of the tunnel.

            The main reasons to use Euston rather than St. Pancras:

            – No more terminal capacity at St. Pancras (which itself was picked for Eurostar because Waterloo East/Charing Cross had no terminal capacity)
            – Euston was rebuilt in hideous modernist style in the 1960s, so it’s legal to knock down the building to create more terminal tracks
            – I suspect the fact that Euston is the historic WCML terminal is also a reason
            – I also suspect that between the language barrier and Britain’s insistence on immigration controls and security theater at Eurostar, Birmingham-Paris and Manchester-Paris are not viewed as big enough travel markets to justify through-service

          • michaelrjames

            Yep, I know GBT is single-track tunnel but as you have pointed out elsewhere the incremental cost of bigger tunnels is not proportional to diameter, and considering the GBT (which are actually quite big diameter tunnels for single track; I believe it was begun with the intention of being two-lane road) was begun in the 90s, and the cost of tunneling has reduced ….

            That’s a peculiar definition of SE England, apparently for statistical purposes. But if you asked any Brit I am sure they would include Cambridgeshire, Essex and of course London (and which has 23m popn.)

            rebuilt in hideous modernist style in the 1960s, so it’s legal to knock down the building

            Hah, plenty of modern hideosity is heritage protected … and St Pancras came within a whisker of being demolished several times. But yes, Euston is a considerably bigger station. And yes HS2 is a duplication of WCML.
            And sure, of course they would argue capacity issues but that only makes the original choice of St Pancras for HS1 a poor one. I see that it is 700m between StP and Euston so it is worse than I thought. Also the plan was to make that linking tunnel a bit north of the stations.

            Birmingham-Paris and Manchester-Paris are not viewed as big enough travel markets to justify through-service

            Don’t go repeating that to any northerners (and remember the north starts at Watford) cos it makes you sound like the London-centric types they have come to hate with a vengence (and which is partly responsible for Brexit). Some trains using HS2 will go all the way to Glasgow and Edinburgh so there is quite a big market. They are specifying two types of trains for HS2, those that will travel exclusively on the new track and those that will travel on both HSR and standard track. Obviously if the demand is lower they will only run a few trains a day that proceed to the continent, like the London-Avignon trains in summer. And part of the concept was to entice those dirty–but rich–continentals to travel beyond London to spend some of their feelthy euros.

            Anyway, what do we (nominal feelthy continentals both) care anymore? Boris or whoever is PM before Xmas might yet cancel the whole thing. A half-built HSR would be an appropriate monument to Brexit!

          • Herbert

            Virgin Rail USA (formerly Brightline formerly All Aboard Florida) is planning to link Orlando to its network soon. Do you think Orlando-Atlanta (a busy air route well within hsr range) would be the next logical step?

          • Reedman Bassoon

            Virgin Rail recently announced its next project — Las Vegas to Southern California.

  7. df1982

    Alon, surely the main problem with Musk-style “numerology” is not the fact that he rounds up from 6-7x to 10x cost reduction, but that he thinks this reduction can be entirely achieved with magical technological innovations, when in reality, as you point out, most of the effort to get New York or California down to Spanish or Scandinavian costs would involve the gritty work of changing a political and corporate culture that fosters cost overruns, over-engineering, outmoded labour practices, pandering to NIMBYs, pointless gold-plating, architectural vanity projects, etc., etc., etc.

    • Herbert

      Well he claims that by virtue of being private sector he can eliminate all those issues without even knowing about them in detail…

  8. Gok (@Gok)

    I don’t get the overpromising section at all. Are we throwing Vision Zero away because it hasn’t actually reduced pedestrian deaths to zero? What’s a case where “only” getting 2x wins when you set out to get 10x wins has actually been bad?

    • Alon Levy

      I mean, in the US, getting Vision Zero just to the point of zero change in pedestrian fatalities since ~2014 would be an improvement.

      But if you want an example of setting for 10x and getting nothing: the tech industry is generally uninterested in improving public transit, because there are (other than NY construction costs) no 10x improvements, just a bunch of 1.2x ones that stack together and don’t let a first(-ish) mover get a global monopoly like Facebook.

      • Gok

        You pretty routinely shit on anyone trying to use tech to make transit better by either a little or a lot, so where are some legitimate places you think tech could improve transit by 20% that aren’t being pursued?

