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 2022, 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.

In Week 5, we revisited one of my favorite findings. We know that early-season overperformers and early-season underperformers tend to regress, but every year, I test the data and confirm that preseason ADP is still as predictive as early-season results even through four weeks of the season. I sliced the sample in several new ways to see if we could find some split where early-season performance was more predictive than ADP, but I failed in all instances.

In Week 6, I talked about how when we're confronted with an unfamiliar statistic, checking the leaderboard can be a quick and easy way to guess how prone that statistic will be to regression.

In Week 7, I discussed how just because something is an outlier doesn't mean it's destined to regress and predicted that this season's passing yardage per game total would remain significantly below recent levels.

In Week 8, I wrote about why statistics for quarterbacks don't tend to regress as much as statistics for receivers or running backs and why interception rate was the one big exception. I predicted that low-interception teams would start throwing more picks than high-interception teams going forward.

In Week 9, I explained the critical difference between regression to the mean (the tendency for players whose performance had deviated from their underlying average to return to that average) and the gambler's fallacy (the belief that players who deviate in one direction are "due" to deviate in the opposite direction to offset).

In Week 10, I discussed not only finding stats that were likely to regress to their "true mean", but also how we could estimate what that true mean might be.

In Week 11, I explained why larger samples work to regression's benefit and made another yards per carry prediction.

In Week 12, I used a simple model to demonstrate why outlier performances typically require a player to be both lucky and good.

In Week 13, I talked about how a player's mean wasn't a fixed target and predicted that rookie performance would improve later in the season.

In Week 14, I mentioned that hot and cold streaks are mostly a mirage and that all players tend to regress strongly toward their full-season averages.

In Week 15, I looked at the disheartening finding that even the best teams only win a title 30-40% of the time.

In Week 16, I explored the tension between predictions that were interesting and predictions that were likely and how I try to walk the line between both.

Our passing yard per game prediction is down to the final week. I predicted that yards per game would be the lowest it had been in more than a decade, and it has been fairly handily. But I also predicted that it would finish below 218.4, and it's looking like it'll come up just short.

Our rookie production prediction cleared fairly handily, though, even with the deck stacked against it. Of the fifteen rookie receivers in the sample, nine saw their production improve over the last four weeks with a median gain of 2 points per game. Here are the six exceptions: Tank Dell, who was hurt in the first game and never registered a reception; Jordan Addison, who had improved by 3 points per game before getting injured on his first reception and leaving early in Week 16; plus Michael Wilson, Josh Downs, Jalin Hyatt, and Marvin Mims.

If we remove Dell from the sample, the rookies rose from 8.56 to 9.61 points per game, more than a full point improvement. Even among rookies who were already productive at the time of the prediction (Puka Nacua, Rashee Rice, Jayden Reed, Zay Flowers, and Josh Downs were all averaging at least 10 points per game) tended to see their production improve, with all but Downs improving by at least two points on their full-season average. As I said, rookies make phenomenal redraft picks because their production tends to peak when you need it the most.

Our "hot" players remain incredibly tepid since our prediction, performing even a hair below their full-season average. This isn't influenced by one or two outliers, either; 17 out of 31 players are underperforming their full season average, which is fairly close to the 50/50 split we'd expect to see if season-long average accurately represented each player's "true" performance level.

Regression and Dr. Ian Malcolm

My first love in fantasy football is dynasty, a format where once players are on your roster, you keep them indefinitely, drafting new rookies every year as a fresh batch of players enters the league. This column naturally focuses on shorter time scales with predictions that are testable over four-week windows, but late in the season, I like to deviate a bit and look at a way that regression impacts my favorite format. By now, the redraft cake has already been baked, so to speak; 80+% of teams are already eliminated from contention, and the ones that are still alive don't have much of a future to look forward to. But dynasty is forever.

In recent years, I've taken the opportunity to look at how incoming talent tended to regress over time, with positions getting younger once a strong crop of rookies entered the league and then older over time as that group aged. The 2017 running back class was the best in history, so top running backs were very young after 2017 and 2018 but much older in 2022 as Christian McCaffrey, Dalvin Cook, Austin Ekeler, Joe Mixon, Leonard Fournette, James Conner, Aaron Jones, and even fantasy-viable role players like DOnta Foreman and Jamaal Williams continued to age.

This year, I wanted to talk about one of my most important beliefs in dynasty and how that belief is shaped by my knowledge of regression. The belief is expressed in some quarters as "talented players tend to shine eventually" or "good players get theirs". Personally, I like to call it the Dr. Ian Malcolm Hypothesis after Jeff Goldblum's character in Jurrasic Park. Confronted by a scientist over his concern that the dinosaurs of Jurrasic Park might begin to breed despite all the dinosaurs in question being female, Dr. Malcolm is asked how that could happen. Malcolm replies that he doesn't know but that "life finds a way".

I often feel similarly about talented young wide receivers. If there's a talented young receiver in a terrible situation, I tend to remain quite high on their prospects going forward. If confronted and asked how I expect them to become productive in such terrible circumstances, I respond that I don't know, but life finds a way.

Why do I believe this? Let's go back to one of the first lessons of Regression Alert. Some factors that contribute to production are intrinsic and remain stable between samples. Other factors are largely noise or luck and tend to fluctuate between samples. Identifying regression is about locating and anchoring to the stable elements while avoiding getting seduced by the noise.

When it comes down to intrinsic factors, there are few things as intrinsic as raw talent. Good players stay good to a greater degree than lucky players stay lucky. Situation changes; talent endures. And to illustrate the Ian Malcolm hypothesis in action, I wanted to look at a single draft class and walk through just how much situations changed over the course of their rookie contract. So let's revisit the top 12 players from the 2020 WR class and compare where they started to where they are today.

• Ignoring for a moment the choices he made and their tragic consequences, Henry Ruggs was drafted in 2020 as a deep threat for a quarterback who wouldn't throw deep. Within his first three years, Derek Carr developed a wicked deep ball, head coach Jon Gruden (who was entering Year 3 of a 10-year fully-guaranteed contract) was fired, Carr was given a monster long-term contract, then benched for Jared Stidham, then replaced with Jimmy Garoppolo, and now Las Vegas is starting a rookie 4th-round draft pick who had 1100 passing yards and 8 touchdowns in college at the time Ruggs was drafted.
• Drafted into a team with Vic Fangio as head coach, Jeudy, in three years, went from Drew Lock at quarterback to Teddy Bridgewater. Then the Broncos fired Fangio, hired Nathaniel Hackett (who has been perhaps the worst offensive coach in the league over the last two years), and traded for no-doubt future Hall-of-Famer Russell Wilson. Then the Broncos fired Hackett and traded for likely future Hall-of-Famer Sean Payton, and now they're benching Wilson.
• It's hard to remember now, but there were questions about how CeeDee Lamb would be able to stand out in a receiving corps that already featured Amari Cooper and Michael Gallup, both of whom were 26 or younger and both of whom went over 1100 yards the year prior. Cooper lasted two years before getting traded to Cleveland, Gallup hasn't been healthy very often or effective when he was, and the Cowboys passing game has been the CeeDee Lamb show for the last two years.
• In just four seasons, Jalen Reagor has gone from Carson Wentz to Jalen Hurts on the Eagles, to Kirk Cousins on the Vikings, and now he has 66 yards with the Patriots.
• Justin Jefferson's situation has actually been very stable during his short time in the NFL, though some of the concerns surrounding him look laughable in hindsight. Some (including the Eagles, who picked Reagor over him) questioned whether he'd be able to transition to outside receiver. The bigger concern was that the Vikings were a low-volume passing offense and Jefferson shared the field with Adam Thielen. Minnesota ranked 30th and 27th in pass attempts in 2019 and 2020, and Stefon Diggs had just forced his way out of town in part over frustration at his lack of involvement; Minnesota has been one of the highest-volume passing offenses in the league, ranking 11th, 3rd, and 4th over the last three seasons.
• Brandon Aiyuk seemed to have a wide-open path to the top of the depth chart in San Francisco, but the emergence of Deebo Samuel in 2021 (1700 yards and a first-team All Pro award) greatly complicated that. Aiyuk went from Jimmy Garoppolo at quarterback to Trey Lance, back to Jimmy Garoppolo, back to Trey Lance, and back to Jimmy Garoppolo, with stops along the way at a rotating cast of 3rd-stringers, before San Francisco took Brock Purdy with the last pick of the draft and turned him into an MVP front-runner.
• When Tee Higgins joined the Bengals, nobody knew whether rookie Joe Burrow would be any good after just one strong season in college. Burrow showed flashes early and Higgins looked set to be his top wide receiver for years to come until the Bengals took Ja'Marr Chase and relegated Higgins to a sidekick role. Then expectations were that Higgins wouldn't be able to produce behind Chase only for the two to develop into one of the Top 3 duos in the NFL.
• Michael Pittman has seen the Colts bring in three different past-their-prime quarterbacks in his three seasons, going from Philip Rivers to Carson Wentz to Matt Ryan. Then the Colts did a complete 180, drafting a talented-but-raw rookie who got hurt within five weeks, then turned to Gardner Minshew, who at the time Pittman was drafted was the starting quarterback for the Jaguars. (I'm not even going to get into that brief stint with Jeff Saturday as the team's head coach.)
• Laviska Shenault was drafted to a team with Gardner Minshew at quarterback, which didn't inspire much confidence. After the Jaguars were terrible in his rookie year, they were positioned to draft one of the best quarterback prospects in recent memory in Trevor Lawrence, which was viewed as a huge boost for Shenault. It might well have been, except Shenault was traded to the Carolina Panthers and instead played with Sam Darnold, Baker Mayfield, and Bryce Young. (Ironically, if you told someone before the 2020 season that Shenault would be playing with Darnold, Mayfield, and Young, they probably would have taken that as a major positive.)
• When he was drafted by the Broncos, it was assumed that K.J. Hamler would have plenty of competition. The team drafted Noah Fant 20th and Courtland Sutton 40th in 2019, then followed by drafting Jerry Jeudy 15th and Hamler 45th in 2020. Fant was gone within three years, Sutton has been constantly injured, and Jeudy has been a massive disappointment. But it wouldn't matter because Hamler wouldn't be able to stay healthy. Today, he's on the Colts' practice squad.
• Drafted to help fill the void left by Antonio Brown, Chase Claypool landed with a Hall of Fame quarterback in Ben Roethlisberger and a wide-open depth chart, competing with Diontae Johnson, James Washington, and JuJu Smith-Schuster. Within three years, Johnson had established himself as a target magnet, and while Smith-Schuster was gone, the team drafted George Pickens to fill the void. Roethlisberger fell off a cliff and is also gone, but it doesn't matter because Claypool is the fifth of the Top 12 receivers who didn't make it to the end of his rookie contract with the team that drafted him. Claypool, like Reagor, is actually on his third team after twice being traded in the middle of the season.

There's nothing really special about the 2020 class; this is just the kind of turnover the NFL sees in four seasons. Three of the twelve receivers have already been traded (one of them twice), and a fourth is in jail. Eight of the twelve played with at least three presumptive starting quarterbacks (not just injury replacements like Jake Browning for Higgins, but actual changes in the preferred starter). Teammates viewed as threats have been hurt, traded, or underperformed. Teammates not viewed as threats have broken out and demanded a larger role.

It's easy to forget all of this when we look back. It's easy, knowing what we know now, to underrate just how much of a threat we thought Michael Gallup posed to CeeDee Lamb's production, how worried we were about Joe Burrow (or, later, about Ja'Marr Chase), how much of a positive many considered playing with Carson Wentz instead of Kirk Cousins to be.

Valuing a dynasty receiver based on what his situation is today assumes that today's situation will be roughly similar to tomorrow's situation. But as you can see, that's just not the case. Situations change significantly more than you'd think. The most talented receivers can ride out those changes, the least talented receivers couldn't succeed even with perfect stability. At the end of the day, it's best to value players based on the intrinsic factors and ignore the situation entirely.

And as the offseason approaches, it's not a bad idea to target a few players you like who are currently in awful situations and trust that everything will get scrambled in a year or two (hopefully for the better). The path to their current Top 15 positions was far from smooth for Michael Pittman and Brandon Aiyuk, but they made it there all the same. Maybe you don't see the route to success for someone like Drake London or Kyle Pitts, but as Dr. Malcolm reminds us, life finds a way.