Purple Insider - a Minnesota Vikings and NFL podcast - Bonus episode: The Ringer's Austin Gayle talks about Football is a Number's Game

Episode Date: October 12, 2023

The Ringer's Austin Gayle, who worked at Pro Football Focus from 2016-2022, talks with Matthew Coller about Matthew's book "Football is a Number's Game" and describes the analytics revolution in football and what PFF does for teams. Check out the book here: https://www.amazon.com/Football-Numbers-Game-Data-Driven-Approach/dp/1637272189 Learn more about your ad choices. Visit megaphone.fm/adchoices

Hey everybody, welcome to a very special episode of Purple Insider. Well, I shamelessly promote my own work. Austin Gale from The Ringer is with me. Formerly and a long time Pro Football Focus employee, rose the ranks there and then got a great opportunity now to work with the ringer and since my book football is a numbers game is out which chronicles the entire journey of pro football focus what it does for teams where it came from how it has influenced the league and the way people view football in a gazillion different ways well we're going to be doing some podcasts about it from time to time and i figured austin that i don't quite want to introduce the world to Neil Hornsby just yet. So you're going to lead off the self-promotion tour is what I'm going to

call it on the podcast. But I'm super proud to see you working for the ringer now and building on your journey, which is chronicled in the book, which is truly one of my favorite parts, all the things you went through to get to this point. So super happy for you. And thanks for coming on to help me push my own stuff. Absolutely. Yeah, of course. I'm happy to come on. I always like talking with you. Yeah. And I think a good place to start is that you had a front row seat for the NFL beginning to adopt analytics. And I think that the biggest thing that people are interested in when they ask me about writing this book is the analytics revolution of football. Where are we in analytics usage and the role that PFF played in that? I'd love to get your opinion on that, having been behind the time that you worked there was the adoption of analytics in ways that it never had before. People thought it could not be like baseball and it started to become that, but I also think it's still very much in its infancy.

Absolutely. When I first started with PFF as a part-time data collector in 2016, I was, you know, I think a junior or senior in college. And I wasn't, I didn't have line of sight to a lot of the decisions that Neil Hornsby was making or other people were making. But when I started as a full-timer on my first or second day, Neil Hornsby took me out to dinner and like, he is one of the more transparent people you'll ever meet. I think that's a positive way of framing it. Like, he'll just tell you whatever the hell you want to hear. And like talking to him specifically about the B2B side of this company, like the business to business, selling to teams, selling to agents, selling to college football teams, selling to the Canadian football league and

selling, selling maybe isn't the right word for it, but like helping, impacting, like delivering to these teams that are making decisions on and off the field and talking to him in 2017, 30 of 32 NFL teams had bought in to PFF and were in some way, shape or form getting that data directly. I think the two teams that didn't have it at the time were the Tampa Buccaneers and the Washington football team. And him talking about getting those other teams was so motivating.

I remember him like, every week you're rooting for these teams to lose. You're like, they don't have it. They don't have the data and all this stuff. I thought it was hilarious. But when you talk to them about what do they buy? What are these teams getting? Why are they buying? And it started with the Giants.

I know you chronicle that in the book and how they wanted this edge, this moneyball edge in terms of evaluating players and on-field decision-making, like fourth downs on some of that stuff. But honestly, where it started, it was like these little small things that data can pick up so much better than your gut feel. And like a lot of these things, I bring it up as like going forward on fourth and 30 or fourth and 10, when you're down 30 is an analytics decision. Now it was a gut decision for a long time. And like, we're down 30 points and it's fourth and 10, when you're down 30 is an analytics decision. Now it was a gut decision for a long time and that like we're down 30 points and it's fourth and 10, it's the fourth quarter. We have to go for it. It wasn't a data decision for a long time, but if you look at any win

probability model, it's going to tell you to go for it every single time. Now it was a gut decision. Now where it gets harder to make those decisions, fourth and four, down 10 with four minutes to go, like that's where this data was helping teams so much. That on-field decision-making, when to punt, when not to punt, was a very big part of that. I think the other pieces of that, too, were where they were really impacting the game was positional value. It was looking at how much of this fixed amount of money

should you allocate to the running back position? How much of this fixed amount of money should you allocate to the quarterback position? And a lot of these decisions were already being made. Well, right. Quarterbacks were the highest paid players before PFF ever came along. That was known. Like they felt like, yeah, these guys are the most important players using not data, but anecdotal points like they touched the ball the most and they, all these different things.

