Goldman Sachs’ sophisticated statistical model to predict the outcomes of the World Cup “did not even come close” to hitting the mark, Leonid Bershidsky writes for Bloomberg. Just to give some perspective: The model thought that Egypt had a 34.4% chance of making it out of the group stage, and that Croatia’s chance of making it to the final match was no higher than 2.1%.

“The failure to accurately predict the outcome of soccer games is a good opportunity to laugh at the hubris of elite bankers, who use similar complex models for investment decisions.” The investment bank continued to update the model as new outcomes continued to emerge, but the model continued to crank out predictions that were completely off, which goes to show that real life and a game like soccer are “infinitely more complex” than any model based on past data and statistics.