While there is no question that the PR6 (punt return for TD) and a subsequent long return for good field position were critical to the final score, I find it hard to believe that our punting game gave away 13 points. Maybe I missed something. Hoping that "fixing" the special teams continue to be a point of emphasis.
When you see an EPA (expected points added) or a PPA (same thing, different initials), they correlate every play with the final score. One of my stats professors brought it up when I was an undergrad ages ago. For example, they’d worked out the correlation for an interception as adding a couple of points to the defending team’s final score, and an interception returned for a touchdown was correlated with more than 7 points for the final score. Why? The explanation they had was that there actually was some momentum to it—you got the 6 for the TD and the defensive team shifted the overall course of the game by a few more points from the interception, including taking a scoring chance away from the other team.
But, the explanation they gave is a rationalization. The big idea is that they've taken thousands of games and hundreds of thousands of plays and mathematically correlated each play to the final score. First downs help. First downs on the opponent's side of the field count more. The closer you get to the goalline, the more that yard is worth in getting points--mathematically. (That's another reason not to punt from your opponent's 39 yard line--the payoff for you getting that first down is worth the risk from not punting).
Even in the last six minutes of the game, there was a punt returned for a TD, a punt nearly returned for a TD that put us on our heels, and another punt that wasn’t great that did lead to a TD. They helped Duke a LOT.
The punt returned for a TD had an expected points added of 5.87 for them--you can rationalize that as "we had a chance to answer back" (we didn't). The second punt, where we netted 10 yards, was worth an EPA of about 2 points. They're "Expected" points--they don't necessarily convert, but the odds of getting points are good when you start on the other team's side of the field.
On the other hand, there were 3-and-outs and other sequences that subtracted from their expected score. If you look at the 3rd and goal that Duke had in the first half, we picked them off--that went from them expecting at least a field goal to getting absolutely nothing--and us getting the ball.
And, if you look at the game, Duke got two touchdowns in the game--an easy one off of a punt return, and a hard one that took forever even with multiple penalties helping them move the ball. Not were kicked off from us being crummy in punt coverage. The punt returns were the easiest and quickest way they had to get points or get close to points.
If someone is looking, it was punt returned for a TD, then three plays by our offense, punt returned to our 44 where we scrapped and got the ball back on downs, then four plays by our offense, a punt that should have been aimed for a sideline but went for a touchback instead, and then a TD helped by penalties.
The 3-and-outs in the late 4th were also hugely negative for us, and let them in the game. Those had a negative EPA for us. It all adds up in the end.