Models for evaluating players part 3: Markov models

แชร์
ฝัง
  • เผยแพร่เมื่อ 11 พ.ย. 2024

ความคิดเห็น • 1

  • @andreastasoulas7939
    @andreastasoulas7939 3 ปีที่แล้ว +1

    Thinking a bit about the latest point, I would say that the model lacks context in terms of players' positions of both teams. Also the objective at each time is not just to naively bring the ball closer to the opponent's goal, but also protecting the ball from the opponents (or just getting the ball out of their own box as much as possible). Teams may just want to play out time (another context depending on the scoreline). How would a simulation that took into account individual player movements, would fare in terms of realism? And how difficult would it be to implement, in the first place?