2022 College Football Returning Production (ESPN)

quote:

---

As mentioned, the formula I use changes from year to year. Here's the current weighting for determining the offensive percentages above:

Percent of returning WR/TE receiving yards: 37% of the overall number

Percent of returning QB passing yards: 29%

Percent of returning OL snaps: 28%

Percent of returning RB rushing yards: 6%

Broken out by position/player, you're looking at approximately 29% for the quarterback, 9% for each of four wide receivers and/or tight ends and 6% for the running back and each offensive lineman.

(Note: Since FBS newcomer James Madison played at the FCS level in 2021, I didn't have full snap count data for the Dukes. I used player starts as a substitute.)

It's a bit trickier on defense -- units aren't quite as strictly defined, and the percentage of returning production is derived both from position units and types of stats (tackles, tackles for loss, sacks, passes defensed). Here's the approximate layout:

Percent of returning tackles: 59%

Percent of returning passes defensed: 28%

Percent of returning tackles for loss: 8%

Percent of returning sacks: 5%

Perhaps surprisingly, turnover in the back of the defense causes far more of a shift in a team's SP+ rating from year to year than turnover up front. By position, defensive backs make up about 51% of the defensive formula, while linebackers are at 32% and the defensive line is at 17%. (Remember: This is not based on my personal opinion of positional importance -- it's all about what impacts the numbers the most. That distribution surprises me too!)

---

quote:

---

Transfers Returning production: Quite crudely, if a player transfers from one FBS school to another, I mash his production from his previous team into the numerator and denominator for his new team. So if your quarterback leaves, and you bring in a transfer who was starting somewhere else -- a la USC, Ole Miss, Cincinnati and others -- that dampens the overall blow significantly. Because the translation from lower levels to upper is extremely inconsistent, I don't do this for players transferring up from FCS or Division II. I used to, but it didn't produce predictive value.

---

quote:

---

Recruiting: As of 2022, I am now attempting to account for incoming transfers' recruiting rankings in a team's recruiting averages. It will not be a significant piece of the recruiting puzzle -- anytime I do something new, I ease it in from a weighting perspective -- but there will be at least a marginal impact here.

---