Teachers often let behaviour influence their grading patterns. This is what leads to male underperformance.
There is little difference between male and female academic performance and cognitive ability. The only difference is behavioural. In elementary school, where behaviour is a big factor in teacher’s grading, boys underperform, not because they objectively perform worse in school, but because teachers are biased and let classroom behaviour influence their marking. 
A study published in the Journal of Human Resources found that even when male students performed as well as their female counterparts on maths, science and reading tests, teachers rated them as less proficient.  This is because classroom behaviour is a big factor in teachers’ grading practices and students with poor classroom behaviour (generally boys) are deemed less proficient academically. Boys might lose focus more easily, challenge teachers more frequently, and be unable to sit still for long periods of time. This causes the teacher to hold a more negative opinion of them, which influences their marking practices. As teacher grades influence college decisions and are viewed before offering university places, this can lead to more females being admitted to top universities. But this is not a systemic issue. The school system and its teaching methods are not inherently weighted against boys. It is just the marking practices of individual teachers that disadvantage male students. The same study also found that when boys have a positive attitude to learning and score the same as girls on tests, teachers actually mark them higher than girls. This suggests that teachers have lower behavioural expectations for male students and when they exceed these expectations, they reward them with bonuses. In this way, individual teachers also discriminate against girls.Boys also seem to have a small marking advantage in mathematics, where they consistently receive higher grades than girls for getting the same objective mark. These findings make it impossible to say that the school system benefits girls more than boys. Individual teachers may grade in favour of girls. Some teachers grade in favour of well-behaved boys. But there is no underlying current of discrimination against one gender.
The teachers are the system. If female teachers, which make up the bulk of the educational workforce, are under-grading boys across the board, this is a systemic problem. It is not an isolated incident but a problem of discrimination playing out within the school system.  If it was a case of individual teachers sometimes under-grading and sometimes over-grading boys, then the trend would not be visible in large sample sizes. The fact that the trend is visible means that our way of measuring performance inside the school system is inherently weighted against boys. By moving it more towards measuring objective skill, boys would be able to close the academic gap emerging between the sexes.
[P1] The subjective beliefs of individual teachers sometimes influences their marking. [P2] This can negatively impact boys' grades. [P3] It can also positively impact boys' grades. [P4] Therefore, we cannot say that the school system is inherently weighted against boys.
[Rejecting P3] More often it does not positively affect boys grades. That is why in large sample sizes, the visible trends indicate boys are under-graded. [Rejecting P4] The way we measure academic performance is part of the school system. If that way is discriminating against boys, then we can conclude that the system is weighted against boys.