Welcome to Regression Alert, your weekly guide to using regression to predict the future with uncanny accuracy.

For those who are new to the feature, here's the deal: every week, I dive into the topic of regression to the mean. Sometimes, I'll explain what it really is, why you hear so much about it, and how you can harness its power for yourself. Sometimes, I'll give some practical examples of regression at work.

In weeks where I'm giving practical examples, I will select a metric to focus on. I'll rank all players in the league according to that metric and separate the top players into Group A and the bottom players into Group B. I will verify that the players in Group A have outscored the players in Group B to that point in the season. And then I will predict that, by the magic of regression, Group B will outscore Group A going forward.

Crucially, I don't get to pick my samples (other than choosing which metric to focus on). If I'm looking at receivers and Justin Jefferson is one of the top performers in my sample, then Justin Jefferson goes into Group A, and may the fantasy gods show mercy on my predictions.

Most importantly, because predictions mean nothing without accountability, I report on all my results in real time and end each season with a summary. Here's a recap from last year detailing every prediction I made in 2021, along with all results from this column's six-year history (my predictions have gone 36-10, a 78% success rate). And here are similar roundups from 2021, 2020, 2019, 2018, and 2017.

The Scorecard

In Week 2, I broke down what regression to the mean really is, what causes it, how we can benefit from it, and what the guiding philosophy of this column would be. No specific prediction was made.

In Week 3, I dove into the reasons why yards per carry is almost entirely noise, shared some research to that effect, and predicted that the sample of backs with lots of carries but a poor per-carry average would outrush the sample with fewer carries but more yards per carry.

In Week 4, I explained that touchdowns follow yards, but yards don't follow touchdowns, and predicted that high-yardage, low-touchdown receivers were going to start scoring a lot more going forward.

If-- as I often claim-- yards per carry is a pseudorandom number generator, then we should occasionally expect the high ypc group to retain a high ypc and the low ypc to retain a low ypc, not because of anything innate to the groups themselves, simply because of a lucky (or unlucky) roll of the dice. So far that's what we've seen; five out of six backs in Group A have remained above 4.5 yards per carry, while five out of seven backs in Group B have remained below 4.0. Overall, Group B has maintained its workload advantage (averaging 25% more carries per game), but Group A remains stubbornly ahead because it's averaging 5.1 yards per carry while Group B is stuck way back at 3.7.

We're still in a great position to win this prediction. The workload advantage is the key, and Group B is doing fine there. The groups' yard-per-carry average might not have trended toward the league average over the last two weeks, but that doesn't mean it won't over the next two.

On my second prediction, I noticed an error in the math last week. I said that Group A was averaging 54.6 yards and 0.72 touchdowns per game; actually, I had failed to account for missed games by Brandon Aiyuk and Jakobi Meyers. I was dividing the group's total by 36 games when, in reality, they had only played 34. Correcting for this error, Group A averaged 57.8 yards and 0.76 touchdowns per game, and their advantage over Group B rose to 7%. (This is good; we wanted Group A's advantage to be larger.)

We predicted touchdown regression, and it arrived with a vengeance. Group A had scored 26 touchdowns in 34 games to start the season, but they combined for just two touchdowns in twelve games last week. Meanwhile, Group B started the year with 6 touchdowns in 33 games; they scored more than that last weekend alone, reaching the end zone 7 times in 11 games. As a result of that touchdown onslaught, Group B staked out a commanding 57% lead in fantasy points per game.

(If this sounds impressive, remember that before we made our prediction, we kicked the five top receivers out of Group B to make things more interesting-- Justin Jefferson, Stefon Diggs, Keenan Allen, Amon-Ra St. Brown, and Deebo Samuel. Those five receivers scored another 7 touchdowns between them.)

Revisiting Preseason Expectations

In October of 2013, I wondered just how many weeks it took before the early-season performance wasn't a fluke anymore. In "Revisiting Preseason Expectations", I looked back at the 2012 season and compared how well production in a player's first four games predicted production in his last 12 games. And since that number was meaningless without context, I compared how his preseason ADP predicted production in his last 12 games.

It was a fortuitous time to ask that question, as it turns out, because I discovered that after four weeks in 2012, preseason ADP still predicted performance going forward better than early-season production did.

This is the kind of surprising result that I love, but the thing about surprising results is that sometimes the reason they're surprising is really just because they're flukes. So, in October of 2014, I revisited "Revisiting Preseason Expectations". This time, I found that in the 2013 season, preseason ADP and week 1-4 performance held essentially identical predictive power for the rest of the season.

With two different results in two years, I decided to keep up my quest for a definitive answer about whether early-season results or preseason expectations were more predictive down the stretch. In October of 2015, I revisited my revisitation of "Revisiting Preseason Expectations". This time, I found that early-season performance held a slight predictive edge over preseason ADP.

With things still so inconclusive, in October of 2016, I decided to revisit my revisitation of the revisited "Revisiting Preseason Expectations". As in 2015, I found that this time, early-season performance carried slightly more predictive power than ADP.

To no one's surprise, I couldn't leave well enough alone in October 2017, once more revisiting the revisited revisitation of the revisited "Revisiting Preseason Expectations". This time I once again found that preseason ADP and early-season performance were roughly equally predictive, with a slight edge to preseason ADP.

