869: Bailing on Unproductive Rookies

00:00:02
Speaker 1: This is the labor of podcast with your host JJ Zacherathas J J Zacher, Retha Jess. What's up everyone, It's JJ Zachary sin in this episode eight hundred and sixty nine of the Late Round Fantasy Football podcast sponsored by Underdog Fantasy. Thanks for tuning in to anyone who pre ordered the Late Round Prospect Guide. First off, thank you, I greatly appreciate your support. But second, the guide should be in your inbox. I send it out on Sunday to everyone and anyone who now goes to lateround dot com to purchase it. You're paying the full price of fourteen ninety nine, but the pre draft version of the guide can be instantly downloaded once the NFL Draft is finished. If you had purchased the guide, you'll get an email with the updated version of the guide. That updated version will go out the weekend of the draft. One of the things that's featured in the Prospect Guide is my Year two model. I feel like I've mostly been talking about the prospect stuff over the last couple of months, but the Year two two models super important too. That model was built with an old Late Round podcast episode in mind. Back in the day, I did an episode called Bailing on Unproductive Rookies. It was episode three hundred and forty three. The concept of that episode was about how we as fantasy managers might want to consider getting rid of second year players who are very unproductive as rookies. And when I say unproductive, that's in the context of other rookies because essentially, when players had bad rookie seasons, when they performed below expectation, they very rarely turned into huge league winning fantasy football assets. Rather than just guessing which second year players we should trade away, I eventually built the year two models. They give me a score that tries to predict how well player are going to perform in year two and year three of their NFL careers. If the score is low, then I know, based on history, that player might struggle. If it's high, the opposite happens. I'm going to be talking about that a little bit more today as sort of a rebrand of that older episode than I once did, because through the years, my analysis has changed, the way I approach things has changed, So it's a topic that I think is important to revisit before get into all that stuff. Though, don't forget to check out Underdog Fantasy. It's the best and easiest place to play fantasy sports. You can draft a fantasy football team in minutes and with their best ball format. There's no waivers, trades or lineup setting. There's no league management. After you draft, you can sit back and relax. Drafts are happening right now over there, and you can use the information that I share on this podcast or in my prospect guide to help you draft. If you're new to Underdog, make sure you use promo code late Round when you sign up and you'll get a deposit match up to one hundred dollars. That's promo code late Round. Head to Underdog Fantasy dot Com or download the Underdog Fantasy app to sign up. I want to start out today's discussion by talking about the year two model at a high level, because I think when I share some of these probabilities, you're gonna be pretty surprised when you really dig into these results. The year two model both at running back and wide receiver. It looks at a player's prospect score, and then it looks at a bunch of metrics from his rookie season in the NFL. It combines all of those things to get a year two score, and like I said that year two score is trying to predict how well a player is going to perform in fantasy football and year two and year three of his NFL career. I've got some cool charts in the Prospect Guide that show historical hit rates at different year two model scores. For instance, of the wide receivers drafted from twenty eleven to twenty twenty one, and I'm stopping at twenty twenty one here because those players have now played three years in the league. But of the wideouts drafted from twenty eleven to twenty twenty one, of the ones who had a year two score between ninety five and one hundred, sixty percent of them gave fantasy managers a sixteen plus PPR point per game season, and either year two or year three of their careers, forty percent of them posted an eighteen plus PPR point per game season, which is a very very good season at wide receiver. When you look at the ninety to ninety five group, their rate of scoring sixteen or more PPR points per game in either year two or year three it drops to thirty two percent. So it's sixty percent for the elite group, and then thirty two percent for this next one. Then that number continuously falls as you get lower and lower in year two score. What's wild is that the year two database has never seen a wide receiver who had a year two score below eighty get to sixteen or more PPR points per game in their second or thirty year in the league. For context, sixteen PPR points is usually a low end wide receiver one high end wide receiver two season. So anyone since twenty eleven who's been below that eighty mark, you can kiss upside goodbye. The biggest name from last year's class that's right there, that's right at eighty, is probably Jonathan Mingo. Trey Palmer is under that eighty mark as well. The chance that these players actually reach some status of being super super usable in fantasy football, those chances are slim. I highly doubt that Jonathan Mingo is gonna be a wide receiver one in fantasy over the next two years in just one of those two seasons. But even players in like the eighty to eighty five range, they have rough odds only four percent of them have gotten to sixteen or more PPR points per game this year. That's toa Mario Douglas, Quentin Johnston, Marvin Mims, and Jalen Hyatt. Again, all of this stuff is in the Prospect Guide. You can get profiles and all of these guys too. I'm just giving you an idea of how clear cut it can sometimes be to bail on these unproductive rookies. Running Back is a little bit different. It's because running back just has more natural volatility. When you're trying to project how well a player is gonna do and his second or third year in the league. Not only are we measuring off of one season, so a player could just see a massive workload one year and sort of skew these results. But running back volume and fantasy points come down to a coaching decision. If the coach wants a split backfield, he gets a split backfield. Just ask b