crs1026
Superstar
The methodology used above has some obvious shortcomings (like using distance-by-rail rather than Euclidean distance), but while it's time for me to go to bed, I'll curiously check tomorrow what reactions and reflections this quick-and-dirty table will provoke...
Interesting. I presume the formula isn’t backed by any documented validation, it's just somebody's intuitive attempt at a formula?
I won’t nitpick the model - we could spend a year debating that. But taking the numbers at face value, it's certainly interesting as a discussion point.
The big takeaway for me is the huge difference in order of magnitude between the various routes.... sure, we *could* run a train Moncton-Campbellton, but is that the best place to spend investment dollars? At the other extreme, every day spent waiting to begin upgrading Montreal-Ottawa is a huge opportunity cost. Ottawa can afford that investment without any elaborate resort to investment banks, etc. so why not just do it?
The one shortcoming I see that could be addressed just by changing the presentation is how the model treats each station pair as a discrete corridor. So Toronto-Windsor stands alone separate from Toronto-London, whereas a business case for investment on that corridor would want to capture both riderships. Similarly, I have always believed that if we ever see Calgary-Edmonton investment, there should also be service (of some lesser frequency, to be sure) to Lethbridge because Edmonton-Lethbridge will have some revenue potential that isn't captured by the E-C number.
Comparing the two Montreal-Toronto options (via Ottawa, or direct via Cornwall) is interesting - same numerator, but denominators favour the direct route. But how significant is the difference?
This feels like you've opened the bag of potato chips, we've taken just a couple, and sealed the bag and put it back in the cupboard. That never lasts for long.
- Paul
Last edited: