This Week’s Bets

Posted by mrformula on October 26, 2008

Record, year to date: 10-8-1

Houston -9.5

NE -7

Pitt -3

Bal -7

TB +1.5

Was -7.5

Also, Tenn is -4 right now, if it moves to -4.5 by tomorrow night, we will bet that too.


2 Responses to “This Week’s Bets”

  1. jonnychu said

    i saw the Tenn/Indy game was at -4 and you guys wanted to bet it at -1.5 or better or -4.5 to 6. On one hand I understand the idea of the ranges and for whatever reason the formula wasn’t interested in taking Tenn from -1.5 to -4.5 but on the common sense side of the equation if you’d take -4.5 why wouldn’t you take -4? So I did and it paid off.

    I’ll probably take those chances when I see em and maybe by the end of the season there will be a sample size significant enough to see if this matters.

  2. mrformula said

    That is a really good question, and to be honest, I’m not sure I’m entirely convinced that you shouldn’t bet it. Common sense certainly says to bet that situation. All I use as an argument is that we have 7 and a half years of data that suggests you shouldn’t bet outside of the ranges. These ranges, where we use a small range (-4.5 to -6 in this case) and then an absolute number to bet anything better than, are from a second model that we just discovered early in the season. To give you the breakdown over the last 7.5 seasons (excluding favorites of more then 10 points, which we never bet):

    When the spread doesn’t reach any part of the range (in this case, Tenn -6.5 or worse): 48.1% (in 736 games)
    When the spread is in the small range (in this case, Tenn -4.5 to -6): 68.6% (in 103 games)
    When the spread is in between the ranges: (in this case, Tenn -4 to -1.5): 49.0% (in 98 games)
    When the spread is at least as good as the absolute number (in this case, Tenn -1 or better): 61.7% (in 48 games)

    So following the data strictly, it isn’t profitable to bet outside the range. The 49% when in between ranges gets even worse when the doesn’t qualify in the other model’s range (44.7%). So if neither model qualifies, we will not bet the game. I just wish I could come up with a good ‘real world’ explanation of why the games in between the ranges are not good plays.

    Having said all that, we are still in the trial phase of using these models to bet games, so if there is a game in between ranges, and you really like it, then by all means you should bet it. We definitely think that there is room for improvement with these models, and any feedback we get we will certainly take into account. While our sample is for the game that fall in between the betting ranges is fairly large at 98 games, I would not say that it is large enough to make any definite conclusions, so keep betting the games that you like and we will follow how these games do for the next few years until we have a large enough sample size to make a definite conclusion.

    I wish I could give a better explanation, but I hope this makes at least some sense and helps you understand why we don’t bet those games.

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: