From here, you can download the predictors which are necessary to implement the equations from my book “MATH to WIN: Football”.
Also, in the second sheet of the file, (NUMBERS talk) you will find the values of the eight factors that affect the outcome of a game. Soon, you can find how to use these factors and many more useful tools in my new book MATH to WIN: NUMBERS can TALK.
You can find a full description of my method along with the England Premier League equations in my article Premier League with optimal equations in Kaggle.
Download the essential information and the predictor values for the games (May 22- May 24)
There are thirteen predictors, four for the home team, four for the guest team, and another five for the whole game.
homeSCFT: Tries to capture the average goals number for the home team in a given game. Its value comes from estimating the home team’s offensive ability in the past games, the guest team’s defensive efficiency, and some other factors depending on the specific division.
homeSCHT: Same as the above but for the first half only.
homeXgoals: Has the same principles as the first one, but I calculate it using the SHOTS and SHOTS on TARGET if provided, trying to measure the efficiency of a team’s offense and defense. You will notice for some divisions has the same value as the homeSCFT; this happens when the SHOTS and SHOTS on TARGET data are not available.
homeSCprob: Combining all available data, I try to predict the team’s percentage to score at least one goal in the game.
The predictors awaySCFT, awaySCHT, awayXgoals, and awaySCprob, are the same predictors trying to measure the guest team’s efficiency.
tGoalsFT, goalDifFT: Expected total goals for the game and expected goals difference based on the predictors homeSCFT, awaySCFT. If the value of the goalDifFT is negative, the guest team has a better chance to win; if the value is positive, the home team has the advantage.
tXgoals, goalXdif: The same as the above but based on homeXgoals and awayXgoals.
scProbDif: This is the difference of the scoring ability; the negative value favors the guest team and the positive the home team. Surprisingly (or not) is the predictor with the most impressive results.
You can find the optimal equations for fourteen divisions in my book, plus extra informative tables for each division and tens or real-life examples to show how you can put this method to action.