It is always difficult to compare such figures. Not only because of different geographic/demographic situations, but simply because of the figures themselves.

It is always difficult to compare such figures. Not only because of different geographic/demographic situations, but simply because of the figures themselves. First, there is the problem of being careful understanding which figure is given to you. For example, for Toulouse, he quoted 125.7 M metro+tram trips. But 125.7 M is not the number of trips, but the number of metro+tram "validations" (each time you transfer to another bus or metro line, it counts as an extra validation (yes, in this city, even between metro lines)). The total number of trips (132 M) is 30% lower than the total number of validation (188 M). There is no detailed by mode number of trips, so you cannot get the number of metro+tram trips. I guess the best you can do is apply the 30% discount to get an approximation of 88 M metro+tram trips [and I shouldn't even write that because that depends how we decide to count mixed-mode trips]. Then, there is the problem of deciding if some assumptions which are valid here are still valid there. He has decided to ignore bus network. Still in the case of Toulouse, the bus network usage is a bit bigger than one metro line, but the thing is that there are only 2 metro lines... Very roughly, each of (metro line 1), (metro line 2) and (bus network) is 1/3 of the total passengers. Worse, if you consider the number of kilometres ran, the bus network represents 73%! Now you may argue that most bus passengers transfer from or to the metro. It could be true. But that would only affect the relevance of the figure of trips/inhabitant. However if you want to talk about the quality of the service, the quality of the offer, I don't think you can neglect something that represents 1/3 of the passengers transported and almost 3/4 of the kilometres offer (those figures mean that the metro lines only cover a very small part of the territory, and that's indeed how it is, the bus network is 50 times longer than the metro network). The frequency and spread of the bus offer has to be taken into account, hasn't it? Maybe in other cities the bus network can be neglected, but I don't think it a good idea to neglect it by default. (One could add that in the case of Toulouse, the bus traffic has grown clearly and steadily, while the metro traffic plateaued years ago.) And at this point you fall back on the first difficulty I mentioned: how difficult it is to compare networks which grew differently and developed different structures.