And of course, as a creature of habit, when October 2018 rolled around, I simply had to revisit my revisitation of the revisited revisited revisitation of "Revisiting Preseason Expectations". And then in October 2019 and October 2020 and October 2021 and October 2022 I... well, you get the idea.

And now, as you've probably guessed, it's time for an autumn tradition as sacred as turning off the lights and pretending I'm not home on October 31st. It's time for the eleventh annual edition of "Revisiting Preseason Expectations"! (Or as I prefer to call it, "Revisiting Revisiting Revisiting Revisiting Revisiting Revisiting Revisiting Revisiting Revisiting Revisiting Revisiting Preseason Expectations".)

METHODOLOGY

If you've read the previous pieces, you have a rough idea of how this works, but here's a quick rundown of the methodology. I have compiled a list of the top 24 quarterbacks, 36 running backs, 48 wide receivers, and 24 tight ends by 2022 preseason ADP.

From that list, I have removed any player who missed more than one of his team’s first four games or more than two of his team’s last thirteen games so that any fluctuations represent performance and not injury. As always, we’re looking by team games rather than by week, so players with an early bye aren't skewing the comparisons (though this wasn't a factor last year, as all byes happened after Week 4).

I’ve used PPR scoring for this exercise because that was easier for me to look up with the databases I had on hand. For the remaining players, I tracked where they ranked at their position over the first four games and over the final thirteen games. Finally, I’ve calculated the correlation between preseason ADP and stretch performance, as well as the correlation between early performance and stretch performance.

Correlation is a measure of how strongly one list resembles another list. The highest possible correlation is 1.000, which is what you get when two lists are identical. The lowest possible correlation is 0.000, which is what you get when you compare one list of numbers to a second list that has no relationship whatsoever. (Correlations can actually go down to -1.000, which means the higher something ranks in one list, the lower it tends to rank in the other, but negative correlations aren’t really relevant for this exercise.)

So if guys who were drafted high in preseason tend to score a lot of points from weeks 5-18, and this tendency is strong, we’ll see correlations closer to 1. If they don’t tend to score more points, or they do, but the tendency is very weak, we’ll see correlations closer to zero. The numbers themselves don’t matter beyond "higher = more predictable".

Here's the raw data for anyone curious. If you're willing to take my word for it, I'd recommend just skipping ahead to the "Overall Correlations" section below for averages and key takeaways.

Quarterback

Running Back

Wide Receiver

Tight End

Overall Correlations

That's a giant wall of numbers, but all you really need to concern yourself with is the very last row of that very last table. The difference in correlation between preseason ADP and early-season performance is just 0.016, a value virtually indistinguishable from zero. Overall, I've run this study eleven times. Preseason ADP has been more predictive six of those times, while early-season results have been more predictive the other five.

The square of the correlation is said to represent how much of the variation in the second dataset is explained by the variation in the first. By that measure, preseason ADP explains 32.8% of the variation in rest-of-year production. Early-season performance explains 31.0%. (The average of the two factors explains 41% of the variation.)

Starting in 2018, I standardized the format that I kept this data in. That allowed me to combine all data from 2017 to 2022 and run correlations on a significantly larger sample, which is a more robust process than merely averaging the yearly correlations together. Across a huge sample of 533 player-seasons, the correlation between ADP and performance after Week 5 is 0.565. The correlation between early-season performance and performance after Week 5 is 0.555. The two are once again virtually identical. (An average of ADP and early-season performance produces a correlation to performance in Week 5 and beyond of 0.625.)

This combined data set also let me start slicing the sample and testing various splits. For instance, I was asked whether early-season performance was especially predictive, specifically among late-round players. And the answer was... not really.

Looking *JUST* at the bottom 50% of the sample in terms of preseason ADP, here are the same correlations—
ADP: .436
Early points: .463
Avg of both: .516

Maybe a teeny edge to early performance? But that’s so close I’d call it a wash.

— Adam Harstad (@AdamHarstad) September 10, 2023

Is early-season performance especially predictive just among players who are early-season disappointments? Again, the answer is no.

Out of concern I’m unknowingly conditioning on a collider, I repeated looking just at the bottom 50% of the sample in terms of Week 1-4 performance. Correlations—
ADP: .487
Early points: .455
Avg both: .526

— Adam Harstad (@AdamHarstad) September 10, 2023

Over the biggest sample available and across all four positions, preseason ADP predicts stretch performance almost exactly as well as early-season performance does. ADP performs as well as early-season performance when you look at early-round players or late-round players. It performs as well as early-season performance when you look at surprises or disappointments.

In eleven years of testing, a period covering thirteen years of data, I have yet to find a split where preseason ADP and early-season performance weren't equally good at predicting rest-of-year performance. The first four weeks of the season feel like they're incredibly meaningful, but the truth is they only tell us as much as we already knew over the offseason.

(Of course, just comparing preseason ADP to early-season results is a false dilemma; in truth, we should be basing our expectations on a blend of the two. A strict 50/50 mix of both sources predicts rest-of-year performance substantially better than either source alone at every position. And when I've tested Footballguys' rest-of-year projections using this same methodology, they've performed even better still.)

There's no testable prediction this week other than just a general reminder that player performance to date will tend to regress strongly in the direction of our initial preseason expectations.