I see where Hare’s going with his idea but I have a couple of reservations.
As I said in my second comment the purpose here is “to build up an accurate picture of how much real, on-track overtaking goes on in a race.”
So what we’re looking for is what Hare calls a “Type C” pass – ‘proper’ on-track overtaking.
I think defining four different types of pass and attempting to count all those up is going to make this a lot more complicated.
At the very least, let’s start off with the more conservative aim of identifying all the ‘Type C’ passes in each race.
Then, if that’s proving very quick and easy, we can expand the analysis and include the many other forms of position change that Hare’s idea would include.
Based on your responses I’d suggest defining a ‘pass’ as being a change of position which:
* Does not take place on the first lap
* Does not happen because one of the cars involved has pitted
* Does not happen because one of the cars has gone off the track
* Is not immediately followed by a ‘re-pass’
I think we should filter out first lap passes because the standing start means everyone has a better chance of overtaking the car in front. By the end of the first lap the field has generally strung out and that’s when we find out if ‘real overtaking’ is possible.
My proposal is we start off by analysing and deciding what counts as a real pass by these terms. Once we’ve got that process running smoothly we can look at expanding it along the lines Hare suggests.