Here’s a better way to measure 2019 college football returning production

My process for releasing each preseason’s S&P+ projections:

February’s Signing Day is a little bit later into the month this year (February 6), so we’re going to run the preview series a bit differently. I’m going to wait until the 11th to begin the team preview series with complete recruiting data built into S&P+, and for the first week of previews, I’m releasing two previews per day.

Two-a-days as prep for football season. Get it?

Over the last few years, I have attempted to move beyond the too-simple “returning starters” figure to measure experience.

We use the best tools we have, even if they’re not that great. We try to derive value from offensive line starts because it’s the only individual measure of offensive linemen that we have. You can’t get much from “he’s been one of Team A’s five preferred linemen 16 times since he started school here.”

It’s the same with returning starters. We use it because it exists. It is fine as a really quick snapshot, but we know that one team’s six returning offensive starters aren’t another’s. What about go-to guys? Returning backups? And quarterbacks are worth more than other starters, right?

In 2016, I began using a returning production figure based on what seems to best correlate with year-to-year performance. With what is now a few years’ worth of data, let’s take a look at some updated correlations.

Returning experience on the offensive line doesn’t have nearly the statistical impact that we expect. But with more data in the bank — and a new set of tweaks to S&P+ that I’ve been unveiling at Football Study Hall — we can see there’s a little correlation.

The higher the number, the more likely returning production in these areas is to coincide with strong offense:

With more data, the offensive line correlations have begun to grow stronger, which makes sense, but the conclusion remains: continuity in the passing game matters a hell of a lot, and continuity in the run game doesn’t have as strong an impact.

On defense, where returning production appears to matter more in general, the correlations are both stronger and more diverse. Since teams use different numbers of defensive linemen, linebackers, and defensive backs, I look at both unit-specific categories and those for defense as a whole.

That’s right, the correlations for defensive back tackles and passes defensed is stronger than the correlations for overall tackles.

The strongest correlations on the defensive line, by the way: 0.154 for tackles, 0.119 for tackles for loss. Continuity in the trenches does not appear to be worth what we tend to think it’s worth. But continuity in the passing game, on both sides of the ball, means a ton.

One more takeaway: tackles for loss are important, but the ability to get hands on passes might be harder to replicate than any other.

As with last year, I used categories like the ones above, weighted for largest effect — so returning quarterbacks, receivers, and defensive backs carry more heft — to create numbers for offense and defense. I have updated 2018’s rosters for 2019, accounting for NFL declarations and, as much as possible (since it’s impossible to keep up with all of them), transfers.

Thus far for 2019, seven teams return 80 percent or more, including Tennessee. (I will release updated returning production figures right before the season begins.)

For now, 13 teams are at 50 percent or lower, including Washington, Texas, and Georgia Tech.

Here are 2018’s top 10 teams, according to the revamped version of S&P+, re-ranked in order of 2019 returning production:

It’s safe to say Alabama will begin second in the preseason polls, and sportsbooks are listing Georgia in the top four. But there could be legitimate reason for putting four SEC teams near the top, including LSU, which overachieved last year’s returning production, and Florida. The Tigers and Gators went a combined 21-6 in 2018 and ranked fifth and ninth, respectively, in the revamped S&P+. And now they’re both going to be projected to improve by a decent amount.

In fact, if we were to use returning production as the only S&P+ projection factor — eschewing recruiting and recent history — here’s how the projected top five would take shape. Again, this would be based on nothing more than last year’s S&P+ ratings and this year’s returning production, not the complete formula:

Auburn, Mississippi State, Missouri, and Texas A&M would all be in the top 15 with this approach, too. Plus, Tennessee, second in overall returning production, could take a couple of steps forward as well.

Finally, the SEC gets its ducks in a row. (Yes, that was sarcasm.)

Let’s keep the “If this were all there was to S&P+ projections” thing going. Here’s how the projected Pac-12 would rank based on returning production:

11. Utah
21. Washington
29. Washington State
31. Oregon
33. Stanford
36. USC

For now, never mind that there would be more Group of Five teams in the projected top 25 (Appalachian State, Memphis, and UCF) than Pac-12 teams.

Let’s just focus on the fact that Utah, which finally broke through and won the Pac-12 South, might take one final step up the ladder.

Washington could catch fire with Jacob Eason at quarterback, Oregon could break through in Mario Cristobal’s second season, USC could [insert things we say about USC every year]. But based solely on who produced for each team last year and who’s returning, Utah might start as the front-runner. We’ll see how that changes as we work recruiting rankings and whatnot into the mix.

121. Texas (48 percent)
119. Georgia Tech (49 percent)
118. Washington (50 percent)
116. Ole Miss (51 percent)
114. Kentucky (51 percent)
112. Kansas (52 percent)
110. Boston College (53 percent)
109. NC State (53 percent)
108. Stanford (54 percent)
106. Duke (54 percent)

Some of those, you could have guessed. Kentucky loses LB Josh Allen and most of the primary reasons for 2018’s breakthrough. Ole Miss’ offense takes a big hit. Washington loses Jake Browning, Myles Gaskin, and most of its secondary.

That top name might come as a surprise.

Fresh off of a 10-win campaign — the school’s first since 2009 — and a Sugar Bowl win, Texas is all but guaranteed to begin 2019 in the preseason top 10. The Horns bring back quarterback Sam Ehlinger and receiver Collin Johnson as headliners, plus the fruits of successful recruiting.

They do not, however, return their leading rusher (Tre Watson), leading receiver (Lil’Jordan Humphrey), three honorable mention all-conference offensive linemen, their top three tacklers on the defensive line, their top two linebackers, and three of their top five defensive backs, including corner Kris Boyd, who led the team in havoc plays (tackles for loss, forced fumbles, and passes defensed).

At 48 percent returning production, the Horns aren’t in the “guaranteed regression” range like, say, UAB and Fresno State. But Tom Herman’s recruiting classes are going to have to break through quickly if Texas is to live up to expectations.