520: Team-Level Regression

00:00:02
Speaker 1: This is the Label Podcast with your host JJ zacher Resa j Jake Zacherson. What's up everyone, It's JJ Zacharyesen, the editor in chief at fandle and at number fire dot com in this episode five and twenty of the Late Round podcast so many shows that are part of the FanDuel podcast network. Thanks for tuning in. I've said it time after time, but regression is everything in fantasy football. The game of football is a pretty small sample game, and crazy things can happen in those small samples. But when fantasy managers are making decisions off of the things that they see in the statistics that they analyze within these small samples, that can cause issues. Now, fantasy managers are getting a lot better, and I mean a lot better at recognizing regression. We see that reflected an average draft position each and every year. Back in the day, a player having some insane outlier season would have almost guaranteed a high draft slot the following year. The public is sharper now though that doesn't happen at the same rate. Even still, it's good to spot where regression can occur year over year because it'll give you a better grasp on both players and teams, it'll make you a better fantasy football player, and the team part is important here. I think most people look at player level regression, but they'll often ignore the team element. But everything on the projections front starts at the team level first, and there are some really easy ways to spot team level regression. I've done this episode the last couple of seasons, so some of this may sound familiar. But on a team level, what I think is easiest to look at when it comes to regression is touchdowns and play calling tendencies. On the touchdown front, you can simply look at total touchdowns scored, or you can look at pass to rush touchdown ratio, or to put that in another way, how those touchdowns are being converted. On the play calling front, I like to look at pass to rush attempt ratio. With both touchdowns and play calling numbers will tend to regress to a mean. For example, from twenty eleven through twenty nineteen to the last nine NFL seasons where we have next season data, I've had twenty eight seasons where a team scored fifty or more touchdowns. Of those twenty eight, only three were able to increase their touchdowns total. The following year, and the average decrease was eleven point four touchdowns. So just from that simple way of looking at things, you can already say to yourself, Hey, if a team was really good at scoring touchdowns one season, maybe we shouldn't automatically project that team to score just as many the next season, because chances are that team did so well because they were a little bit fortunate. Whether that was because of their schedule, their lack of injury, whatever the case, they were likely a little bit lucky, and we don't want to bank on something like that. The same thing happens with play calling too. Over the same nine year timeframe, there have been thirty teams to finish the season with more rush attempts than pass attempts. Of those thirty teams, only three of them actually became more run heavy the following season. You can probably tell why this would be helpful knowledge to have in your brain before your fantasy drafts. If a team did really well or was placed at the extreme in one of these categories, it can change the dynamic of the players on those teams. In fantasy football, I focus on touchdowns and play calling because those two areas are the two that drive fantasy football, the most touchdowns are the biggest scoring boosts that you can get, and play calling can dictate volume or at least how volume gets divvied up in an offense. So on today's show, I'm going to walk through some regression candidates at the team level from a play calling and a touchdown perspective. Next week, then I'll dig into individual players before we get into the data. Basketball fans, the NBA Conference Finals are happening right now, and Fandel is offering a special new user bonus. Place a bet on any single team's money line and you'll receive exclusive plus three thousand odds for that wager. All you have to do is head over to the Fandol's Sportsbook, create an account, and take advantage of this amazing NBA playoff odds boost. Maximum bet of five dollars. Must be twenty one years or older in President of Colorado, Iowa, Indiana, Michigan, New Jersey, Pennsylvania, Tennessee, Virginia or West Virginia. New users only, first row money wager only ten dollars first deposit. Require restrictions apply see fall terms at sportsbook dot fandle dot com. Problem called one hundred and fift too forty seven hundred Colorado, one hundred bets off in Iowa, one hundred nine with it in Indiana, one hundred two seven zero seven one one seven for confidential health Michigan. One hundred Gambler in New Jersey, Pennsylvania, and Virginia the Tennessee redline at one hundred and eighty nine nine seven eight nine, or visit one hundred Gambler dot net in West Virginia. Let's kick things off with touchdown scoring first, because this is easily one of the more difficult things to figure out. Going from twenty twenty to twenty twenty one. Last season, we had a huge uptick in scoring across the league. There were over fourteen hundred offensive touchdowns scored, the highest from twenty eleven through twenty nineteen. So the nine seasons prior to last year was twelve hundred and eighty six touchdowns. So the question is is this a trend or is this something that's just naturally going to regress. And I think it's a little bit of both. I think high scoring is here to stay to some degree because offensive holding penalties were down last year that helped contribute to it. According to data from NFL Penalties dot Com. We had four hundred and seventy seven offensive holding calls last year. The year before that it was seven hundred and thirty two. Those calls can completely blow up drives and that limits scoring. Teams are also just getting smarter. They're becoming more aggressive on fourth down, they're using more play action, they're doing things to help their offense. Now, it's hard for me to just say that