BahnMining - Pünktlichkeit ist eine Zier (David Kriesel)
media.ccc.de2 HN recommenders · recommended in 2 months, 2022–2026
talk · general · 2022–2026 on HN
David Kriesel's 36C3 talk, in German, analyzes a year of scraped data on every stop of every long-distance train in Germany. It shows how Deutsche Bahn delay statistics hide cancellations and delays.
Recommended in threads on German rail as the data-backed explanation of how reported punctuality understates real delays.
Up next
14 mo on HNA Better Default Colormap for Matplotlib | SciPy 2015 | Nathaniel Smith and Stéfan van der Walt
Enthought
15 HN recommender · 2015–2026
11 mo on HNDON'T PANIC — Hans Rosling showing the facts about population
Gapminder Foundation
8 HN recommender · 2017–2021
8 mo on HNIs Most Published Research Wrong?
Veritasium
8 HN recommender · 2017–2025
6 mo on HNHow not to be ignorant about the world | Hans and Ola Rosling
TED
5 HN recommender · 2015–2023
6 mo on HNI don't like notebooks.- Joel Grus (Allen Institute for Artificial Intelligence)
O'Reilly
6 HN recommender · 2018–2025
6 mo on HNVisiData Lightning Demo at PyCascades 2018
Saul Pwanson
6 HN recommender · 2018–2025
6 mo on HNEmacs + org-mode + python in reproducible research; SciPy 2013 Presentation
Enthought
5 HN recommender · 2014–2018
5 mo on HNI Like Notebooks
Jeremy Howard
4 HN recommender · 2020–2025
4 mo on HNInteractive notebooks ~ Pluto.jl | Fons van der Plas | JuliaCon 2020
The Julia Programming Language
6 HN recommender · 2020–2021
6 mo on HNYou Suck at Excel - Joel Spolsky (Reupload)
Cool Bug Facts
7 HN recommender · 2024–2026
5 mo on HNMichael Stonebraker | Big Data is (at least) Four Different Problems
Michigan Engineering
5 HN recommender · 2015–2024
5 mo on HNSpecial Topics - The Kalman Filter (1 of 55) What is a Kalman Filter?
Michel van Biezen
5 HN recommender · 2017–2025
7 comments from Hacker News
I think they want to get rid of the trains going to Interlaken directly. At least that's what I read in the SZ [0]. If you haven't seen it, there's the great talk from David Kriesel on the 36C3 [1] about how Deutsche Bahn is... "doing statistics" to hide their delays. It's well worth the hour! [0]: link [1]: ▶ this video
That's why you should hire the PR companies they're using, not the train companies themselves. Partially joking, I think the image stems from the past, but the system degrades when it's not maintained appropriately. There was an entertaining talk at a CCC event a few years back which looked at DB reliability and accuracy (in German): ▶ this video
For context, David Kriesel gave the infamous talk called “BahnMining” at 36C3 highlighting this. IIUC it’s only available in German: ▶ this video
Quite some years ago there was some data backed evidence of this I would say. link Here is the link to his talk directly ▶ this video Its true that the network is quite dense and used by also cargo trains, but there is no denying that things got worse and worse. I constantly experience delay to do some stuff not working. I forgot even the minute threshold when a train is still punctual according to DB. I believe it's 10min by now, which can be deadly if you need to switch trains :)
8 years, wow. Try 30 and actually growing up in this country. School is letting people down, rent prices are exploding. Public transportation is a complete joke ( ▶ this video ) Sorry, but with how rich Germany is I can and should expect more from this country.
> Here are the official statistics: link ... A few years ago David Kriesel did some independent data mining on the reported delays. One point that came up is that official statistics do not even cover train stops that where cancelled and I think it was also mentioned that cancelling stops was one of the ways used to make up for lost time if needed to avoid knock-on effects. German presentation: ▶ this video
This kind of hits the core. As the DB (Deutsche Bahn) is a private company now (although they are 100% owned by the state), they have to make a profit. How do you make a profit? Either raise prices or lower costs. Doing both is the best and that's what they do. Raising prices is easy, it just gets more and more expensive, even the "cheap" Deutschlandticket is too expensive, experts say it should probably be around 29 Euro/month to make a real change. How do you save costs? You do only the bare minimum of maintenance, don't invest in building more tracks and instead reduce tracks and track switches. That's what they've done the last 20 years or so. [0] This leads to ICEs being stuck behind slower trains as there are simply no more tracks to switch to and pass cargo trains etc. So, if some train gets stuck somewhere or has a defect, every other train is stuck behind that for some time. Now, as the state subsidizes some parts of Deutsche Bahn, building something new is also cheaper than maintaining what is there. So they let some things rot down so far that they have to rebuild it completely. As the subsidization is then higher, that's better for the turnaround numbers. There have been thoughts about splitting off building and maintenance of the tracks completely of the rest but that didn't work so far. In that scenario the state would take care of the infrastructure and DB would just "rent" it to drive on it. To add something else to it: I think David Kriesel has it in his talk