        • Alon Levy

          A couple examples, all of which are technology, of which only one seems familiar to Silicon Valley:

          1. Various track geometry trains. They do track maintenance by traveling over the tracks at low speed and washing the ballast, sometimes replacing the ballast, replacing the ties if necessary, changing the superelevation angle, etc. They allow track renewal on a line to be done in a few weekends’ worth of closure.

          2. ETCS on mainlines, and various moving-block signaling systems on subways (“CBTC”). Of note, Cuomo tried getting some untested alternative to CBTC in the genius grant, which Byford wisely ignored.

          3. Driverless metros, and potentially driverless mainline trains.

          4. Better fare payment back end – some systems spend a large fraction of their revenue just on fare collection, e.g. some of the smaller systems in the Bay Area. This may involve working with mainstream tech firms for stuff like ApplePay or some WeChat clone.

      • adirondacker12800

        They’ve been deeply interested in 2x every two years since Mr. Moore observed it happening. It’s been slowing down lately.

  9. fjod

    Birmingham-Paris and Manchester-Paris are not viewed as big enough travel markets to justify through-service

    The original purpose of Stratford International station was to act as the Eurostar station for East London and East Anglia, with through trains on the West Coast Main Line to Birmingham and Manchester. Running on HS2 instead of the WCML, this would begin to make logistic and political sense, and may well come close to breaking even. Manchester–Birmingham Interchange–(Stratford–)Paris would be competitive with flying (just over 3hrs Mcr-Paris) and could probably merit five or six trains a day. Existing passenger demand for flights is 1.05m/year from Manchester and Birmingham, which is enough to fill 3-4 Eurostar trains per day without any induced demand, and without accounting for the fact that many prospective Birmingham-Paris passengers currently use Eurostar from St Pancras.

    The infrastructure needed to link HS1 and HS2 is relatively minimal if you permit trains to go at ~70kph over the short connecting section rather than at HSR speeds. It essentially requires a flyover at Primrose Hill, a small bit of viaduct widening west of Camden Road station and the reinstatement of Camden Road’s northerly track pair. Probably the best UK comparison (urban, new short viaduct, 2 new platforms) is the construction of platforms 15 and 16 at Manchester Piccadilly, which should cost £300m or so.

    • Alon Levy

      The infrastructure and frequency you’re suggesting showcase serious limitations in French and British HSR scheduling. All of the important cities are stub-end terminals: London, Paris, Lyon Part-Dieu (not literally a terminal, but is used as one by TGVs), Marseille, Birmingham, Manchester, Liverpool. To serve more than two cities on the same train, it’s necessary to use outlying stations like Birmingham Interchange, Stratford, CDG and MLV-Chessy, and Saint-Exupéry. CDG and MLV are both half an hour from Les Halles, and a few minutes farther away from most hotels. This is why German is spending so much money on converting stub-end terminals to through-stations, first at Stuttgart but now at Frankfurt.

      • Herbert

        Thankfully Berlin built its Ringbahn when that was doable for somewhat plausible prices and together with the Stadtbahn had a site ready for its main station that required only high, not outrageous investment…

        Linking Paris’s dead end stations with a Paris Ringbahn or establishing a Centrally located main station in Paris would probably burst all budgets

        • michaelrjames

          Linking Paris’s dead end stations with a Paris Ringbahn or establishing a Centrally located main station in Paris would probably burst all budgets

          Other than that they already have about half of it built (LGV Interconnexion Est, and I recall the under-construction new station at Nanterre-La_Folie for RER-E + M15 +SNCF was originally planned for a La Defense TGV terminus too, now abandoned to keep them at St-Lazare) it is only 5.6km from Gare du Nord to Gare Montparnasse (4.4km to Gare du Lyon but little point as InterconnexionEst serves this market). The under-construction 8km tunnel for RER-E western extension (St-Lazare to La Defense) was costing about €4bn.

          No, I suppose they are not going to do it but just sayin’.
          Gare Pleyel will take pressure off Gare du Nord and allow thru-pax to avoid central Paris (and have impeccable crosslinks via M14, M15, M16, M17, T1 & RER-D).
          Unlike London & UK, Paris is benefitting from long-term planning and implementation. For the sake of a rounding error on the cost of HS2, they will forever regret dropping that 1km link between HS2 and HS1 (which of course is also a link to Stratford so would be a viable convenient link to Eurostar for northerners).

