It seems the network of roads built in the 40s, 50s and 60s just can no longer be done efficiently.
Not uniformly. New York's LIRR (90 to 95% [1]) and Metro-North (99% [2]) feature on-time rates that rival the Swiss (93% [3]).
[1] https://www.osc.ny.gov/files/reports/pdf/report-9-2025.pdf
[2] https://wpdh.com/metro-north-on-time-reliability/
[3] https://reporting.sbb.ch/punctuality?=&years=1,4,5,6,7&scrol...
China’s next-gen is being deployed with goals of a 400kph service speed.
If an ICE is, say, 15 minutes late, they cannot just drive faster to catch up. The schedule went on, and at that point there may be a much slower regional or intercity train on the same trajectory.
This is why ICE delays tend to cascade. It starts with a short delay, the ICE gets stuck behind a slower train, increasing the delay, etc.
The solution is better maintenance of tracks and trains, adding more rail capacity, adding redundancy, etc.
Of course, these are all much more expensive than an ICE speed experiment for PR.
> A commuter train is considered on-time by the LIRR if it arrives within 5 minutes and 59 seconds of its scheduled arrival time.
The second source doesn't say, but let's assume it's the same as for LIRR, i.e. 6 minutes.
It's also unfortunate that the SBB doesn't immediately tell us the metric, but I happen to know it's 3 minutes (more specifically 2 minutes and 59 seconds).
In other words, the LIRR permits a delay of twice the time as SBB for it to constitute late. The S-trains in Copenhagen now has a punctuality of 97% using a 3 minute metric.[0]
[0] https://www.dsb.dk/om-dsb/virksomheden/rettidighed/s-togs-re...
Of course they can, we have not lost that capability. It's not a matter of efficiency but effectivity. Road constructions of the 1960s are not effective for 21st century traffic demands. Today's level of traffic far exceeds the anticipated level of traffic at the time of construction. Germany sees this all the time, esp. with regards to bridges. Maintaining a road or a bridge to be effective at supporting the original traffic levels is easy but under today's load would require constant maintenance to not deteriorate immediately. Those constructions need to be upgraded which is hard to impossible to do in situ.
Let's take a well-known construction in Germany, the Leverkusener Brücke an der A1. It was originally built in the 1960s for a traffic level of 40,000 cars (and trucks) per day. It was upgraded and refurbished over the decades (meaning almost constant construction work happening) to a level of 100,000 cars per day. It wasn't enough, in 2016 about 120,000 cars crossed the bridge per day. At the same time trucks got about 30 % heavier from 1960 to 1990 and we all know that passenger cars got heavier, too.
So the whole bridge was replaced, which took 7 years, ending in 2024. During that time traffic was rerouted over two nearby bridges in Cologne and Düsseldorf. The Cologne bridge was so badly damaged by the additional load that it had to be partially closed down and now is up for refurbishment or, maybe, replacement. Network effects at work ;)
Anyway, what I'm trying to say is: we are actually better at building stuff than our predecessors but the demands put on our constructions are much, much higher. I don't dare to say if our capabilities have grown as much as the demands require.
On some Swiss lines it's fairly common to have trains recover their delay at departure (which also avoids cascading impact).
DB is doing their best work for having people reaching out for cars.
While you might get stuck in traffic on the motorway, there are usually workarounds as soon as you get the next exit point, while being stuck on an train stopped in the middle of nowehere with a full train excedding passanger capacity because "pick random excuse", and reservations being optional, isn't really motivating to keep traveling by train.
Citation needed.
As I see it, the US is still riding on the coattails of 1960-s road construction. We should be doing more of it, in fact, not sabotaging it with bike lanes and road diets.
