The goal of the year awards are already wrong. They always are. By the time the panels vote, the narrative has solidified around three or four goals that looked spectacular on television. The actual difficulty of the finish, the conditions under which it was scored, and the number of players in the world who could have replicated it are all secondary.
This is the correction. The model has been running since January. These are the 25 best goals of 2025 by pure difficulty score.
The top 5
1. Marcus Rashford vs Manchester City, March 2025. A solo run from the halfway line, beating three defenders with close control at pace, and a finish from the tightest of angles with the goalkeeper set. The difficulty is the sustained pressure: each defender commits fully, and Rashford has to maintain control while accelerating. The finish is not spectacular. The creation is. Difficulty score: 9.34.
2. Vinicius Junior vs Barcelona, January 2025. A bicycle kick from a cross behind him, with his back to goal, at full stretch, with a defender attempting to block. The contact is perfect, the power is generated from core rotation, and the placement is into the only corner the keeper cannot reach. The fact that this is the El Clasico is irrelevant to the score. The technique is not. 9.28.
3. Erling Haaland vs Liverpool, February 2025. A first-time volley from a cross arriving at pace from the right, struck with the instep while the ball is still rising, into the top corner with the goalkeeper at full stretch. Haaland generates power from a standing position with a defender closing. The coefficient for first-time volleys on rising balls is high. 9.21.
4. Mohamed Salah vs Arsenal, April 2025. A solo dribble from the edge of the box, beating two defenders with quick feet and a body feint, and a curled finish into the far corner with the goalkeeper partially unsighted. The difficulty is the combination of close control under pressure and the need to place the shot precisely. 9.17.
5. Florian Wirtz vs Bayern Munich, March 2025. A 25-yard half-volley from a cleared corner, struck with the outside of his boot while the ball is dropping, swerving away from the goalkeeper and into the top corner. The contact type is rare, the distance is significant, and the goalkeeper is set and sees it. The swerve is what makes it unsaveable. 9.14.
6 through 15
6. Bukayo Saka vs Chelsea, May 2025. A solo run from midfield, beating three defenders with pace and close control, and a finish with his weaker foot from a tight angle. The model values sustained individual creation highly, and this is a textbook example. 9.08.
7. Kylian Mbappe vs Atletico Madrid, February 2025. A 40-yard sprint with the ball, beating two defenders with pure pace and a step-over, and a powerful finish into the roof of the net. The difficulty is the speed at which everything happens. The model accounts for the fact that Mbappe is moving faster than the defenders can react. 9.05.
8. Victor Osimhen vs Juventus, January 2025. A header from a corner, with the ball looping over defenders, and a finish into the far corner while being pushed by a defender. The header itself is not difficult. The positioning and the contact under physical pressure are. 9.02.
9. Phil Foden vs Real Madrid, April 2025. A first-time strike from the edge of the box after a quick passing move, with the ball arriving from behind him, and a finish into the bottom corner with his weaker foot. The first-time contact and the weaker-foot finish compound the difficulty. 8.98.
10. Jamal Musiala vs Borussia Dortmund, March 2025. A solo run from his own half, beating four defenders with dribbling and acceleration, and a finish with his right foot from a tight angle. Similar to the classic Messi runs but with more distance covered. 8.95.
11. Martin Odegaard vs Manchester United, February 2025. A 30-yard free kick that dips over the wall and under the crossbar, with the goalkeeper at full stretch. The distance and the dip make this a high-difficulty set piece. 8.91.
12. Lautaro Martinez vs AC Milan, April 2025. A bicycle kick from a cross at the back post, with his back to goal and a defender closing. The technique is exceptional and the defensive pressure is high. 8.88.
13. Rodri vs Tottenham, January 2025. A 25-yard volley from a cleared cross, struck on the half-volley with his weaker foot, and a finish into the top corner. A defensive midfielder scoring this type of goal is rare, but the model does not care who scores it. 8.85.
14. Declan Rice vs Liverpool, March 2025. A solo run from deep, beating two midfielders with power and close control, and a 20-yard strike into the top corner. The model values the combination of dribbling and long-range finishing. 8.82.
15. Bruno Fernandes vs Chelsea, May 2025. A chip from the edge of the box while running at pace, with the goalkeeper rushing out and expecting a low shot. The decision difficulty is high and the execution is perfect. 8.79.
16 through 25
16. Alexander Isak vs Manchester City, April 2025. A first-time finish from a cross at the back post, with the ball arriving behind him and a defender closing. 8.76.
17. Cole Palmer vs Arsenal, February 2025. A solo dribble past three defenders and a curled finish into the far corner. 8.73.
18. Jude Bellingham vs Barcelona, January 2025. A 30-yard strike from a loose ball, struck first-time with power and swerve, and a finish into the top corner. 8.70.
19. Antoine Griezmann vs Real Madrid, March 2025. A backheel from a tight angle while being marked by two defenders. The creativity and the spatial awareness are exceptional. 8.67.
20. Rafael Leao vs Juventus, January 2025. A solo run from the left wing, beating two defenders with pace and skill, and a finish into the far corner. 8.64.
21. Julian Alvarez vs Liverpool, April 2025. A first-time volley from a corner, struck while the ball is dropping and with a defender attempting to block. 8.61.
22. Pedri vs Real Madrid, February 2025. A solo run through midfield, beating three defenders with dribbling and quick feet, and a finish with his weaker foot. 8.58.
23. William Saliba vs Manchester United, March 2025. A header from a corner, with the ball looping over the goalkeeper and into the far corner. The positioning and timing are perfect. 8.55.
24. Khvicha Kvaratskhelia vs AC Milan, April 2025. A solo run from the halfway line, beating four defenders with pace and dribbling, and a finish from a tight angle. 8.52.
25. Bernardo Silva vs Real Madrid, April 2025. A chip over the goalkeeper from a tight angle after a solo run. The decision and execution under pressure are what make this goal. 8.50.
What the 2025 data shows
Three things stand out from this season's data.
First, solo runs are back. After several seasons where team goals and set pieces dominated the top of the rankings, 2025 has produced more individual dribbles and solo runs than any season since 2015. The model does not know why. It only knows that the inputs have shifted.
Second, Premier League goals dominate the top 10. This is not Premier League bias. The model does not know which league a goal is scored in. The defensive pressure input is higher in the Premier League because the defenders commit more aggressively and the pressing is more intense. Higher defensive pressure means higher difficulty scores.
Third, the goals that will win the official awards are not in the top 5. The model does not care about narrative, context, or how good a goal looked on television. It cares about difficulty. The official awards will almost certainly pick a goal from the top 10, but not the top 3. That is the gap between punditry and methodology.
The full methodology
The model runs live. Every goal from the top five leagues is scored within 24 hours of being scored. The inputs, the weights, and the validation framework are all documented in the companion piece. This is the output for one season. The full historical ranking is in the 50 greatest goals piece.