I read the article on his use of a group of ratings and have to say I found it a fascinating approach. It makes more sense to compare a reliable horse which returns a similar rating all the time against a patchy horse with good and bad runs using this method. It uses the good runs and the bad runs instead of an average of those runs to create a set of prices. I haven't used the method yet but intend to try it very soon.
His example using the dice is fascinating too and as soon as I fashion a set of these dice and learn how to use them I think I'll be able to grift a lot more winnings from my friends than I ever make at the TAB. :-)