A couple of casual thoughts to get the ball rolling.
1) This is really cool work. I did note that at this stage the key/only metric involved seems to be Time, unless my quick look missed something. By itself, isolated, I'm sure quite a few conclusions can be drawn from that data, but even moreso, I think that collation with other key data areas such as passenger numbers, traffic intensity data, even things like weather - and I'm sure, plenty of other things I've not thought of - could greatly bolster the data's analytical potential.
2) This is subjective of course - but something feels off (by which I mean, not an entirely neutral presentation of data) to me, about the presentation's critical or punitive framing : things like the celebratory "leader board" of largest delays, complete with gold medal emoji, or the site's name seeming a sort of play on "where's that late bus?!".
