479: Mailbag 2/5/21

00:00:02
Speaker 1: This is the lad Podcast with your host JJ Zacher Resa Jay Jake zacher reson What's up everyone? It's JJ Zachares and the editor in chief at fandel and at Number five dot com in this episode four hundred and seventy nine of the Late Round podcast so many shows that are part of the FanDuel podcast network. Thanks for tuning in as all of you listening to this. No, the Bucks face the Chiefs this weekend, and just as a reminder, the NFL Big Game Bowl is live on Fandel. For just four dollars and forty four cents, you have a chance to compete for a share of the three point five to five million dollar prize Bowl, including a whopping one million dollar first place prize. For more details, visit FanDuel dot com or download the Fandel app Today. Eligibility restrictions apply. I've got to show this week with fewer questions but longer and more detailed answers. Let's get to it. The first question this week is from at setaz Ff. I see many people disliking Wattle Rugs and DeVante Smith due to their lack of early breakout ages and low dominators. If they all end up being first round picks a very unique circumstance. I think it will corrupt their analytics. What split of analytics slash film slash narrative do you use? Sal later followed up the tweet by saying, corrupt meanings. Splitting the targets with four top tier wide receivers and a solid running back will skew the numbers against them. Dominator is largely a team market share stat and breakout age is based on that. I'm gonna give a long answer to this one, so strap In we do this every year. Every single year. There are players in every single class who have high draft capital, but they didn't necessarily have great college production. Sometimes that was because of injury, sometimes it was because they were in offenses that spread the ball around a lot. Sometimes they were in offenses that didn't fit their skill set. And sometimes NFL teams were just really objectively batted about and they just overdraft a player. And then obviously, in the case of Alabama, sometimes there are just really strong teammates. I think what's really really important for anyone listening to this show who maybe have their doubts about scouting through data. If you're kind of like, eh, I don't really think that you can do this through looking at numbers alone, because there are so many details to a player's profile that you'll miss. You're not wrong, You're really not wrong. My ability to scout players through data has flaws. It has holes, but so does a person who's only watching film. I never understood the idea that tape never lies. When everyone has differing opinions on players all the time when they watch them play. It's not some guaranteed method of evaluating. The reason that I and others look to data for scouting is because it allows you to have a pretty level headed process for figuring out which players are good or not. And remember, my goal with doing this isn't to find the best player necessarily, it's to find the best fantasy player. So attributes that I look for may not be the things that a typical scout looks for. Now to this point, when we're scouting through data, we're finding general trends that enable us to hit on players at a higher rate than our competition, our league mates. This doesn't mean that we ignore any information outside of what a model says. That may sound crazy to some people, that if you have a process, if you have a way of scouting talent through data that you believe in and that you've tested, then why would you stray away from that? And the very very simple answer to that is we're wrong about prospects all the time, So you don't have to just follow a model, even if that model is good, Like I think my model's pretty good. It predicts the first three years of fantasy production far better than draft capital alone does. It gives a lot more context to draft capital, but it's still not this full proof thing. It's impossible for that to be the case. I mean, just look at the NFL draft, look at what teams are doing. Teams have millions of dollars backing them, and they still get a wrong all the freaking time. So my point is when it comes to scouting through analytics, you're going to be wrong, just like you're going to be wrong if you only prospect these guys with your eyes. But because you know things aren't close to perfect, you can subjectively stray away from what a model is saying. And the perfect example of that is with these Alabama wide receivers. All of those guys were playing with other good players, so their numbers likely took a hit as a result, and Waddle in particular got hit with an injury during what would have been his most impactful year. But market share is in relationship to someone's team. So of course, if there's really really good competition on that team, a player's market share numbers won't look as good. I've got a question later on today's show about the too many mouths defeat argument, and I'll talk about this a little bit more. Then, if you've got a random wide receiver who went to Slippery Rock University, then he better have crazy high market share numbers because the quality of his competition on his team is probably lacking. If he's going to play in the NFL, he should have dominated that competition. But with the Alabama guys, you can probably be a little bit more lenient. And that's why in my model, I've got a teammate score, and my model adjust for conference as well. But going back to my original point, the model is there to find trends. If a large percentage of good NFL wide receivers, just as an example, if a large percent of them had good breakout ages, good sophomore year production, a high yards per team attempt rate, then we need to use that as a baseline. If a player doesn't hit that baseliner come close to it, you then ask questions. Those questions were asked last year for Henry Ruggs. Why didn't he produce? What were the reasons his stats score was so low in my model? And some of those reasons were competition based, but some of them weren't. If a player is really special, he's going to beat his competition. That's why the numbers generally show what they show. Alabama's