But PFF is actually able to tell you what different players are worth at specific positions. Obviously there are outliers to that. Obviously there are, you know, teams that like the 49ers who over-prioritize in terms of spend the fullback position. And that fits the different offensive philosophies. It's not black and white. It's not like don't bunt versus bunt. There are a lot of situations that data, you know, not necessarily doesn't see, but like that the gut opinion still affect. I think the last piece I'll bring up in terms of like what I saw change so much was player evaluation in the draft. And one of the biggest booms in the B2B

business when I was there was colleges buying into the data and then NFL teams buying that college data because they wanted to see does production at the college football level, sacks, pressure rate, yards after contact per carry, does this data at the college level actually translate to nfl success because what the league was doing was how big are you how fast are you let's go like essentially like and obviously they would look at like passing touchdowns and success and those things and like but they didn't really care about production at college football level what we're seeing and what the nfl saw is wait a second pass rushing success specifically in the power five, has translatable qualities to the NFL.

I'm like, we need to pay attention to that. And again, like the NFL just continues to buy and buy and buy and like take up on this data to just make better decisions. And I think, I don't know, we could keep going long on like how much teams use it and how much, why teams bought into it. But so much of it was the market was so saturated. Everyone's trying to do the same thing, win the games, win the Superbowl, that eventually when you get into any market where it's that saturated to our common goal, the data is what's going to bear out. It's what

Moneyball is. It's what all of that stuff was. And people previously thought the NFL, because of its complexity, because of the moving parts, it could never get to that point. There weren't a lot of fixed events. In baseball, there's so many fixed events. There's only so many things that can happen with a pitch. There's only so many things that can happen with a pitch. You know, there's only so many things that can happen with a bat swing. It makes it easier to make data because the events are fixed in the NFL. It's not, but guys like Eric Eager, guys like, you know, George to hurry. Those guys are like really, really smart. And they're like, wait, we can figure this out. And now you have next gen stats and the chips and the helmets and all that stuff.

It makes it to a point where like, if you're ignoring the data now you are not behind you're five ten years behind because of what pff has done over the last 10 years and when we talk about neil hornsby he's the main character in the book he is the founder of pff and the one who built the original team that started collecting data and got contacted by the new York Giants and started selling to teams. And that's kind of how this whole thing took off. And boy, is he one of the most interesting people you'll ever meet. So he's going to appear here at some point, but his drive to get more and more data is really what's behind all of this. All NFL teams and all college teams are using now at this point and i'd love

to speak you to speak to what you're collecting as a data collector because some of the things that you touched on there i think that people are either really aware of if you like analytics or like at least vaguely aware that you're always seeing these discussions that running backs aren't worth it or whatever yeah and i like that you mentioned the draft because that one I think is still blowing up for how much data can mean to decisions where it used to be analytics was just the NFL combine. And now it goes like a hundred layers deeper. Kevin Cole also did amazing work from a data science perspective on what translates from the combine and things like that. But NFL teams, what they told me was the biggest area where this impacted them, where PFF has impacted them, is the game planning.

I mean, because of course the tentacles reach out to player value, and once the data science comes along, that changes everything. But from the very start of PFF, what was going on with teams is getting all this information was helping them understand very quickly what coverages opponents are playing, what routes other teams are running, what personnel packages they're using. So now every team is logging in and they are studying the tendencies of other teams within seconds as opposed to having quality control guys spend hours charting all of this stuff. No. And as a data collector and before ultimate, which is, I think something a lot of people know of, but maybe not enough people know of. It's like a database that takes a lot of what PFF had started doing all the way back when the company started,

like puts it all in one, right? It puts it all in one thing. It makes all that, those processes quicker. It's one link. It's one thing, video access, all that stuff. But before that, when I was a data collector, what you could also do is put together PDF packets for every freaking team. And you'd send them to the teams with like their upcoming opponents, different threat rates. And like threat rate was a big thing where you're looking at when the opposing team is an 11 personnel and it's third and six or less, and they have this running back at this part of the position, they're at 37% of the time they are running this play to the right side and like other threat rates.

And when you get to like anything over 35%, anything over 45% in certain personnel usages or certain down in distances, when this guy's laying in the slot, when he's not just lined up in the slot, he's lined up in the slot on the inside of the other slot receiver, when he's off the ball, on the ball, because PFF is charting all that stuff, right? Like when you are a charter, you're looking at who's on the field, where are they lined up? What is the down and distance? When you do like meticulous charting in that way, the data is going to bear out more fixed events. We go back to those fixed events, right? It was like, wait a second, this is fixed. When they do line up in this personnel, with this personnel in this formation on third and five or

less on their side of the field, on the right side of the hashes, they do throw the slot receiver 56% of the time. There's my fixed event. That's how I game plan. Now, there's stuff that's less minute than that, where you're able to identify concepts that they go to routinely and maybe not as specific as when they get into this, they run this play. But that is like what the data really had helped people with, I think, before Ultimate. Now they're not getting PDFs, I'll tell you that.

