Can Qwen2-VL AI Model Improve Man United's Game Plan?

In Manchester United 's latest clash against Liverpool Football Club in the The Premier League , a 3-0 defeat at Old Trafford, tactical issues became glaringly apparent as United struggled to contain their bitter rivals. Liverpool capitalized on key moments, with goals from Luis Diaz and Mohamed Salah, ultimately highlighting the gaps in United's game. In this article, I explore how cutting-edge AI models like Qwen2-VL by Alibaba Cloud , a newly launched Vision Language model, can provide valuable insights for coaches like Erik ten Hag to refine post-match strategies.

I used the Qwen2-VL model to analyze specific moments from the match, focusing on player positioning and defensive lapses. The model's ability to extract visual data and deliver tactical suggestions can be a game-changer for post-match analysis. This level of AI-driven insight could be essential for clubs like Manchester United, who already employ advanced data analytics but can always benefit from additional perspectives. Below is an analysis I performed using the Qwen2-VL model on one of the images:

Here are the insights generated by the model:

The image shows a red circle highlighting a specific area on the field. The red circle appears to be placed over a player who seems to be in a vulnerable position, possibly due to being too far from the ball or out of position for the play.

Tactical Issues
1. Positioning
The player in the red circle is positioned too far from the ball and is not in a good position to receive a pass or make a play.
This can lead to poor team cohesion and a lack of support for the player.

2. Defensive Positioning
The positioning of players around the red circle suggests a potential defensive issue.
The players in the yellow circle (defenders) are not in a good position to support the player in the red circle, which could lead to a weak defensive line.

3. Team Dynamics
The player in the red circle is isolated and not contributing to either offensive or defensive efforts.
This could lead to a breakdown in overall team performance.

4. Execution
The execution of the play by the player in the red circle could be a concern.
If the play is not executed correctly, it could result in a missed opportunity or a poor outcome for the team.        

AI models like Qwen2-VL are transforming the way we analyze match footage. While Manchester United, like other top-tier football clubs, already integrates AI and data analysis into their tactical framework, there's always room for enhancement. By leveraging Qwen2-VL, coaches could better pinpoint areas of concern, whether it’s poor defensive positioning, ineffective team dynamics, or lapses in execution. Erik ten Hag can utilize these AI-driven insights to address specific weaknesses, improving both individual player performances and overall team cohesion.

Here are the suggestions by Qwen2-VL AI Model for the scenario on how they can improve:

Tactical Improvements for the Scenario

1. Optimal Positioning
Key Focus: One of the critical errors visible in this footage is the poor spacing between the players. The player in the red circle is not positioned optimally to affect the play, either defensively or offensively. In modern football, the concept of positional play is crucial to maintaining structure. The player should occupy a space that creates passing triangles, offering a passing option while also being able to transition into a defensive role. This could be achieved through positional fluidity, which requires constant movement to be in the right spaces between the lines.

Solution: The player in the red circle needs to move vertically into the half-space to offer support to the player on the ball, or horizontally to stretch the defense and provide width. This ensures the team’s attack doesn’t become isolated, keeping the lines of connection between midfielders, attackers, and defenders intact.

2. Defensive Depth and Structure
Key Focus: The lack of support from the surrounding players (particularly the ones in the yellow circle) indicates a breakdown in defensive shape. Modern football defends as a unit, and the gaps between lines must be minimized to avoid the opponent exploiting open spaces. In this instance, United’s defenders are too static, and their horizontal spacing is too wide, making it easier for Liverpool to progress through the central channels.

Solution: The defenders need to form a tighter defensive block. Utilizing compactness, where players close down space both horizontally and vertically, could prevent Liverpool from exploiting those gaps. The defensive line should push higher up the pitch, compressing space, and forcing Liverpool into wider areas, where they are less dangerous.

3. Team Dynamics and Cohesion
Key Focus: Teamwork is essential for any successful tactic. In this scenario, the player in the red circle is isolated, indicating poor team compactness. This is a common issue when lines (defensive, midfield, and forward) get stretched, leading to large distances between players. This not only hampers passing options but also increases the vulnerability of a counter-attack.

Solution: To fix this, the entire team needs to maintain tighter connections, especially between the midfield and forward lines. The use of counter-pressing (Gegenpressing) could be an effective strategy in regaining possession when the ball is lost, requiring immediate pressure on the opposition to win the ball back. By doing so, United could reduce the risk of transition play, especially against a counter-attacking side like Liverpool.

4. Execution of Plays
Key Focus: The player in the red circle is not in a position to either support the attack or shield the defense, suggesting poor execution of tactical instructions. Manchester United’s build-up play from the back should focus on breaking lines through vertical passes or moving the ball quickly through the flanks to disorganize Liverpool’s defense.

Solution: For better execution, players must adhere to tactical discipline. This means understanding when to engage in high-pressing, when to drop deep, and when to hold the ball for transition play. If players like the one in the red circle fail to position themselves correctly, the entire team’s rhythm and flow are disrupted. Emphasizing quick transitions and positional awareness will allow United to adapt more fluidly, whether in defense or attack.        

Check out my YouTube video on AI Anytime where I walk through different use cases of Qwen2-VL, including how it can be applied for match analysis: Qwen2-VL Video

In conclusion, the use of AI for football tactics is no longer a futuristic concept but a present-day reality. The Qwen2-VL model offers an additional layer of depth that could help teams like Manchester United stay competitive by improving their post-match analysis and decision-making on and off the field.