First of all, this is not a critical topic....more of an observation.
The Haskell writeup gives a cogent insight into TG's handling of 'ground loss' as an important handicapping factor. It is very consistent with the approach I have used for years, with some success.
However, the terminology is really offkey from the scientific point-of-view. What happens to a horse on a turn is not related to the extra ground it travels as much as the momentum a horse loses when it is further and further away from the rail. If you can get this into your mind's eye think about the fact that EVERY stride on a turn involves a change in direction. Horses have to turn left continuously with every stride on the turn, which takes more energy than running in a straight line. Every stride must find a new direction.
Something similar, but less affecting, happens when a horse switches lanes in the straightaways. I try to account for momentum loss in turns as well as in the stretch and backstretch runs (but as I said, it doesn't mean nearly as much in the straightaways).
This is really a small point, but if the loss of ground only involved the extra ground traveled, Trakus would be a lot more important handicapping tool than it is. It's almost a non-factor to me.
To summarize: the loss of momentum equates to an exertion of energy. Running turns is tiring, whether a horse is in front or behind.
