Here is a short explanation and it is not scientific as I created this a few years ago in about 4 hours time.
There is a total of 100 points possible each week.
50 points for the score. Basically, you lose .5 for each number away from the score. Let's say you predicted the Redskins to have 30 and the Saints 24 and the actual result is Redskins 31 to Saints 24, then your score total would be 49.5. If you miss badly on this either way, it could be a large hole in your score.
25 for predicting the winner. 0 for getting it wrong.
12.5 each for the tie breakers and these work similarly to the score but use smaller increments.
There were a few players that hit the score exactly last year but I don't believe anyone has registered the perfect 100 score.
No one has ever seen Arthur Blank and Snidely Whiplash in the same room at the same time.
I'll second the "Just Beat Bill" theme this year. I also use the KCBaghead method of picking scores (he used the same one for years in the old contest) which is surprisingly close many times; I just have to remember to play every week. The one week I forget and the Saints win big, I stop playing. Don't want to jinx it from there on.
Well, I'm sure that someone this week predicted with their head rather than their heart. I wanted to do that... by my heart wanted this game too badly to let myself pick the Redskins.
This "Us Against The World" stuff doesn't work in the NFL. You have to 'bring it' every week on offense, defense, and special teams. Right now we have taken a few steps backwards from last year. We have a tough road this year.
I will be adjusting my future predictions accordingly.