~snip~
Well, let's take the example of football. in football, it takes 11 players and each player plays their respective roles. when the two teams compete, strategies and coaches are needed. So, in theory, football involves many parties. one mistake made by a player, can be fatal for the team as a whole. as well as performance, performance plays an important role to be a measure in every consideration of our analysis assessment. besides that, the absence of several players can also have an impact on team performance.
Well, the question is, can AI explore all of that? Can Al predict accurately, I say no for now. why, because football doesn't only talk about data and statistics, but it's more complex than that. and it seems, the machine will not answer what we need regarding 100% correct predictions. if only to find data, information, and everything related to the two teams that will compete, it looks like we can rely on Al. yes, because, they are more efficient. to predict, I don't think so. so, overall I agree with what you said.
The thing is that you can combine a statistical AI that has all the previous information about the matches, players, etc, and starts the match with an idea, say team A has 80% chance of winning.
Then, while they are playing you can have an AI trained with computer vision to actually look at the performance of each player and see how things are going, in real time, and update the odds in real time.
Something like that would be really powerful and I think it might be already out there somewhere.