Since the Kansas City Chiefs have been on their bye week, I decided to compare our weekly Madden simulations to the real-life 2025 Chiefs to see how they stack up against each other.
So Madden has been pretty good at picking the winner of each Kansas City game this season, but not very good at predicting the final scores of those games.
The Madden-simulated Chiefs entered the bye week with an 8–1 record. Madden’s predicted loss actually came true in Week 5, when the Jacksonville Jaguars scored exactly 31 points — just as the simulation had predicted. That simulation also came the closest to matching the 31-28 result — and was the only game this season in which the Madden prediction was closer than those made by all our AP panelists.
Of course, Madden also predicted three Kansas City wins that didn’t happen.
Let’s look at some season statistics.
At the end of the season, we’ll see how these comparisons change.
*To calculate a prediction’s points of error, the differences between the prediction and the actual score in point spread, home team score, and away team score are added together. For example, a prediction calls for a 17-10 Chiefs win. They end up winning 16-10, so there were two points of error: the point spread was off by one point, the Kansas City score missed by one point, and the opponent’s score was predicted correctly. But if the Chiefs lose the game 17-10, there were 28 points of error in the prediction: the point spread was off by 14 (the difference between +7 and -7), and both scores missed by seven.