case is more of an outlier, but just generally, if someone's making this argument, do they think that every productive college player had no NFL caliber teammates with him? There are productive college receivers with NFL caliber teammates all the freaking time, all the time. So with these Alabama wide receivers, I'm doing all of the above that I just talked about. I'm plugging them into my model, and then I'm digging into each individual guy. And keep in mind, draft capital is part of this evaluation. I don't know why I wouldn't want to utilize that information. It's important. Draft capital matters to anyone saying right now that Jalen Wattle is gonna suck because of his market share numbers. That's silly. That's an incredibly black and white way of viewing this, especially because data itself can show you that Wattle was good. Wattle as a freshman basically outproduced Henry Ruggs, who was a sophomore, and Henry Ruggs was a first round pick in the NFL draft. Wattle was posting great numbers before his injury this year. If someone's knocking him for that, because their model understandably knocks him for that, then they're not approaching this with a very open mind. Not only that, but Wattle's likely to go pretty high in the NFL draft. Draft capital should be part of that evaluation because players who were drafted early are likely to get more of a shot, and there's embedded evaluation in that too. A team is saying this guy is good. We've put a lot of resources behind this, and we're saying this guy is worthy of a first round pick. That matters. Like I said, my model didn't like Rugs pre draft. Any data related model didn't. Post draft, though, because of draft capital, Ruggs was only ranked behind in no particular order, Ceedee Lamb, Jerry jud T Higgins, Jalen Rager, Justin Jefferson, and he had a very very similar percentile ranking as Brandon Ayuk. So if you know a player is going to be drafted high, you shouldn't write him off. But I think what data is helpful for is showing us that, in conjunction with the draft capital, we can still be lower than the market on those players, or at least some of them, when you can't easily explain away their lack of production. Like I wasn't drafting Henry Ruggs close to the wide receiver one last year in rookie drafts because there were clear red flags to his profile, but he went as the wheide receiver one in the NFL draft. With Wattle, I'm not writing him off at all, even though his numbers don't look great in my model. I mean, he does have a better stat score in my model than Ruggs did, But I think Wattle is a much much better prospect than Henry Ruggs was, and DeVante Smith is on another level. His breakout age is an ideal, but he more than makes up for that elsewhere. I know I went in a lot of different directions with this answer, but I really wanted to talk about these battles that we see all the time on Fantasy Twitter about scouting through numbers. I don't think the right approach is to look at this in such a black and white way, even though I see a lot of people doing that all the time. I just don't think that's how evaluation works at the NFL level. But to answer your question now, super super directly, the data isn't corrupt because of team competition. It's not corrupt because it can't capture Jalen Wattle's college journey well enough. What's corrupt is the lack of open mindedness that people have when scouting through data. This has questions from at fresh Underscore Bentley, who's projected better Davante or Chase for the wide receiver one? So I think Jamar Chase is the wide receiver one in this class. Davante Smith actually has better numbers throughout my model, and the overall stat score does favor him, but Chase has the edge in size, and we can't forget that his numbers are coming from his sophomore year, where he also played with good proven talent, and where he played with arguably the best talent on either Alabama or LSU's teams competition wise over the last few seasons. And that's justin Jefferson. I say arguably because DeVante Smith played with Calvin Ridley in twenty seventeen when Smith was a freshman. But we don't really have an apples to Apple's comparison of DeVante Smith versus Ridley because DeVante Smith was a freshman with Chase he was a sophomore when Jefferson was a junior and Chase balled out. Depending on what's important to you and how you evaluate, you could argue that Chase had better numbers than Jefferson did. My stat score between the two guys is almost identical, and that's only because receptions per game is one of the factors in my model. But my point is, I do think that right now chases a safer bet. That opinion can change, and I still like DeVante Smith a lot. I'm not one who thinks that his size is going to necessarily hurt him at the next level. It's just one of those things that if you're looking at two very study prospects, you can knock one of them for that size. This last question says dynasty theory, if you owned an arbitrary amount let's say four to six of picks in the second and third round of rookie drafts. What would your ideal strategy be with those picks. Would you package them to move up, throw a bunch of darts on guys, or use them as throw ins to trade for vets. The answer to this has to be that it's team and draft dependent. Like, we can't answer that question with any sort of certainty. If I'm rebuilding, I'm probably not trading my picks, at least any of my second rounders. You could argue you could trade the third rounders for upgrades, which plays into what I talked about on Monday show. But I wouldn't be dealing my second round picks very easily unless I was getting more picks. If I'm competitive, it's a different story. I'm more open to trading up to get an obvious impact player like a running back, or I'm open to dealing a pick for a proven veteran who can get me production right away. It seriously depends. I don't think anyone should be viewing Dynasty Strategy without in It depends type angle. This has questions from at Brett Goold for what metrics does your model favor for incoming college running backs, breakout age, BMI, teammates score, et cetera. So I'll be going over this as I