After 3 rounds of the Premier League, 0 goals scored, 9 goals conceded, 0 points, the 20th place in the Premier League, and the start of Arsenal's worrying 21/22 season.
After 28 rounds of the Premier League, with 54 points, the premier league's fourth place result has become the best team in the season of the fourth team.
What kind of change is it that the same head coach Arteta and the winter window transfer that can only go in and out have made Arsenal change from the laughing stock of the fans at the beginning of the season to a big favorite for the Champions League seats?
Defensive end - The main team returns and gradually improves
The three consecutive defeats in the first three games are largely due to the lack of manpower in the rear defense. Arsenal's starting line-up includes Pablo Mari, Callum Chambers and Seyard Kolasinac, all three of whom have only completed 2 league appearances each before leaving the team, and are clearly not Arteta's main candidates. Right-back Kenyo Fuan joined, centre-back Ben White recovered from the outbreak, Gabriel returned from injury, goalkeeper Ramsdale grabbed the starting position, plus left-back Kieran Tierney, who had been alone. After the five defensive combinations joined forces to start, they quickly left the dismal record of 9 goals in 3 games completely behind.
As the season progressed, their performances became more stable. In the first 14 games, the average xGA per game (expected number of goals conceded) was 1.70; in the last 14 games, the average xGA per game was 0.82, a huge gap of direct halving. From the moving average line chart of xGA in the chart below for the last 10 games, it can be seen that Arsenal's expected number of goals conceded has gradually declined over time. (xGA data taken from understat)
In terms of defensive arrangement, Arsenal's high-pressure pressing system is not their strong point, a data used to observe the team's aggressiveness of pressing - PPDA (PASSES per defensive actions, the number of passes that the team performs defensive actions such as steals and interceptions, the lower the number of passes that the opponent completes, the less frequent and more active), the arsenal season average is 12.99, ranking 16th. The reason why this figure cannot be depressed is mainly because their main strikers (formerly Aubameyang or now Alexander Lacazette) are not good at grabbing, and the midfielders and wingers are more inclined to maintain the formation and protect the space between the two lines of defense from the opponent, which is also the importance of the combination of Thomas Parte and Granit Xhaka. However, with the rise of a new generation of players, such as Martin Erdegaard (averaging 18.8 shots per game) and Gabriel Martinelli (16.4 times), the team's defensive system is gradually changing, and from the moving average of the PPDA below, you can see a slightly lower trend. (PPDA data taken from understat)
Diverging from the topic to talk about, high PPDA does not mean that the defense is not good, this season's defensive Wolves PPDA is the second to the bottom, while leeds United, which has opened its doors this season, is high in the 1st place, this data presents only the style of the team's defense. The gradual decline in the PPDA of the Arsenal brings more ball power conversion in the frontfield threat area, and even suppresses the opponent's dominance, I believe that from these perspectives, Arteta will be happy to see such a change.
Offensive end - the whole team shares the organizational burden equally
Ramsdale, who has secured his position as number one goalkeeper this year, has the ratio of executive passes (over 30 yards) that is the biggest difference between him and Baind Leno. The former had 54 percent of all passes this season long, compared to just 37 percent last year. The accuracy of long passes is 41.4% and 46.0%, respectively, and Ramsdale is slightly worse. The biggest threat to Ramsdale's long passes is to go straight to the front arrow, averaging 0.92 passes to the front court every 90 minutes, well above Bernd Leno's 0.14 passes. So Ramsdale's contribution is not a safe transmission at the back (of course, it's not worrying if it's done, and finding a defender to pass in the backcourt is also his usual way of passing), but the way he helps the team advance is more direct. A few comparisons in the chart below give a clearer picture of Ramsdale's style, and if Ederson (backcourt passing) and Pickford (Bigfoot balling) are crudely used as the two extremes of the spectrum, Ramsdale will be closer to the latter, but his footwork - whether it is the accuracy of the long pass in the figure, or the decision-making and stability of the short pass will be more trustworthy than Jordan Pickford. (Data taken from fbref)
The change came from a multi-point blooming striker, Martin Erdegaard averaged 2.19 passes into the box, SCA (Shot-creating actions) averaged 4.63 shots before shooting, passing, etc.) were the first in the team; Smith Rowe averaged 1.68 successful passes per game, 6.77 effective ball-carrying advances, Bukayo Saka 1.65, 8.44, the two accounted for the top two of the two data, Bukayo Royo. Saka also averaged SCA in the second place on the team with 4.17 per game.
At the forefront, the main striker was replaced by Aubameyang at the beginning of the season to the current Alexander Lacazette, who pays more attention to the style of connecting with his teammates, which is exactly the type that head coach Arteta likes, averaging SCA at 3.71 times per game second only to the aforementioned Martin Erdegaard and Saka. Alexander Lacazette sometimes retreats too far, with vacated areas in the penalty area being exploited by teammates, Smith Rowe tied for first in the Premier League with nine goals throughout the season, and Gabriel Martinelli, who has been reused recently, is also good at looking for opportunities. Alexander Lacazette averaged xG 0.42 per game and Gabriel Martinelli 0.38 points per game in the top two of the team.
In the attacking plan, Arteta did not fully replicate Guardiola's football philosophy of persistent passing, with an average of 51.8% possession only ranking 7th in the Premier League. There are probably two major influencing factors, the first is the defensive end mentioned in the previous paragraph, the high PPDA makes arsenal unable to recover the ball as immediately as Manchester City to continue to attack; the second is the difference in offensive rhythm, Manchester City is good at slow testing, pressing the opponent in their own half and then slowly looking for opportunities to break through, Arsenal is more tiki-taka flavor in the 2010s, deploying troops between the line in the opposing backline. Then through the smooth fast short pass with the instantaneous breakthrough, will be their main means of attack.
Finally, let's look at the results they have brought out. Similar to the way defense is compared, the chart below is viewed using a moving average line chart of xG for the last 10 games. Although the amplitude is not as obvious as the xGA (the last two hard battles against Liverpool and Villa have also caused xG to drop slightly), there is also a slight upward overall trend. (xG data from understat, the rest from fbref)
epilogue
Looking at the entire season, Arsenal's xG ranked 6th, xGA ranked 5th, and xG-xGA - used to compare the number of goals that a team "expects" to score more goals than its opponents throughout the season - ranks 5th, behind Manchester City, Liverpool, Chelsea and Tottenham Hotspur, of which Tottenham Isthuis is the closest opponent to Arsenal in the current competition for the fourth rank. With three points and the advantage of one game less, coupled with the recent gradually warming offensive and defensive performance, Arsenal will be a favorable candidate for fourth place.
But regardless of the season's ending, the young manager gradually learns the skills of "belting" and "prying" from the Ozil and Aubameyang incidents; the young defender gradually reduces the number of low-level mistakes as the experience grows; the young frontcourt continues to grind and expand the skill pack with time; and the ideal striker that looks forward to next year's transfer window - a striker who combines Alexander Lacazette's withdrawal organization, Aubameyang's penalty area grabs, and highly aggressive strikers. The growth of this arsenal continues.