          • Herbert

            When I look at trains from Germany to Spain, they include a one hour trip on local transit in Paris every single time. In Berlin I can just connect everywhere to everywhere at Hauptbahnhof with maybe ten minute walk if it’s from the lowest to the highest level…

          • Alon Levy

            What you’re missing is that these interconnections split frequencies between different services. There’s no train that stops at Lille, Paris, Lyon, and Marseille. There isn’t even a TGV that stops at Paris, Lyon, and Marseille, if you define Lyon to mean Part-Dieu (i.e. the CBD) rather than the airport. The French paradigm, copied by Britain for a combination of understandable and stupid reasons, is that TGVs should connect two cities and run nonstop or at most with 1-2 low-ridership intermediate stops.

            Also, you’re overstating the Saint-Lazare to La Folie tunneling cost. The tunnel as I understand it is 2-2.5 billion – the higher cost includes other things like surface improvements. (It’s the same with Crossrail: 21 route-km of tunnel don’t cost 15 billion pounds, not even in London.)

        • michaelrjames

          Herbert, 2019/10/29 – 08:52
          When I look at trains from Germany to Spain, they include a one hour trip on local transit in Paris

          Reminds me of the joke about the Texan cattle rancher visiting an Australian cattle ranch, having been shown part of it by light plane. Texan says to the Aussie cattleman that he can ride his horse all day and not reach the end of his property. The Aussie tells him laconically, “yeah, I once had a horse like that”.
          Dunno what you are spending that hour doing in Paris. I can walk across Paris in that time:-)

          I don’t know the likely routes from Germany but if you really wanted to avoid Paris you could presumably change (Lille, maybe Brussels?) to one of the TGVs that takes the InterconnexionEst line to the south. They built this line specifically for you!
          But of course, Herbert, you don’t really want to avoid Paris, do you? (These days I recommend people bound for the UK–from Asia–take the plane to CDG, spend the day in Paris and take the last Eurostar to London … it actually costs less, is supremely hassle-free and you get a free day in Paris).

          Also you’re comparing a journey of thousands of kilometres across the largest country in the EU ….

          BTW, as you well know, there’s a single stop between Gare du Nord and Gare de Lyon on RER-D. A 7 minute ride. You need a new horse.

          • Alon Levy

            Trains from Spain don’t go to Gare de Lyon, they go to Montparnasse (enjoy M4 with your luggage) or maybe Austerlitz (enjoy M10-M4 with your luggage), and trains to Bavaria don’t go to Gare du Nord, they go to Gare de l’Est (technically RER-connected, but it’s a looooooong walk).

            P.S. Both of you, bug me sometimes to finish my series about different national traditions of rapid transit and write about France and Germany.

          • Yom Sen

            The issues are the low frequency on intersecteurs TGVs and the lack of interest for connections in France: Only 2 trains a day from Strasbourg to Bordeaux and to Montpellier, TGVs Frankfurt Paris don’t stop in Lorraine TGV station. The only option I see is a train from Strasbourg to Montpellier leaving at 09:03. You can leave Frankfurt at 6:58 and arrive at Barcelona at 19:30 and Madrid at 23:10 but that’s not possible from Nuremberg for example. All other options are through Paris which is really painful when you have luggage and are not used to the Metro. Note that SBB website gives 55 min needed for a connection Gare de l’est -Montparnasse. By foot it’s 6km, clearly more than 1 hour.

          • michaelrjames

            You’re making the same comments as Alon. Clearly those long-distance services are reflecting demand. And I’ll repeat my guess for why: a lot of travellers, especially train travellers, might think a day, or even a few hours, in Paris would be a bonus instead of a chore (unlike London where it is definitely a chore and a bore). Especially if they are familiar with it and thus have confidence of getting around efficiently via Metro + RER. (I’m told some people even prefer buses because they get a bonus tour of Paris.) Thus whether it is 7 minutes (RER-D travel time) or 1 hour by whatever means is kind of irrelevant. IMO Paris is one of the easiest (big) cities to get around, partly because it is so compact and partly because of transit options are so good.
            Travellers with a different attitude are much more likely to be using some crap LCC to fly from Germany to wherever in the south or Spain. Train travellers always consider a big-city stop to be a bonus of such travel (exception might be London, as I say).

            Note that SBB website gives 55 min needed for a connection Gare de l’est -Montparnasse. By foot it’s 6km, clearly more than 1 hour.

