It’s a funny thing how the narrative of a season can be shaped quickly. Take Clemson’s passing game, for instance. In the early weeks of the season, Deshaun Watson struggled to connect downfield, his receivers dropped too many balls, the offensive line didn’t protect particularly well and the story became pretty simple: The Tigers’ offense isn’t meeting expectations.

The tough thing about narratives, however, is they don’t change nearly as quickly as they form.

Take a real look at Clemson’s passing attack, however, and it’s as good as anyone could’ve hoped for this season, and it’s gotten better nearly every week.

Remember all those drops the first couple games? Even coach Dabo Swinney still brings them up as a measure of the self-inflicted wounds his team endured this season. But here’s a surprising stat: Clemson finished the season with the fifth-lowest drop rate among Power 5 teams. From Week 3 on, the Tigers dropped just 2.7 percent of their targets.

Or what about the narrative surrounding Watson targeting Mike Williams too often? That certainly seemed to be the case against Auburn in the opener, when Williams was targeted on 42 percent of Watson’s throws. Certainly the Tigers were trying to work Williams back into the offensive scheme early, but that, too, shifted quickly. From Week 2 on, Williams saw more than 30 percent of Watson’s targets just once, and was on the receiving end of just 22.6 percent for the season following the Auburn game.

Or how about those deep balls? Watson was perhaps a bit overzealous with them early. Thirty-five percent of his attempts against Auburn were for 15 yards or longer, and through Week 7, 18.5 percent of his throws were for 20 yards (in the air) or more. From the NC State game onward, however, he didn’t have more than 15 percent deep balls (20-plus air yards) in any game, averaged just 12 percent throws of 20-plus and saw his overall completion percentage rise from 61 percent in the first half of the season to 72 percent in the second half.

A quick way to measure the overall effectiveness of the passing game for Clemson is to look at the percentage of Watson’s dropbacks that resulted in a forced scramble, a sack, an off-target throw or a drop. We’ll call them “wasted dropbacks.”

In Week 1 against Auburn, 31 percent of Watson’s dropbacks were wasted.

In his next four games versus FBS opposition, he was over 23 percent every time.

In the final seven games of the season, he had just one more game higher than 23 percent: 23.54 percent in a blowout victory over Syracuse.

Overall, those first five games resulted in 26 percent of Watson’s dropbacks resulting in an off-target throw, a drop, or a scramble. In the final seven, that rate was just a tick higher than 16 percent.

All of this is particularly relevant in regard to the Dec. 31 matchup with Ohio State in the College Football Playoff semifinal at the PlayStation Fiesta Bowl, because while Clemson’s inconsistency on the ground has been a season-long affair, the growing improvement in the passing game stands out as a marked advantage for the Tigers.

For the season, Clemson averaged about 8 yards per attempt, about a 13 percent improvement over what its opposition allowed against other FBS teams. Ohio State’s pass defense, on the other hand, was essentially right at the FBS average, given its competition. Long story short: The Tigers can win matchups in the passing game, and the Clemson passing game is playing its best football of the year.

None of that changed the fact that Watson will need to be more dynamic with his legs than we saw him for much of the season or that Wayne Gallman needs to be more effective than we saw him for much of the season, but it’s a good indicator that the easy narrative for 2016 surrounding Clemson’s aerial attack hardly stands up against the Tigers' performance down the stretch.