John Robinson. Those decisions are hopefully driven by talent, yes, but it's not as seamless as it is a wide receiver. Running back. Year two scorers sort of fall into three clusters. Any player over a score of ninety is in a pretty good spot, especially if they're over the ninety five mark. Players over the ninety five mark like Bjean Robinson and Jamier Gibbs. Those running backs have scored eighteen or more PPR points in either year two or year three at a forty four percent clip. A player with a score from seventy to ninety, that's the second cluster. He's more up in the air. These are your standard RB twos in fantasy football. Generally speaking, we rarely will get a highly highly elite player from that group, but we can definitely get RB one seasons from this cluster. We just saw that with Rashad White this past year. And then a score under seventy, a year two score below sete that's where we don't see many hits. To be more specific, from twenty eleven to twenty twenty one, only two running backs with year two scores below seventy gave us a sixteen plus PPR point per game season and year two or year three. Those were CJ. Anderson and Miles Gaskin. Now, if you expand that and you look at twenty twenty two and twenty twenty three as well, you get Kyrien Williams as part of that group two. So there's three of them, three running backs who were below that seventy mark to give us an RB one season in year two or year three, when you're looking at fourteen or more PPR points per game more RB two type seasons, you add Damien Harris and Chris Carson to the list, and Damien Harris was at fourteen point zero one points per game. Needless to say, none of those running backs, aside from maybe Chris Carson, had year in, year out high end production. And we'll see what happens with Kieren Williams. The models aren't going to capture everything. I'm not trying to sell it that way, but the year two model is done a pretty good job at capturing which running back should at least be in our consideration sets. And this goes back to what I talked about a couple of months ago about Chase Brown. Tank Bigsby is under the seventy mark this year, which is no surprise, but Chase Brown is two. I don't think it automatically means you have to fade the guy, at least in redraft in early underdog drafts. His ADP has been reasonable enough when you consider everything that's been up in the air with a Cincinnati backfield, and now Joe Mixon was released, so i'd imagine that ADP is gonna rise. It's also a little different from a best ball perspective when you know Brown has some big playability, but regardless of that, in terms of investing in him long term, he'll need to be somewhat of an outlier in order to give you higher end production even without mixing around. Zach Moss is there anyway, and Zach Moss might be better than Joe Mixon. Chase Brown isn't the first running back in history to have an opportunity in his backfield with this low of a score. Anything is obviously possible. That's not the point here. The point here is probabability. The year two model can help us spot these unproductive rookies and it can make us feel more confident in moving on from them. That's one of the hardest things to do in fantasy football is to move on from players because you're afraid that they're going to blow up on someone else's team. And it's not going to be one hundred percent perfect, but it's definitely going to help us across multiple moves. Probability is on our side when we use these models. Now. Something I only started researching recently about this is the degree in which a player's score falls from year one to year two. Let me explain a player gets some prospect score when he enters the league, he gets a ZAP score. Because the model I use is called the ZAP model. That ZAPP score will either increase or decrease when compared to a player's year two score because both scorers are out of one hundred. So, for example, Zay Flowers enter the league with a ZAP score of ninety point four. His year two score is now ninety sive six point eight, so he saw an increase of six point two points from year one to year two. Clearly, someone with a high ZAP scorer will typically drop in year two score because it's difficult for him to get any better. If he has a ZAP score of ninety nine, it's really hard to get better than ninety nine in a year two score, even if that ZAP score is part of the year two calculation. Jon Robinson ejamieer Gibbs had ZAP scores above ninety eight. They had good rookie seasons, so their year two scores are actually just as high, but it's a lot harder for them to maintain their score versus someone who had a ZAP score of like sixty or seventy. What I was curious about was what happens when a player sees a fairly dramatic shift downwards in ZAP score to year two score. In other words, what happens when a player really underperforms during his rookie season versus how his prospects score expected him to perform. Let me give you an example. This new ZAP model it liked Jackson Smith and Jigba even more than the last model Itteration did, which, Yeah, looking like a not great call right now. That's gonna happen though, And fortunately my rookie rankings last year had Gibbs above JSN. But regardless of all of that, JSN had a pretty good score in that ZAP model. It was ninety eight his year two score ninety two point eight, so his score dropped by over five points. That type of dip in ZAP score to year two score is not very encouraging at all. Actually, from twenty eleven to twenty twenty one, there were twenty one wide receivers who had ZAP scores above ninety who then saw that score drop by five or more points in the year two model. That may seem like a lot of wide receivers, but keep in mind what I just said. A high score is really difficult to maintain. But get this. Of those twenty one, only one of them scored fourteen or more p PuO points per game in a season and either year two or year three, that's a rate of four point eight percent. Of the wider who didn't see that big of a drop, their rate of getting fourteen or more PPR points per game was almost seventy seven percent. Now you might be thinking, well, that player's year two score is now lower. Jackson Smith and Jigba his year two score is now ninety two point eight. Other players who had a similar ZAP score as JSN, their year two scores are probably higher. That's why we have the year two model in the first place. There's also a big