741: Running Back Prospects Who Produced Like Studs

00:00:02
Speaker 1: This is the lated podcast with with your host JJ zacher rethas JJ zacher reson. What's up everyone, It's JJ zacharyesin in this episode seven hundred and forty one of the Late Round Fantasy Football podcast sponsored by FanDuel. Thanks for tuning in. It's officially prospect season. I haven't done any episodes of the show dedicated to this year's draft class, but here we are. It's the first one. Now. I do this exercise each year a sort of a primer for running back and wide receiver. As many of you know, I've got prospect models for those two positions. I know how sick some of you get when I reference my model, but it's because it's a backbone for all of the prospect work that I do. A lot of effort has been put into it, and it's proven to be a helpful tool. That model has a lot of different inputs, which you can learn about in the Late Round Prospect Guide, which, by the way, you should pre order that right now over on Lateround dot com. You'll get an outline of what goes into those prospect models, and then there are profiles for every single running back and wide receiver who are going to the NFL combine this year. But there are a lot of different inputs you can learn about in that guide, and some of the inputs are production related. On today's episode and then on an episode next week, I'll be focused on those production inputs, hence the title of this episode. I'm going to look at running macks and wide receivers who have been successful in the NFL. I'll look at how they produced within that model in college, and then I'll compare this current draft class to see how players stack up. In other words, I'm going to see which running mas and wide receivers produced like studs in college. My models help predict how a player is going to perform during the first three years of his NFL career from a fantasy perspective. So clearly, if these production metrics are in the model, they should be somewhat predictive for future NFL success, and they are. There's a lot more that goes into these models, and these metrics aren't close to everything standalone, but they at least give us a high level idea about these players. Today, we're gonna focus on running backs, so let's get going. So again, we're gonna look at this year's draft class and compare this class to successful NFL running backs. That way, we can see which players in this year's class produce like studs in college. Will be analyzing through the lens of the three main production metrics and the Z prospect model. That's the name of my prospect model because I'm super creative. The three main production metrics for running backs are best season total touchdown share, best season reception share, and best season total yards per team play. I think all those are pretty self explanatory. When I say best season, I'm just looking at a player's top number within a statistic. So best season total touchdown share is a player's top total touchdown share season, and total touchdown share is simply total touchdowns scored divided by how many touchdowns that player's offense scored. Reception share is the percentage of receptions that the running back had within his team's offense. And best season total yards per team play is taking a player's total receiving and rushing yards and dividing it by the number of rush and pass attempts that the team ran that season. All of these metrics are market share statistics, similar to something like target share that I talk about. All the time on the show or running back rush share, those are market share statistics. The reason they're helpful is because every team is different, and that's especially true in college. You get a wide range of team offenses. Some are really really pass heavy, some are really really run heavy. So this gives those offensive situations a little bit of context. And no, these numbers are not perfect. There are going to be times where a player doesn't see the right volume in his offense because his competition is just absolutely insane. Sometimes the coaching staff is just stupid. But these numbers work in unison to help spit out a stat score within the model. And I can't emphasize enough that this is not everything, especially at running back. The truth is draft capital itself. That leaves the charge with a lot of this stuff. If a running back has an insane production profile, but he's a sixth round pick, he's something that we should have our eyes on as a potential value in dynasty, but you need to adjust expectation. That's what the Z prospect model helps us do. We're just focused on production today. This year's running back draft class as twenty six running backs who are headed to the NFL combine. So those twenty six backs are what we're analyzing. Their average total touchdown share is thirty one point two percent, So the average player in this year's class had a best season thirty one percent total touchdown share. With regards to reception share, the average is ten point six percent, which is actually really good and pretty surprising. That's a really good average for an entire draft class. And then with best season yards per team play, we're at one point sixty one. So thirty one point two percent for total touchdown share, ten point six percent for reception share, and one point six y one for total yards per team play. You're probably listening to that and you have no idea what that means. We still need context for what's actually good, right. That's where the NFL stud sample comes into play. We need to compare these numbers against something. Now. To find that sample, I had to set some arbitrary parameters. What is an NFL stud I looked at all running backs since twenty eleven who had at least two seasons where they scored fourteen or more p Piero points per game. Why fourteen because it usually represents a mid range RB two season. If I went any higher, our sample size of running backs doing well for multiple seasons wouldn't be very large. If I went lower, then we'd be capturing players who aren't actually studs. Now, that