1088: Mailbag - De'Von Achane's Outlook, ZAP Model Methodology, and More

00:00:02
Speaker 1: This is the Late Podcast with your host JJ Zacheratha j J zacher Reef. What's up everyone. It's JJ zacharyesin in this episode ten eighty eight of the Late Round Fantasy Football podcast.

00:00:18
Speaker 2: Thanks for tuning in.

00:00:20
Speaker 1: Before we get into the mail back questions, I just want to send a reminder to check out the Late Round Prospect Guide.

00:00:24
Speaker 2: You get info on all the zapp models.

00:00:26
Speaker 1: There's player write ups, rankings, bad dad jokes, and more, all for nineteen ninety nine.

00:00:31
Speaker 2: It's over one hundred and seventy pages of content.

00:00:34
Speaker 1: Learn more over on lateround dot com. The first mail bag question this week is from Twitter's from at Hannibal Lekta. It says possible Jalen Waddle implications willis running Moore draft a top wide receiver e Chan to the moon. Okay, so I talked about this a little bit in the Late Round show on Wednesday, and I'm sure some of you listen to that in the podcast feed yesterday, or maybe you just watch it on YouTube. Regardless, let's talk about this trade. I think Jalen Waddle is a very good wide receiver. He's been atter above a two point one to nine yards per out run rate and three of five season. That's no easy feat. Anything above two is usually really strong. At wide receiver. He's typically been in that twenty two to twenty four percent target share range. His targets per out run that's been good too. He can line up all over the field when he's healthy. Jalen Waddle is really good, and when you look at Denver's wide receiver room, you get a lot of okay to good players. None are as good as Waddle, even Courtland Sutton. I mean, you can just look at yards per out run to help tell that story. Sutton has been on teams with weaker competition than Jalen Waddle at least you can argue that. And he's been above a yards per out run rate of two just once in his career, and that came all the way back in twenty nineteen. I genuinely believe that Jalen Waddle is in a different tier, especially at this point in their careers. Sutton's gonna be thirty one In October. Last season, Troy Franklin started to lose a little bit of work to Pat Bryant. And while I did think Brian was a trade throw in type at the start of the offseason as a by candidate. It's not like he's proven to be special. My thinking at this point is that Brian can rotate in a little bit and play the slot, Sutton on the outside, Wattle can line up everywhere, and then Troy Franklin he might be screwed the most out of all these wide receivers. Regardless Waddle steps in, he's the best wide out on this team. He can see a twenty two to twenty four percent target chair because good players create opportunity, and Denver's been above zero in pass right over expected over the last two years. They were fourth in the NFL last year in pass attempts. It's a good place for Jalen Wattle to land. Now, when you look at Miami, I have seen a lot of these takes Like this question is asking of eight chan to the moon.

00:02:45
Speaker 2: I'm not really there.

00:02:47
Speaker 1: Eight chance coming off a season where he had a nineteen percent target chair that was in a Mike McDaniel offense that definitely favors his type of play style. It's just straight up not easy for a running back to hit that high of a target chair. We at least wouldn't lock them into that. We've seen twenty seven running back seasons since twenty eleven with a nineteen percent target share. A better Christian McCaffrey owns five of those seasons. Alvin Kamara has five.

00:03:11
Speaker 2: More of them.

00:03:12
Speaker 1: It's just not a common thing to have happen. And in today's NFL it's happening less target share, it's shifting more to tight end than it is to running backs. So to me, the departure of Mike McDaniel was bad for Devon h Chan. Maybe this waddle move helps things a little bit, but I don't think we should lose sight of that greater picture. Not only that, but how good is this offense actually going to be? We want our running backs and our players in general to play in good offenses. The Dolphins have wide receiver threes and wide receiver fours as starters right now, which hopefully will change. But it's gonna be a questionable offense. And maybe you feel that way about last year's Dolphins, but they were twenty second in touchdowns per game. That could be worse. Malik Willis gets a downgrade here too. Can It'll be plenty fantasy relevant because of his rushing, but losing Jalen Wattle is definitely tough because they won't be adding anyone more than likely who's as good as Waddle here in twenty twenty six unless someone really hits from the NFL draft.

