The Athletic Football Show: A show about the NFL - The state of the analytics conversation in the NFL with Amazon's Sam Schwartzstein; Plus, Ty Dunne on the beauty of the tight end position

Episode Date: October 19, 2022

The analytics in football discussion has reached a point where the opposing sides seem to be shouting past one another, lacking the nuance necessary to such a discussion. So, what should we talk about when we talk about analytics. Sam Schwartzstein, Amazon's analytics expert and producer for live events focused on analytics and insights, joins Robert Mays to dive into that on this episode of The Athletic Football Show.Then, Ty Dunne of Go Long hops on to talk about his new book about the history of the tight end position, 'The Blood and Guts.'Subscribe to The Athletic Football Show...AppleSpotifyYouTube1:34 Most frustrating part of the analytics discussion13:30 The details of modeling26:30 What is and is not an analytics decision30:30 The football guy mentality35:50 Presentation of analytics in NFL broadcasts53:54 Ty Dunne on 'The Blood and Guts' Hosted on Acast. See acast.com/privacy for more information.

This is the athletic football show. Welcome to the athletic football show. Today's Wednesday, October 19th. I'm Robert Mays. Fun show for you guys today. Tyler Dunn from Go Long is going to be joining us a little bit later to talk about his new book, The Blood and Guts, kind of about the history of the tight end position, the guys that have really shaped that position, really how it's shaped modern football,

because that spot is so central to kind of how we understand the game. Before we do that, though, I want to talk about analytics on this show. I think it's time. I think it's time we have the analytics discourse discussion. Analytics has been a hot button topic earlier this year, just like it was at times last year. You know, the conversation around how analytical thinking has kind of influenced NFL decision making has been kind of maddening at times. You know, with both sides really shouting past each other and I think poorly characterizing the other side's arguments pretty often.

Even the term analytics often changes depending on who's saying it and how much poison is dripping off of it. So today I wanted to talk about how we talk about analytics in the NFL and what that really means. And here to help us do that is Sam Schwarstein, the prime video sports analytics expert, somebody who has shaped the way that you're watching Thursday night football this year and is somebody that I'm very excited to dig into this with. Sam, thank you very much for taking the time to do this. Thanks for having me. I'm really exciting to hop on here. All right. So like so many of us, you're a terminally online person.

So in your mind, what has been the most frustrating part of the analytics discourse and where it sits right now? I think for me, it's when people make the straw man arguments. Like after a game, people just go, why? Why? And then it's a roundtable of everyone going, why, why, why? and then it's someone go, and it doesn't even matter, right? And then you go to people who are saying, you know, the model can't account for weather. That's actually one of the things it's best at, right?

That's an actual number of weather, like what the wind is. It's one of the best things it can do. And the models can't account for certain players been on the field. It can, right? And so there are things that people assume a model cannot do. The model cannot play football, right? It's different than poker where a predictive model from a poker or from chess, Those types of algorithms would do better because there is no air injuries.

Your night will always move in an L shape. Like those are the things that it can count for, but the stuff that I can't for is execution on the field. And so that's where there needs to be that kind of that bridge of being like, yes, there's a percent likelihood chance and things like that. But someone needs to be able to go on TV and help people understand, here's how we derive these things from a model. Or what's this information being derived from a model?

So I want to dig into all of that as we keep going here, but I want to start at a pretty simple place. To me, on a lot of the broadcast recently, like the ESPN broadcasts, I don't want to have you besmirch other broadcasts. I'll do that for you. When the ESPN guys are sitting there and, you know, Joe or Troy saying, like, you know, the analytics says this or the analytics and we treat it as this monolith, I think that can be problematic. And I think that for people who want to look at this in a nuanced way, it can be a little bit frustrating. So what I say to you, analytics as it relates to the NFL, what does that work? word mean to you? Well, this is funny because to me, analytics is like anything you're deriving an answer from numbers or a deriving decision making from numbers. And so the first

sport to ever use analytics was football. In seventh grade, I had breakdowns of what fronts I would see and what's the percent likelihood a team would blitz in a certain front. That's been going forever. When you talk to, you know, I talk to analytics teams every week as we prepare for each game and talking to Paul DiPodesta, you know, he was like, yeah, like this is the original analytics sports, right? He's one of the people who implemented it in baseball. And so it's just this ethereal kind of overarching next level analytic is what people I think when they really reference analytics. It's EPA. It's things that we cannot see on the field. It's not volume stats. I think that's the difference. But any number that helps you derive decision making or

help you derive your gameplay style, that to me is an analytic. That's why the, that's why the is so strange to me. Literally five minutes before we started recording this, Antonio Carmardi tweeted something, and it was about how he took notes while watching film. And in the right corner of his notebook, there are percentages for how often the offense is in a particular formation. That has been a part of this forever.

And it brings me back to that very famous Bill Belichick press conference moment when he was being asked about analytics. He's like, absolutely not. It doesn't influence our decision making whatsoever. And if you look at the way and how often the Patriots go forward on fourth down. The modern understanding of analytics, maybe he's actually being honest. But the Patriots have used this kind of stuff famously for so long. Ernie Adams is sitting up there in his like dark shrouded office helping make these decisions. So I'm so interested in how did we run into

this roadblock where a team like the Patriots has used analytics forever, but now this modern way that it's framed are just nerds on computers somewhere so it has no place in how decision making is made in football like where did the translation become screwed up i think to me football is such a physical sport that people don't want to get rid of some of that part of it right and that's where analytics is like yeah you should go for it and then that's that fear-based thing of hey i if i go for it on fourth down like these yards are hard i think every offensive coach would say getting yards as hard in the NFL, as easy as people might think it is from a schematic or from a rule standpoint, right?

Oh, everything's made easier. Yards are really hard to get. I think they're not getting that. I think what they're also not getting is what this framework is scoring is. One of the things with analytics and what makes our game unique is two field goal drives is worse than one touchdown drive because your likelihood to get a touchdown and the extra point so much are. So that's kind of where there's a lot of this derives from is like, yeah, you're going to

get down there again, you're going to get an opportunity, and then also where we have the extra point kick or the new way that the kickoffs are, you start at your own 25 predominantly because of the touchback rule. And so you're giving away yards when you kick a field goal oftentimes, right? You're making it easier for them to score. And so I think a lot of this stuff is where people get rid of that physical nature, and then they don't understand what the second, third order effects that are going to happen, or high likelihood to happen. And that scares them, right? They only see what happens in this one moment versus seeing what's going to happen throughout the game, which like we've said before, teams will run power in the first half to run power pass in the second half.

So it's very similar. You're showing something in the first half, and then you're ultimately using the second half. You're gathering information. That same thing is happening with going forward and forth down. So you mentioned something I want to dig into a little bit. It's kind of a little bit of a sidebar. I don't want to get too sidetrack, but you're talking to analytics people as you're going through this.

Is that part like the same way that an announcer would have a production meeting with coaching? is you're talking to analytics people at most of these stops? Yep, absolutely. That's what we're doing. It's nothing to be made to air right now, just to get a better understanding of the ecosystem. How I frame analytics, and this is how a lot of teams do it, is there's four main aspects of analytics. And again, for me, it's numbers driving decision making.

It's talent acquisition. That's where you get the draft salary cap. Talent development, that's health and safety, making sure what players, assets you have they play, and then also getting them better. and then it's game planning, which is the traditional one, what's the likelihood of their and that's in-game decision-making, right? And certain teams use it in different ways, right? And you can often see how they're housed in an organization, whether they're in football

ops or if they're in the front office or they're in the video department, how teams will utilize analytics. That's why it's so different how these teams. You can see the Rams aren't the best team on fourth down, but they are the best team at sports science. Yeah. That's still changing how they do game, a game, planning that's changing how they do practice based on data they get from a practice field,

which you wouldn't ever, I could tell you how far wide receiver runs in a game. I can assume it, right? Because, you know, how I log into a huddle, how many go routes, all these different things we can assume it. But now we know the actual data from practice. And the Rams are doing that, but they'll, they'll won't use analytics to make a lot of their decisions. But so some departments are great with it. Some departments they're not. And so my job is trying to figure out what each department does well with each team. So again, going further down this side road that we may not need to go down, but I'm interested in it.

