Model

[Professor1215]

I have a question about how to develop a model.

I have a certain number of "factors" that I believe tell a pretty accurate story as to the outcome of an event. I have done a regression analysis on these numbers and feel pretty confident about them.

My only problem with this is, I still do not understand how those factors can combine to get my prediction.

I am not going to guess and say the "magic formula" is (X1+X2*X3/X4)+(X5....), but every time I try using the factors to get a prediction, I get some whack numbers.

I have always hesitated from even asking this question and I am not even sure that I am asking it right, but I am just becoming extremely frustrated. Again, I do not mean to sound ignorant, but I feel I am out of options as to which direction to turn next.

Is there anywhere I can go to get an example of how a game is handicapped, theoretically? I do not want another capper to share his secrets or stats, I just need to know how it is done using a specific set of factors.

Thanks to anyone willing to reach out

[Professor1215]

Couple of things from Advanced Search

FROM CURIOUS:

"Creating a predictive model requires determing which variables are important and assigning them the proper weighting. Some of these variables are objective. Some of them are subjective. Predictive power is not directly related to the complexity of the model or the number of variables. Personally, I think there is in inverse correlation to the complexity of the model and the number of variables with prediction accuracy. I have seen models that used a dizzying array of variables, and they were not very accurate. And I have seen models that use 3 variables and are accurate."



FROM DATA:

"I think that developing profitable model in general takes a few steps:
1) acquiring historical data
2) understanding of scores distributions (Poisson for soccer and hockey, Gaussian for basketball, etc)
3) understanding of the statistical research that has already been done (SABREmetrics for baseball, APBRmetrics for basketball)
4) using and refining methods and models from 3) with offensive and defensive efficiency being the cornerstones.
5) formalizing you proprietary angles and including them in the models from 4).
6) testing with data gathered at 1) and were not use in 5).

This is not a Holy Grail but something that would help me two years ago and saved me some time. Come on guys, let's share."

[pouroupoupou]

So, you have the variables and you want to find the exact weight of each variable and the relation between them ?
Why don't you use a neural network to do that for you?

[Tommy_de1st]

I think the weight of each factor and the corelation between them is different in every sport and league.
Im talking about the soccer here. Some leagues are typical home leagues, meaning home teams have more chances to win the game (even if they are lower in the table) coz of menthality of guest who will be happy with a draw. In other leagues historical data is irrelevant.
This is just an example but IMO you need good knowledge of every league or couple of good native tipsters lol