Shemar Stewart, Ted Karras, and Orlando Brown Jr. before a preseason 2025 game against the Philadelphia Eagles
Getty Images
I have long made the argument that a team with a great offense and bad defense can usually land a playoff berth. But last year’s Bengals were the exception. When the defense is as bad as Cincinnati’s was in 2024, it’s hard to win close games. Cincinnati made a coaching change in the offseason, replacing coordinator Lou Anarumo with Al Golden, who’s jumping to the NFL from Notre Dame. As mentioned above in the Colts section, Anarumo produced an above-average defense just once in six seasons. Having said that, talent was and continues to be the big issue for this group. The Bengals finished 27th in efficiency last year, and what makes that especially alarming is that they faced backup quarterbacks at the third-highest rate of any team.
How often did defenses face the opposing team's preferred quarterback last year? League average: 76.4%.
The Titans, Colts, Bengals, Commanders, and Jaguars faced backups most often.
The Browns, Rams, Vikings, 49ers, and Chiefs faced the opposing team's #1 QB most often. pic.twitter.com/ISjdSlixor
— Bill Barnwell (@billbarnwell) August 15, 2025
There aren’t a whole lot of reasons to feel optimistic about a major improvement. Since 2022, the Bengals have spent 10 picks on defensive players in the first three rounds of the draft, and the bottom line is that they need those young players to be productive. For our purposes, we’ll assume that edge defender Trey Hendrickson, who is unhappy with his contract, won’t miss regular-season games. If first-round pick Shemar Stewart can contribute right away, maybe the Bengals pass rush can have some juice. But ultimately, their success might come down to whether the young defensive backs—Cam Taylor-Britt, DJ Turner II, Dax Hill—can perform. If I were Golden, I’d try to manufacture turnovers and negative plays. That’s a high-variance style of play, but it’s just hard to believe that this will be a sound unit on a down-to-down basis.