Not sure how you compare a small simple system like Lord which is pretty much one line with a few branches with an integrated multi-national system like the entire Swiss railway either
The main cause (often somewhat hidden behind the term "decayed infrastructure") is that there are too many trains on too few tracks. There are many reasons for that. I think the main ones are:
* Political pressure to have more trains, without an adequate increase in infrastructure capacity (trains are cheaper than tracks and can be delivered faster). For example, political pressure utterly destroyed the reliability of the local rail system in our area, because the number of trains per hour was increased by a factor of 2-3, with only a minimal amount of new tracks (the majority of the network is still single-track). Apparently, the system worked in simulations under near-perfect conditions (no delays, few passengers, no technical problems). So let's build it! The chaos that ensued during the first few months after the network opened again made national headlines. Another example: the highly overloaded Rhine valley line between Mannheim and Basel was proposed to be upgraded to 4 tracks in 1964. In 1970, the project was scheduled to be finished in 1985. Currently the (ambitious) goal is to finish the project in 2041 [0]. The original line (270 km) was finished after 17 years in the 1840ies.
* On regional and local lines, a tendency to increase train frequency and to decrease train capacity (more trains, but shorter ones). I suspect this is also because of political pressure ("your station now has 4 trains per hour!!"), but it doesn't make any sense. A short train which can hold 150 passengers occupies exactly the same amount of "space" (blocks) on the tracks as a full-length train with a capacity of 1,200 passengers, and they require exactly the same amount of personnel.
* Privatization of DB on the early 90ies, with political pressure to be profitable. Tracks are expensive to maintain, so those parts of the infrastructure that could be classified as "redundant" were dismantled. Now they have a network with little redundancy, which is great from a short-sighted business standpoint, but terrible for reliability.
[0] https://en.wikipedia.org/wiki/Karlsruhe%E2%80%93Basel_high-s...
Counterintuitively perhaps, bike lanes, road diets and public transport actually work better. Give people other ways to move around, and you take pressure off the road system, making traffic better for everyone. And that includes the drivers.
Switzerland SBB/CFF and the German DB can not be compared, not even from far.
The Swiss trains are amongst the best in the world in term on punctuality. Delays barely exceed few minutes most of the time. Every connection is scheduled to be done < 5min. The usage is smooth like butter and It works like a Swiss clock.
At the opposite, German trains in the eastern part are barely on time and give you an almost Soviet experience for the regional one: The trains are old, poorly maintained, like the track itself and the service suffers of it.
The only place in Western Europe I experienced train to be worst than in Germany is currently in Hungary where there were actual soviet trains.
Even the freaking French SNCF with their legendary strikes tend to be more punctual than the DB.
As for the LIRR, it seems it's only the terminal station that counts:
> Thus, a train is considered late only if it arrives at its final destination six minutes or more after its scheduled arrival time.
Do you have a citation for this?
The people I know from the DB bubble are telling me that while some places have not enough track (e.g. the infamous Frankfurt-Mannheim/Riedbahn), but the everywhere problem is that there's just fault over fault over fault in tracks (often switches, but even tracks themselves) and trackside equipment.
> […] the number of trains per hour was increased by a factor of 2-3 […]
Even this I'm not quite willing to accept without citation; the railway timetables in the 70ies and 80ies, especially after the oil shock, were quite dense.
As someone from the region, since when is Hungary 'Western' Europe? We are happy when acknowledged as 'Central' but this is new to me.
I don't, this is just my personal experience with passenger rail in southwestern Germany. The smaller lines with 1-2 trains per hour are usually extremely reliable, while the lines on crowded tracks are usually delayed, or cancelled completely. Note that many of the reliable lines I regularly used over the past 10-15 years ran on track equipment from the 19th century, some still with wing signals and switches operated via pulleys. (Anecdotally, I never experienced any technical problems with this old equipment as a passenger, the technical problems usually started after modernization).
> Even this I'm not quite willing to accept without citation; the railway timetables in the 70ies and 80ies, especially after the oil shock, were quite dense.
The factor of 2-3 was for my local network (which was converted to an S-Bahn network a few years ago, and a 30 minute frequency was introduced, with 15 minute frequency during peak hours). Some parts of that network only had 1-2 trains per day from the 70ies to the mid-90ies.
Right here. Those slower (older) trains don't magically get any faster just because the train behind it sped up.