this type of scoring is here to stay when there was a very obvious thing that happened in twenty twenty that didn't happen in other seasons, and that's the pandemic. Should we just shrug off the increase in scoring as coincidence that it coincidentally happened during a season where there was a pandemic, where there wasn't any real prep for these teams in July and August, and where there weren't any fans in the stadiums for the majority of the season. The lack of prep intuitively can hurt defenses. They need to work more cohesively as a unit because they're not the ones who know what's coming next. And the lack of fans can't be understated either. Opposing quarterbacks don't have to deal with noise at the line of scrimmage, and more control would hypothetically lead to better numbers. So I think we're in store for some regression over last year. Maybe we don't get to twenty nineteen levels, but I'd be surprised if we saw as many touchdowns in twenty twenty one as we did in twenty twenty. Now I bring this all up because last year, nine teams scored fifty or more touchdowns. As I mentioned earlier, from twenty eleven to twenty nineteen, that happened twenty eight times, So we went from it happening about three times per season to nine times last year. This doesn't mean the data that we're working with is irrelevant. It really is just another example to show you, guys how absolutely insane twenty twenty was, and if you're building projections off of that twenty twenty season, that seems a little naive. One team that stands out in particular is Green Bay. Though they scored sixty four times in twenty twenty, that was five more than the second place Titans. Now, like I said, we shouldn't necessarily expect the same type of regression that we've seen in history to happen to Green Bay. What I mean by that is it still seems logical to believe that the NFL will see more scoring than it did pre pandemic, because there are reasons for it. So looking at history and saying Green Bay will regress by X amount of touchdown, that's not necessarily fair. But but I will say only the twenty thirteen Broncos and the twenty eighteen Chiefs scored sixty four more touchdowns since twenty eleven. Those teams saw their touchdowns decrease by sixteen and twenty the following year. And if you want to adjust that for the year to be more precise with Green Bay, we could do that, and twenty thirteen, that Broncos team scored about thirty three more touchdowns in the league average. That's insane. The following year, though, they scored roughly eighteen more touchdowns in the league average, so there was a decrease by about fifteen touchdowns there. With Kansas City, they went from outperforming the league average by twenty six all the way down to seven, so it's a decrease of about nineteen. Last year green Bay outperformed the league average by twenty touchdowns. We've had fourteen teams from twenty eleven through twenty nineteen outscored their league average by fifteen or more touchdowns. Of those fourteen teams, just one was able to increase on that differential the following season, and the drop in differential was by for fourteen touchdowns. In that case, the expectation would be green Bay outperforming the league average by about six this year instead of twenty, which means even if we're adjusting for the changes that we saw offensively last year, it still may not be easy for the Packers to get to fifty touchdowns this year. And this is obviously with all the Aaron Rodgers drama pushed aside. Now, obviously there's an extra game in twenty twenty one, and when I do this analysis next year, I'll have to start looking at it on a per game basis. But for now, let's just work within the sixteen games, because the concept is just easier to grasp and it's all relative anyway, me saying green Bay will decrease in touchdowns means that I'm expecting that to happen on a per game basis for the Packers. I actually don't think they'll reach sixty four touchdowns in twenty twenty one, even with that extra game on this episode, though, it's more about the idea, it's less about the exact projection, if that makes sense. Now, if you're curious, the nine teams that scored fifty or more touchdowns last year were Baltimore, Kansas City, Minnesota, Seattle, Buffalo, New Orleans, Tampa Bay, Tennessee, and Green Bay. The teams on the higher end that I'd be a little bit more concerned about from a regression standpoint, or Tennessee, Tampa Bay, and New Orleans than obviously the Packers. The Packers are the one to watch most though. Now on the complete opposite side of the spectrum, you've got the low touchdown scorers. In particular, you have both New York teams. The Jets and Giants, both scored just twenty five offensive touchdowns last year. That happened just twenty one times from twenty eleven through twenty nineteen, and in all but two of those cases, the team saw either an increase in touchdowns or the same amount of touchdowns the following year, and the two teams who saw a decrease saw just a one and a two touchdown decrease. If you want to set the parameters a little bit more conservatively, you can look at thirty touchdown teams. We had seventy two of those from twenty eleven to twenty nineteen. Only eighteen percent of those teams saw a dip in touchdowns the following year, and none of the teams in that eighteen percent decreased their touchdown total by more than four We've had nineteen teams over this timeframe with touchdown totals from twenty four through twenty seve so right in that twenty five range. And among those teams, the average increase from one year to the next in touchdown total was eleven point seven. Now, one of those teams was actually the twenty nineteen New York Jets, who went from twenty five touchdowns to twenty five touchdowns. That's really not a common thing to happen, and I'd expect both the Jets and the Giants to see an uptick and touchdowns for a lot of reasons. Natural regression is one, but so is effect that they'll see regression enhanced. Because they scored so few touchdowns in a really inviting season for touchdowns, it was a high scoring year, and there were changes there too. This offseason. The Giants