An Update To The Formula

Posted by deck104 on October 16, 2008

In an effort to make our betting model better and more innovative, the Wizard has been trying to tweak some stuff. Because we’re actually investing a lot of time and money into the model this year (where in previous years we would talk about it on and off, but it wouldn’t dominate our time like it does now), we’ve been able to actually see how things work on a week to week basis and make some observations. For example, I’ve postulated in both weeks about why the model would have us take a team giving nearly a touchdown to the underdog, but when the line is that same team getting points, we wouldn’t take them. In addition to that, the Wizard really didn’t like the ways some of the components of the model were weighted and wanted to play around with parameters and see what he could come up with.

The results were oustanding. Where our original model won 119 of 186 (64%) games over the last five seasons (between weeks 5 and 14 only), the new model picked far less games, but won 67 of the 91 games picked for a 73.6 win percentage (the new model picks games most successfully between weeks 6 and 15). Not only that, but the new model provides a much smaller spread range in addition to a minimum betting range. So for example we might take a team at -5 to -7.5 but also at anything better than +1 (so we wouldnt bet between an EVEN pick and -4.5 but within our range or our team getting at least 1).

The Wizard thinks that Mr. Formula 2.0 working like this could take into account things like major injuries to major players. Not in the way it outputs the model, but in the way that it eliminates wonky spreads that are better than our range, but not so good that the percentage signifies a payoff. So in the above example, Mr. Formula would love us taking a team giving between 5 and 7.5 but anything better than that could signify there’s something going on that can’t be explained mathematically, and then once that team starts getting a full point, screw what linemakers think– this team can win anyway.

With all that explained, things get even more out of hand. In games that BOTH models picked over the last five years the formula has won 80% of it’s games. Yes. 80%. Now it’s only picked 30, but it’s 24 for 30. Thats 6 games a year, winning 4.8 of them. So if you bet $1100 to win $1000 on those games, you’d have won $19,800 since 2003. And that’s not including the New Orleans/Oakland game in week 6 that we would’ve given you.

So from now on we’ll be presenting games with a confidence number from 1-3. The games with a confidence of 1 will be the games the original model outputs. This model gives us the most action, about 40-45 games per year, and we recommend betting all of these if you like making bets. If you’re going to bet games from this model we recommend betting every game we post because it’s designed to give you a profit over the course of a season and are not by any means considered “locks.” We will assign a confidence of 2 to games the second model picks. Again, the success rate of this model is higher than the first model but it produces about half the games the first model does. A confidence of 3 will be given to the games that both models pick.

Most people that I’ve talked to about this project ask me what the big money games are to get rich quick. I remind everyone that the goal of the model is to bet what your comfortable with but bet all the games if you want to win money. People can be underwhelmed by that because too many people look at gambling as a get-rich-quick type thing. If you want to put big money on any of the games, clearly I recommend those with the confidence 3 label. When the Wizard told me about the 80% thing, 24-6, that’s just pretty high.

Anyway, we hope you all like this new way of doing things, hopefully it makes us all a little richer.


One Response to “An Update To The Formula”

  1. […] Comments (RSS) « An Update To The Formula […]

Leave a Reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: