Looks cool, but I can't even try this -- the search won't load any results for me.

I discovered this a few days ago and shared it with some friends. It worked quite well for the few routes I tried.

It provides percentage of chance of delay and average delay for specific routes. Each time slot has different stats. So for example the 13:00 train from Copenhagen to Hamburg is much less reliable than the 17:00 one (random example, not sure if accurate).

Very cool overall.

I would be so curious how they got to the data!

Is data coming from country-wide apps? For example my SBB app (Switzerland) shows this. Annoying when you want to take a train in switzerland, look it up in the app and find out it's a French or German train - you know that it'll be late and unclean. And the app will confirm the lateness.

Got this error for a Uk->UK station search:

  Uh oh!

  We couldn’t find what you were looking for. Try searching with some different criteria.

  Error message: e is not iterable