00:04:14
Speaker 2: So those are my overall thoughts. I don't have season.

00:04:17
Speaker 1: Long rankings done quite yet, and we're working on projections, but I think this is a good thing for wattle sustainability. It's a more pass heavy offense, the weapons are beatable, and he's just really good and I don't want to over count it on the volume men for Devon a Chan. This next one's also from Twitter, is from at GH Horsefield. It says which running back room is a sneaky good landing spot for a non Jeremiah Love draft pick. Seattle, Minnesota, Washington, maybe Baltimore.

00:04:46
Speaker 2: Yeah.

00:04:46
Speaker 1: I mean the obvious ones with openings should be on everyone's radar. Seattle, Minnesota, Washington. Those are the teams with immediate needs. And then I look at other teams who may not have absolute studs, who may at least consider doing something in the draft potentially to add depth. Chicago's won if they want to get a little bit more depth and competition. Carolina the exact same reason I think Chase Brown is good, but you never know. With Cincinnati and Dallas could definitely use some help behind Javonte Williams. And I still think, despite what they did this offseason, that Denver and Houston should both be on our raiders right now too. And then I look at the teams who have running backs who are set to become free agents next year after the twenty twenty sixth season. The problem is, this isn't the kind of draft class where you really want to hold on to that player into the twenty twenty seven season when you know that he'll do nothing in twenty twenty six because this draft class doesn't have a ton of talent usually if we're looking at Day three picks. I think that's with the majority of these running backs where they're gonna land. We want to find better situations for that immediate upside and that value jump in dynasty. But there's some players set to be free agents next year where their teams might be looking to add. Specifically, you have the Jets and the Colts Breesaw and Jonathan Taylor their contracts are up, But that's not to say that I think those guys are actually going to be free agents. So those are some of the teams where there are some possibilities. It sounds like half the NFL, but really it's just a lot of these NFL teams could go either way given what they have right now, like the Bears, that's a great example. They could very easily just go after more depth the sas question, says hey JJ. As a first time buyer of the Prospect Guide, I found myself wondering about the historical comps of this incoming tight end class, specifically Stours and Sadik. Would you say this could be a Loveland and Warren situation in terms of expectations for these two and the gap between tight end one and tight end two, how much of their combines warping market perception and how early would you be willing to take each of them in rookie drafts without tight end premium, without superflex andrew. So all my rookie rankings are in that guide. You can see the rankings at the end. This is a very weak class overall in my opinion, like at all positions combined. So I think both Sadek and Stowurs they're first rounders in single quarterback leagues in non tight end premium leagues because things are just rough out there. The tight End ZAT model likes Sadiq more than Stours, and it's by a relatively decent margin. That's because Sadik is projected to go well ahead of Stours in the draft right now. Realistically, if you look at their breakout scores, which is the same methodology that I use at Wide Receiver, they're not that far off. Stours actually has a better breakout score. I think Stowers has a better production profile overall. But this is not a Tyler Warren Colston Loveland situation, because I don't think either of these guys are as good as them. Maybe Sidik, but when you combine his player comps with his production profile, it's a little bit scary. I'm not gonna lie. Last year's tight End ZA model it liked Colston Loveland. Most you guys saw that in the prospect guide. This new model that Brandon Godoola and I worked on, it likes Loveland arguably more than last year's model. Loveland's breakout score is elite, whereas Sadiq's is more in line with Tyler Warren. So maybe the proper way to think of this is that Sadik is more like Warren than he is Loveland as a prospect just from a ZAP score perspective, but as a pairing, Stowers just doesn't match up because he's not projected to go in the first round. Draft capital doesn't mean everything, but when modeling, it still gets signal, significant signal. But that doesn't mean that I don't like Stowurs.

00:08:25
Speaker 2: I do a lot.

00:08:25
Speaker 1: He's basically the fourth best tight end out of these last two draft classes according to the model, though he's not that far from Terrence Ferguson and Harold Fannin, and that's with mid second round draft capital. Sadiq just scares me more subjectively. His breakout score is fine, but there are other metrics out there like yards perou run that don't paint him very favorably. And as comps in the ZAT model, they're fine, They're not elite Noah fen Oj Howard Evan Ingram.

00:08:52
Speaker 2: So overall, I'm definitely giving the edge.

00:08:54
Speaker 1: To last year's top two tight ends over this year's and that's not based off of their NFL. This isn't a hindsight take. That's just based on them as prospects. This best questions from Mike on Patreon. It says the Rams and Patriots have been rumored to be in TALX with the Eagles for AJ Brown with his contract making this more of a post six to one thing. What's the possible ceiling on aj and should we be buying low as he continues to fall down startup and best Ball draft boards. Brown had a down year by his standards and just through film evaluation. I do think he took a step back, but the market is saying all that right At this time last year, AJ Brown was a top fifteen Dynasty wide receiver cornet keep trade cut right now. According to that same source, He's wide receiver twenty six. Brown has seen yards per out run rates over two in every season of his career. That's wild, that's amazing, and that still includes last year. Despite all the crap that people were saying AJ Brown had a two point zero six yards per out run rate. Brown didn't even have that bad of a year in PPR points per game. The problem and the difference is that he'd been above two point five in yards per out run every other season of his career, and Brown did that with a very similar catchable targets rate. I don't think this is purely a Jalen Hurts thing last year, but I do think it could be an offensive system thing. The offense itself didn't set him up to do a whole lot, the lack of motion, the lack of creativity, But seventy percent of Brown's targets were catchable. That was seventy one percent in twenty twenty four, seventy eight percent in twenty twenty three. That's according to PFF data. What I don't want to do is be unfair to AJ Brown because of his baseline, because he's usually just really good, and he wasn't as good last year. Like if this were any other wide receiver at this age, and he's coming off a year with that kind of yards per out run rate and a very reasonable twenty four percent targets per out run rate, I would say he's doing pretty well for a guy in his late twenties. But I do think that downward trajectory has started with AJ Brown. It's just that trajectory is beginning at a very high place because AJ Brown is really good. And it's not just because those metrics are lower than AJ Brown's typical averages, but he also had the second lowest AID out of his career last year, which can often lead itself to better yards after the catch numbers. But Brown had his worst yack per reception season of his career. Just don't confuse that with being bad. Like I've said for years that Perfect Dark is a better Nintendo sixty four game than GoldenEye. It's literally just a better version of what GoldenEye was. You can debate me all day on that, but that doesn't mean that I think GoldenEye is bad, just like AJ Brown had better seasons than his twenty twenty five season, but that twenty twenty five season wasn't bad, and per route run metrics, they're not everything. ESPN's Open Scorer actually put Brown ninth last year. His overall score, according to their methodology.

00:12:00
Speaker 2: Was sixth.

00:12:01
Speaker 1: So I think what this is all saying is that if you're rebuilding, you probably don't need to take this on. Generally speaking, when a player is around that ajapex and things start to fall off, I don't want to buy into that this time of year. I'll do it more mid season, when I know where my dynasty team is at and if I'm truly competitive, and that way, you also know how effective that player is actually performing. There's really not a need to buy into Brown right now because even if he does get traded, do we think his cost is going to substantially be higher than it is right now? And I feel pretty similarly about best Ball and from a redraft perspective, his early ADP is in that high end wide receiver two range. It's wide receiver thirteen over on DraftKings right now and their best ball contests. I really don't think that's a clear by situation. And yes, keep in mind I haven't done my rankings yet, we're just starting to build projections.

00:12:53
Speaker 2: We're working on it.

00:12:55
Speaker 1: But there are some very good wide receivers with a whole lot of upside being drafted around, like wide receiver twenty five. So his ADP might be slipping, but I don't think it's slipping enough. And by the way, if you look at early ADP, it's shaping out to be similar as last year, where you get a running back or two early and then there's that group that range of wide receivers where there's just a lot of upside that we're gonna want to choose from. But this is just one of those situations where I don't think you need to be aggressive either way right now, especially on the by side. I wouldn't be actively trying to be overweight with AJ Brown at this moment. His cost is still relatively high in redraft, and I don't think the Dynasty price is gonna rise that dramatically even if he is traded. And some of these destinations don't guarantee a better outlook for AJ Brown versus his outlook in Philadelphia in a new offensive system. Keep that in mind too. Now. The last one today is focus more on my methodology, but I think it needs to be said.