left me with thirty NFL running backs. Like most of the data that I work with, the Z Prospect model goes back to twenty eleven. There were some testing with data pre twenty eleven, but the database that I focus on only consists of players who were drafted or went to the NFL Combine from twenty eleven onward. It's a lot of running backs, but players like Adrian Peterson, Marshawn Lynch, they're not in the data set. So when you eliminate players like that from our thirty running backs, you get twenty two of them. Those twenty two backs, those are the ones that we're gonna focus on today. That's our NFL stud sample. Let's look at those players' averages. We already looked at the twenty six running backs that are going to this year's combine and how they fared within the model's three main production metrics. But let's see how these twenty two stud running backs did in terms of best season total touchdown share. These running backs average about a thirty six percent rate, so roughly five percentage points better than this year's class. This year's class had an average reception share of ten point six percent. Like I said, that's quite good, but it's not as good as the NFL stud sample. They're at twelve and a half percent. And then total yards per team play this year's class average a one point six to one rate, the NFL stud sample one point nine to nine. So unsurprisingly, these successful NFL running backs had better production in college within those three main metrics. The question is, though, which players in this year's class have good looking production profiles. So let's break this down by each metric and see if there are any higher profile backs in this year's class who have red flags within that metric. We'll start with best season total touchdown share, which is honestly the least important of the three metrics in the model. Just three running backs and our stun NFL sample had a best season total touchdown share below twenty percent, and there have only been four running backs since twenty eleven within the Z Prospect model who have had one season of fifteen or more PPR points while finishing below that touchdown share threshold. So only four running backs had a fifteen plus PPR points season while having under a twenty percent total touchdown share mark. And keep in mind this model consists of every drafted running back and or running back who went to the NFL combine during this time. There are one hundreds of running backs in his database. Only four of them had below a twenty percent total touchdown share with one season during their first three in the league of fifteen or more PPR points. The three running backs in the stud sample that were below this mark Devonte Freeman, Josh Jacobs, and Chris Carson. Within this year's draft class, Jamier Gibbs is a little bit over that twenty one percent mark, so he didn't get below the twenty percent, but he's awfully close. One player who definitely has a red flag here, though, is Zach Evans. Evans went from TCU to Ole miss He played his final season in Mississippi this past year. In every season, he was sharing a backfield with a talented running back and that can help explain his mediocre production profile. But the production profile is absolutely lacking, and that's the focus today. It's production. Since we're on the topic of Evans, i'll share this about his production profile. Within the model, he failed to reach a twenty percent touchdown share. We already know that he also failed to reach a six percent reception share, and he failed to reach a one point one total yards per team play rate. Chris Carson is the only real hit from a player who was under those marks. I'm not writing off Evans at all, because we've got a long way to go and he has a lot working in his favor outside of production, but it's at least a flag. It's at least something that we should keep in mind. Roshan Johnson is the running back with the worst best season total touchdown share in this draft class. He also played in a backfield with Bejon Robinson, who's probably the best prospects in Saquon Barkley, So can we really expect that much? Now? I'll have more of a breakdown on Johnson within the prospect guide. There's actually a lot to like about his profile. When you give that profile the context that it needs, he some of the model will inherently not like that much though. Moving on, let's talk about reception share. I've gone over this a lot of times in the past on this show, but reception share gets signal likely because it shows versatility. It doesn't mean the running back can't catch passes. There's nothing more tilting on fantasy Twitter than when a guy who had a low reception share in college or a bad receiving profile in college, when he makes some great grab in an NFL game and then people are out there and they're like, but I thought he couldn't catch passes. No, that's not the point. The point is not that a player has bricks for hands. The point is that players with high reception shares are just good receiving profiles are naturally more well rounded and better at the game, and when you're better, you have a higher chance of sustained fantasy production. Not only that, but it's rare for a bad receiving profile to turn into a good one. At the NFL level, we'll see guys go from really good collegiate receiving profiles to mediocre ones. As a pro that happens, but the opposite doesn't happen that often. Within our NFL stud sample, only Chris Carson and Derrick Henry had a sub six percent best season reception share. Josh Jacobs was in that six to seven percent range, and then everyone else was over the eight percent mark. This class does have decent receiving profiles, which is why the reception share average isn't bad. But the player with the worst best season reception chair is Dwayne McBride out of UAB. McBride is a beast on the ground, like one of the best rushing profiles that you're gonna find, but the receiving profile is just not there. He caught a grand total of five passes during his three seasons at UAB five his best season reception chair was just one