difference between a player who had a ZAP score of ninety one and one who had a ZAP score of ninety eight. The players above the ninety five ninety six range usually have pretty awesome hit rates. This is just an exercise about understanding how players produce once they had disappointing rookie campaigns. But let's focus on players who had a ZAPP score of ninety six or above before seeing that five plus point drop. Those players at wide receivers since twenty eleven include Corey Coleman, Corey Davis, Rashad Bateman, and Jamison Williams. Those are the guys that are grouped up with Jackson Smith and Jigba right now. Those are the players who had ZAP scores above ninety six and then they dropped in year two score by five or more points. As we know, it didn't happen for any of those guys. Maybe it happens for Jamison Williams, but it's not looking great for him, especially because this is trying to predict year two and year three. So if it does happen for Jamison Williams, it has to happen in twenty twenty four. Now you can look at the same type of thing at running back two from twenty eleven to twenty twenty one. So again that gives running backs three years of NFL experience. But from twenty eleven to twenty twenty one, we had twelve running backs who had a ZAP score above ninety then fall by five points in the year two model five or more points of those twelve. None of them had a season with fourteen or more PPR points per game in year two or year three. So they entered the league as pretty good prospects, and then they had bad rookie seasons, and then in year two and in year three they did not perform. Now a couple of them did perform after year three, like mark Ingram that's one of them. Dereck Henry's another. People forget that Dereck Henry had a pretty late breakout. But we're talking mckel, Lashore, Ryan Williams, David Wilson, Darius Geis, Rashad Penny, Ronald Jones, Royce Freeman, Sony Michelle, AJ Dillon, and Cam Akers. Now, if you expand that sample to beyond twenty twenty one, you do get one more player, and that one player did buck the trend, and that's Travis Etn. But obviously circumstances were way different with Etn because he missed his rookie season. When higher end prospects end up giving us a bad season, and those prospects don't hit on the right peripheral numbers either, which is what the year two model is looking at. Those prospects usually don't work out, they're usually the ones that we should bail on, because those are the prospects that still hold some value in Dynasty too, and you do see similar trends at all levels, not just for players who had ZAP scores in the nineties. Every single level of ZAP score shows that when a player sees a significant drop from year one score to year two score, that player has a worse hit rate than others who have similar year two scores. Now, like I said, this stuff is pretty new. The drop and score isn't part of my evaluation right now, at least mathematically. It's more subjective. The year two model is largely gonna catch this stuff. But it's more evidence that when a player fails to meet expectation in a big way as a rookie, it's evidence that you might want to move on from that player as long as the market is giving you the right price. Now, I'm sure a lot of you are wondering what this means for the twenty twenty three class, and I can tell you that at running back it's basically Tank Bigsby. Bigsby's the big disappointment. He went from having a ZAP score of eighty two point nine to a year two score of sixty three point four. He obviously might have the ability to have some spiked week. Maybe Travis Etn gets hurt, but I'm not convinced that it'll ever be a league winning player. Anything can obviously happen, especially at a more volatile position like running back, but we don't see it from players like this. We generally don't see a tank Bigsby like player do very very well in the NFL after such a bad rookie season. Some other one off running backs who saw their ZAP score drop by five plus points compared to their year two score include Duz Vaughan, Izzi, Abana Kanada, Sean Tucker, Evan Hall, and Zach Evans. Now, I understand that injuries play a role here, but injuries do set players back. That's just the reality of the game wide receiver. To me, is more interesting. Again, the year two model is going to help you understand range of outcomes Jackson Smith and Jigba having a ninety two point eight score means there's still some shot that he's a big producer. It's just that when you layer on top the fact that he saw his score drop by so many points, things get shakier, but beyond JSN, you also have Quinton Johnston. The ZAP model actually didn't like QJ nearly as much as the market. He had a ZAP score of eighty nine point seven, which is low for a first round pick, but his year two score is eighty one point six. That drop and score is a really bad sign. You also get big drops from Jalen Hyatt and Cedric Tillman. The model definitely overrated Hyatt given his crazy final season in college, but I had written about his volatility entering the league. He had a really high slot rate, but he didn't profile as a slot guy in the NFL, and he landed in a spot in New York where his team had plenty of slot players. That forced Hyatt to the outside in year one, And even if he does move more into the slot in year two, it's tough to be overly optimistic. Cedric Tillman is just dust. The Browns is traded for Jerry Judy, and Tilman had some of the worst route run peripherals that we've ever seen. He was horrible in twenty twenty three. Sure he's a decent guy, but he was not great on a football field. Turns out those tennis sea wide receivers just weren't it in the twenty twenty three class. No, I don't think all of these players are must sells per se, because anyone you buy or sell that should be related to what the market's saying. But we need to be cautious with these guys. There's no doubt about that. All of those players that I just mentioned have some very very big red flags. You might want to consider bailing on them before it's too late. That's it for today's show, though, thanks to all of you for listening. If you get subscribed to the Late Round Fantasy Football podcast, make sure you are by searching for it pretty much anywhere podcasts can be found. Inter for it to follow me on Twitter at Late Round QB. It's a Late Round Perspectives week and that episode's going to drop Friday morning. I'll talk to you then