> pack the schedule
You actually can't just double it up because faster trains need bigger gaps between them, just like driving on a motorway. If the train in front needs to slow/stop for whatever reason, you don't want the train behind smashing into it at 400km/h because it was tailgating.
The most efficient use of the tracks would be to limit all trains to the speed of the slowest train that travels a route. Then you can have minimal distances between trains.
Also, faster trains are less fuel-efficient, quadratically with speed. So a slowdown would help the environment and the throughput. The only thing it wouldn't improve is passenger happiness ;)
Switzerland doesn't have high-speed trains, only low-speed ones. And all timetables are carefully tuned to half- or full-hour station distance intervals that all trains on a track take at the same speed ("Taktfahrplan").
Germany has the worst of all those worlds: No Taktfahrplan (because it probably would be impossible due to the larger and far more complex network), high- and low-speed trains on the same tracks, and only some sections of dedicated high-speed rail that drive up the cost but still have shared stations and sections with low-speed rail so that punctuality goes down when the tiniest thing goes wrong.
It is just that trains and bikes are much more efficient in terms of land use.
The 3 lane road in front of my house is "good" for 16,000 cars a day. The 2 lane train line a 5 minutes walk from my house is "good" for 120,000 passengers a day. A train line can carry about 10x the traffic of a car lane (in practice) with similar ground usage.
So when a train system has more demand/use than expected (e.g. leman express in the geneva region) there are more options to increase throughput (in the leman express case double level trains) that require less new infrastructure to be build.
When new infrastructure is required, limitations of space mean that a 15 year period from plan to implementation is normal. Which means infrastructure which has more head-room is preferred over quickly saturated ones.
To add the adding of one lane to the A1 for 18KM costs half the total of the leman express infrastructure. But has significantly less benefits in total transit capacity.
Yes 100% right. Japan has the same where the Shinkansen uses dedicated track. If I do not say a mistake, China does the same too.
The fact it never has been done in a modern country like Germany with a dense high speed train traffic like the ICE is clearly a sign of planning deficit from the authorities.
And yes, in Germany this happens and is reflected in rising "cancelled train" statistics. KPIs for Deutsche Bahn managers haven't caught up yet ;)
The Shinkansen initial network and separated tracks have been built at an age where JRails was still a single centralized company.
The main reason was to create a network with a focus speed and punctuality. And to be fair, it was the right choice and pretty revolutionary at the time.
Btw. Wooden sleepers laid in the 19th century have completely rotted away by now; it's impossible for the line track itself to be that old (except very slow speed shunting tracks, maybe.)
That is not true, some TGV dedicated stations have been put in remote area (mostly in far right areas), but there are lots of cities that have high speed train stopping at the central station in the center.
Train frequency is the most important aspect of transportation. It's probably more important than max speed or even ticket price. A train every 15 min means you don't have to make an appointment because one will be coming soon. At half an hour you will need to study the timetable and plan your life around it.
Bikes are used there ONLY because there's no alternative to them. Transit takes too long, and there's no space for cars. And yet still around 20% of commutes in Amsterdam are by car.
> not every country has to be a nightmare for life quality like the US aspires to be.
The US is far, far, far ahead of Europe in urban quality of life that it's not even funny, if you disregard the dense hellscapes of SF and NYC.
This is easily seen in the number of children per capita. In modern societies, two groups of people tend to have more children ("inverted J-curve"):
1. Happy content people.
2. Desperately poor people.
Now look at Europe and the US, and I suggest looking at the US suburbs and not the dense cores.
No. I'm well aware of induced demand. Now apply the same argument to _housing_.
And the fix for housing prices is to reduce the housing _density_. We already have more housing than needed (there are 1.1 housing units per family in the US), we just need to make sure all of it is viable.
There's even a Wikipedia article with some photos: https://en.wikipedia.org/wiki/High_Five_Interchange
See: https://www.theguardian.com/cities/2017/sep/19/britains-1960... Another example is China, it had bikes as the main transportation mode in many cities, but it abandoned them as soon as they could.
Bikes are a freaking _miserable_ transportation mode. Try biking in subzero temps, or during rain/snow.
