And to play devil's advocate myself...
I was at a boxing match at White Hart Lane in London in the early 90s between two British middle-heavyweights fighting for the WBC world championship.
One was Chris Eubank, an absolute technically perfect boxer, against Michael Watson, a street-fighter per se.
Eubank would win on every metric had a computer programme been assessing the fight. But Watson mauls him for ten rounds.
In the 11th, Eubank, barely on his feet, came out and his technical prowess shone through, and his ability to punch like a superlative pro knocked Watson out....
If an analytics programme had been running that fight, it wouldn't have just looked at the fact that Michael Watson was mauling Eubank for nine rounds. It would have flagged something deeper.
Watson was operating at an unsustainable output....
He was walking through fire, expending a massive volume of energy, and taking heavy, accumulative strain to do it.
The computer would have looked at Eubank's cold, clinical precision and calculated his punch efficiency and the fact that Eubank wasn't wasting energy, while Watson was burning through his gas tank.
The metrics would have whispered to Eubank’s corner:
"He’s overextending. His defence is leaking when he crowds you. Keep absorbing, preserve your energy, and the opening will statistically widen by the championship rounds."
And that’s exactly what happened. Eubank didn’t win on a random miracle, he won because his technical perfection meant his punch economy was superior.
When Watson inevitably tired, the data's baseline prediction, that precision and power eventually punish exhaustion it proved completely, brutally correct.
So even in a sport as raw and savage as boxing, the metrics didn't fail. They just saw the ending long before anyone in the stands could.
Edit) And just in case anyone is wandering what this has to do with football?
Well Michael Watson was a well known Arsenal fan and Chris Eubanks supported Brighton.....