Why do you think, how do you think that starts? How do you think we get to that point where you have these organizations that in one silo of the organization, they're totally willing to embrace this data? It's like, okay, like the numbers make sense. Like we're going to be forward thinking and cutting edge in some of the sports die and stuff. Even with talent acquisition and with resource allocation, I think the Rams have probably looked at some underserved markets or what they think are kind of undervalued assets with the way that they've thought about draft picks. But on fourth down, this shit is just completely out the window.

So do you think that is a messaging from certain members of the analytics departments of coaching staffs? Why does some of this stuff get through to certain staffs and why do you think some of it is still a little bit harder to embrace? Like all companies everywhere, it starts with a decision maker, right? So this is a little longer a football talk and just business talk. If that decision maker is buying in and willing, then they'll change, right? And so if Les Need is the one that's running their talent pool and Sean McVey runs what happens on Game Day, and he's buying into it, hey, and be using that, utilizing that team, then they'll use it. So some teams, if it's used in the video department, which is a lot of places that's been, which is who's my biggest nerd on staff, they're now going to be in charge analytics. And then they're going to the

The dude who's in the AV club in high school is the one who's now running analytics for our NFL team. Yeah. You know, it's, and it's, it's, it's, it's,

it's, it's, it's, it's a natural progression because numbers go into, and computers, they're tied together. I have tons of stories from the XFL about how I had to explain that your IT department doesn't build your app.

But, you know, there's different organizations that how they, how is that product, that's where you're going to get buy in, right? That's where you're going to get it. And there are often different bridges. So how your organization's built and how,

and how, who your decision are in within each organization, they're then going to dictate how much you leverage analytics, right? If Billy Bean doesn't want to buy in 100%, it's not going to work with Paul DiPedesta, right? He's got to get a buying from that top decision maker. For me, at XFL, Oliver Luck, bought in, Vince McMahon had bought in to we were going

to be different and use analytics and data to make decisions, largely the same way he does it in WWE. You might think it's crazy that that's how he was doing it, but that was. So, like, that made a natural progression. If I didn't have buy-in from the top, it wouldn't have worked. I'm curious, do you think that there are some throughlines with the communication skills at the people who are best at selling these ideas? Like, what do you think are the tricks of the trade of kind of getting that immediate penetration to allow somebody like a Billy Bean or a forward-thinking NFL general manager, a Brandon Bean, to buy into some of this stuff? What are the best ways to communicate these ideas and have them get through.

You want to make sure that we're all on the same page and then help make your job easier. That's what a lot of teams have done, right, is you break down film. Let me do it for you faster. You draw cards. So one of my favorite quotes when I was trying to build a product for football right out of college, I asked my offensive line coach, what's your biggest, you know, problem? How do I solve it? He goes, I get paid $400,000 and I play with crayons 90% of my day because he's drawing cards, right?

And so it's how do I make your life more efficient? If you're going in and looking at, hey, I want to look at every screen pass more efficiently than going in and breaking it down myself, creating automated reports. That's the first way to do it. And then once you've made that person's lives easier, and then you can show them what I can also do in the same method is make this for you. And then it's also, hey, you're trying to find better markets like you've talked about everything in this world is just market. So where draft picks were valued now, free agents were valued now. It's probably shifted, all that stuff.

It's huge different ways for how you think about it. Change it. Don't go, I've created this model that now will make us better at decision making. You should implement it. It's, hey, there's an opportunity for us to have an advantage in fourth down because they don't have the best interior deal offensive linemen, right? Or they line up in an eagle front. If we're in 11 personnel, we can rush the ball and run these things certain ways. That's your speaking the coach's language, right?

if you're not speaking the coach's language and you're just saying here's this number, you're never going to get through to them. So let's talk about modeling because I feel like that's at the center of a lot of the friction that has existed here over the last couple weeks and a lot of the conversation that's been had. The common refrain from the old guard, you alluded this a little bit earlier, is that modeling can't take into account all of the factors that should be considered when you're making these decisions. So as someone who looks at this stuff closer than we do, what actually is,

actually gets taken into account with these fourth down models that we're hearing so much about now. I sound like I'm 80 years old. It's like, what's this modeling I've been hearing all about? Like I'm like some grandpa coming at this. I'm 35. I was like raised in the football internet. But whatever. It's a whole different thing. Yeah, the new wave that is models. Yeah, I think, I want to make sure it's clear. What you'll see on Twitter and the internet is going to be different than what you'll see each team come into each game with. So each team comes. in the game with what's, they often call the book, right, which is they're situational. There is no technology used in stadium by these teams.

So it's not live and dynamic, right? If you're coming into the game where they're, hey, we're going to produce me a 2420 game, right? Like our boy Tony Romo was able to figure out for that Chiefs' Bill's game this week. I can't believe you predicted that correctly. But it's, you know, if you're coming in and you're predicting it and now it's 3-6 in the end of the game, right? Your model was predicting the certain scoring pattern and certain things. it might not be as quality, right?

And so you're in-game, they don't have that. So if all of a sudden the wind changed and there was an act of God that changed how everything worked, then your information on what the likely or the wind percentage or the wind miles per hour would be different than what would you have pregame because in-game, those teams do not have a dynamic model. Interesting. They have more information, right? They have more information about their own team than we, than a model ever could, right?

an internet model ever could, right? And so I don't think a team would ever do this, but let's say a team did not put a guy in a scouting report or an injury report. And, you know, there was no injury information. And we don't, there is no understanding that he might have a bum hamstring right now. And so they're not going to utilize him on deep passes

because they don't want him to overextend, right? There would be no way for a model to understand that if there's information that cannot be plugged in. What dynamic models that you'll often see on the internet, they'll have updating information based on what's happened so far in the game. The players that are on the field, more importantly, the quarterbacks that are on the field, the defensive players on the field, the next-gen stats model powered by AWS, the one that I largely utilize right now during game day,

is incorporating both the time EPA at the end of the game, so EPA is different if there's 40 seconds with no timeouts, right? What your likelihood to score is is different. And so you have to update for that. But all this stuff is not done in-game in the stadium, but they have more information about their team, right? Weather information are, because AWS has built a lot of great models. You know, we have our index gen stats has the field goal model, catch probability model, all those other models, over-expected models that play into what those decisions are being made. I think something that's also interesting is what you'll see, and I call it Dactronic information because a lot of the scoreboard technology is from a DACTR.

machine that's plugged in. You've heard long ones and short inches and different things. Those change dramatically on how what decision should be made. And so if you're seeing fourth and one, but it's really fourth and one in point seven yards. That's the exact situation that arose in the Browns Chargers game where the Chargers went for it on what was characterized as a fourth and one by a lot of people on the internet. And it was fourth and one point seven. And what you can do on that sort of play, now we're getting down the rabbit hole a little bit,

but if a quarterback sneak is 80% successful on a fourth and inches play, you have far fewer options on 1.7 yards to go to have an 80% probability of success with any given play. So those little tiny differences between what is actually happening and what the models can say if you're just somebody who's made your own watching the game, I can see how big of a chasm that can create. Right. And your model is only as good as the inputs, right? And that's really important is to be able to get what those inputs are.

Now, here's what the questions I have for teams is, is that coach considering that? Or, you know, if our running clock situation, they have 40 seconds, they have 18 seconds to call that or get that play call in to make sure they have the right personnel in. That's a very fast call. Are they looking up at the scoreboard and I'm looking down at the field going, that's not right? And then they have to change their whole mindset. Then they have to go to their play card or if they're not the offensive coach, they have to signal somebody else. Those are really hard decisions. So, you know, they might make the decision off of Dachronic information.

I'm not a head coach. I don't know. And so what you might be seeing inputs in stadium might help drive decisions because also the analytics experts are often upstairs. Who is the best view of that true yardage on these short yard situations other than the head coach? So I guess that this brings me right question. Like this kind of, in my opinion, gives credence to the fact that a lot of these decisions have to be made more by like gut feel than they do have to be by pure analytics just

because if it's not updating in real time and you have such a limited amount of time to convey or communicate that information, then these decisions can't be as analytically driven as we want them to be, just practically. It depends. So I'll talk out of, so that, so here's how I look at it. I played center in college. And so my job was to call the mic and call the protection on all these plays. And people are like, wow, center's such a hard position.

You don't really know, you have to make all these calls. you have to me, well, actually, it's only half the playbook, because 96 power I block back, 97 power I block back, right? So it's actually not that hard. But what is hard is memorization of all the possible situation that would come up. And so I would have to memorize, okay, here's what they like to do in first and second down in the field. Here's the like to do again start when we go formation to sideline. I have all these situations I have to memorize to then be I could call the front and predict what my team's going to do. Here's the likelihood they pirate, all these different things. And so then on game time, it's, I'm in flow. I'm just

practicing these things. And so what I want analytics to do is you're going to use fourth downs more. How do I make this decision faster? And it's not just clearing comms. It's not just saying, we're going unless we're not. Those are all valuable things, but it's how do I change practice time? Because practice time is organized. Here's our first and second down in the field, third and short or short, yard, goal line situation, red zone period, two minute period. There should be a fourth down period where teams are practicing not just the plays that get into fourth, third in fourth down, but also. the plays, here's what we want to do. We want to rush the ball better. Or we're going to design plays like the Philadelphia Eagles that have designed a play. Right now with them pushing Jalen Hertz

from behind and doing the Bush push on every play, it's a high conversion rate. I think they're 15 or 16. That play will likely be ruled out next year. I'm sure competition committee is going to look at it. They're going to ignore a lot of other mistakes that they might have to fix, but they're going to look at this play because they're pushing it and we're creating. And we saw late Van derrisch trying to do the Lavar-Aryant and Joe Brow the top. That looked like a dangerous opportunity for a dangerous play, right? But it's converting so well, they're going to continue to do it. So they're actually using what in-game mechanics you can have. Now, you need to be able to do that faster. You can tell the Philadelphia viewers are practicing different

mechanics pre-game so that that decision-making gets easier, right? And so if you feel more comfortable fourth and second down, or your defense now knows they might have to run on the field pretty quickly, right? They're going to have to be able to defend short fields. What's that different mindset? How do you do momentum shifts, right? Which I just said the evil word in analytics momentum, but players still have emotions. You can still take advantage of it, right? And so, you know, there's different things that you'll have to then account for. So we now have information on how to make better decisions in game.

How does it help our game planning? How does now our game planning then go back into what we do in game? It's all that memorization to be able to make those decisions faster. It's such a fascinating way to frame it because I think you see that all the time. The Eagles are a perfect example. I think the quick math that I did before we started doing this, the Eagles have added like almost two touchdowns worth of VPA and fourth down conversions this year. It's right near the top of the league, which when you watch them play makes total sense.

Their offense is built to go forward in four downs, to convert in four downs, and they have a clear plan. There are so many moments where you'll see their third and fourth down decision-making paired together, and that's the way that they attack these problems, because their processes have been clearly laid out and clearly distilled. The Browns are another perfect example of this. example from last week. They had a third and five, I want to say, from about the 30-ish yard line. They ran the ball, and they only got a yard and they ended up kicking a field goal. I'm sure that people watch that.

Be like, why are you going to run the ball on third and five? Well, it's because they're going for it on fourth and two or less, and they think they can get three yards here. The Giants are another team that I think you've seen the process come up this year where it's like, all right, we know what's going to happen in these moments and we're planning on four downs. So this almost feels like because so much of it has to be done in advance, another place where you can gain an edge solely by preparation and refinement of process is what you're telling me here. Absolutely. Yeah.

And it's trying to frame it as a coach already does it. They already do this. And, you know, when they're going to their play card, it's not mad at where they go to, you know, they just pick a random play or the suggested play for them of the three plays on what to go in a certain situation. They have the play card that gets them go down into situational football. They have their scripted plays. They have their situational plays. And they be able to pick it and go.

That same concept that's to come with territorial decision making, right? We have an advantage in inches, right? Well, some teams don't have that advantage, right? Jacoby Brissette has an asset in converting four thousand of QB sneaks. Tom Brady has it. Russell Wilson doesn't. He's very good at other things. And so they don't utilize it.

But that to me is where I see a deficiency, and what the Broncos are doing is they saw that they had a deficiency in interior line, and their fourth down decision to go for it near a goal line on Thursday of football. And the best part of the Washington commanders team is their interior defensive line. And they run with three covered guards, full back belly. Right. You should know, hey, we're going to run full back belly. We feel confident this week. But if they show up in barefront, we have to get out of it. Right.

That's what analytics tells you to do, is it tells you to drive that decision making, where are the differences, the deficiencies you have versus the other team, and how do you gain an advantage? And there's some teams that just say, oh, I've now need to make fourth down decisions based on a model. No, it has to change how you look at the entire game. So I guess that brings me to my next question. What is and is not in your mind an analytically driven decision? I think the best recent example is probably what Josh McDaniels did in that two-point try against the chiefsumina-9 football, right? It's decried as this analytics-driven choice when it fails, when in reality, if you look at models in real time, it was not an analytically driven decision.

It was something that was based mostly on a gut feel and the flow of the game and all that other stuff. So that example you just shared about the type of front. How are you trying to bucket decisions that you see as analytically driven and decisions that we shouldn't color that way and we shouldn't blame analytics on when they fail or succeed? So something I'm learning this year, and this is what great about analytics is the iterative process, right? and it gets better over time, right? And so to me, because it's a model, you can update it. To me, one of the things I'm learning about is what end game situation is four and a half minutes left, right, when it's a back and forth score. So if they had converted that drive, right, and it was a toss-up on the next-gen stats model, right?

So that's the one I utilize that they had a toss-up. So it had no decision one way or another. And so then you look at it from an analytics point of view. This is if, you know, you get it. now you have the lead, you'd then have to just stop Patrick McHolmes one time and you get to win the game. The other part of it is, why would you ever give Patrick Mahomes four downs? To me, it was almost like it was better to not convert and then you get four downs because there's four and a half minutes left, right?

So four and a half minutes with all your timeouts, there is a high likelihood you get the ball back, right? And no one, everyone knows doesn't happen. If you have a top 12 quarterback in the NFL, your ability to come back in those situations is good, right? Now, that's not analytics telling you that. That's just being able to see how teams have progressed. Kickers have gotten so good. There's opportunities, you know, to be able to come back and beat the team with the ball last because then you get an extra down.

You have better decision making. And that was a little large situation of what, you know, John Harbaugh did against the bills is, you know, they did not plan on getting a touchback. But if they, you know, scored a field goal there, then you give an extra touchdown or an extra down to Josh Allen to get down there, right? And that was a large part of the decision making is how do I influence their decision making with mine? So, you know, sometimes teams will do gut decisions. Sometimes if I were a team, I would always blame analytics for every problem I had, even if I made a gut decision because then I don't have a scapego.