point sixty five percent. Here's a list of players in the Z Prospect model to score ten PPERO points in a single season over his first three in the league with that lover reception chair. Yeah, that's the list. We haven't seen it before. The best player with that lover reception chair was probably Jonas Gray. Again, I'm not totally writing him off, but I'm a sucker for a good receiving profile. McBride's probably not someone I'll aggressively be on this year. Total yards per team play gets more weight in the model than something like touchdown share does, and really there aren't any real problems among the higher end running within this metric. I've already touched on Roshan Johnson and Zach Evans, and they're towards the bottom of the list in total yards per team play. But otherwise there aren't really high end running backs that are that low. I will say only three studs were below the one point four to five mark within this metric, Josh Jacobs, Chris Carson, and Alvin Kamara. Kamara was lower in college, largely due to the fact that he wasn't that much of a rusher. His receiving profile made him decent in the model. There's one running back who's just below that one point four to five mark, and that's Tank Bigsby. He's at one point four to four, so he's right there, but let's talk about him. Bigsby had solid production at Auburn, but he wasn't seeing the type of volume on the ground that some higher end running backs tend to see. The good news is that he had an eighteen point nine percent best season reception. Chair that's not an easy thing to hit as a bigger body back at least one that's over two hundred and ten pounds. He wasn't super efficient through the air, though, which is why his total yards per team play rate isn't all that high. We've I've seen two hundred and ten pounds running backs get to a seventeen percent eighteen percent reception share in the past. But the ones who really thrived, the ones who really hit, were better in total yards per team play So on one end, for Bigsby, you could look at someone like John Kelly as a comp. Kelly was a later round pick who I really liked. But another player who shows up as a comp for him is DeMarco Murray. Murray was explosive. So we'll see how Bigsby tests. Essentially, he has a three down skill set, but we'll see how it all translates to the next level. Overall, the Z Prospect model does like him. Now's the fun part. Let's take the averages from our stud sample. A thirty six percent best season touchdown share, a twelve point five percent best season reception share, a one point nine to nine best season total yards per team play rate. Let's take those three numbers and let's see which running backs from this year's class exceed those marks. When I filter things out, I get two players, and one of those players is no Bjhon Robinson. Because Robinson's best season in reception share was twelve percent, not twelve and a half percent. The first of the two backs is Sean Tucker. He had a thirty seven percent best season total touchdown share, a fifteen point seven percent best season reception share, and he had a two point twenty six best season total yards per team play rate. Now, he did have some issues with pass catching despite that strong reception share, and his valuation seems to be all over the place right now. So draft capital, like with any running back, is going to go a long way. My prospect model is built with draft capital in mind because it's got an inherent evaluation in it. So if Tucker slips and goes later in the draft, his prospect score will reflect that. Even if he has good production, We've seen that type of profile be hit or miss in the past. Aaron Jones a hit, Jeremy McNichols not a hit. Jay A Jaii had a great production profile and he was sort of fine. But we've seen this kind of profile before, and we've seen a variety of results. But this goes back to how I view this stuff. It's something that I've said for years. Good fantasy assets in the NFL were almost always productive in college. But productive collegiate players are not always productive in the NFL. And people love to write off the production thing when they see productive collegiate players not do well in the NFL, But that's not how it works. We just know that production does matter because successful players in the NFL were typically productive in college. The other player to hit all these thresholds is a much more difficult valuation. It's duce Von. Quite simply, he's tough to evaluate because he's five foot six. We just haven't seen a ton of guys that size enter the league. My model's top comparable for deuce is Tarreek Cohen. Vaughn was better in college, though, and he has a seriously insane production profile, like insane, but you have to wonder if he'll be utilized in a similar way as Cohen at the next level. And I would imagine that once rookie drafts hit, people will dismiss vaugh because of his size and understandable, so so I would assume that size, that lack of size will end up being embedded in his price. But he, like Sean Tucker, he hit all the marks. Now, if you were to loosen some of these parameters a bit, that's when you'd get the players like Zach Charbonnet and Devon a Chin. So don't think that those guys are in bad spots or anything. They're the next guys who show up in the model. Production wise. Now, there are obviously other players who are worthy of shout outs on today's show, and I'll talk about those players not just in the Prospect Guide, but on this show in the coming weeks, and you all can ask whatever you want to ask me for mailbag questions each week two. My overall feel is that this is a pretty strong running back class. We have an elite prospect at the top two. It's a good year to need a running back for your Dynasty team. Next week we'll see what the model says about wide receiver. 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, and don't fit to follow me on Twitter at late on QB. Remember to check out Lateround dot com and pre order that Prospect Guide. Thanks for listening to everyone. I'll talk to you later in the week.