Reggae-Boyz said:
Scott,
How would you breakdown the game tonight (Louisville/Pitt) using the Sagarin ratings? I am trying to understand how to read this Sagarin rating stuff. I am a little confused as to which numbers I should be using. Thanks in advance for you help.
The #'s you use is up to you. Sagarin himself says to use his predictor model to determine future outcomes. He has 2 other ratings called Power Ratings (one I use) and Elo Chess (the one BCS uses)
*
In ELO-CHESS, only winning and losing matters; the score margin is of no consequence, which makes it very "politically correct".
*In Predictor the score margin is the only thing that matters.
*Sagarin Power Rating is a combination of ELO-CHESS and PREDICTOR.
I use the combination because IMO the predictor favors teams with high powered offenses or teams that run up the score. I like having the combination of the 2.
So what I did was take the Sagarin Power Ratings and put parameters on it. In essence, I look for false favorites. Teams who Vegas has favored but Sagarin has the underdog winning outright. On top of that, it has to be a difference of 5 or more pts. For example, if Vegas has Team A -3 and Sagarin has team B -3, I make team B a play. (difference of 6pts) So tonights game would not be a play in my system using Sagarin Ratings. I also do not factor in pts for home field advantage. You could argue my paramater of the Sagarin line having to be 5pts or more from the Vegas line covers home field advantage "if" my team is the road team. If you designate 3pts to home team, worst case scenario I still have 2pt advantage over Vegas line.
Lastly, this is not a proven system but has produced 4 straight winners and 3/4 games were outright upsets! So far so good but this is more of an experiment for me but it makes logical sense. If you look at my original post, you will see exactly how I calculated it and came up with the #'s. Sagarin Ratings are located at
www.usatoday.com which Madjack has links too in his handicappers guide.