That's a valuable thing to do. It's like, oh, analytics made a decision. And so there's situations where, hey, last year, fourth and one, or maybe it was fourth in inches, Baltimore Ravens went for it on their own 30 with, instead of putting the ball to Patrick Hohms at 40 seconds left, right? That was a large analytics decision, but we all see the video of John Harbaugh going, do you want, Lamar, you got it,

when they're making a decision. I think that's still an analytics decision. There's a lot of fanfare that you can do if it fails, blame the thing that you can't fire, right? You can't fire the model. So that's what the beauty of blaming analytics is. But also, I think that's part of the framing and it's kind of selling the human element of it, right?

It's the same way we talk about how you communicate these ideas to coaching staffs, how you communicate certain mindsets to players, saying we're going to go for this because the numbers say that we should go for it. I assume you feel more confident about the way Lamar Jackson feels about it if he's the one making this decision. And I think that brings me back to the human element of this in a way that I've always been kind of confused and perplexed by. This idea that if you go forward and fail, you've incurred so much risk. Rather than the risk, that is realized if you don't go for it and you give the ball back to a Patrick Mahomes or one of

these offenses. And I think that's what is one of the most confusing elements of this for me, is that we have these kind of football guy types that are making these decisions that are actually the more conservative risk-filled decisions rather than the more aggressive I want to take the game into my hands decisions because in their minds they're coming down from some dude with a calculator on high. And I've just never understood that part of this conversation. Yeah, it's very weird with all the offensive gurus we have hired, they often are some of the most conservative people. That is a wholly different thing. That will always baffle me.

So it's funny. I'm trying to think, like, you know, I have to reframe a lot of my thinking. So Brandon Staley was one of the best for going for it. And then the first Thursday night football game we had, he stops going for it as much. And so I'm trying to think, like, what happened? Oh, we spent $100 million in defense. He's a defensive coach. He's going to trust his defense. So maybe he was a football guy after all, right? And I still consider myself a football guy, just, you know, maybe a de novo, Nuevo football guy.

But, you know, there's, it's the, okay, at least I get it, right? And so that's how I look at all this stuff is like, analysts can tell us so much. There's so much variance in all the different models that exist, right? So it's not, you know, we talk to ESPN. There's Ben Baldwin's model. There's an extra chance of stats part of an AWS model. you know, the first people who created models are sports books, right? And so why should we trust models?

Because sports books still exist. Some sports books don't even use models. That's what the story for another episode. Sportsbooks not only exist. Sportsbooks are popping up every 18 seconds because of suckers like me. Like there's a reason that those models are working and continue to work. We can do a whole episode on the business liability of sports books in general, but let's not go into that too far.

But the whole idea is these. models are different across the board. So if you have what I call like insider information, right, if you have access to information that the model can't possess, like certain player injuries that where they're playing in the game, right, or personal relationship issues, or, hey, you know, this, we don't have a play scripted for this situation that we feel confident in, especially early game season because you don't know what your team is, right? So you can use a lot of models use historical data from the previous season. So you don't know how your team is. Officiating changes, right? What certain officials do. Always go with that field. If you have a reason,

if your reason is, oh, I can't believe the model would suggest that, that's a problem, right? You're not listening and you're not trying to get better, which every time. So I went from a player to then, I coached coaches because I was running the football ops department at the XFL. And it was really hard for your job as a coach is to push your player as far as you can. And that's what I like coaches to do with themselves. Push yourself. You ask your players to do the unthinkable on situations. You should try and do that thing.

Push yourself to learn analytics. Push yourself to learn this world. It's only going to make you better. Even if it's that I know that John Harbaugh, when I play him, he's going to use analytics. And if I don't believe them, I know I can use it against him, right? At least learn it from that perspective. Don't just, you know, ignore it. And I think that is a huge point in the argument.

for offensive-minded head coaches specifically giving up play calling and taking on that game management role. The two examples we used earlier, Brian Dable and Nixiriani, have something very specific in common. They have given that up, and I think it's allowed them to really focus on these sort of processes, because it's not just what you're doing in real time on game day. It's the amount of space and attention you can provide that stuff from Monday through Saturday that you can't do if you're the figurehead of the offense. And the Browns are a different kind of example because the Browns have 17,000 people that can work on this throughout the entire week. So I think that Kevin Stefansky can call offensive plays and make these decisions

in a sound way because he's got a lot of help where other teams might not. Yeah, and that's a huge thing. The salary cap only applies to player compensation. Yep. And fans need to know that. Because, you know, when Cincinnati played Alabama last year in the playoffs, that was a closer matchup than the Bengals. and the Rams to be completely honest, right?

Like that, you know, the amount of what he's spent between these places on different parts of it. Now, Sam Francis from the Bengals is awesome at what he does. He's the analytics guru there. He does a lot of great work for them. And so, you know, but it's a different world. He is the analytics department. Right.

Right. But what it's easy with one person, so is you can get integrated. If you're a good personality, if you're someone that's going to be willing to do it, you know, he's integrated into the staff. Yeah. And so if you have 50 people working on new computer vision models and technology that you're to innovate, and it goes nowhere, it doesn't matter, right? And so it's all about getting outputs and getting people to be more advanced and getting

a change in behavior versus, you know, even having the most elite group of people working on these things. If there's no, you know, outcome from it. And it also helps when you hire someone that you know is a true believer. in it. If you have that huge staff, like Kevin Stefansky is the Cleveland Browns head coach in part because they knew he believed in this stuff before he walked in the door. Like, that is not an accident. I would be more than willing to say that. So let's take this to the framing part of it because unlike me, I guess that's sort of me. I talk about this all the time. But you have a

hand in how these broadcasts are going to be put together moving forward. What you guys have done with the Amazon broadcast, I love so much of it. So what? What is a lot of it? So what is a is what is the conversation right now with you guys and how you guys can work on framing these conversations? I think the typical refrain that I've seen here recently is, all right, we have a Mike Pereira. We have a Gene Stereator who has come on and explained some of these officiating decisions and some of the rules to help us better understand what's happening on the field. Are we going to get to a point where there is an analytics expert that can come on and we can tap them during the broadcast and they give us that sort of clarity?

Yeah, so I'll give you a little bit of the mindset that Amazon has with our Prime Vision broadcast, specifically the one that does integrate analytics into all 22 field into the game, and that's to super serve fans, right? And we're in the business of giving fans what they want and making sure they're better, smarter football fans. That's what my main role is. If you want to look at my SOW, you can erase all the fancy words and just make smarter football fans at the end of the day. and know the why on a lot of the situations.

I think what we're going to do is we're going to continue to iterate. You've seen the go-for bar on the right based on what down a distance or what's the distance it would go. We've changed how we present that. We've changed a lot of the stuff that we present on the bottom of the screen. And we're refining what we're doing. And if it comes to an opportunity to bring someone on air to help explain those things, it is. It is a difficult thing to do, not just because it's different than a rulebook where a rulebook is something that's written And this is dynamic in real time that we have to then interpret what the best thing for a fan to hear is.

Right. And so we're not in the business of saying, in hindsight, you should make this decision. We want to hear what the fan to hear, what the coach or see what the coach hears in their headset. Right. We don't want it to be, hey, this is a wait until after they get the results and be like, yeah, that's what you should have done. Because clearly the results dictate a lot of what you should have done. but we want to make it sure it's more what the fan here gives them exactly what they want but then how do we make sure we can present that information in a thoughtful, clear way

that doesn't confuse the fan, right? Because by giving them information that and trying to explain EPA to the average fan, right, we're Thursday night football, you know, it's late at night and I'm explaining expected points added to a fan, it might be a little bit more difficult than just trying to give them, you know, win probability or a certain situation that doesn't. And then there's other problems is these are machine learning models. Machine learning models are derived from machines. So for us to go in there and just explain, here's exactly what the model was thinking,

the different ways it was going, that can be difficult too. So we're looking through different options of how you can best explain this information to fans where they can understand it. Who do you think is the best messenger for that? I think it's somebody that's kind of lived it and kind of can understand both sides of it where you can, how do someone understand analytics and how it gets into a decision-making process and then how do you actually execute on the field? Because those two pieces are tied together, right? Like we started earlier,

you cannot make a decision not knowing how you can execute, right? I can tell you with Jim Harbaugh, on our first two-point play of the year was going to be Q8 or, you know, F-salsa Q8 and we're going to have Doug Baldwin going to go into the flat, we're going to run pick routes on the side, right like those those things are gimmies right well later in the season you don't have that play as much anymore people are going to defend it right and so when you may explain forth that decisions explain not only what the decision would have been but how a coach can react to that call right here's what i've seen from them in short yards goal line which will be a lot of the situations or this is their their defense isn't holding up you have to add in that caveat to try and

predict what that coach is thinking not just give that day that number out because The other part of this that we haven't talked about is the strength of the number is so important. It's not just, oh, go for it by 0.7%. If anyone thinks their model has 0.7% is so, so fine-tuned that it can account for 0.7% that they're fooling themselves. So you need to make sure you have not just the information and you have a strong enough go-for, a strong enough suggestion to then make sure a coach can feel confident in their decision-making, to them feel did they execute and practice this situation well enough to ultimately put on the field.

And it's a really good point to point out the differences between this and the officiating because it is much more black and white as it relates to the rule buck. And this is much more nuance and bringing somebody on to discuss all of what you just said in 30 seconds would be hard to do. But the pursuit to me is still worth it when you consider what this stuff sounds like right now. because if anything, the lack of nuance and the way that we're presenting it in black and white terms now is misleading fans in the other direction, which I don't think is helping anybody. But I understand that it's a lot to wrap your arms around. And this is why you're getting paid the big bucks here because you've got to figure this stuff out.

Yeah, it's definitely a fun problem to solve for because there's a pathway where I think every coach is using it in years. You know, every coach is buying in. And I think as teams that have coaches, there are teams that buy in and then those coaches get hired elsewhere, they're going to do it. You see it with Frank Reich going from the Eagles to the Indianapolis Colts, and they have a great team there with analytics. And, you know, read the articles about what Frank does

and how he trusts John and his team and George. John Park. Yeah, he's awesome. And, you know, when you read the reports from them, it's the analytics said to go, we had this play in practice. it didn't work. I should have followed the game flow

and gone up the middle, right? Toss worked in practice. That's a very different type of coach who's talking about it, right? Analytics to go. He's really saying that my preparation with John

this week gave me confidence that we had a good play at this situation. I didn't call the right play or my team didn't execute. That's taking the analytics out of the conversation. That's a better way to think about

how they're trying to get better instead of just trying to have this department did this, this department. You wouldn't hear them say, oh, the old line was crap this week. It's also, we're very far away from having those people be in a place where they're ready to be broadcasters. This is in such the nascent stages with how many coaches are thinking that way, how many can talk about it fluently that way, that there just aren't that many options for being able to present this information in a concise way that is adding benefit to the broadcast. So I can understand kind of the reticence there.

You can't make the game more confusing. And that's what's really hard. I mean, I'm confused by this, and I spend most of my time thinking about it. So somebody that just is turning on Thursday night football, three beers in in their basement this week, I think is going to be in a little bit of a different place than you or I or somebody who, again, this takes up a lot of space for us. The last thing I wanted to ask you, you had such a fascinating role when you're with the XFL, kind of director of football operations. You worked in a lot of strategy, innovation stuff, really just seeing what ideas you could incorporate into football as a whole. So I wanted to ask you before we got out of here, what to you is the most practical addition to the sport that you think could make football more fun and entertaining? I think there's two big rules that I was really proud of, and that was, well, there's three main rules.

But the two biggest ones I think the NFL could implement is the coach to player communication going from just the quarterback to all players and offense. I even didn't have it in the offensive lineman's helmet because it's very easy to communicate between. but it speeds the game up. It makes the game faster as well as Robbie Anderson's coming from the Carolina Panthers to the Cardinals this week. If he had coached a player in his headset, you could tell him specifically what to do on each route. We could see him this week.

I think everyone would be happy or seeing the best players on the field, which I don't know if he gets that much playing time this week. And so there's different ways that you can expand it. Is that even position coaches? Is any coach to any player or just the head coach or just the main communicator? So we had it. The system we had built was making it go into every position player, right? So I was solving for game speed and both getting subs in and out,

and then as well as being able to communicate after the, after the, you got a plan, right? So I had players coming in either from November, which were good. And then I had players join in January and then I had players joining the week. So for the XFL, I wanted to coach them up. So we have a guy named Armani Edwards who was able to play right away because we did one channel, we did single channel communication, just like the NFL, but with everyone, and you can call play. So how long would say six? And then he would say, Armani, run a go, right?

Or Armani run this. And everyone hears that. There was a world and a pathway to make it so that every team, every position coach could have their own mapping to their position group. Right. We didn't necessarily, we didn't, you know, that would take a lot of testing. So one of the big things that is I tested all these rules over two years, six different live testing sessions. and, you know, that was something that was really difficult to kind of get through. It was first telling the coach, hey, you're going to talk to every player on offense,

and every coach hated it until they finally got their hands on it. And then they go, oh, okay, now I can utilize it. Hot routes, I'm now hot routing everything. I'm doing Madden, right? And so that was a big one. I think it helps with player safety, too, because eliminating the huddle, the wear and tear that offensive players get running in and out of a huddle is unaccounted for, but over a season, it's miles and miles.

Yeah. It's one and a half miles a game for a wide receiver to run in and out of the have a huddle that that's just wear and tear that we can take off their bodies um the next one's the kickoff so the kickoff was making the game more exciting it's definitely a weird look and play i was at a bar recently and i saw a european football league doing our kickoff and i was like what is this and i go oh wait this is the thing i invented okay now i like it again but but the point is is that it makes the the kickoff safer um and it makes us see more action i think that very rarely do we have hundred yard plays and the kickoff is the only kickoff return is the only hundred yard play and i think by making

that play safer, making that play a play again, not just a line and I'm on an Excel spreadsheet. I think that's a really easy. Easy is the hard word. It's an innovation I think the NFL should make, not necessarily. And they would need to test it. They would need to look at it from their players' point of view. But I think that's a good one. What is the dumbest idea that you came up with that you think could potentially benefit the NFL game? So, okay, I had, this is a long convoluted one, but I really wanted to get rid of the chains, right? There we go. I'm into this.

Let's go. This is it. Every fan, what do you want to do? You want to get, what's the number one thing? Let's get rid of the change. It makes no sense. We have this very sped-up game and this very high intensity game of inches and then just the two guys holding chains.

They're actually paid like $10 an hour. So it's like there are these people that are now mandated the most important part of our game. And so what I was going to do is I was going to use the chip in the ball, which again, you have to have a patent that's now in the center of the ball. It's five inches from either direction, likely, that you have to then code into your ball, right? And I have to make sure it works all the time.

They have to make sure it works in real time. All these things are hard. And then I was going to put a ribbon board on the opposite side of the field as the chains, right? I could sell ads on it. Okay, that's a million dollars that I got to set for me. But I go, okay, now I have to put each team on their own sideline. Well, I looked at baseball games and baseball, and when you play in a baseball stadium, like the Shamrock series,

oftentimes the teams on the same sidelines. So I was like, oh, speed up the game. I'll have a sub zone. So I'll have everyone recessed like in soccer that's not on the field because defensive players won't go. And I'll have the both teams side by side moving all the way 100 yards up and down the field. So you don't have to stand over the ball to sub because people don't realize that rule. When you stand over the ball, do you know how much time an umpires supposed to stand over the ball when it offense subs? No, no idea.

It's whatever they choose. The one, whatever they believe is the right amount of time to let the defense sub. What a stupid sport. It's a beautiful sport. So the long story short of it is we were never going to solve it. It takes it's 0.75 times a game for chain measurements. What we decided to do is I just started every new set of downs on a yard line. So you always measured from a yard line, right?

So if you got six and a half yards for a first down year and a half yard nine, we just started moving up to the next yard line. Over time, every team's going to get those half yard decimal place and it's going to round out as long as you keep getting first downs. And the team that's getting first downs more often likely wins the game. So if one team got more half yards than the other, they were just a better team in general. So it was a long thing to do. And I couldn't.

And the funniest thing, the hardest part was not actually the technology. It was getting the coaches to agree to it. You know, because that's all these ideas, you know, you come up with these things. And, you know, of course I thought of, you know, what if you, you know, make offensive linemen eligible, covered players can go out for passes because I want to see more innovation offense, all those things. But getting rid of the chains is a very hard problem to solve for not that much value added. And then just to speed up the game as well, it's not going to speed up the game.

Well, speaking of value added, before we get out of here, I wanted to commend you on what the NGS broadcast has looked like for your guys' first five games. I don't know if it's just for sickos like me or if you're getting some traction with like the general football audience. It has changed the way I watched that Thursday night game. The fact that I can have it next to me while watching the main broadcast on my TV and I can watch every play twice and have the All-22 angle. It's incredible. Like, it is such a value added to a real-time football experience.

It's been a long time coming. And I'm very glad that you guys have gotten the playground to be able to incorporate and try something out like that because I look forward to it every single week. Like, even last night, I'm sitting there watching my football. I was like, I miss this. Like, I just, I wish I had my little All-22 broadcast. next to me here. So kudos on you guys pulling that off in year one and what the entire Amazon operation has looked like because I think you've done a phenomenal job so far. Yeah, the Amazon teams are working on this for a long time before I ever joined.

And Alex Strand and the team, you know, they've done an awesome job. And it's still, what I love about is it's a very inner process, right? We're very happy we heard from fans that they really love it. Some fans love the 22. Some fans love the stats, some fans, and we're just trying to make it a better broadcast every. We are super serving. all fans, even the sickos like you, because I'm one of those sickos, right? To better, you know, better watch the game. And it's awesome to hear that people like you are enjoying it. Sam, really, really appreciate the time.

Really appreciate you helping us dig into this conversation. Hopefully we can do it again some time. And best of luck making football weird. Yeah, weird, fun, somehow more fun, and smarter fans. That's the goal. But we're definitely involved. I'm smarter now than I was a half an hour ago. So thanks a lot, man.

I am thrilled now to welcome my good buddy. Ty Dunn from Go Long and the author of a wonderful new book, The Blood and Guts, How Tighten Safe Football. Ty, it's good to see you, man. I appreciate you coming on to do this. Great to see you, Robert. My God, I feel like I have so many Robert Mazes in my life. There's the version that I've had beers with on the road. There's the version that I listen to when I mowing the lawn out there.

And now we have this version via Zoom. the yes and about tight ends. It's weird. I feel like you're in my life in so many different ways, but it's an honor to be sitting here with you in the virtual world. That sounds horrible. I feel so bad for you that I'm affecting your life in all these different ways. But I appreciate it nonetheless.

Let's dig into this. Huge undertaking reporting this book. And for people who don't know, it's really about the history of the position and then digging into the individual guys who have shaped that history. You're going all the the way back to the Mike Dickas and the John Mackey's and a lot of the big names that we know about.

And you trace kind of the course of the entire tight end history. I wanted to ask you just first and foremost, why tight ends? Like, you're somebody who has done this for a long time and we've known each other for, I don't know, at least a decade now back when you were covering the Packers. And now you're doing a lot of long form features. That's what you do. You could have written a dozen different books about the NFL. Why tight ends? You know what? It really stems in probably what drove everybody nuts last week, right?

Where you're watching an NFL Sunday and Chris Jones, Grady Jarrett get flagged for basic football plays and we all lose our minds and we all wonder what the future of this sport is even going to look like next week, next year. And what happens on Tuesday?

We all put our waiver claims in. We plug our fantasy football lineups in. We tend to forget, right? But it's shot into the ether. and I feel like the game itself is changing for worse. Hey, I'm an old soul. I love football the way it was intended to be played. And that's with an element of risk, an element of violence, that attrition, that moment, okay, is this for me or is this not for me?

Like, we've all played the game. We all kind of had that feeling. And the NFL is that times 100. I mean, these are gladiators in a profession that we can't even wrap our minds around. My God, you go up to training camp, you're beating the hell out of each other, 95 degrees, right? You're literally throwing punches, and then you're in the cold tub talking about each other's families and kids. And that's not normal, right? And that's pretty damn cool.

Like, you can't just, in baseball, you know, you pick up a bat, you join some buddies and you play a game out back. Basketball, you can, you know, round up some friends and play at the park. Football is just different. And that's why we love it. So it kind of started in that frame of mind of like, how do you save this? I mean, I could have done a book about lineman in the trenches, life in there, but usually if you're doing your job right, you don't even hear about an offensive lineman for three hours. That's quite a sexy answer. Selling that idea, I think, is a little bit more difficult than like, I'm going to throw Grunk on the cover, you know.

That's so true. But you have, the Granc has to do it, right? I mean, he's literally breaking a dude's neck in his rookie season without even really intending. Like he's got that element to his game, but also it's third and eight. You're going to him down the field. He's parted his ass off in the end zone. You've got the character. You've got the authenticity.

You've got the personality all there distilled to the tight end position. So, I mean, that's how it started, Robert. But honestly, when I started traveling the country and hanging out with Jackie Smith, with Tony Gonzalez, Mike Dick, Jeremy Schott, he came down at the bar in Miami Beach. It kind of took on a life of its own where, like, these dudes were, uniquely qualified to save the sport because of their lives and everything they've been through. And it's a book about life as much as football, honestly. I really appreciate it.

Even in the intro to the book, you talk about how tight end is really the distillation of football in its purest form. And I think that's exactly right. And there really is no position like it that combines the grace and the violence that we get when we watch football. What these guys are asked to do in terms of, the actual refinement of their skill sets as past catchers combined with just the nitty, gritty, disgusting elements of having to sit in there when you're in max protection and take one to the chin from a 310 pound defensive alignment.

These are the guys asked to do that. And I think that really does make them so ripe as characters to explain what the sport really is at its most basic level. So I want to talk about that, just how these guys have shaped the game. we know and the way that we understand it. When you're going back through it, who do you think among the guys that you talk to or is the guys that played this position? We don't think about enough in how they shaped our modern understanding of what tight ends are. Great question, because you could go in a lot of different directions.

You know, to take it way back, it's so wrong in criminal and unfair that Jackie Smith is defined by 5.5 seconds of his life, right? when there was so much that went into that drop in the Super Bowl. I mean, that this guy has such a remarkable story from Kentwood, Louisiana, to redefine the position in his own right. But I know that the name that keeps popping up in my head, Jeremy Shockey, because here's a player who,

honestly, when he came on the scene, I think we probably expected more, right? I mean, after watching that Hall of Fame game, and he's just blasting through the Houston Texans defense, and Ernie of course he's talking about him being the next John Mackey, and we all saw what he did at the U. And he had a really good career. I mean, I think he had four pro bowl appearances.

He won a Super Bowl with the Giants. He won a Super Bowl with the Saints. I mean, he can stand on his numbers, but it wasn't to the level of Gronk, of Gonzalez, of Kittle. And I'd come back to Jeremy Shockey, though, because he really was Mike Dicca, what, 40 years later. And he approached the position with, that apex predator mind frame that is football.

I mean, if you said a bad word to Jeremy Shockey, he made you pay. He didn't forget at the U. He's just starting up fights for the hell of it because that's what you needed at the U to build what they built. And he did the same thing with the Giants. He breaks down the story of, you know, day one in training camp when him and Brandon Shore are absolutely going at it because he doesn't want to go through the rookie hazing ritual of, you know, reciting the fight song and telling him

everybody, you know, his signing bonus. He's had enough of the shit. So they can start punching. And Jim Fossel's at their ankles, like Jeff Van Gundy. But then guess what? Fossil loved it. He encouraged it. Brandon Short loved it the next day.

He's like, man, I love this dude. And I think that that just can change not just the position, but your entire locker, your entire team, when you can kind of take on that persona. And everybody noticed. I mean, Rob Grunkowski here in Buffalo is just a kid and sends him a letter and just tells him how that you're my idol. I want to be you one day. And obviously Rob Grinkowski took this whole concept and just shot it to another stratosphere. So I just feel like his play style, his authenticity to just not give a damn, you know, what he said and who he said it to,

calling Tom Copland an asshole of the media. I mean, his players the next year are like, Jeremy, what are you doing? He said, oh, it's America, First Amendment, I can say what I want. He just didn't get a shit. At that moment, for him to come onto the scene, as people will read, It was really important. And then Rob Grinkowski and George Kittle kind of took it to another level because obviously Garanx is doing it in New England where we all assume everybody's a robot. Not so. It's fun because you have that kind of mindset and feel around these guys.

And obviously Rob becomes like Yo, Soi Fiesta. And like that's a huge part of his entire personality. And you dig into that in the book, which is very fun. But there's also these moments where these guys are involved in kind of these inflection points with the way the sport is played. and the Kellyn Winslow was the guy that when I was reading the bit about Kellan Winslow in that chapter, that really jumps out where you have Don Coriol kind of at this pivot point of what the modern passing game will become and Kellyn Winslow becomes such a huge part of the way that they shape that.

And that's kind of why tight ends are so interesting because they are in the middle of all of this. It's a combination of the run game and the past game. It's the way that you play and how you structure an offense. It's why I'd be curious of more tight ends, coaches gain more opportunities to be offensive coordinators and keep kind of going because they see everything. So the Kellynne-Win-Lis chapter really jumped out to me for that reason just because,

man, this guy is like at the center of the evolution of the way that we understand how the modern passing game became what it is now. That's such a great point because the tight end, I mean, next to the quarterback, nobody has to know more and do more than the tight end. You have to join those inline blocking drills in practice that everybody hates. I mean, like Tony Gonzalez said, you got to do the shit you don't want to do. That's why tight end forces you to be a better human being. We all do the things that we all have things in life that we just don't want to do, pay bills, do chores.

But you have to do it. Yet also, you have the glamour side. You have that ability to just absolutely break a game open late in a fourth quarter. So I think it's that combination. But yes, schematically, with what Callan Winslow did with the San Diego Chargers, I think that's really what kind of showed the world what that tight end can do. Because it took the innovation of Don Coriol, like, all right, we're going to put this freak out wide. Somebody who Hank Bauer thought was straight from the Los Angeles Lakers.

And we're going to see how you react. Because if you send a bunch of attention that way, we've got West Chandler, we've got Charlie Joyner, and we've got a guy in Dan Fouts who's going to push all the right buttons. It just kind of, it tilted coverage. in a way we had never seen before. I think before then, yeah, I mean, the tight end was used to an extent. I mean, Dicka and Mackie, they really put the position on the map in terms of, all right, NFL films is playing, John Fissenda's voice is in the background,

and these are two superhuman dudes who we can all just root for. And what is this player doing, catching a short pass and running over everybody? And then Jackie Smith is doing it deeper down the field. and then Callan Winslow is in the epic of Miami, a whole country watching, barely able to move, barely able to breathe, making the plays that won them the game. So I think that he absolutely did kind of shoot it just to a whole new level. You're talking to all these guys. I mean, how many tight ends specifically did you talk to you for the book? Oh, man.

That's a great point. I would say at least a dozen, right? Oh, yeah. So there's 15 chapters. And I was only unable to get Shannon Sharp and Colin Winslow, but like, you know, kind of forces you to talk to a million other people. I think I talked close to 100 people in all for the book. So I'm curious, from those guys specifically that you talk to, the tight ends that kind of form

the meat of the story. What is the most surprising conversation that you had, the one that kind of snuck up on you a little bit? Maybe non-jerry Shockey Division. You don't want to hear about his bar fights down in Brazil? Listen, those are not some. surprising to me, I guess, is how I would frame that. Drinking 100 beers.

They did sneak up on you a little bit. It's Tony Gonzalez, right? I think that when I met up with Tony Gonzalez, there's a lot out there on him, right? He's been on Fox. Now he's on Amazon. We've heard from him. We've probably seen his documentaries, all of that. And his story is remarkable, right?

You can take it all the way back to childhood when he's, he's bullied. He's scared of his own shadow. He has to kind of slay that fear from within. it's unbelievable how far he's come. And he really does kind of, he kind of breaks it down in terms of the Norwegian way. And HBO Sports did a great special on how Norway ended up dominating the Winter Olympics it was. Yeah.

They made sports, right. You saw it. It made sports fun. They wanted kids to organically learn to love, you know, whatever sport they get into without peer pressure, without parental pressure. He kind of had that in his own way because nobody was forced him to play anything. He kind of overcame this fear himself and all of a sudden he's dominating. But what surprised me was how he put up these incredible numbers in Kansas City in Atlanta.

But it was kind of in spite of the schemes that he was in. You know, initially Jimmy Ray, the offensive coordinator, he got it. He knew he had a freak of nature at tight end. And he went out of his way to move Tony Gonzalez around, create those mismatches, just feed him relentlessly. You know, year three after he had the drive. in year two and overcame that depression and all that. But then they just couldn't win game.

So Jimmy Ray, that staff was fired. In comes Dick Vermeal. In comes the greatest show on turf. And yeah, he put up great numbers. But Trent Green sat down with Tony and was like, look, this isn't a tight end friendly offense. This is about the wide receivers. This isn't really about you.

Dick Vermeel pretty much said the same thing. So he kind of put up numbers in spite of this offense. And then I think the climax of the book is really the showdown he added with Mike Malarkey where he's about his old. school as it gets at the position. I mean, he had the guys like Mark Bruner in Pittsburgh. Tight end should be tight as he told me. Like, you need to be in line blocking. So what's he doing day one with Tony Gonzalez? He's pulling up clips of Art Bruner, just, you know, shortening people's necks and saying, this is what you need to be for me. So Tony Gonzalez

is sitting there thinking, why in the hell did you guys trade for me? Mike Mularky's thinking, why did we trade for Tony Gonzalez? It was this arranged marriage that was doomed from day one. And they went at it. I mean, we get into the detail where from day one right on through, they never really saw eye to eye. You know, Malarkey has Gonzalez out there after practice, bashing into a D-Lignment buried on the depth chart. Gonzalez says, after this, this isn't me. What are you doing? So eventually they quit that.

And then the showdown is in Tampa Bay. He's going for his 1,000's career catch. And he's stuck on 999. Tony Gonzalez says that Malarkey froze him out, refused. to call a play for him until the last drive when Matt Ryan was pressured and he couldn't even get him the ball and they both have their own versions of what happened in the locker room and in so many words both assured that they were ready to beat the hell out of each other. I mean, Malarkey goes around, he's shaking everybody's hands.

Atlanta has winning seasons for the first time in back-to-back years and in franchise history. He's happy they got the win. Tony's fuming. Tony wants a piece of him. Malarkey comes up to him with his hand outstretched, and that's where I should leave the cliff, the cliff hanger, right?

