To break it down in to somthing you will understand...they have taken what has happend in the past and want to apply it to the future....such as we know 30% of favs win they add a value you to this fact and that becomes a neural but with in this fact there might be 20 diffirent reasons why that horse as fav won so they compared all the reasons....it is a little complicated...it is nothing new it is just possible with the speed and information that computers can produce that they are able to do this....this is kinda the way i do my ratings....and like i have said in the past s more information better...i don't think so...because you can't factor in the training that the horse has had or the feed it is eating or maybe the raw guts it has....like never give in eaven if the odds are against you....i am not saying they are no good....they are just a different way to sort the race form out.