You are here:Football Prediction Site > Football Analysis >
At The End of The Football Season a Team Analysis Mean
More and more contemporary football analysis articles use statistics more frequently, using different indicators to highlight the strengths, weaknesses or patterns of a certain team or player. We are already very familiar with many indicators, while others may have some unpopularities that require a little explanation. Today we are trying to introduce some of the most common football indicators and how to use them. We will divide it into two parts and share the first half today. , We will start now:
1."Per 90" indicator
This is an important starting point when introducing indicators in football analysis. At the player and team level, the simple use of a summary indicator can be misleading because it does not take into account the time it takes to reach this number.
Background is the key to analysis. We need to find a fair way to compare the performance of players or teams, that is, look at these numbers in the context of the entire game (90 minutes per game).
Since the start of the 2019-20 season, Liverpool's Mane has scored 29 goals in the Premier League, while Abraham has only scored 21 goals for Chelsea.
At first glance, Mane seems to be a more convincing player in front of goal, but the additional background is that Mane played more than 2,000 minutes than Braham during this period. When we look at the number of goals scored in 90 minutes per game, we see that Abraham’s goal rate is 0.58 goals per 90 minutes, which is actually higher than Mane’s 0.47 goals per 90 minutes. Given their different roles in the forward line, This makes more sense.
Let's take a closer look at the top scorers list for the 2020/21 season below. It will be very interesting based on an average of 90 minutes per game. If you only consider players who have played more than 900 minutes in the Premier League, the highest scoring rate is Bale on loan from Tottenham. Although he may not have accumulated as much playing time as other players, Bell averaged more than one goal in 90 minutes per game, which is a very high return when he is on the court.
XG is no stranger to many people. Simply put, XG is a way to measure the probability of a shot becoming a goal.
Not all shots are of the same quality. A shot may be a shot from 40 yards away, or it may be a shot within two yards. Therefore, XG will measure the quality of each shot before the player shoots. Many factors need to be considered, including: the distance between the angle of the shot and the goal, whether it is a header, a weak foot or a dominant foot, and whether the shot comes from a cross, penetration, or Whether the short pass is blocked by multiple defenders.
The XG value is a value between 0 (no chance of scoring a goal) and 1 (a certain goal). For example, the following scene is a shot by Fernandez against Southampton, with an XG value of 0.3, which means that considering his situation, this shot has a 3/10 or 30/100 possible goal.
The XG value is calculated by using thousands of previous shots in similar situations and seeing how many of them are scored.
The above example uses a single shot to briefly summarize the meaning of the XG indicator, but if we want to study XG shot by shot, it is often inaccurate. What we have to do is to accumulate the XG of a player or team in a game, a period of time or a season, according to the quality of their shots, to have a clearer understanding of how many goals they should have.
Of course, when we have a larger sample, in a longer period of time, the conclusions drawn from this information are more reliable. When we do this, we can use XG to explore whether a player or team's performance in front of goal is insufficient or excessive compared to their expectations.
On the basis of XG, we can dig deeper into the possibility of a shot becoming a goal. XG provides the value before the player shoots, and xGOT provides the correction value for the player to hit the target after shooting.
The displayed value is the same as XG, and it is also a number between 0 and 1. But it further adds to the context, distinguishing between shots to the upper corner of the goal and shots to the middle of the goal. As you can imagine, this XGOT model only calculates the value of the target, so it depends on the player to at least make the goalkeeper act accordingly-of course, if the shot goes off the target, your chance of scoring is zero (unless there is an impossible Rebound).
Let's take a look at the following example from Kane last season. Before he took a shot from outside the penalty area, his xG was 0.03, which is a very low value opportunity. It is estimated that only 3 of 100 shots will be a goal.
However, the quality of Kane's shot was excellent because it flew into the upper corner. After the shot, the XGOT value of this opportunity now jumps to 0.54, which means that in 54% of the cases, the position of the strike will result in a goal. In the end, this is indeed a high-quality shot.
