The goal prediction indicator (xG) helps to tell match history better than other shooting stats, as it is understood that two shots do not always provide the same risk to the opponent, such as one made from midfield and one penalty kick. Therefore, each end gets a different value, according to its properties, but mainly taking into account the distance and angle of each of them (see Methodology at the end of the text).
a Goiás He finished the match 14 times and produced enough to score 0.9 goals (near reality, one record). There were eight shots from inside the box and six from outside the box, with a negative focus on the shot from behind by Felipe Bastos’ midfield, resulting in less than 1% of goal conversion expected. Of the more than 84,500 shots that Espião Statístico have scored since the 2013 edition, only 40 have been taken from the field of defense: none of them have converted into a goal.
a Bragantino It has finished 13 times: one from inside the small area (Arthur’s goal), four from inside the big area and another eight from outside the area. Just looking at the kick’s distance, a A shot from inside the box has a 35.5% chance of converting it into a goal (5,079 rounds from inside the small area and 1804 rounds from the net). The probability of a shot from inside the penalty area is 11.4%. (38,995 shots from inside the penalty area and 4,479 shots from the net).
a Sao Paulo He finished 12 times in the match (six from inside the box and six from outside), produced enough to score almost a goal (0.98 xG), was effective and scored two goals. However, he was incompetent in defense: he suffered from six requests from car In the game (one inside the small area, three on the inside of the big area and two on the outside), which generated an expected target of 1.12 xG and ended up getting two targets. Similarly, the same is true for car.
The difference is that among the six shows car In the match, three of them were face to face with Jandry: two turned into a goal and one saved the goalkeeper well.
Each team finished 13 times in the match. a Energy He finished five from inside the box, two from half moon, four from outside the frontal area and two from greater distance and/or angle, producing a prediction of 1.01 xG and scoring one goal. Already youths He scored seven from inside the box and five from outside, producing slightly lower expectations (0.67 xG) largely due to the angle of these shots from inside the box, but he also scored a goal.
Expected target (xG) of saints ahead of expectations Palm trees During most of the match (shown in the lines in the chart below), but then came the penalty kick for Verdau, which was missed by Rafael Vega. a A penalty kick is the type of shot most likely to convert into a goal: 76.4% of the time. Of the 984 penalty kicks scored and collected by Espião Statístico since the 2013 edition of Brasileirão, 752 have converted into a goal.
Curitiba (2.40) 1-0 (1.67) Botafogo
Team Goals Predictions (xG) indicates that we could have seen more goals in the game. a Curitiba He produced enough to score 2.40 goals and score one goal (practically 1.0 xG comes from three consecutive shots in the 17th minute of the second half). Botafogo produced enough to score 1.67 goals and scored none. Glorioso’s best chance was with Erisson from inside the small area in the 18th minute of the first half, when Matthews Alexander saved him.
a Fluminense Produce an expected target (xG) higher than that of flamingo (1.76 xG vs 1.28 xG), but it stopped most of the time on Hugo. It is not always the team that produces more and better chances in the game that wins.
There were 16 rounds of Flu (one from within the small area, seven from within the large area and eight from the outside) against seven rounds of Fla (four from within the large area, two from the half-moon and one from a long distance).
Interesting to note that even the goal America MGHey Corinthians It barely produced chances (see the black line in the chart below). There were only seven shots in the match by Timão and Mosquito’s goal was the most dangerous: from inside a small area without a goalkeeper. On the other hand, Coelho finished 12 goals (half from inside the area and one from outside) and after Aloysio’s goal he practically made no production (see the gray line in the chart below after the goal).
There were nine endings in cuiaba Most of them were out of the zone (six against three from out of the zone), so they produced low goal expectations: they finished short distance and long distance (0.56 xG). Already Atlético PR She finished 20 times: two from the inside of the small area, seven from the inside of the big area and 11 from the outside.
while the Atletico MG Producing enough to score 2.36 goals (and in fact score two), Avaí produced a prediction of 0.36 xG and still managed to score. There were 26 shots by Gallo in the match (half from inside the area and half from outside), five of which were when Kevin was already in goal after Douglas Friedrich was sent off. Avaí only finished four times (two from inside the area and two from outside).
Despite the draw, the visiting team came closest to winning, which produced enough to score almost two goals (1.90 xG), but scored only once. The dragon ended twice like international : three from inside the small penalty area (one of them comes out the goal), six from inside the penalty area and nine from outside.
a international I finished nine times (four on the inside of the box and five on the outside). It is interesting to note that Inter was practically not produced in the second half (the black line on the chart below is from the 45th minute): only two letters were presented in the final stage.
Simulation 10,000 times every match ends, the international He will leave with three points 22% of the time, a tie will happen 22% of the time and a victory Atlético jo It will happen 56% of the time. With this, it is possible to calculate the points prediction (xP): for Inter it would be 0.88 xP (3 points x 22% + 1 point x 22%), while the truth was the break-even point.
These probabilities can be found to the right of each graph, and finally, we can find out which teams are most efficient and least efficient in terms of points, goals scored and goals scored in the table below. It should be noted that football is a sport in which success is rare (the goal in this case) and this means that the end result of the game does not always correspond to the performance of the teams on the field.
While Atletico MG And the Palm trees are the teams that have won the most points, and Corinthians They are the most efficient team in terms of points (taking 6.1 more points than I would have expected for the shots made and suffered per game).
Galo and Verdão are also the two teams that have produced goal-scoring expectations in attack, while Timão is also the most efficient team in attack (they have scored 3.2 more goals than the quality of their finishing would suggest).
a Energy He is the player who has had the least amount of shots (he was expecting to have conceded six goals, but he had already conceded ten), while Palm trees And the cuiaba They are the most efficient team in defense (they have conceded fewer goals than the quality of shots received would suggest).
You can find graphics and analysis for other games at Stats Spy Twitter
The “Expected Goals” or “Goal Prediction” index (xG) is a standardized measure in data analysis that contains for reference more than 84 thousand shots recorded by Espião Statístico in more than 3.3 thousand matches in Brasileirões since the 2013 edition. Distance and shot angle relative to goal, The part of the body used to complete it, whether the shot is first time or repaired, whether the shot is with the player’s good or bad leg, is considered as such. As well as the origin of the shot (whether from accidental, direct foul, theft, etc.). The market value of the teams in each season is also taken into account based on data from the Transfermarkt website (such as agent team quality), playing time and difference in score at the moment of each shot.
For example, out of every hundred shots made by Half Moon, only seven turned out to be a goal. Therefore, given the distance of the shot only, the half-moon shot’s target (xG) is about 0.07. Each position on the field has a different expectation that a shot turns into a goal, which increases if it’s a counterattack because there are fewer opponents to avoid completing play. Each shot for each team receives a value and is added throughout the match to arrive at the team’s xG total in each game.
* Bruno Imaizumi is a partner economist at the Espião Statístico team, which consists of: Felipe Tavares, Guilherme Maniodette, Guilherme Marsal, Joao Guerra, Leandro Silva, Mateos Pinheiro, Roberto Malleson and Valmir Storte.
“Internet Teacher. Certified TV Enthusiast. Beer Teacher. Unconditional Pop Culture Enthusiast. Web Scholarship Holder.”