Jacksonville Jaguars vs Cincinnati Bengals — Win Probability

Jacksonville Upsets Cincinnati as Model Underestimated Jaguars

Jacksonville's 22-17 win over Cincinnati on Sunday qualifies as a clear upset—the model gave the Jaguars only a 43% pregame win probability and projected them to lose by 2.5 points. The market agreed, favoring Cincinnati by 2.5 points. Neither read the game correctly.

The model's expected final score was Jacksonville 24.5, Cincinnati 27. The actual result was Jacksonville 22, Cincinnati 17. Jacksonville won by 5.

That is a 7.5-point swing from projection to reality in Jacksonville's favor. The Jaguars not only won the game the model thought they'd lose—they did it by a margin nearly three times larger than the model predicted in the opposite direction. This was not a narrow miss. It was a substantial miss.

The gap between the model's -2.5 expected margin and the actual +5 margin reveals a meaningful underestimation of Jacksonville's game plan or execution. Cincinnati underperformed its projection by 10 points (expected 27, scored 17), while Jacksonville underperformed by only 2.5 points (expected 24.5, scored 22). That differential suggests Jacksonville's defense or Cincinnati's offense broke down in a way the preseason simulation did not anticipate.

Upsets are not failures—they are facts that occur at measurable rates. A team projected at 43% to win will lose sometimes and win sometimes. But a 7.5-point swing from expected margin to actual margin is large enough to warrant examination of what Jacksonville or Cincinnati did differently than the model assumed. The Jaguars improved to 3-1 with a performance that significantly exceeded their pregame probabilistic standing.