In order to master the football prediction, we would need to learn a bit of mathematics, to be more precise, the Poisson Distribution. It is a mathematical concept that helps to outline the possible outcomes converting the historical mean data. To make it less complicated, let’s use a simple example, assume that team A scores 1.7 goals per match. By putting this data into the distribution formula we will get the following results:
- Team A scores 0 goals 18.3% of the time
- Team A scores 1 goal 31% of the time
- Team A scores 2 goals 26.4% of the time
- Team A scores 3 goals 15% of the time
How does it work? Let’s find out.
First things first
Before you can predict the score you have to make a couple of additional steps. First of all, we have to find out the average number of goals each team can score. For that, we would need to “Attack Strength” and “Defence Strength” data.
This data plays a vital role in our calculations. If our data range would be too long the results won’t satisfying, and if too short, it could happen that some data would be beyond our designated rage. You can get this data from the seasonal statistics of any football league.
To calculate this part we would need to find the average of home and away goals scored by each team.
This is very easy to do. You just have to divide the total amount of goals by the number of games played.
Season total goals scored at home / number of games (in season)
Season total goals scored away / number of games (in season)
For the example purposes, let’s use the 2015/2016 English Premier League data. By putting all the necessary information into the simple formula, we would get:
Average number of goals scored at home: 1.492
Average number of goals scored away: 1.207
Attack Strength is the ration between the league’s and the team’s averages.
Good news, you don’t have to do much more calculations here, as they have already been done. All you need to do it reverse the previous results because the goals a home team scores are conceded by the away team and vice versa. Therefore, our results will look the following way:
Average number of goals conceded at home: 1.207
Average number of goals conceded away from home: 1.492
Once again, the Defence Strength is the ratio between the league’s and the team’s averages.
Once you gathered all the above information, you can try to predict the goals. Just apply these formulas following these easy steps:
Step 1 - You have to take the total home goals scored by a team and divide it by the total number of home games.
Step 2 - Divide the answer you get by the league’s average home goals divided by the total home games.
Team’s home goals - THG
Team’s home matches - THM
League’s home goals - LHG
League’s home matches - LHM
(THG/THM) / (LHG/LHM) = Attack Strength
Step 3 - You have to take the total goals conceded by a team while away and divide it by the total number of away games.
Step 4 - Divide the answer you get by the league’s average goals conceded while away divided by the total away games.
Team’s conceded goals - TCG
Team’s away matches - TAM
League’s conceded goals - LCG
League’s away matches - LAM
(TCG/TAM) / (LCG/LAM) = Defense Strength
Step 5 - Now all you left to do is to calculate the possible number of goals. To do so, just multiply the home team’s Attack Strength by the away team’s Defense Strength and by the league’s average number of home goals.
Attack Strength - AS
Defense Strength - DS
Home Team - HT
Away Team - AT
HTAS x ATDS x LHG = Possible home team goals
ATAS x HTDS x LHG = Possible away team goals
How does it help us?
If you have done all the above calculations, you are all set to predict the score. Of course, you can go ahead and utilize the formula created by the French mathematician Simeon Denis Poisson but since we’re kind of advanced on technology, you can use one of the Poisson Distribution Calculators online. It will do most of the work for you in a matter of seconds.
Poisson Distribution formula:
P(x; μ) = (e-μ) (μx) / x!
Just input the goal occurrences and team’s scoring probability and you all get the chances of each possible case. Let’s show you an example.
Assume that Team A’s scoring probability is 1.623 and Team B’s - 0.824. Looking at the matchup between two, we are interested in 0-5 goals for each one. Using the aforementioned tools, we are going to get the following results.
As you can see the most probable outcome is going to be Team A - 1, Team B - 0. You can also multiply the possibilities of each score together and get the chances of that particular score (1-0). In our case, it will be 0.3202 x 0.4386 = 0.1404 or 14.04%.
Congratulations, now you know how to calculate the score lines. Compare your results with the odds offered by the top online bookmakers and see if you can find anything you can take advantage of. Adds some personal knowledge to it and no bookmaker would be able to stop you.
A couple of things before you can go, it is not enough to know just the most probable outcome. What about the draw scenario? Don’t worry we got you covered here as well. Instead of finding the chances of each draw separate, we can calculate the possibility of a draw occurring regardless of a score.
Of course, by doing so we are exposing ourselves to the infinite amount of options but how often have you seen a football match that finished at 6-6 or 10-10? We bet you won’t even be able to come up with one right now. The chances of such scores happening are close to zero, therefore we can easily neglect them.
In order to the chance of the draw happening between Team A and Team B, we need to calculate each draw scoreline chances separate and add them all together. By using the data from the table above, we would get the following result, 0.2472 or 24.72%.
Pro tip: convert this into odds and compare them to the ones your bookmaker has to offer. In our case, the draw odds in such a matchup would be 4.05 (61/20). Use this advice, to find value.
Poisson Distribution is a very handy tool. Unfortunately, it is a simple prediction model, which does not work for many factors. Such things like game status, locker-room environment, coach influence on the player, etc. are beyond the formula’s reach. It fails to factor in the physical condition of the player as well as home-field advantage.
The formula is not designed for all of that. Though, it finds its use in lower leagues, where it is easier to gain an edge over a sportsbook. You will struggle to beat the bookie by using just this distribution for major leagues as bookmakers have more advanced tools and resources at their disposal.
Remember, the odds you find via this method also do not include the betting margin, which has a huge impact on finding value. We suggest using this technique as a great add-on to your assessment and analysis arsenal. Happy betting!