Like those PDFs aren't coming in and it's not me like writing all this stuff down in Excel. Ultimate is able to tell you and show you the play diagram, the next-gen stats dots, and the video of any play you ever would want. What do they run when it's second and 10 or more on their own side of the field when DeAndre Swift isn't on the field? What do they run?

Show me every play. And you're able to see that in minutes. And that speed, I don't think the speed gets talked about enough. It is an efficiency tool. It makes what could take a week's worth of time 10 years ago into not days, not hours, like minutes. You can get minutes of work that used to take weeks of time. And that has allowed for teams to do shit they probably never could have done. They never could have done some things. They never could have been prepared for some of these things with what PFF did. And like, that is like an answer that I think some people aren't super

excited by. They're like, well, they just made things faster. They just made things quicker. Like essentially like they scraped things and they made a speedier version of what already existed. Yeah, sure. But like, you don't understand like the depths beyond that, like what PFF is doing to like take that data and then make fourth down models, win probability models, like different positional value models and all that stuff. How much is this player worth to your team versus the contract you'd have to sign him for? And like, that's where you get into the money ball stuff. And it just goes on and on. I think that at its base level, at its worst level, it's impacting your game plan to a point where you're saving time at thousands of percentage points.

Like you're saving time, infinite time. Then beyond that, it's informing decisions on the field, when to call timeouts, when not to call timeouts, all that other stuff. Yeah. When I was first shown what they call the PAT tool, which was one of the early inventions that was really taking the, it took everything that was being gathered and then put it into play diagrams. I was like, wait a minute.

So every team has this. They have all the play diagrams of everything that was done on the field at their fingertips. Like they're not on a chalkboard, like still drawing it out. And I think this is really interesting because teams bought into this and started using it and then started asking for more and more and more data. And the tendency stuff is huge because people who were tasked with the early on selling to teams for PFF going in and saying, here's why you need our product. I think that those coaches thought that

they were going to get a presentation on why their decision-making was wrong. And then they went and told coaches, actually, we can save you thousands of hours over a year of you drawing it on a chalkboard or you charting it out in a notebook or whatever else. And then when PFF was able to tie the video element. So now you can look at these tendencies and everything else. If you're the Vikings and you want to know how Justin Jefferson is being used in the red zone and how it's working or whatever, you can call up all those plays in an instant

and study them as a coaching staff and take notes and look at why the data says what it says, how are opposing defenses playing against Justin Jefferson inside the 10-yard line, bang, which I think has been part of a result in, in my opinion, the complexity of the game increasing. And now the chess game is not any longer. Here's how we play offense.

Here's how you play defense. Let's see how this all plays out. It's more of let's see how our coaching staff prepares. Let's see how your coaching staff prepares. And then who can execute these game plans? Because everybody is going to know everything about everybody else. And I think that is a fascinating change in how football operates. And here's PFF like causing this.

And you have, and look, it's not why I wrote the book, but you have people on the outside who don't know this, who are like PFF, what do they even do? Like, well, this is what they do. And it's insane. And I have to tell you, I didn't even know this when I started writing the book. But the positive impacts on the product is something that we don't even talk about. Like obviously changing the game. Like it's changed the NFL in a lot of ways in terms of making things faster and like

improving game plan and all that stuff. The product is improving by a lot of the things that PFF has given, the tools that PFF has given to NFL teams. I think it was a quote from either Kyle Shanahan or Mike McDaniel this year who said, if you aren't breaking the rules, you aren't innovating. You aren't, like, if you always play by the rules, you aren't innovating. And, like, you can identify what those rules are so much faster and so much easier with the tools that you have with PFF.

Like, this whole cheat motion thing that's happening now in the NFL, like, the most adopted principle, like Mike McDaniel went into the lab and he's like, I'm going to start doing this and I think it's going to be sick. And now every fucking team is doing it. It is because you're able to see how defenses react to when you,

like, so you can go into ultimate and you can say, when we motion into bunch or we motion into stacked, how do defenses react? How often are they passing off players? How often are they doing off players? How often are they doing these different things defensively to react to these rules? I'm going to break that. I see how they do it. Now I can break it. That to me is like, again, just makes the game so sick.

