Prediction Results!

Fooch your ability to predict scores and outcomes of games is similar to the passing ability of Alex Smith.  The potential is there but in the end are his hands just too small?

On achieving a low score in an equation that measures your accuracy on predicting both total pointage for a game and point differential for a game.

The idea is that these are the two main factors that determine the total math of a single game's score (basically, the sum of the scores and the difference).

The equation works this way: You add the total score for each team in your prediction and find the difference between that total and the total score of the real game.

For instance, in tonight's game I guessed a 20-23 outcome for the Niners. The final score was 17-20. My predicted total points were 20 + 23 = 43 and the actual total points were 17 + 20 = 37.

The first part of the equation measures how close I was in this aspect of the game. So, in this instance, I was 6 off (43 - 37 = 6). The closer your prediction here, the closer you get to zero.

The second part of the equation measures your accuracy for point differential for the game. In my prediction the differential was 23 - 20 = 0, and the actual PD was 20-17 = 0. So, in this case, I was 0 off (0 - 0 = 0). Obviously, the closer to zero, the better.

Then you take how far off you were in each category, and add them. For me this would be 6 + 0 = 6. So my total "score" for the game is six. Naturally, a perfect prediction here would score 0 points, for the win.

What this does is it prevents somebody from winning the game today with a 0-3 prediction (which would score a zero for PD, but a 34 for total points) or with a score of 2-32 (which would score a 0 for total points, but a 27 for PD).

It essentially forces you to maximize your accuracy in both areas, getting closer to a perfect prediction as your "score" goes lower.

From there, we can get our rankings on who won for the games. The prediction with the lowest score who predicted the correct winner wins the game and receives 3 points, unless they were perfect - which would net 10 points. 2 points go to second, and 1 for third. To differentiate this scoring in the lexicon, I will call it a "rankings score."

Every week, Fooch will select up to six games for the community to predict. Each game will be awarded points for first through third, with ties. Ranking Scores accumulate week to week, so by the end of the year we will have our Top Predictor.

This indiidual will be rewarded by Fooch. My memory tells me that he promised a Full Front Page Article to the winner for the regular season, but it is certainly a prize awarded at his discretion.

We're planning on having two of these during the year: a regular season on (which we're doing now) and a post-season one (in which every game of the playoffs will be predicted).

Mostly, it's just a fun community competition. Scoring is pretty easy thanks to MS Excel, so it shouldn't become a hassle.

Please, enjoy the game.

I was 3/6 today, and just got very lucky on two of them.

It just goes to prove: it's good to be lucky, and it's lucky to be good, but it would really kind of kick ass to be both.

is not to predict the winner in and of itself, but the winner and the score. We use an equation to determine whose predictions were empirically closest to the actual score, and then we award individual points based on that. Even if my explanation of the math doesn't make sense, I hope this clears up the basic workings.