Back when Caesar was a corporal and I was doing my first NBA projections, I thought it was a snap. The first couple of months, I was really nailing it, and I thought I was really onto something. But then the number gods stepped in and said hold your horses young man, and kicked me in the teeth with something like reversion to the mean; but ever since, I’ve kept punching; revising the algorithm, adding and subtracting variables, changing weights, etc. until I now have achieved perfection!
Well, maybe not – but I’m still trying.
As some of you may know, my NFL model needs 4 full weeks of data to run, so for the first month of the season, I’ll be using year-old numbers (along with a handful of team unit updates), just to get the model started. The algorithm uses team and player stats and team unit matchups for each game. Also, as the NFL runs on injuries (or doesn’t), I’ll be posting several times a week: Tuesdays, after the prior week’s stats are in, Wednesdays after the first injury updates are posted, Saturdays for further updates for Sundays games, and then late Sundays or early Mondays for MNF.
I grade the model’s ATS efficacy by using the last posted MOV (projected margin of victory) and the Closing Vegas lines as reported by VIsN. That means that pick-flips happen; it all depends on those pesky closing lines, but it’s the best way I know to grade the model. It should therefore be apparent that the model’s posted projected ATS winner is NOT a recommendation (I am not a tout), but more like a suggestion, or a guide to your own judgment.
Now, as perfect as I and the model are, sometimes my staff can make mistakes, so don’t hesitate to point out errors or bugs when you see them.
And, I hope you all have a good season.
WJCJR
PS. First projections tomorrow.
Well, maybe not – but I’m still trying.
As some of you may know, my NFL model needs 4 full weeks of data to run, so for the first month of the season, I’ll be using year-old numbers (along with a handful of team unit updates), just to get the model started. The algorithm uses team and player stats and team unit matchups for each game. Also, as the NFL runs on injuries (or doesn’t), I’ll be posting several times a week: Tuesdays, after the prior week’s stats are in, Wednesdays after the first injury updates are posted, Saturdays for further updates for Sundays games, and then late Sundays or early Mondays for MNF.
I grade the model’s ATS efficacy by using the last posted MOV (projected margin of victory) and the Closing Vegas lines as reported by VIsN. That means that pick-flips happen; it all depends on those pesky closing lines, but it’s the best way I know to grade the model. It should therefore be apparent that the model’s posted projected ATS winner is NOT a recommendation (I am not a tout), but more like a suggestion, or a guide to your own judgment.
Now, as perfect as I and the model are, sometimes my staff can make mistakes, so don’t hesitate to point out errors or bugs when you see them.
And, I hope you all have a good season.
WJCJR
PS. First projections tomorrow.