That's where you got to buy Blake. The Tony Gonzalez stuff was fantastic. I mean, he was so incredibly open with that relationship, just kind of his own, I guess reservations about playing the game a certain way, the fact that he, when he first started playing, he didn't even enjoy some of the physical aspects of it. I mean, he's like a different sort of breed when it comes to these guys.

The one thing that jumped out, Apparently he rented a haunted apartment at one point in his life. He moved into an apartment and the guy who rented it to him was like, yeah, the lady just died in there. I mean, it's the amount of Tony Gonzalez stories you got out of that conversation were fantastic. I'm so glad you found that part. It was freaky. Yeah. He said it felt like there was somebody just, you know, in bed with him and just wreaking havoc and making his life a living hell.

And the last day after the chiefs drafted, if he's walking out to his car, and, you know, the landlord, you know, the manager of the place who he said is like your, your quintessential, like, Berkeley looking grad in his words, who might have taken a few, a little LSD in his day. So, oh, yeah, by the way, that, that house is haunted. The lady died there. So, yeah, Tony was unbelievable. He, I guess that's what was a surprise.

I thought I knew a lot about Tony Gonzalez and he kind of, kind of blew my mind. And, you know, it's really, he forced the league to evolve. I think that probably Rob Bernkowski is the best. tight end ever because of the whole package. But Tony Gonzalez can really state his case because, you know, he really forced the NFL's hand to basketballify this position. He forced you to look in new places for tight ends, to think about the sport in a completely different way to where Antonio Gates doesn't even play football in college.

He's getting a shot. Jimmy Graham, right? He's a college basketball player, plays one year football, and he's coveted, who, by the way, was coveted by Bill Belichick a year before that. Right. He doesn't even play it down to football yet at Miami. And Matt Patricia is putting him through a workout because Belichut's thinking,

I want to sneak this guy in the practice squad and develop him as a tight end, which speaks to Belichick's genius. So Jimmy Graham stays in school for a year and continues to kind of build upon what Tony Gonzalez started. And that's really what led to today where every team pretty much has an athletic freak at tight end

one way or another. If you don't, you're losing. The other guy I wanted to ask you about, before we get out of here. Ozzie Newsom is somebody that has become kind of an object of fascination in modern football, you know, a whole fame, tight end,

which I'm sure there's some 23-year-old kid out there or somebody that just remembers Ozzie Newsom from his final days as the Ravens GM and has no idea that Ozzie played in the NFL, which is incredible to me, but I guarantee you that person exists. So when you talk with Ozzy, I'm so curious, because I don't feel like for how central he is to,

the way the Ravens organization has operated, how much success that they've had. And the fact that he kind of is the gold standard for how you build a front office. I feel like we don't know a ton about him, the person, or the executive. When you sat with him, what really jumped out to you about

what aspects of him or his personality might lead to him having the success he has in Baltimore and the way that he shaped that franchise? I'm sure you hear it all the time too, Robert, around the NFL, when you ask, okay, who does it right in the front office? Like who is the gold standard? Because you hear about everybody who does it wrong.

Like this GM doesn't know what the hell he's doing. This coach doesn't know. These scouts are, I mean, it always comes back to the Baltimore Ravens that they just kind of do it right. They don't panic. They take it everybody's opinions.

They have that voice at the top, who obviously was Ozzie Newsom for a long time, who does make that final decision. But it does stem from Ozzie Newsom's upbringing first, you know, growing up in the segregated South, and wanting to go to a white school, wanting to prove, like, I can compete with anybody

academically, academically, kicked everybody's ass, gets to Alabama. By the way, not long after Alabama was T-segregated and dominates there. And obviously has the career he had with Cleveland,

all the playoff heartbreak. And I think that it starts with Sam Ritigliano, knowing that this split end, this receiver in Alabama, could be a tight end. He sends Rich Coat tight down, just to look at, his ass just to see if it was big enough for him to be a tight end,

comes back and says, yep, he's got the frame, he can be a tight end. And it kind of takes off from there. He never forgot that this team and this coach thought outside of the box. Like, he wasn't the prototypical tight end. Nobody was really transitioning wide receivers to tight end. If anything, it was offensive linemen to tight end at that time. So you fast forward and he makes that connection between then and trashing an offensive a playbook, forgetting everything you thought you knew about the quarterback position and building

an offense completely around Lamar Jackson. He never forgot that the Browns did what they did and thought how they thought. And that's what they, he always did with Baltimore. But in addition to that, just that calm and chaos, right? When they move from Cleveland to Baltimore, I'm kind of jumping all over the place here. But when they move from Cleveland to Baltimore, I mean, their roster's gutted. They're working out of a police barrett. You know, they're barely have a personnel staff. You know, the tapes are kind of lined up around the boundary. And it took an Ozzie Newsom to kind of be that, that calming voice, that calming presence

to lead. He was the de facto GM that Art Modell trusted to really build that thing up. And Art Modell and Ted Marchabro, the head coach, they wanted Lawrence Phillips. They did not want John of the Nagden. But Ozzie Newsom was calm enough and took in everybody's opinions and knew that the scouts did a year of work on all these prospects and said, look, Ogden is at the top of our board, we're going to take him. And that decision, coupled with Ray Lewis later in that first round,

just set the course for everything where, yeah, they're not going to panic. They're not going to act out of emotion. They're going to take all the opinions into account and do the right thing by and large year and year out. So it's not bombastic. It's not crazy. He isn't a GM that's front and center. He barely does any interviews.

I was lucky enough to talk to him for an hour. That's kind of why I'm so interested in it is that he's not somebody you hear. hear a lot from. And the fact that you got him for this, I think is, is such a fascinating part about his story about the story of the NFL. And that's what this book does. It tells the story of the NFL through the lens of all of these guys who, whether it's schematic innovation, the way that we understand the mindset and the personalities that drive the people who dictate this sport, all of this stuff, I think can be wrapped up in the way that we understand tight ends and their history. So if you guys love football, you guys should absolutely go check out

and pick up the blood and guts for my buddy Ty done here. Ty, I really appreciate the time, my friend. It's really good to see you. Great to see you, Robert. Man, thanks so much for the love and thanks so much for having me on. Absolutely. We'll talk to you soon. You got it, man.

Thanks so much. All right, guys, that's all we got for today. Thank you so much to Sam and to Ty for their time. Really enjoyed that conversation. Just to let you guys know, didn't mention it at the top. Worth mentioning now, doing a little bit of a schedule tweak moving forward on the show Thursday morning tomorrow. You guys will hear Mike Sando and Randy Mueller doing the football GM.

That is going to be coming to you every single Thursday morning from now on through the end of the season. So be on the lookout for that. We'll still be doing our Thursday night recaps on YouTube. Be ready for that. Still doing the weekly preview with Nate. Only thing that is changing schedule-wise is Mike and Randy going to that Thursday morning slot and replacing the typical show that we were doing on that day.

So that's all we got. We will be back a little bit later this afternoon, 4.30, excuse me, 5, 4.3. We will be back 3.30 p.m. Eastern time. Thursday. Me and Nate will be doing our week seven preview. God, it's a disgusting week of games, but we'll make it work. In the meantime, please subscribe to the show on YouTube. If you would, the YouTube link is in the description of this podcast. It's the only place you can catch our Thursday night recaps.

We're going to be doing a couple Monday recaps here over the next few. few weeks. So please go subscribe on YouTube. Please rate and review the podcast on your podcast platform of choice. Give us five stars on Apple. Tell us why you like the show. It does help us. So it would mean a lot to me.

I'd consider it a personal favor. So really appreciate that. Also, please subscribe to The Athletic. Theathletic.com slash football show where you can read all of our fantastic NFL coverage. That's all we got for today. Appreciate you guys listening. Talk to you soon. This was the Athletic Football Show.