00:13:53
Speaker 2: It's an email that says, HEYJJ.

00:13:55
Speaker 1: As someone who has purchased your Draft guide every year Prospect Guide, I noticed the grading methodology has changed this year and is particularly noticeable on the high end. For example, you list Romadunza as eighty point two, whereas he was previously a ninety six point five in your model. Could you elaborate on this new scaling a bit. I am curious where Jeremiah Love would have landed in the previous methodology. Similarly, where would some of the high end prospects from previous years land in the new methodology. Thanks Alex, so I've gotten more responses about this new score scaling than I anticipated, So I figured I can just throw everything out there right now on this show. We're all on the same page. Hopefully this helps, and I can reference this episode when those questions come in, and I think a lot of you are going to be interested in this. And look, I understand the questions, I understand why I'm getting them, but let's get some clear cut answers here.

00:14:50
Speaker 2: First.

00:14:51
Speaker 1: If you have the Prospect Guide, there's a section in it on interpreting the data where it talks through some of this. But basically, the way the models always work that it spits out some sort of raw number, some score, but that score doesn't make any sense to even me, because just based on the inputs and the proper weight for those inputs, it might be like twenty three for an elite running back. That's not easily digestible. So what I've done in the past is I've made those scores into percentiles. If you go back to the very first late Round Prospect Guide, I actually call those scores percentiles. When you transfer scores to percentiles, you make them on a scale from zero to one hundred. The problem in doing that, though, is a percentiles. They're linear in nature, right, it's one better than the next better than the next, and that can create a lot of problems that I've tried to address, but I didn't fully address until last year. Last year, for example, I started creating categories. The categories would group prospects by their raw prospect score, not their percentile score. And the reason for doing these categories is that you could group players up based on that raw score and they wouldn't be skewed by the ranking system of percentiles. You get a better idea of where drop offs might occur. It could help with interpreting some of these percentiles. But the true problem with all of this is that percentiles, like I said, they're linear in nature. Adding categories doesn't change that. And when something is linear, you're not properly seeing where the gaps exist from one player to the next. Not only that, and more importantly, if you were to only look at percentile scores, they're not nearly as predictive as the raw output that the model was generating per player. So I just decided to fix all of it this year. Instead of percentiles, you're getting more accurate numbers of what the model is actually saying about these players, and what that means is there's a larger separation between the elite players. Roma Dunday in that twenty twenty four draft class was the example that I showed in the guide. There's a little graphic in there. Under the previous methodology, a Dounesay had a score of ninety six and a half, which was three points off of where Marvin Harrison was and where Molik Neighbors was. Xavier Lagette had an old score of eighty two, Mileikue Washington fifty three, all in the same class. But now a Dounzay's score dropped from ninety six and a half to about eighty. Leget went from eighty two to under sixty, Washington's goes from fifty three to twenty one. With percentiles, Before this change, too much real estate was taken up by the undrafted. In day three picks, you weren't really capturing the true value differences from one wide receiver to the next, or one running back to the next. Because again with percentiles, every player gets their little piece of that pie from zero to one hundred. The worst is zero, the best is one hundred. So if we have a ton of these low tiered players That means a Day three pick like Maleikue Washington will still get a score over fifty because there are so many players who are making up that zero to fifty range. Really, when it came to relevant players, the ZA model was scaling those players from fifty to one hundred, not zero to one hundred, because there are so many players who go undrafted or who get drafted on Day three who just don't matter that much from a fantasy football perspective. It wasn't that the results were any different or my interpretation of that data was any different. In the guide, I still feel the same way about those players today that I would have in twenty twenty three or twenty twenty four, though the model has changed because I try to iterate that model year in and year out to make it better. It's just a slight rebrand. It's like when a company changes their logo. It's still the same company. People are just interacting with that company in a different way. The actual calculations of scores didn't dramatically change year over year. The raw numbers, they're mostly the same, aside from those changes that I made in the models, which are all listed throughout the guide, but this allows us to better interpret the results and see those gaps. For instance, Malik Neighbors is now just one of a handful of wide receivers with a score of one hundred in that model. His old ZAP score was ninety nine point five, but now there's a cluster of prospects with a score of one hundred, and that's a twenty point gap from him to a Dounesay, which allows you to really see that there's a pretty significant difference.