aded Kenny Galladay, most notably, they get Saquon Barkley back. The Jets have a completely new coaching staff and team. There are a lot of reasons to be far more bullish on the New York teams in twenty twenty one versus how they performed in twenty twenty. Now, like I said before, you can also look at how teams are scoring touchdowns, and what was interesting about twenty twenty was that with the increase in touchdowns scored, we actually saw a decrease and passed a rush touchdown ratio. To put that in another way, we saw teams running for touchdowns at a higher rate than we've seen before. In twenty twenty, the average pass to rush touchdown ratio was one point seven to seven. That's the lowest since twenty eleven by far. What that means that we don't have really obvious regression candidates on the high end like we have with Jacksonville in twenty nineteen. Let me explain what I mean by that. The Jags two years ago had twenty four passing touchdowns to three rushing touchdowns. There was obvious regression coming, and it happened. They increased their passing touchdown total by one, but they increased their rushing total by six. That gave them a much lower pass to rush touchdown rate. But I'm more concerned about the team's last season who are on the lower side and passed to rush touchdown ratio. Specifically three teams with the ratio under point nine. Those three being the Patriots, Panthers, and the Washington Football team. The Panthers are an interesting case because they were a positive regression candidate in this category last year and nothing changed despite a new coaching staff and no Christian McCaffrey. For most of the year, the Panthers were very run oriented with their touchdown scoring. Sometimes that happens. Sometimes regression doesn't fully hit, but I'd still expect that regression to hit at some point. So the Panthers, who had a point eight to four pass rush touchdown ratio last year, should see that number increase in twenty twenty one. Why is that the case, Well, we had thirteen teams from twenty eleven to twenty nineteen with a ratio under one. They scored more touchdowns on the ground than through the air. The only team to become even more run heavy at scoring touchdowns year over year was last year's Panthers, and they went from a point eighty five ratio to a point eight four ratio. In fact, they're the only team with a pass to rush touchdown ratio under one point one to not see an increase the following year. To put this another way, those teams, the Patriots, Panthers, and the football team, they're going to see more touchdowns come through the air than on the ground in twenty twenty one when compared to twenty twenty. I'll get more into this on next week's show with the individual player regression candidates, because there's one in particular who really stands out. Now, lastly, let's touch on the play calling side of things. There were three teams last year who had a pass to rush ratio under one, meaning they ran the ball more than they threw the ball. Those teams were Tennessee, New England and Baltimore. No surprise, I would still expect those teams to try to run the ball a lot in twenty twenty one. But for the record, as I stated earlier, among the thirty teams that ran the ball more than they threw the ball from twenty eleven to twenty nineteen. All but three increase on that ratio the following year, So twenty seven of thirty teams became more pass heavy the following season. And guess what. All three of those teams that didn't see an increase in pass to rush attempt ratio had Greg Roman coordinating them. It happened twice in San Francisco and once in Baltimore. I do think Baltimore is going to be a little bit more pass heavy given their offseason moves, but I'm certainly not in the camp that thinks that it's going to be significantly different than the last couple of seasons. To be honest, I'm more interested in the one team that was super pass heavy last year, and that's the Jacksonville Jaguars. We know that bad, bad teams are generally more pass heavy than good teams because bad teams play from behind, and when you're playing from behind, you're throwing the ball more to catch up. Last year, no team had a higher pass to rush attempt ratio than the one win Jags. Their ratio is one point eight three, so they threw the ball one point eight three times for every rush of the twenty six teams with a ratio of one point seven or greater since twenty eleven, just three of them saw that ratio get even higher the following year. This also makes some sense because Jacksonville didn't want to be as pass heavy as they were last year, and that's the case for a lot of these teams. And I know that because in neutral game scripts their ratio is ten highest in football. Their overall ratio was a lot higher than their neutral script ratio or when games were close, and that neutral script ratio is much more predictive for year end plus one play calling than the overall ratio is. So I would absolutely expect Jacksonville to lean more on the run this season, just naturally. They've got enough change on that team that we can't just purely blame any sort of decrease on regression, but I'd expect that to happen. Nonetheless, don't just use last year's numbers to project the Jacksonville Jaguars this season. That's a terrible idea on so many levels. Now. I know I threw a lot of data at you in this one, but the data was more so proof than anything else, proof that if you have a team that did something at the extreme, whether that's the number of touchdowns a team scored, how they scored them, or how they called plays. When you have a team at the extreme, that team tends to regress the mean the following year. Use that to your advantage, and always remember whatever happens at the team level will trickle down and affect the player level. I'll get to that next week. That's going to do it for today's show, though. Thanks to all of you for listening. If you've got subscribed to Late Round podcast, make sure you are by searching for it pretty much anywhere podcasts can be found, and don't forget to follow me on Twitter at Late Round QB. Thanks everyone, I'll talk to you later in the week.