How do you use advanced statistics to outsmart your competition in football predictions? This question is straightforward and we will be answered in this article.
In Football Prediction, everybody wants to be right. And given the increase in attention and money, the difference between a shot taken outside the post vs one inside the post can be as big as £200 m. This is why you should analyze every single turn taken, run, pass, and shot in every football match.
Football was classified in the past as an ‘immeasurable sport’. But data analytics and statistics gained more traction over the years, even in the conservative world of football.
Statistics and data analytics in football offer metrics that can be used for a variety of things like player scouting, opponent analysis, performance analysis, and injury prevention/fitness.
Traditional football metrics shown during and around football games you’re probably familiar with are number of shots, shots on goal, corners, and team possession. However, many other metrics have gained more in-depth insights.
This post will discuss those advanced statistics that you can use to analyze football and of course outsmart your competition in football predictions.
See: Using Statistics For More Accurate Football Predictions
Article Content:
- 20 advanced statistics for outsmarting your competition in football predictions
- How to use advanced statistics to outsmart your competition in football predictions
20 Advanced Statistics For Outsmarting Your Competition In Football Predictions
1. Expected Goals (xG)
2. Passes per Defensive Action (PPDA)
3. Possession Adjusted Passing Accuracy (PPA)
4. Dribbles per Game
5. Shots per 90 minutes
6. Tackles per 90 minutes
7. Key passes per 90 minutes
8. Total shots ratio
9. PDO
10. Game states
11. Expected assists
12. Expected threat
13. Possession ratings metrics
14. Expected possession goals
15. Possession value and possession value added
16. Expected goal chains
17. Defensive coverage
18. Shots on target
19. Deep
20. Non-shot based xG
Related: Analyzing Offensive Versus Defensive Stats For Improved Football Predictions
How To Use Advanced Statistics To Outsmart Your Competition In Football Predictions
There are lots of advanced statistics you can use to analyze football performance and in turn predict matches.
We have listed them out here and how you can use them to outsmart other bettors in football predictions.
1. Expected Goals (xG)
Expected goals or xG is used to measure the quality of a team’s chances and predicts the number of goals a team should score based on the chances they created.
This statistics is the industry standard for predicting football matches. Here, rather than count shots, give every shot value by reflecting the possibility that it will be a goal.
Look at the expected goals taken and conceded. This does a good job of predicting future outcomes. To build team ratings in attack and defense based on some type of expected goals is best practice.
For shots focus on details like where the shot was taken from, the distance and angle. The part of the body the player took the shot with. Headers realize less goals than kicked shots do. Also if the shot was from open play or set piece.
Smart xG models use factors like speed of attack, the kind of pass that led to a shot and if the player dribbled another player before a shot to get information about the surrounding situation.
Not every xG model is created the same, and the best ones integrate all the data they can find. In terms of predictive modeling nothing tops xG.
2. Passes Per Defensive Action (PPDA)
Passes per Defensive Action (PPDA) measures the number of passes a team allows before they take defensive action, and it can be used to assess a team’s pressing and defensive performance.
PPDA is a great metric that decides a team’s pressing intensity. The higher a team’s pressing the lower the PPDA metric as the opposition may not get the chance to play much passes without facing defence actions.
Similarly, the PPDA value of a team is high, when they play a low block and do not pressure the opponent actively.
3. Possession Adjusted Passing Accuracy (PPA)
Possession Adjusting Passing Accuracy (PPA) is an advanced statistics that adjusts the passing accuracy of a team for the amount of possesion they have and can be used to evaluate their passing efficiency.
4. Dribbles per Game
Dribbles per Game is another advanced statistics, it measures the number of successful dribbles a player makes per game and is used to evaluate the ball control of a player and ability to beat opponents.
5. Shots per 90 Minutes
Shots per 90 minutes is an advanced statistics that measures the number of shots players take per 90 minutes and can be used to evaluate a player’s offensive productivity.
See also: Football Prediction And Statistics – A Partnership With Incredible Results
6. Tackles per 90 Minutes
Tackles per 90 minutes measures the number of tackles a player makes per 90 minutes and can be used to evaluate the defensive productivity of a player.
7. Key Passes per 90 Minutes
Key Passes per 90 minutes is an advanced statistics that guages the number of passes that lead to a shot, and can be used to assess a player’s offensive creativity.
8. Total Shots Ratio
Total Shots Ratio is an advanced statistics that measures a team’s ability to take more consistent shots than their opposition.
TSR is a good metric to use during your Football Prediction to understand the extent of a particular team’s strength.
This statistic doesn’t just relate well with the number of points a football club scores in a season but it is an metric you can repeat as much as you want.
9. PDO
PDO is an advanced statistics that shows a team’s shot and save percentage. It differs over a short period, which makes it a good method to measure a team’s relative fortune or misfortune.
