One of the simplest observations that one can make about artificial intelligence is that in essence, it is simple: it measures trends, and then, based on those trends, it makes predictions. But football managers don’t always do what the predictions say they will do, not least because sometimes, despite their best efforts, they can’t actually buy the players they want, or can’t get them to play in the new tactical approach that they have devised.
So the managers of Premier League clubs, sometimes by design or sometimes in desperation, then attempt to outplay the trends and the predictions by doing something different. And sometimes it works,, but sometimes it doesn’t.
One of the earliest variations we saw of this type was that of playing a right-footed player on the left wing – it can result in the player tripping himself up, or it can leave the full back marking the player totally off balance and easy to bypass.
But now of course we have AI and AI is starting to tell managers what tricks their opposite numbers got up to in their last game, and what they are likely to do in their next game. Indeed, since AI has no limitations programmed into it, it is already predicting who is going to win the league in the next quarter-century. But it is not yet predicting how AI predictions can be beaten, or how AI can be used against itself.
But in the more intelligent parts of the community, such thoughts are entering the managers’ brains. Club X has player Y, so let’s ask AI how club X will use this player, and then let’s ask AI how to overcome this ploy….
That sounds good except… the word is spreading around as to which managerial and tactical teams are known to be AI enthusiasts, and of course, AI then builds that into its models. Thus, AI then change its predictions, and so it goes on.
But what AI is also being used for is to predict what ManC will do next in their battles against the Premier League, and what other clubs, that have been watching the ManC manoeuvres, are planning to do
Second, take new patterns and approaches into account just as ManC did by bringing in more money than was ever thought possible and then threatening to close down the league through court cases if anyone challenges it. Best to ask AI a) what ManC is most likely to do, and what the other clubs should do in response and b) if that response is followed, what will ManC do after that..
As you can see, the whole process gets more and more complex as the days go by because each club is raising more “what if” questions, and AI is making more and more guesses, and at some stage, reality is going to be left behind.
But what the process does tell us is that one thing is sure: while in the old days clubs used to buy players with the thought “he’s going to make our left wing a lot more tricky for opponents,” team selections are now based on what it is thought the opposition’s computer might be telling the manager to do.
Of course, Arsenal do have an extra trick up their sleeve, which most other clubs don’t have, and that is the extraordinary ability of the club to bring in genius youngsters. Obviously, everyone is out there looking for young players of quality because not only will the opposition not have much idea how to handle them, but there is every chance that one day the player will be sold, and that income will be pure profit.
At the same time, young players and their parents and advisors are looking for clubs that will give them the chance to shine in the first team at the earliest possible moment. So they look at Arsenal’s record with young players and realise that this is a club that will help them on their journey.
No one is suggesting that AI can help clubs see a perspective first team genius at the age of eight… that is still down to the coaches of the juniors, but there is always the chance that AI can suggest a slight change of tactics to make the most of the youngster’s ability when he is ready to take the great leap forward.