It's why the NFL, it's one of the major reasons why the NFL is so popular. It isn't as simple as what it used to be in terms of we're going to line up, you're going to know what we're calling, and we're just going to beat your ass or whatever, like this toughness piece of it. There's still elements of that. Everyone knows there's still elements of that. But it's not even chess anymore because you don't even know what all the moves the pieces can do.

It's like chess, but they go behind it. They put it behind their back and they're like, the queen can now do this. They can fly. The queen can fly. And the rook can do jump and backflips. You're like, wait, whoa, whoa, whoa. How do we do that? And then you go back and like this constant changing of like what the pieces on the chessboard can do to me is still like going to continue to

change because of what we have access to the film access and and the data access that we have access to is just like making things that much easier to innovate and that's not just that's not that's not independent of football like that's every business ever. Every business ever collects more and more data, identifies opportunities in the market and then makes changes. And that's exactly what's happening with the NFL. The data is just late because for a long time,

the market and the conversation was dominated by people who like legitimately just put eye black on and punch each other in the face. Like eventually some nerds got hold and they're like, wait a second, this can be cool. And like that has like, again, like a big part of why the game is changing. Make little Caesars, the official pizza sponsor of the NFL part of your game day order online

during our pizza pizza pregame one hour before NFL games and get ready for some football and fun. Choose your favorite little Caesars pizza or pick the toppings you crave. Either way, you win. And speaking of winning, everyone scores with convenient delivery or our in-store pizza portal pickup. So grab some friends and enjoy a few slices during the tastiest hour before kickoff. Folks, we are going all in on prize picks this football season.

Every week we are playing and testing out our skills here on Purple Insider to see if we could predict what numbers players will put up every Sunday. If you haven't heard of it, trust me, you're going to want to check it out. Prize picks is the easiest and best way to play daily fantasy. Instead of battling against thousands of other players and people who spend their entire lives doing fantasy, all you do is pick more or less on between two and six player stat projections. So say a quarterback's number is 250.5 yards,

go more or less and bang, you are playing. And you can pick from hundreds of players and numbers this football season. The cool thing is that it's quick and easy and does not cost an arm and a leg. You can turn $10 into $250 just like that. Again, the perfect way to fit it into a busy day. Click, click, and you're playing. This isn't just something that I like. You're going to hear us doing every single week prize picks on the show on

purple insider. So go to prize picks.com slash purple and use the code purple for a first deposit match up to $100. That's prize picks.com slash purple with the code purple daily fantasy sports made easy. And what you're seeing now is teams have like, they got a sniff of the data and then now they're saying, well, what else can we learn from this? What else can we do? And so I told the story about the big data bowl, which is at the NFL combine, a bunch of people are given data who are data scientists,

and you can just sign up for this and submit a project. And then the best ones give presentations to NFL teams. It's a whole room of people, but it's also a lot of NFL teams, if not all of them, represented in this. And then a lot of the data scientists get hired for these teams. And even my former intern, also former PFF intern, Haley English, she worked on a project and she's now with the Detroit Lions. So we're seeing teams add and grow these analytics departments, which makes it feel like they're just starting to scratch the surface of the data science. Now, I wondered about your opinion, though, having seen it all develop. In your mind, was there a big bang? Like, was there a moment that made teams turn on to this

as opposed to it's helping us save time? Sure. But to go all in to the point where we're at now, which to me is teams have fully, fully bought into this. The sad part is that big bang is, is, is a result reaction. It's an outcome reaction rather than a reaction to inputs and a reaction to the work that was put in. If the giants don't win that first Superbowl, I don't know. I don't know that because they were the first team to adopt it. They obviously won the Superbowl against a Patriots team. They would not have beat like they should not have beat.

And then teams were like, wait, what did you have? And, like, it's that simple, right? It's like, wait, I swear. Like, it kind of sucks. Like, it sucks that that's the reason. What should have happened is teams have been like, wait, this is actually really good and invested it through themselves and, like, all that stuff. But, like, sometimes that's how it works. Sometimes people are reactionary, especially, like, a league that has been doing the same thing for so long, reactionary to a result, a result that in not every way is random, but in some ways is random,

right? There's different variables that come into play, but like the Giants win that Super Bowl and every team's interested. Neil, that's how Neil tells the story. Like the Giants win that Super Bowl and everyone's interested in that point. Like then it gets to a point where like, oh, eight teams have this, we're going to get it. it and like we talked a lot about in the office about when we were selling the college teams it's like dude 10 of the 12 sec teams or well 11 of the 14 like how are these other teams surviving and eventually that pressure is just like they have an edge that we don't and that i think is a big part of that big bang as well like the competitiveness of this is a factor it's like how it impacted baseball as well like if the a's don't go as far

as they did when they're first doing that stuff, like it gets killed. Like you do have to have some results. You do have to have like, it has to work essentially. Now I think it would work whether or not the Giants won the Superbowl or not, but I do think that that was a big bang. So to speak. Yeah. Shout out to a Minnesotan read Elbergati who wrote the original article for

wall street journal. And I interviewed him in the book, but he wrote the article that sort of unveiled who Neil Hornsby is and where the giants were getting their secret data. And that played a big role in the world finding out because what I think is a fascinating element of this is I had a friend text me after reading the book and say, I never would have guessed Tom Coughlin, like would have been the coach at kind of the center or the start of an analytics revolution because he was such an old school seeming coach, but they also had a long time analytics guy who had actually started under Bill Parcells, who was the one to buy into this data.

And then they started getting their hands on it and seeing where it could impact them now the part about the giants at that time was they were not interested in the grades in fact offended by the grades and yes during that time when people started to become first aware with pff we saw a lot of that we saw a lot of coaches saying, you can't grade my players. In fact, I'll fight your graders right here. That is also changed. And I see that trope come out from time to time. Now, I don't believe every single team uses the grades in the same way, but I had teams in writing the book tell me that they would use them in a lot of different ways. One is for data science and study. The other way was to compare their own grades because one of the problems is in-house you have bias sometimes you have coaches who like a player more than another and a lot of these things do come down to a subjectivity which which i gets

talked about a lot with the grades but what they would do is they would look at our grades pffs grades and if there's a big difference then they wanted to find out why like what is it that they missed what is it that we missed and i think that that's a really interesting way to use them the grades conversation fires me up it fires me up because i do think that one of chris collinsworth's key points when he had first bought into the company, the grades were done on their raw plus minus. So like every player, you know this, but like every player is graded on a scale of positive or negative to positive. And then they would some trickery in the math. They would aggregate that and kind of average it to like a player like Aaron Donald had a big game plus 14.4.

It's like, OK, no one knows what the hell this means. And Chris, who works on the biggest show in NFL, literally is like, well, no one can get this. What if we normalize it to a zero to 100 scale? Like letter grades. And that had a lot of faults, right? Like it's hard math. It's hard math to turn stuff into zero to 100

where it makes sense, especially across positions, especially one position has more positively graded plays than another position because they're more often involved in one-on-one. I don't want to get into that, but the grades conversation fires me up so much because the consumer-facing product is a 0-100 scale that is admittedly, and I think even PFF knows, flawed in a lot of ways in terms of how you're trying to create a 0-100 letter grade scale for grading every single player. What doesn't get talked about enough is PFF is trying, like they're actively improving the grades every single year,

trying to actually tell the world at the college level, the Canadian Football League level, even the AFL and XFL level, and obviously the NFL level, how good every player is. That is hard to do. And it's flawed. And it's not perfect. And no, the grades aren't great every single time. But God, someone's trying. Someone's trying to actually do something that feels impossible. You can look up right now on Western Kentucky, a backup guard that has only played 30 snaps, how good he's been, subjectively in a way, but charted with film that you don't tell somebody that five years ago 10 years ago they'd be like what are you talking about they don't even know

if that player played that to me i think is just like now i'm not saying trying is a full tip of the cap but the whole grades conversation is just like yeah these aren't good yeah no no these aren't good and like there are some truths to that like they aren't perfect they're the zero hundred scale has fucked up things like that is like a fact but dude do you understand how cool it is that someone is trying to get to the point where you can quickly say if a player has performed well or not and it's in a data set that i mean is a worthwhile pursuit and that is something that neil has always banked his chest on it's like we're gonna try to do this knowing it's hard knowing it's not going to be perfect because we think it's going to be sick.

And like, I don't know. I do miss that energy. I miss that energy with Neil a lot. It's getting me a little sad. It's like, it was cool what he tried to do. He tried to do something that no one in the world did. And everyone in the world will tell you is impossible. That is crazy.

That is crazy. That is crazy. The most interesting thing about that is that when Neil would go to teams and they would talk about the grades, they would all say the same things. You can't do this. You can't grade that. And what he would do is he would ask, well, why? Or how could I? Or how do you guys do it and I think what the outside world doesn't understand is that the grades are forged by the league they're forged by executives coaches uh several I mean one coach in particular who's in the league now but actually no two Bobby Sloat the

offensive coordinator of the Texans was in PFF and was influential in sharpening the grades and Zach Robinson Robinson was the quarterback coach for the Los Angeles Rams. Both of those guys came through. And also Paul Alexander was a longtime offensive line coach. Like these guys helped them sharpen the grades. So when they felt like things were not correctly attributing certain grades, they said, okay, here's how you need to do this. And particularly like with coverages and stuff like that, that someone like Zach would attributing certain grades they said okay here's how you need to do this and particularly like with coverages and stuff like that that someone like zach would understand as a former nfl quarterback much better than the outside world and so over the years they have worked

with the feedback to build this book of how to grade but an nfl gm said to me about the grading system he said the thing is that it's hard to grade corners. It's hard for us to grade corners. It's not just hard for them to grade corners or linebackers. It's hard for us to know whose fault it was unless it's our place. can look at and compare our ideas versus their ideas. And then Kweisi Adafomensa told me that he used this to create data science from where he would look at the grades and try to see correlations and things like that to what would succeed and learned so much from that. And he said in the book, he's quoted saying that he uses PFF grades probably more than anybody else so i think that it was a take once upon a time where people would say oh no one in the league cares about these grades but what data science was able to do is show actually there's plenty of reason to do it and why would you ignore an independent look at your players i guess is is my thought as well but certainly if you think that

you know, every grade is the perfect representation of every player, well, that's never what they claimed. Right. And one thing that is hard, Austin, you know, this from writing about it, I use PFF grades all the time for players. I cover to give, to sharpen my eye is that nobody thinks you should use it without context, just like interceptions. Does anyone think you should use interception? Because I saw Kirk Cousins' arm get hit last week and the ball pop up. That's not his fault.

That's the right guard's fault. But we do this in our mind with the traditional stats that we're used to that it's almost like, oh, that's not allowed with the grades. Of course it's allowed, but it is hard, I think, and it takes some more effort than just looking at yards touchdowns and fantasy stats you you bring up a great point on like multiple levels i think one and and and quezi knows this eric eager knows this george shahari knows this anyone who's worked with the raw data at mass levels i'm talking not like you know like just

tj watts grade or cleo max grade like, like massive amounts of data at the college level and the NFL level knows that PFF grades, even in their subjectivity are stickier or higher correlators to future success than traditional stats, like interceptions, passer rating, yards per carry, sacks, all these things, fumbles lost, all that. Like the data that the PFF is collecting is stickier and more predictive of future outcomes than other stats that are more commonly known. Right. Now more,

those stats that are more commonly known are also more commonly understood. Like, so at that point, it's easier to look at what Devon a chain or Devon a chance yard through carriers right now, which I think is probably like 10.4 or something. And you're like, I know that's not going to hold. Cause I understand that stat, but PFF grades are so new and there are they are a little more convoluted they are more subjective that the average fan is like no idea what this means i have no idea how to progress this forward so i'm going to discredit it and i think over time

it's going to change i think it will and like i also will say the tip of the iceberg that mean where like there's an iceberg up top and then like there's a bottom of the iceberg in the water the tip of the iceberg is like you graded tj watt 84.1 like the grades were not meant to tell you how good tj watt is okay we know how good tj watt is the grades are meant to tell you his backup at charlotte had the highest passers win rate when he was there of any non power five player and he was top 10 with power 5 players. And he was someone that PFF identified as like, this guy's winning at a level that we think he can win at in the NFL. And we would specifically tell teams about Alex Highsmith.

And now, I'm not saying every player is like that. PFF misses too. But like, those are small things in the data that if Highsmith did not produce at the level he did at Charlotte, I do not think he's in the conversations that he was or in like the viewpoint that he was in the NFL when he was drafted because he produced well and because he was grading well and you were able to get line of sight to him in ways that other players weren't.

That's an example of someone that is winning because of the grading system. And now there are certain positions that are easier to grade. Because he's right, cornerbacks are hard. Any off-ball player on the defensive side of the ball is very difficult. The further you get away from the ball, the less one-on-ones you get, the less fixed events you get, and that makes things difficult. Does that mean we shouldn't try? No. Does that mean we should just continue forever to measure defensive backs

by how many picks they got? Freaking no. Let's try and do something else. And next-gen stats is trying things. The FF is continually trying things. All this stuff is changing, and I think you have to respect that pursuit, even if it's not perfect yet. Purple Insider is brought to you by BetterHelp. Friends, have you ever had times where you felt like your brain is getting in the way? Like maybe your brain is racing and you can't fall asleep or you can't slow down negative intrusive thoughts and you know you

should look for help, but you haven't brought yourself to do it yet. Well, therapy can help you figure out what's holding you back so you can work for yourself rather than working against yourself. Therapy can be as simple as finding training methods for yourself to learn positive skills to improve your everyday mental health. It isn't just for people who have gone through especially hard times, though of course it can help there as well, but even with just dealing with the day-to-day can be vastly improved with someone helping on your side. If you think that might be something that they would be helpful to you, try BetterHelp. It's entirely online, designed to be convenient, flexible, and suited to your schedule. Just fill out a brief questionnaire to get matched up with a licensed therapist and switch therapists anytime for no additional charge make your brain your friend with better help visit betterhelp.com slash insider today to get 10 off your first month

that's betterhelp dot com slash insider. Chris Collinsworth, one of the things he said to me is that his sort of main mantra was, we've never got it all figured out. We need to keep trying and we need to keep reassessing at all times, like how we're doing in figuring it out. And some things that we think we know might change based on new information. We saw this in baseball with Babbitt. Remember people were really into batting average of balls hit and play. And then new data came in that said, not surprisingly, the harder you hit the ball, the better,

you know, you've got a chance to get a hit, but we can also measure how hard everyone hits the ball. And then you don't really use Babbitt anymore, but this is But we're only just starting to go down that road with the tracking data. But there's also something that, and this is another interesting subject to me that I tried to cover in the book, is there's going to be other people who are trying different ways to infiltrate the NFL to try to supplant PFF and say we have better data. And I think that there's going to kind of be an arms race here as we go into the future of just how teams are influenced by data and by companies.

So you got PFF really at the start of this and sparking it in a lot of ways. And then now you have a whole bunch of other people who are saying, how can I be the next one to create the company that every team signs up with and so forth. But I think that what makes it unique is that you do have actual eyeballs on these things. It's not trying to have the dots of, Hey, well, this defender was really close, right. But he had his head up his butt. Like, you know what I mean? Like he was looking at trying to have the dots of, hey, well, this defender was really close, right? But he had his head up his butt. You know what I mean?

He was looking at somebody in the stands as the receiver caught it, but you gave him credit for being right near him. You know what I mean? We're an actual person watching it. So I think it's going to be this interesting competition from their product to grow versus how other people are going to create new products for the NFL? Yeah, I think it's already happening. I think that companies and teams are all like all kind of pursuing the same problem in that we want to have these kinds of tools that make game planning

more efficient, right from an hours to minutes from a days to minutes type of thing. And that will be continually pursued. That market is dominated by PFF right now. Like they have like dominated the efficiency market. Like it is hard to find a product that is better than ultimate in speeding that process. Where I think there are opportunities and maybe getting too business speak here, but like where I feel like there are opportunities

into breaking that market into like competing with PFF's first advantage is how that data is collected. Because can you look 10 years in the future and say hiring human beings to watch all these games is the fastest and most cost-effective way when you have chips in the helmets, when you have video tracking technology that you're seeing, where like, if you know, if I, well, here's a good example. If I go like this on my camera, it tracks that I just did two peace signs in the air and throws from petty. People are doing things with AI. People are doing things with video tracking technology that can change the game that humans can't, right? Like you're going to be able to identify based off broadcast film. I feel like very soon where a player was lined up, who exactly it was, what they did,

and maybe even get to the point where that video tracking technology can tell you if that player won versus his assignment. And that I think is the innovation necessary to breaking into what PFF's first move advantage and all that stuff. That's just in like the business to business model. Like there's a consumer facing model as well. And like, in terms of what challenges you're solving for there it's betting it's fantasy that's the stuff that a lot of people use pff for now because their data predicts future outcomes what is fantasy football predicting

future outcomes what is betting predicting future outcomes now there's this like fan service piece to this where there aren't like problems you're solving for with pff data outside of like awareness knowledge information gathering you know barstool not barstool but like bar fighting and like oh this guy's good kirk cousins is better like there's that but there's other problems and challenges that are honestly like more cost interested like betting and fantasy that pff is still trying to solve for on the nfl side like selling the team selling the agents like agents i'll never forget i won't name the player. There was a player that wasn't that good in my opinion, in everyone's opinion. His agent asked us to put together data that should not, not saying like,

can you tell, can you lie? But it was like, what are the best pieces of data you got? Oh my God. Can you put together the best piece of data? Never like manipulating the data, but like, can you put the best presentation together using your guys' data that shows that he is actually, there are some things he's doing well. We do that. He signs a contract that, oh my goodness, no one saw coming. And we're like, what do we, we just got this guy so much money. And the agent's jumping for joy. And like, there's like this business model to like some of that stuff too. I don't know how that's connected, but like there are people who are using PMF data to like help inform some of those decisions as well it's like