It is the total of a team’s shooting percentage (shots / shots on target) and its save percentage (saves / shots on target against). It gives every shot an equal opportunity of being scored – no matter the location, the player who shoots it, or the identity or position of the goalkeeper and defenders.
You can PDO to know which teams are over performing or underperforming based on metrics like TSR. You can use this metric to examine the last 10 games of a struggling team, calculate their PDO, and compare their TSR to know if the team may be suffering an unlucky streak.
10. Game States
Game States are a set of effects on a team’s behavior when they are tied, a goal ahead, or a goal behind. It explains why some metrics don’t give the whole picture. For instance, a team may have a low TSR but still win the league with a good points total.
11. Expected Assists
Expected Assists is an advanced statistics that measures the likelihood of a pass becoming an assist. It considers factors like the finishing location of the pass, the type of pass, and other factors too. It doesn’t care if a shot was taken from a pass and credits all passes, whether they become a shot.
You can use xA to assess a player’s ability to break matches open with precise passes and create scoring opportunities. Other simple metrics like assists and key passes exist. But, with xA you don’t rely on the finisher’s ability and can differentiate in quality of key passes.
12. Expected Threat
Expected Threat is the more comprehensive and complicated version of Expected Assists. xT quanitifies passes and actions that move the ball into dangerous areas.
Just like xA, this framework rewards action regardless of the outcome of the possession.
However, the difference is that xT doesn’t just provide the opportunities for a high Expected Goals scoring, it rewards players for taking the ball into ‘threatening positions’ too, which leads to high-Expected Goals shooting position.
13. Possession Ratings Metrics
Possession Ratings Metrics allows you quantify the contributions of more defensive players. This is opposed to other advanced statistics like xG, xA, and xT that gives you information about attackers and attacking midfielders.
14. Expected Possession Goals
Expected Possession Goals answers the question ‘What is the value of a possession within a match?’ xPG builds on xG by calculating the Expecte Goals for every football event like pass, dribble, or shot on specific action coordiants as if there had been a shot.
15. Possession Value And Possession Value Added
Possession Value and Possession Value Added are the same as xPG because they rank each action on the ball before and after the probability that a goal will be scored. After the difference is calculated, you’ll find the Possession Value Added.
Read: Uncovering Unexpected Insights For Football Matchups For Predictions
16. Expected Goal Chains
Expected Goal Chains or xGC is another advanced statistics that recognize players for the attacking contributions they make besides shots and assists.
xGChain takes the xG and gives to players who played a role in creating the shot that xG measures. It corrects one of the major flaws of xG as an advanced statistics that is helpful only on a player level for the players who take the shots.
For every shot, xGC adds the Expected Goals of that shot to the players who assisted in the momentum leading to it.
17. Defensive Coverage
Defensive Coverage is another defensive advanced statistics. This one measures the area of defensive responsibility caused by a player’s defensive actions during a match.
The corresponding output of Defensive Coverage includes a series of coordinates that define a polygon of the defensive zone of a player, including the area of that zone.
18. Shots / Shots On Target
Shots and Shots On Target are perfect for getting a deeper understanding about the quality of chances.
For example, if two clubs had the same number of goals or XG. May be they score on average 1 xG per match. One team takes 10 shots on average and the other just 5. The second team has the capacity to create more dangerous scoring chances, because they create 0.2 xG per shot and the other one only 0.1xg per shot.
19. Deep
Deep measures the number of completed passes within 20 yards to the goals. This is every pass in and around the penalty area. It is a good advanced statistics that shows how dominantly a team played.
The closer successful passes are to the opposition’s goal, the higher the chance of that pass causing a shot. Possession isn’t a very good indicator of how dangerous at team is, but a possession in and around the 18-yard-box is dangerous.
20. Non-shot Based xG
One problem with xG models is that they are built solely on shots, making them ignorant of possessions that end with incomplete passes, turnovers, or any other actions that happen in run of play.
xG works well despite those actions, but it is an area you may want to use other modeling solutions to bring more information for your Football Prediction.
There are xG models that augment their core information by including other factors that decide if a team is better or worse at turning possession into shots that end in goals.
Here you would want to find teams that put up higher or lower xG numbers than their other statistics show they should and predict correctly how those clubs will perform.
This primarily involves modeling passing and the volume and type of passing that clubs do in the final third of the field, and then predicting how often that pass should bring shots against how often it has been ending in shots.
This now informs the modeling of the values of shots the club took; it is an additional layer of complexity.
An xG model predicts the number of goals a football club ”should” have scored from the shots it took. A non-shot based xG model on the other hand, predicts how many goals a club should have scored from the shots it took and the pass it made.
There are a vast range of advanced statistics that make football easier and more analyzable. As data analytics in football become widely adopted, these statistics will be updated and newer frameworks developed.
Related Articles:
5 Tips On Winning A Football Prediction Contest
How To Develop A Winning Strategy For Football Predictions
Analyzing Football Betting Trends To Develop Winning Predictions