The main content provided by XGOT is to have a clearer concept of a player's shooting execution. If a player's XGOT value continues to be higher than their xG, this tells us that their shooting percentage is better than the quality of their chances.
It is worth noting that these data only include shots that were on target, not blocked shots. Therefore, there may be some players who think they are unlucky. They executed a perfect shot, but the defender blocked their shot.
Then, we can use Opta’s "shot added value" indicator to calculate the difference between a player’s xG and xGOT. Below, you can see Tottenham’s Sun Xingmin added the most value to his shots last season, and he improved his chances of 3.8 goals from these shooting positions.
On the other hand, Leicester City’s Vardy has the biggest gap between opportunity and execution opportunity. His performance is 3.4 goals worse. This is due to his poor technique in hitting the goal.
Similarly, we can think of XGOT as a good way to measure the performance of goalkeepers. This can more accurately reflect the goalkeeper's performance, because it can predict how many goals the goalkeeper will concede based on the quality of the shots they face.
Then we can use this data to measure the number of "blocked goals". This data compares the actual number of goals conceded by the goalkeeper and the expected number of goals conceded. The higher the number of goals blocked, the better the goalkeeper's interception performance.
Last season, the best blocking shot performance came from Tottenham’s Loris, who blocked 5.1 goals more than expected throughout the season.
4.Non-penalty goals（Non-penalty goals）
This indicator hardly needs to be explained, but the important thing is why the number of goals considered is often evaluated by subtracting penalty goals from the total number of goals. The penalty kick itself does not reflect the player's ability to create opportunities for himself. Of course, you can have a player who can both score goals and make penalties, but usually, the penalty kicker will take the penalty kick even if he is not involved in the moment that caused the foul.
The changes in handball rules and the introduction of VAR also mean that the number of penalties awarded last season has reached a record high, with a total of 125 penalty kicks, an average of one penalty kick in every three games.
Those who take penalty kicks can significantly increase their scoring stats from unsustainable goals. Looking at the top scorers in the Premier League last season, Manchester United’s Fernandez and Leicester City’s Vardy both doubled their number of goals. Without such high-value scoring opportunities, they are far from being ranked in the top. 10 people.
With this in mind, it is more reliable to evaluate the chances of scoring by observing non-penalty goals or non-penalty XG in the game. In this case, it can provide more level playing field to explore the team or player’s performance. Chance to score.
5.Expected assists (xA)
Assists by themselves do not reflect a player’s creativity very well. A player can simply pass the ball to his teammate and hit the ball into the goal, while another player may run a long distance and send the pie to his teammate. , Let his teammates score, these two situations are almost incomparable. In addition, in order to get an assist, the passer needs to rely on the receiving teammate to complete the scoring opportunity, but this does not always happen.
Therefore, another extension of the expected goals framework is expected assists (XA). It works in roughly the same way as XG, but slightly different in giving points to players who pass before hitting.
Simply put, XA measures the expected goal value of an assisted shot, and it is also expressed as a number between 0 (no chance of assist) and 1 (definite assist).
XA can provide credit for creative players and give a clearer concept of how many assists a player should have given the quality of their offensive output. We can also add up all the values to explore whether a player is creating a very valuable opportunity or passing the ball in order not to lose the ball.
Looking back on last season, it is not surprising that Manchester City's De Bruyne became the most creative player. Considering that the Belgian missed quite a few games due to injury, he once again highlighted his ability in the team on the basis of an average of 90 points per game. On average, he can provide more than one assist every three games (average of 90 minutes per game, 0.37XA).
Of course, just as a penalty kick can exaggerate the number of goals scored by a player, being a free kick taker can also increase a player’s creative output, because they have more opportunities to send the ball to a dangerous area without an opponent. , Let more players attack the ball. Therefore, the XA performance of all players in the game last season was slightly different, which is worth seeing. Here, Fernandez is almost ahead of De Bruyne at the top of the list, and players such as Odoi and Maximan have also shown their creativity to pass the ball to find teammates in dangerous positions in the game.