do each year when I do my yearly episodes on running backs and wide receivers who perform like studs in college. But I figured I could talk about it a little bit on today's show before digging into the specifics of this year's class. My running back model has athleticism measurement thresholds BMI matters a lot more at running back than it does wide receiver, and then there's a teammates score as well, and then the production metrics that I focus on are total touchdown share, reception share, and total yards per team play. This is why I generally make a pretty big deal out of receiving backs in college. It's not that I expect them to be good receiving backs in the NFL. It just shows you that teams were using them in a lot of different ways, which tells us intent they want to get them the ball. And yards per team plays slightly skews the players who are able to catch passes out of the backfield because you're not just looking at rushing you're looking at total yards but yeah, that's how the models built. I found that age in school and conference. They're not as big of a deal at running back as they are wide receiver. I think it's more of a tiebreaker thing than anything else. And in the end, of course, draft capital is king, even more so at running back than at wide receiver. That's another piece of the puzzle. It's in the model that I didn't mention draft capital is huge at running back. I've got one more question this week, and it's a good one. It says, hey JJ. Anytime I hear the phrase too many mouths to feed, it makes me cringe. I've heard that targets are earned, and presumably it's easier to earn them when there is less competition, But does less competition mean fewer warm by or does it mean the caliber of their teammates is lower. Obviously, things like quarterback tendencies in play calling will impact target distribution beyond relative talent, But does it really matter to good players if there's a glot of average wide receivers on their team? Thanks Kevin, Okay, So let's dissect this question. First, targets are definitely earned, at least to a decent degree. A player needs to be open to see the ball, and he also needs to be a worthwhile player to throw to. He also needs to be on the field to see a target. A lot of things need to happen for a player to be targeted, and those things that are happening they all surround skill level. Second, you ask if too many males defeat has to do with fewer warm bodies or the caliber of teammates being lower on a team. I think the answer to that in how people reference too many males defeat. It's a little bit of both. Both of those things are happening. People get excited about a player because his competition either sucks or his competition isn't very deep where it's just him and another teammate gobble up a lot of looks. That's how the too many males defeed argument goes. Now, this is going to anger some people, but to me, there's at least something to this argument. There's something to the too many males defeed argument. I sort of alluded to it with the first question on today's show. Let's just run through an example hypothetical. Let's say there's a team out there with DeVante Adams, Stefan Diggs and Robbie Anderson. In twenty twenty, Adams led the NFL and target shair despite missing two games. Diggs was third and Robby Anderson was fourth. This is only at the wide receiver position. Do you think that Robby Anderson sees almost twenty six percent of his team's targets if DeVante Adams and Stefan Diggs are on his team instead of DJ Moore and Curtis Samuel Hell no. So it's impossible for the too many moles defeed argument to be completely irrelevant, at least for players who aren't super elite. Someone who's in the Devonte Adams tier, someone who's just really, really really good, that target shair is generally going to carry over year in and year out. But we've seen many of spiked seasons from wide receivers because of a lack of competition. I wouldn't say that Robbie Anderson was that. That was just an example. But Pierre Garsan once had a thirty percent target shair. He saw one hundred and eighty one targets in twenty thirteen for Washington. His biggest competition on that team for targets a thirty four year old Santana moss A rookie, Jordan Reed, and Leonard Hankerson. You think Pierre Garson's going to see a thirty percent target share if he had DeAndre Hopkins on that team, of course not. So you can't say team competition doesn't matter. If you build out projections, you see it altering target share all the time, and if it's altering target share, it's altering overall targets. Now, if the question was is the too many moles defeed argument complete bs for upper end wide receivers, then I would say yes, targets are indeed earned, and elite wide receivers are going to see a lot of looks in their offense no matter where they go. It might fluctuate slightly depending on team competition, but you can bet that they'll have a high baseline. Each talent leads the volume though. I think that's what the argument really should be focused on at wide receiver. Like, sure, some players are going to see a lot of volume by default because their situation lacks any real alternative, but that's not going to happen across thirty two teams. That may happen only three or four times a year, and even when that happens the player who appears to be average, the player who looks average but is getting volume, he's usually better than you think. Like Perr Garson, It's not like Garsan was trash. So yes, too many males defeed is a real thing if the competition is very legit. But if a player is very good and you believe in that player, it's unlikely that player is going to widely fluctuate in volume and performance from one year to the next. And if there is a lot of variance, maybe he's just not as good as you originally thought. That's going to do it for today's show, though. Thanks to all of you for listening. If you've got subscribed to the Late Round podcast, make sure you are by searching for it pretty much anywhere podcasts can be found, and never forgets follow me on Twitter at Late Round QB. Thanks everyone, enjoy the game this weekend, and I will talk to you next week