00:19:15
Speaker 2: In their outlooks.

00:19:16
Speaker 1: But before, someone might read the guide and be like, oh, there's only a three point gap between elite neighbors in romadunsay, that's not that big of a deal.

00:19:24
Speaker 2: But that is a big deal.

00:19:26
Speaker 1: One of the toughest parts of my job is interpreting this data to thousands of people. I'm not going to say that it's easy, because it's not.

00:19:34
Speaker 2: But it's also part of my job.

00:19:35
Speaker 1: It's on me to properly communicate these ideas so that we're all on the same page. It is that you feel comfortable with how to take the information that I give you and act on it in fantasy football. Now, stemming from these changes, I've gotten a lot of requests from people to see scores from previous classes, and again, I understand anyone wanting that, I get it. It's a reasonable request. I do want to say that you can I see last year's class scores. In the year two section, each player's profile has their zapp score listed. But when it comes to previous class data, I don't publish it for a reason, just to reiterate and to be clear, a lot of people are consuming this content, and I'm unbelievably grateful for that. But if I'm being completely transparent, and I'm really being vulnerable here, if I were to just say, oh, here's the database of scores from previous classes, the floodgates would open my inbox, my DMS, the late round discord. It'll just be people asking about why this player had this score or why this player had that score. And I know that to be the case because that already happens anytime I post my rankings. Literally the instant I update my Dynasty rankings, I'm flooded with responses, which is great. It's a necessary part of my job. I'm not complaining, I'm just being open and honest. I can't take that on. When it comes to prospecting. I can do it with rankings because those rankings are immediately useful. If I did it with the prospects, we'd be dissecting players who were drafted four years ago, when that score isn't even relevant anymore. I just don't have the capacity to manage all that, not even from a time perspective, but from a mental health perspective. It's not like I'm hiding something here, you guys. If you guys want to talk about a specific player what his score was in the zapp model, I'm more than willing to talk about that. I do it all the time on discord. We talk about where that player ranked to ZAP score, all that good stuff, and I share screenshots of the spreadsheets.

00:21:34
Speaker 2: But having just eight hundred different.

00:21:35
Speaker 1: Wide receivers and their scores just out there, I can't do it, man, I just can't do it. That takes a ton of management capacity that I cannot do. And one of the main reasons that I priced the guides where I price them, which I think are very fair prices for what you get, is because I feel like I'm giving you guys a very fair amount of content for that price, and I'm not going to come out here and say, hey, if you want the full database, give me two hundred dollars. That's not what this business is about. That's not how I operate. This is not some money grab. Maybe people would buy that, but that defeats the purpose of what I just said, which is the management side of all of that and the mental health side of all that. You're more than welcome to ask about individual players, though, But like if anyone's trying to compare a prospect to one in the past because you want to take some action on your dynasty team, like create a trade based on those scores, I wouldn't even do that. The ZAP scores are how we viewed those players as prospects entering the league. We shouldn't view those players the exact same way after they've played in the NFL. That's why something like the year two model exists. So look, I know I got deeper than anticipated here, but I wanted to be open about why I make the choices that I make. But the bottom line is that I'm trying to make things as consumable and easy to understand as humanly possible for you guys. That's on me, that's my job. But I also can't just publish everything in the database because it genuinely would probably destroy my day to day work flow. And it doesn't mean that I don't want to be transparent. It just means that I can't focus all my time and energy on the ZAP model and the prospects in that model.

00:23:10
Speaker 2: That's it for today's show, though, thanks to all of you for listening.

00:23:12
Speaker 1: If you had subscribe 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 and on Blue Sky at Late Round QB. Thanks for listening everyone, I greatly appreciate all of you. Have a great weekend. I'll catch you next week.