> Bikes are a freaking _miserable_ transportation mode. Try biking in subzero temps, or during rain/snow.
Your entire argument seems to revolve around car vs bikes. I am arguing for prioritizing human-centric infrastructure over car-centric infrastructure. This simply means not prioritizing cars above other modes of transit. For urban environments (high density areas) cars are sub-optimal at best and other alternatives like rail, buses, bikes, and trams are simply better. Granted, sometimes cars are appropriate, but I would not use those minority of cases as an argument to justify having a city designed around cars.
> Another example is China, it had bikes as the main transportation mode in many cities, but it abandoned them as soon as they could.
Because Chinese cities now have better public transit. Just look at the size and ridership of their suburban rail and metro networks.
But that's not true. Your chances of living within 5 minutes of a train stations are slim, unless train stations are spammed everywhere. And if stations are spammed everywhere, then they become inefficient.
Meanwhile, cars are only mildly affected by additional 400-500 meters of distance.
There's a great resource: https://www.geoapify.com/isoline-api/ - it shows isochrones for different commute methods.
> A train line can carry about 10x the traffic of a car lane (in practice) with similar ground usage.
In practice, a train line effectively is only slightly better than cars, unless you enshittify your city into a Manhattan-style dense hell.
Moreover, self-driving cars with mild carpooling (think 4-6 people per vehicle) blow ANY transit mode out of the water in speed and efficiency. It's not even close. A good approximation of this are airport pickup vans (the ones that you arrange in advance).
> To add the adding of one lane to the A1 for 18KM costs half the total of the leman express infrastructure. But has significantly less benefits in total transit capacity.
Yeah. Imagine that instead of wasting money on useless transit (see: Seattle ST3), we used them to incentivize companies to build more offices outside of dense city cores.
Then these lanes wouldn't even be necessary!
And here comes the propaganda. Why is your (failed, btw) idea of infrastructure is called "human-centric"? What makes it _human_?
I fail to see anything human in dense cities like Tokyo or Manhattan.
> Granted, sometimes cars are appropriate, but I would not use those minority of cases as an argument to justify having a city designed around cars.
If cars lead to better outcomes (and they do, once self-driving is deployed), then why not? Why force people into densified anti-human hellscapes?
> Because Chinese cities now have better public transit. Just look at the size and ridership of their suburban rail and metro networks.
Yes. And once people can, they opt out of transit as well.
Great question. No strict definition as such for me, but in the context of transportation, it loosely means that it must be 'easy' for most people to get from point A to point B, using the least amount of time. Ideally it must must be accessible for all people (so teens without a licenses, legally blind people, or anyone without the means to get a car etc). From a non-transportation context (just adding, not very relevant to the answer), places with no loud traffic noise and just a nice general atmosphere of people around. Think streets with trees, and lively public spaces for people to hang out.
> If cars lead to better outcomes (and they do, once self-driving is deployed), then why not? Why force people into densified anti-human hellscapes?
No one is being forced. If anything, these measures only makes the cities with high urban density more livable for people. The 'hellscapes' have more to do with the housing crisis and zoning laws than transit infrastructure.
Idk why most people think urban density is a problem to be solved (using tech like self driving for eg), because it ultimately always circles back to the fact that density is the only sustainable way for cities to develop and grow in the long run. Even if it is not perfect.
> I fail to see anything human in dense cities like Tokyo or Manhattan.
It would be even more dystopian if Tokyo didn't have a great metro and people had to wait in traffic for hours. Imagine the increased stress, lesser free time, and lower quality of life from all the noise and pollution.
> Yes. And once people can, they opt out of transit as well.
When exactly will most people have the option to opt-out, and how likely and how exactly will it happen?
I also feel like you're arguing just for the argument's sake. If not, we probably have fundamentally different views on what's 'better' and I don't think any amount of convincing will change either of our minds.
In or near Amsterdam I've seen and/or used busses, trains, ferrys, trams, bicycles, cars; and aircraft on approach to Schiphol looked like they almost flew past the window.