            Closer to 5.6km. My natural walking speed is precisely 6km/h (averaged over hundreds of 3 km trips to work, it was 30m, with best time 26m) so I could walk it in 1h. Probably not with luggage (but James’s first rule of travel is to only take luggage you are happy/capable of carrying for long periods without exhaustion, and while keeping at least one hand/arm free). The 55 mins is to allow for getting from/to TGV platforms etc.; M4 journey would take 20m to Montparnasse.
            I’ll admit that I don’t think much about Gare de l’Est but ok that is where trains from Germany and eastern-France arrive. Also far more travellers will be heading to Gare de Lyon for the south of France or even Spain, and for that you’d stroll (450m) across to the Metro/RER entrance in front of Gare du Nord (a 9 minute walk, though at my natural pace it would be only 5 minutes so 9m is very slow, with luggage I suppose).
            Here’s seat61:

            The Gare de l’Est is my favourite Paris terminal station, an elegant station that’s much calmer than the hectic Gare du Nord next door, or the bustling Gare de Lyon a mile or two to the south. … The Gare de l’Est is a terminus with easy, flat & level access between the taxi rank, concourse & all platforms, so painless to navigate with wheeled luggage.
            (For) the nearby Gare du Nord – I always walk via the Rue d’Alsace, but this involves steps*. For level access, go via the Boulevard Magenta or the rue du Faubourg St Denis. It is a lot easier to walk than to take the metro one stop!

            (*these are the steps featured in the movie where Amelie chases the scooter when the rider looses his binder of photobooth photos that she picks up).

            And yet again, you along with Alon are missing my point: Paris has 5 mainline stations that have TGVs all of which used existing ROW tracks converted into LGVs. London will have only two LGVs one of which serves the continent and both have had to be newly and expensively built by tunnel into the station, and one could envisage a significant number of pax arriving from the north on HS2 transferring to HS1. Which is why they proposed that 1km link between them. In an extraordinary but entirely habitual bit of short-termist penny-pinching they have abandoned that plan. In fact I think it even makes sense so they can run trains from the north direct to Stratford which is going to be a mega-transit point, including for Canary Wharf. Maybe there is already a level walk between St Pancras and Euston, I don’t know but it’s already a lot longer than Est-to-Nord and it’s London, not Paris, that 4 decades of experience tells me to worry about. Maybe they’ll build a pedestrian tunnel linking HS2-HS1, ie. the two stations, to make it painless for travellers but I doubt it.

        • Alon Levy

          Not would, but did: the RER A’s central segment cost, in today’s money, around $750 million per kilometer, the bulk of the budget presumably going to the construction of Les Halles. It was most likely plausible to include provisions for a two-track mainline tunnel connecting Gare du Nord and Gare de Lyon in addition to the RER tunnels (the Les Halles-Gare de Lyon section was eventually built as the RER D); I suspect this was not done because at the time France wasn’t yet thinking about high-speed rail, and at legacy rail speed nobody would ever ride between Lille and Lyon.

          • adirondacker12800

            According to WIkipedia they settle on a plan for the RER in 1965 and the first hovertrain tests are in 1966. They were thinking about it.

          • adirondacker12800

            They decided in 1973 to use electricity. After deciding the hovertrain idea wasn’t panning out and using turbine powered trains would use too much imported oil. 90 seconds of Wikipedia. I don’t really care all that much, I’m not the one trying to make a living at this.

          • michaelrjames

            I was speaking of the TGV service, Paris to Barcelona, which I believe leaves from Gare de Lyon. You must be thinking of Atlantic coast, Montparnasse to Bordeaux and on to Basque country, which is a much longer route. The old night train to Madrid actually left from Gare d’Austerlitz (but it no longer runs) and its replacement was shifted to Gare de Lyon because there’s no LGV into Austerlitz. Too bad, no excuses for splurging at Train Blue.
            So, you know, arriving at Gare du Nord, or Est, or Gare Pleyel you’d use RER-D or M14.

            Re InterconnexionEst etc, sure there are limited options today, though I guess at peak times they might run trains, or at least sell tickets with appropriate changes, from Germany to the South of France to bypass Paris. But its equally true what I wrote: a lot of that traffic might still choose to pass thru Paris if only for the day. But the point is that the bypass option exists.
            On the other points, I can’t imagine they are going to build TGV tunnels under Paris. And this discussion tracks back to the option of an approx. 1km tunnel linking HS2 to HS1, which absolutely I think they should do. Unlike Paris, London can be a real drag to cope with if you are really aimed somewhere else, and especially for northerners who would rather avoid it. Heck, never mind northerners, when I lived in Oxford I would prefer to avoid London.

Leave a Reply to adirondacker12800 Cancel reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.