I think it's especially interesting because the data is fictional, based on a few simple rules to simulate horse behavior (that's right - this is just like a single realization of a Monte Carlo process!). Andrew has provided the complete R code to run the simulation on his blog.
It's an excellent first stab at the problem of vizualizing this complex dataset. Countless records are kept on the outcomes of races, but few document the dynamics of the race itself.
If anyone does have intra-race data for the Kentucky Derby, please forward it along.