08-24-2016, 06:35 PM
(08-24-2016, 05:48 PM)Boston Card link Wrote:* This is a bit new this year, and means that someone who picks the correct team, but totally messes up the score (for example, guessing 52-30 when the score ends up being 7-3) still ranks higher than someone who picks the wrong team but is close on the score (for example, guessing 7-10). To figure out how off someone is from the true score, I calculate the difference between the predicted and actual score spreads (so, the aforementioned 52-30 results in a predicted differential of 22 versus an actual differential of 4, for a difference of 18). To that, I add the difference between the predicted and actual mean score (so 52-30 results in a mean score of 41 versus an actual mean of 5, for a difference of 36). That results in a difference in spread plus difference in means of 54.
Sorry to cause trouble, but I don't quite understand how this works.
I get the calculation of a score of 54 for the person who guess 52-30 with the correct winner. But how does this system make sure that someone who picks the wrong team at 7-10 comes in behind the former?
Is it merely that all the people with the wrong winner fall in a separate category that rank behind all the people with the right winner?
