354: Mailbag 3/6/20

00:00:02
Speaker 1: This is the Label podcast with your host JJ zacher Retha JJ zacher reson What's up everyone. It's JJ Zacharesen, the editor in chief at FANDEL and at number fire dot com. And this is episode three hundred and fifty four of the Late Round podcast so many shows that are part of the FanDuel podcast network. Thanks for tuning in. That was quick. Before I dig into this week's mailbag questions, I wanted to remind you guys to check out the XFL contest this week over on FANDEL. Just go to fandle dot com, click the XFL tab and find the best contest for you. And as always, you can use the helper that I write on number fire dot com each week to help build your lineups. But go to fandle dot com, click the XFL tab and start building your lineups. The first question this week is from Twitter's from at James Underscore AG one. I'm curious on your thoughts on which league requires more skill A ten team one quarterback, two run back, two wide receiver, tight end two flex league or a fourteen team one quarterback, two running back, three wide receiver, a tight end, a flex and a super flex. I'd argue the first one, since your weekly lineup decisions require more thought the fourteen team, you essentially start anyone with a pulse, So I couldn't disagree with that argument more. When you're making a start sit decision, you're looking at a single game for two players, or maybe three players, if you were to assign probability to those players, Like let's say that you're looking between Tom Brady and Derek Carr. If you were assigning the probability that one would outproduce the other in a given week, if you had a sound process and you were pretty good at fantasy football, you may go with Tom Brady and maybe assign him like a fifty five percent probability of outproducing Derek Carr. Like. The entire reason that there's even a decision is because the range of outcomes for these players is close. So with these start sit decisions, you're focused on a small sample, which is one game, and you make a decision based on a very small difference in probability. Even if that probability isn't super mathematical, it's like a mental probability in how your mind is looking at things. Meanwhile, every team in your league has a pretty strong roster, So, like you're saying in the question, the focus of the league becomes my weekly decisions are more important than they would be in a league where teams aren't as strong, because when teams aren't as strong, making a start sit decision is easy. But you're ignoring two things when you assume that. The first is that a deeper league with more teams in roster spots is then more focused on team building in the draft, which in and of itself takes some amount of skill. And I'd strongly, strongly argue that the skill in drafting a good team and having good game theory when doing so is much much more in depth than making a start sit decision that are based on slim probability margins. My seven year old niece can make start sit choices. My nineteen month old daughter can do that. She can realistically look at two players and pick one blindly, and she be right with her choice close to the same number of times that I would be right, because again, the only reason that it's a decision is because it's close. My niece and my daughter would have zero clue how to draft in a fourteen team superflex league if it was a ten team league, they still wouldn't have that much of a clue, but there's a lot more room for air. So that alone tells me that there's far more skill involved in the deeper league. But then, probably the easiest way to think of this, and this is the second thing being ignored, is that it's not like you're not making start sit decisions in deeper formats. Just because players are scoring fewer points doesn't mean the points that you're potentially missing out on are less significant. Let's use an example. If you're looking between Lamar Jackson and Patrick Mahomes. Because you're in a really shallow league, you may feel like the weight of the world is on your shoulders, because how in the world can you choose between Patrick Mahomes and Lamar Jackson. But realistically, from an outcome standpoint, the two quarterbacks may not give you drastically different outcomes in a given week. Just because they're good doesn't mean one is going to perform really really well while the other one flops. The same type of difference in point output can happen in a deep league. Your start sit decisions aren't any different. Even if you're looking at Cole Beasley versus Hunter Renfro. I think that's a key to always remember that. I think that we'd all agree that with a deeper league, the draft does matter more. Team management and game theory matter more, but the start sit decisions that you make are still just as important as they would be in a shallower league. Just because you're not dealing with studs doesn't make the choice less important. This has questions from at at that hat, it's very difficult for me to say. You talk a lot about draft capital and your model, at what point in the draft does a prospect really start to fall off in the model? For example, when does Tyler Johnson need to be picked for your model to project success? Love the show, keep upping your long pause game. So it's weird about questions like this, and it's a really good question. But what's weird about these questions is that it's not like Tyler Johnson is any different of a player. If he gets drafted in the first round or the fifth round, He's still Tyler Johnson. So realistically, it seems like it shouldn't matter because Tyler Johnson is going to be as good as Tyler Johnson is no matter where he gets drafted. But it does matter because as of today, I don't know how good Tyler Johnson is, and when we have more information, we just get a better idea of how good a player actually is. My model, in particular, doesn't say this guy will be or won't be successful in a binary way, meaning there's no threshold that he's got a hit in order for my model to feel confident that he'll be successful. Instead, it looks at all these players who have been drafted and who have gone to the combine, and it gives them a percentile rank within the position. So let's say that Tyler Johnson goes two hundredth overall in the draft. If that's the case, he's still looking decent as an eightieth percentile player. But that percentile is really dependent upon where other players rank in this class, because you're utilizing this information to see where he'd go in a rookie draft. What I utilize the information for, in other words, is to exploit the mark. If Tyler Johnson is ranked post draft as the wide receiver seven, but my model has him as the wide receiver nine, then I may not be aggressively going after Tyler Johnson. But if he goes two hundredth overall, and his ADP falls dramatically as a result, and my model sees him in the eightieth percentile, which would be well ahead of others being drafted ahead of him, then sure, I'll go for Tyler Johnson. As a fantasy analyst. My job isn't to just say I like this guy or I don't like this guy. That information isn't all that actionable without context, giving it context relative to the market. That's what's important, and that's what the model tries to do. This has questions from at PDX Troy boy. I read on Twitter that the top five rookie ranks have reshuffled since the Combine, and I am surprised at a forty time vaulted Jonathan Taylor over Swift and Dobbins. What metrics slash measurable? Should I weigh more heavily than others when evaluating running backs? Thanks for letting me tune in. I wouldn't be surprised if it's not necessarily the forty time that did it. But you're just hearing more from people who already had Jonathan Taylor as their RB one and he then slayed the combine, so they're just more excited about it like me, Taylor was my RB one pre combine, and that was just reinforced post combine. But like I said on The Running Back Show last week, I look at three main production metrics in my model. I talked through all of that on that episode, and regarding the combine, I'm just making sure that these guys are competent athletes and hitting thresholds like an eighty five or ninety ish speed score, which is really really low. But that's sort of the point. There's not crazy correlation with these measurables and how they translate to the next level. And like I said on the show earlier in the week, some of that could be due to draft capital capturing a lot of it. Draft capital may be capturing the athleticism metrics because teams care about them. So I don't think that they're irrelevant per se, just that when I'm looking at a model, they're not as important. I'd rather input draft capital and have fewer things to worry about. So, to answer this more directly, I don't think a forty times should drastically drastically change the way that you're viewing a running back unless number one that forty time was an outlier and the wrong way or number two it was so good that you're of the belief that teams are just going to value that player a lot more. This last question is an email. It says, hey, JJ, new listener this offseason, have been loving the show. Thank you. In Dynasty formats, what do you do with running backs nearing the end of their first contract like Dalvin Coke, Leonard Fournette, and Aaron Jones? Is it best to try to get younger with early rookie picks, especially in a year with a strong running back class such as this one. Thanks Jesse from Minneapolis. If I had to lean one way between buying and selling these players, it would definitely be to sell. Just anecdotally from looking at adp over the years, it does seem like the market, at least in startups, is sometimes a year late with these players. But I'd also be cognizant of where my team's at if I'm a true contender, and then what I could get as an alternative if you just compare running back to running back. I do think that this is a great time to take one of those players and trade them for an early round one rookie pick plus some because we should feel confident that players like DeAndre Swift and Jonathan Taylor are going to be hits. Draft capital alone pretty much dictates that, So I'm cool with selling those players. I at least would lean that way over buying them. This next question is from at I Chris Lawson. With so many possible changes at the quarterback position next year, do you have any fear that these quarterbacks will have down years since they'll be going into completely new situations and will you be adjusting your strategy at all because of it? I think we still have to be player and team dependent instead of assuming the quarterbacks are going to have down years. But no, my strategy isn't going to adjust all that much from a quarterback standpoint. Not only am I going to be getting a quarterback value later regardless, but the quarterbacks that we're talking about weren't even that good in fantasy in twenty nineteen. Maybe Jamis Winston aside Tom Brady and Philip Rivers likely weren't on your radars to begin with. I think maybe you can argue that some of the weapons on those teams will take a hit, but I'd also argue that the one spot may be Tampa Bay just because you know how fantasy friendly Jamis Winston was. Otherwise, the reason teams are changing quarterbacks is because they won an upgrade, so wide receivers hypothetically they should be getting an upgrade as well. But I think that you should take it case by case instead of trying to make some generalizations. This last question says Marquise Brown and Josh Jacobs were the top rookie wide receiver and running back off the board in twenty nineteen, where would they rank by position if they were coming out in twenty twenty. So I get questions like this a lot because they're fun and they're interesting. But the hard part about answering them is that I'm solving a different question than teams are. It's the Henry Ruggs dilemma. His production profile and profile in general usually doesn't translate to fantasy success, but it can absolutely translate to on the field success. But I can say at least that as a running back prospect, Josh Jacobs was basically only good in my model because he was a first round pick. He was fine as a receiver, but he was in a split backfield and he wasn't used a ton in college, so his production profile was naturally lacking. Now again, some of you are gonna say, well, that's why you can't prospect through data. That's not the point. We're finding trends to be more accurate, and we're not going to be one hundred percent accurate, and this is why draft capital is important, especially at running back. Jacobs was still good in my model because draft capital matters. But yeah, profile wise, I wouldn't be ranking Jacobs ahead of basically the top five or so running backs in this year's class. Post draft. That would likely change because teams really seem to like Josh Jacobs and I do think that he's good, but running backs are pretty boring to prospect because of how important draft capital is. Hollywood Brown wasn't nearly as bad in my model as Jacobs was. When you ignore draft capital. He had a stat score that was actually pretty strong. I don't know if he'd be drafted or looked at in the same tier as Lamb or Judy, and I do think that Brown is one of those players who's probably better in real football than fantasy, but I think that he would be in first round conversation. Still, i'd probably have him close to my top five entering the draft, I did rank him lower than I should have last year based on landing spot in the run first offense that he was going to play in, But going into the draft, I'd be more okay with him and feeling confident with him strictly as a player in my prospect model than Josh Jacobs, because, like I said, objectively, without draft capital, there were some scary signs for Jacobs. The draft capital is what made me say I don't care. I don't care that Josh Jacobs lacked in production because that's how the running back position is. This next question is from at Scruffy Man seventeen. Does your model happen to look at tight ends any differently as opposed to other positions? And do you think that there's a particular reason they take longer than others to break out? Thanks? JJ, loved the show. Thanks. So, I actually don't have a tight end model right now because it's a hard position to prospect and it matters a lot less in fantasy football. But what I do know, what I've tested, is that college production matters a lot less at that position, and athleticism matters a lot more so when I'm trying to draft tight ends. I'm looking at athleticism testing a lot more than I am at wide receiver and running back, and I think the reason they take longer to break out and develop is because they do more. Tight ends, have a lot of different jobs on a football field, and more things to learn, so naturally it just becomes difficult to be good at all those things right away. This last one's an email that I thought was really interesting. It says, this is a dynasty related question and one that I have been wrestling with for pretty much the entire offseason. With the amount of variance and even luck that is involved in fantasy football, wouldn't it make sense to try improving your odds of being successful year over year by selling stud players for less shore performers at the same position and acquiring rookie picks or young ascending dynasty assets in the process. What I mean is that if you have DeVante Adams, let's say, who's widely considered a top three dynasty wide receiver, wouldn't it make sense to trade them for a Courtland Sutton or DJ Moore and get three to five years younger in the process, all while acquiring another late round, first or even second round pick on top of it, even moving down to a DK Metcalf and then making sure that you get a higher rookie pick seems logical. Reason being that variance you have talked about before. There is a universe where Adams doesn't outscore or only slightly outscore Sutton in more next year. And if that's the case, and you have added a Terry McLaurin through that extra rookie pick in the process, then that's a massive win. And even if Adams finishes as the wide receiver two next season, one could argue that having DJ Moore and Terry McLaurin is still the better side to have. Of course, you would have to pick the right guys, as is the case in every situation with fantasy football, because trading down from Michael Thomas last year to Brandon Cooks and using that extra pick on jj R Sega white side would have backfired about as much as you could imagine. But that's where being a good fantasy player comes in. If you think you are good at fans fantasy football and you believe in your player evaluation, this seems like the best way to build a true fantasy football powerhouse in dynasty, and this is from Martin from Berlin, Germany. Okay, so this is a long question, But at first I was going to answer this question by talking about variants and how it's important to not have a bias when we're looking at variants. That it's easy to feel excited about a player like Courtland Sutton and only see upside, and that inherently shifts the way that we view his season. From the perspective of variance. It becomes harder to see any downside for a player, but that downside still exists. I love DJ Moore, but he still has some sort of downside, something bad could happen. Instead of going down that route, though, I think the last paragraph in this email and in this question is what's most important, because you say, but that's where being a good fantasy player comes in. If you think you were good at fantasy football and you believe in your player evaluation, this seems like the best way to build a true fantasy football powerhouse in dynasty. So a few things about that. First off, if you're really good at fantasy football and you believe that you have a significant edge and player evaluation, then you shouldn't have to trade away really good players in order to get really good players. Terry McLaurin was not a top two round pick in rookie drafts last year. Devin Singletary consistently was drafted in the mid to late second round of rookie drafts. You didn't have to trade DeVante Adams to get pieces like that. You could have acquired those pieces very, very easily. So that tells us that we're not as good as we think that we are at player evaluation, because if we really thought that we were good, then we wouldn't be trading away elite dynasty assets for players below those assets. We'd be keeping them, and then we'd pluck away values in future drafts. Now, if we take a step back and we say, maybe we're not perfect, but we're better at player evaluation, which is probably the case for a lot of you listening to this podcast, then sure, I think a lot of you are probably above average and finding the right guys. But let's not overstate what that means. A huge reason that I think that game theory is so important and fantasy football is because player evaluation is everywhere. Ninety nine percent of content around fantasy football is about whether or not a player is gonna be good there's nothing wrong with that, But when there's so much information on something, then it's easy for a bad fantasy manager to take that information and utilize it. In other words, if everyone's saying to draft Kyler Murray next year, then it's not very difficult for Randy and your fantasy league, who knows nothing about fantasy football, to then just draft Kyler Murray. He just has to hit draft player because he read something that Sigmund Bloom said and shout out to sig I love you man. So even though in your brain you have a much better idea of who's good and who isn't, rankings and lists have shrunk your ability to have an edge and player evaluation, it still matters and you can still be better than your league mates at it. But it's very hard to be so much better. That's why game theory is then important, because it helps you understand the things that aren't easily taught, like positional importance, when to trade for and trade away play and all that kind of stuff. So with this question, don't overstate your ability to be better than your league mates at finding player values in drafts, not only because it's really freaking hard to prospect players accurately, but because your league mats can still come close to your evaluation very very easily. But the second part of my answer is when do you stop? When does the powerhouse building stop. You can say that you want to gobble up more and more draft picks and pieces in order for them to hit. That's fine. That's especially fine if you're rebuilding a team because you want more chances to hit on players and younger players as opposed to having all of your value held with one or two players. But when does it stop? When are you satisfied with your roster? You have to win at some point. It's fine if you want to trade DeVante Adams. If you can get Dj Moore and Terry McLaurin for DeVante Adams, like you said in this question, then you should go for it. But if the rest of your roster is relatively loaded, then you're not necessarily doing yourself any good by trading for depth. You need your starting lineup to be optimized in the end. So that's how I look at a question like this. First, as I always say, we think we know more than we actually do. Not only are we worse at evaluating than we probably think, but we're also not that much better than our league mates and player evaluation, even if we know these players way better than they do. So as a result, sometimes it's good to have the sure thing on your team and then second you have to win sometime. DeVante Adams still has the best average outcome among almost any wide receiver next season. He's going to help your fantasy teams a lot if he's healthy, so if you're trying to win, trading away that piece is then less attractive. Thanks for this question, though, or should I say Donka shund I took five years of Germans, so I feel like I can say that that's going to do it for today's show, though. Thanks to all of you for listening. If you get to subscribed to the Late Round podcast, make sure you are by searching for it pretty much anywhere podcasts can be found, and as always, don't forget to follow me on Twitter at Late Round QB. Thanks everyone, enjoy your weekend, wash your hands, and I'll be back in your ears early next week.