I also kind of like New Amsterdam (New York). It's cozy!
Can't argue with taste I guess. %-)
[1] Intercity trains: Long, Yellow, with electricity, free wifi, a place to sit, and you can bring your own Starbucks on board. Do try to avoid rush hour, or you'll be taking "standup" a bit too literally.
And, a lot more stations are within say a 15 minute reach if you use a bike O:-) https://www.youtube.com/watch?v=1UxCbmT9elk
So did I. And I had to be at the office at 7:30am. Even when the weather outside was "bracing".
> [1] Intercity trains: Long, Yellow, with electricity, free wifi, a place to sit, and you can bring your own Starbucks on board. Do try to avoid rush hour, or you'll be taking "standup" a bit too literally.
So basically, you wasted around 2 hours a day on the commute?
And then what? How do you get there?
Your time budget is 30 minutes (the average commute in the US). Go on, try to play around.
> that or telecommute
Yes. But if you telecommute, then why bother with all those trains and dense offices?
> And, a lot more stations are within say a 15 minute reach if you use a bike O:-)
That's already too much for commutes and will result in commutes inferior to the current status quo in the US.
OK. Drop 10 random points in a city and then map routes between them using transit, cars, and bikes. You'll find that cars are around 3 times _faster_ than anything else on average. Even in dense cities. I'm going to call this "Cyberax's constant"!
> Ideally it must must be accessible for all people (so teens without a licenses, legally blind people, or anyone without the means to get a car etc).
This means that bikes are right out, they're horribly anti-human. And transit is barely OK.
> From a non-transportation context (just adding, not very relevant to the answer), places with no loud traffic noise and just a nice general atmosphere of people around.
So, EVs are perfect? They are quieter than buses.
> No one is being forced.
People are. By economic forces. Japan is a GREAT cautiounary tale here.
> If anything, these measures only makes the cities with high urban density more livable for people.
No, they make it more inescapable.
> It would be even more dystopian if Tokyo didn't have a great metro and people had to wait in traffic for hours.
Wrong. A Tokyo without transit wouldn't have become so dense.
> I also feel like you're arguing just for the argument's sake. If not, we probably have fundamentally different views on what's 'better' and I don't think any amount of convincing will change either of our minds.
You simply have never thought about the _effects_ of urbanism. Go on, think about it and actually investigate your assumptions.
I got my driving license at the age of 27 and I lived in multiple dense cities.
Of course I do have a slightly different set of requirements; since I've always lived out in the countryside. You trade in a longer commute for more elbow-room at home.
The trains generally run on time, so that's what I often used to use if I needed to get into a dense town.
That was before COVID. Post-COVID, telecommuting has become available to more people. In my opinion, that's the best solution where possible.
At the very least telecommuting and trains gets the OTHER cars off the road when I need to physically be at factories, labs, or workshops.
From some level of experience doing just this all my life, this probably depends a lot on which city, how it's designed, and the time of day you do it at. There's some cities where you can actually get better results I'm sure! And there's some locations where I bet you'll get drastically worse results, especially during rush hour.
I'm really curious where you're living at the moment! And have you considered making counterpoint videos vs NotJustBikes? ;-)
This gives off "I have experience, so I must be right." vibes. There's no reason you can't be, but this goes against what most other urban planning experts and studies have to say.
> Go on, think about it and actually investigate your assumptions.
I feel like _you_ are the one who is only addressing the points that are the easiest to argue using unverified numbers and hypotheticals that would never be practical, just to support your claims, instead of addressing the argument as a whole.
> OK. Drop 10 random points in a city and then map routes between them using transit, cars, and bikes. You'll find that cars are around 3 times _faster_ than anything else on average. Even in dense cities. I'm going to call this "Cyberax's constant"!
I would like to see you prove it. Even if true, no one is actually going from a random point to another random point. Most people have overlapping commute routes, which are anything but random. It is proven mass transit reduce collective travel times, and which is why I mentioned the least amount of time for _most_ people, which you conveniently didn't account for.
> This means that bikes are right out, they're horribly anti-human. And transit is barely OK.
Again I never argued for _only_ bikes, you're dragging your own assumptions and hatred for bikes into the argument. I also don't know why you think bikes more anti-human than cars. Anyway, how is a city that needs cars to get around better than a city that doesn't, better in terms of accessibility? And again, "transit is barely OK", but you conveniently did not mention about the accessibility of cars.
> So, EVs are perfect? They are quieter than buses.
EV buses have all the advantages of being an EV and none of the disadvantages of being private cars (aka more efficient, cheaper, and more accessible). Idk why you would use EV for cars, but not do the same for buses when comparing them (if not just for pushing your agenda). If anything, I would guess the percentage of EV-to-ICE ratio for buses is higher than for passenger cars in most high-density cities (except for the US, ig idk).
> People are. By economic forces. Japan is a GREAT cautiounary tale here.
Economic forces will always end up pushing for more and more density. Pushing for car dependent cities may mitigate this to a point, but it does not create better cities. Good zoning laws have a much higher probability doing that.
> No, they make it more inescapable.
As mentioned, I fail to see how good public infrastructure is stopping a person from leaving a city more than other economic forces, if it is so insufferable.
> Wrong. A Tokyo without transit wouldn't have become so dense.
Great point. It would have become LA. Where your ideas (at least from what I understand about your ideas) have already been implemented. A place known for zero traffic, that consistently comes on top for best cities to live.
I think I am done with this, have a good day bro
The commutes in large cities (New York is a bit more nuanced) in the US are still faster than in _any_ large European city. Mostly because of cars.
> Of course I do have a slightly different set of requirements; since I've always lived out in the countryside. You trade in a longer commute for more elbow-room at home.
My favorite city from the urban design standpoint is Houston (I hate its climate and Texas that surrounds it). People there can have beautiful and spacious single-family houses with backyards, and yet still have short commutes because it doesn't have a well-defined city core.
So it lacks the obvious traffic magnets, and people tend to chose jobs near their housing. This is the model that needs to be promoted, and it can solve housing issues.
It doesn't depend on the city, if we're talking about cities with population more than 2 million!
It's just the result of the inherent inflexibility of fixed routes. If you pick random points, most likely there won't be any direct route between them. So you'll have to do at least one transfer, and this just kills the average speed.
That's why large cities tend to gravitate to the "hub-and-spokes" model. It's simply the only model that works. Look at the transit map for large cities: Tokyo, Berlin, Moscow, Beijing, and you'll start seeing it everywhere.
BTW, it's also the reason bikes work well in Amsterdam.
> And there's some locations where I bet you'll get drastically worse results, especially during rush hour.
For this particular test, it actually doesn't matter that much, due to randomness of the chosen points. It's likely, that at least some routes will bypass the traffic magnets and chokepoints.
> I'm really curious where you're living at the moment! And have you considered making counterpoint videos vs NotJustBikes? ;-)
Yeah, I'm thinking about it.
You know, I'm not saying _anything_ that experts disagree with. And I talked at length with more than one expert, and I had taken a university course in urban planning and demographics.
I'm just not considering the outcomes of urbanization to be positive. And I try to highlight things that urbanists tend to sweep under the rug. E.g. that bike lanes or rapid transit lines do not decrease congestion in most cases (see: https://www.nber.org/system/files/working_papers/w18757/w187... ).
> I would like to see you prove it. Even if true, no one is actually going from a random point to another random point.
Will write a blog post. I've plenty of notes and code for the experiments that I need to organize, instead of doing flame wars. Although sometimes I do hear new arguments.
> EV buses have all the advantages of being an EV and none of the disadvantages of being private cars
Buses are terrible for commutes. They are rolling lifetime destruction machines, with more than one lifetime wasted on commute, every day, for large cities.
They are also terrible for roads (look up the "fourth power law"), and are way too expensive.
Self-driving personal EVs and mini-buses will kill all the rest of transit.
> Great point. It would have become LA.
Or Houston (population 5 million) that still has a faster average commute than ANY large European city.