agents and all that stuff and uh just to circle back by the time that ai can figure out whether football players are doing the right things the machines will have taken over exactly yes it'll only be john connor and so forth yeah the terminator Terminator fans maybe. But it is, it is a fascinating element because the AI there's sort of a boom in that, in what AI can do right now. And there was very strange to see on your video, the confetti come down as, as you just move your fingers. So that's proof that PFF is done with. No, I'm just kidding. But if they don't stay up, you know, if they don't stay up you know if they don't stay competitive other people will come into that space and that's going to be fascinating for me to watch uh after

having written all of this you know book and how they got here but how it grows and builds off of them what did it mean to people inside pff when quasi adolfo menzo was hired because i he told me a story that at the combineine, after he was hired, he went to PFF. They meet with all the teams there and said, this is for you guys because he really made his bones

early in his career as a researcher using a lot of that data. And I think even though there had been data-driven GMs and data-styled GMs, and there are some that are still beating everybody like howie roseman but i don't know there was never an analytics gm where this was truly his background and now he's in that prominent position to me that was another

sort of landmark type of moment no i think it was huge i think one of the biggest reasons it was huge and this maybe gets back to like the personal element is like a lot of the people in the building knew him and we're excited for him like obviously there is some excitement around like what it means for the league to have someone like was he going to a gm position with the background that he did have right and i think other parts of it were like we've hung out with this guy and talked to this guy and he is someone that we believe in and like like, it was like when Zach Robinson got hired as a coach or Bobby Slot got hired as a coach or Bruce Gretkowski and what he's, what he was doing in the USFL or XFL or wherever he was coaching.

And we've always rooted for the people that have like come through PFF and like done things with PFF. I don't know if Mike McCarthy is in that same way. He had a couple of couple of coffees with us, but like, I don't know. I don't know. Like we like, we don't, we don't haveffrey as much mccarthy as much as like quezi and not ownership but you get what i'm saying like a legitimate connection with these guys where we felt like there was good things happening to people that were like thinking the ways that we thought about i i remember a very similar moment baker mayfield gets drafted number one overall and neil hornsby because this is someone that we you know you know i think there's

some flaws in some of the data that we saw but like it was someone that we really liked from a data perspective at the time and now i think our data at the college level has gotten a lot better or their data has gotten a lot better but i remember neil hornsby was like do i'm getting the jersey i'm getting i'm getting everything he's like this is my guy i'm a brown so now it hasn't panned out but it is examples of just like you, there's been a lot of excitement in that room. It was a lot of fun. One thing, just from your own personal perspective, I don't want you to tell the whole story because I want people to buy the book and tell the whole story. And let me tell you, it's going to blow you away. Your personal journey to PFF through PFF and to where you are now. What do you tell people about it i mean it's

how and how much do i want to say on this podcast because it's like an insane story of how much you went through growing up and then where pff was for you coming out of college and then how you had to fight your way all the way through pff um to get to where you are but what do you tell, like, as part of your life story? And this is for a lot of people in the book, and this is what I try to tell people. It's about football data, but it's also a lot of life stories of people who were in different spots in their life who found PFF and were able to create careers out of their passion,

and that certainly describes you. But I am curious about how you sort of view its space in your life story. It's interesting. You know, in terms of like the background and stuff, I do think it's something that I just don't talk a lot about. I know it's not something that, you know, and whether that's right or wrong, I can talk to my therapist about that. It's not something I talk a ton about. Now, in the PFF of it it all and I do think it's connected and this is very cliche and I'm apologizing in advance but dude never give up I swear I swear you got you got to keep going and I do think that there are things in my life that have helped me keep going

and I think there haven't been called advantages but maybe not but like there are things that have like forced me to keep going and I think that has been that was a big part of PFF. It's a big part of my life story now. I think that how PFF played a role in it, I think they gave me a lot of opportunities. And it was a place where if you didn't give up, and I think this is a big credit to Hornsby, it's a big credit to him. It's a place where if you didn't give up, it would work. If you didn't give up and you kept showing up and you kept trying, you'd get rewarded. Now, was it as fast as you wanted every time? No.

Has anyone been as fast as I wanted? Probably not. Like I run at a speed that, you know, is a little crazy. But there are times where obviously there are frustrations. But I do think that time and time again, I found myself, you know, there were times where I was looking at other positions. And there were our conversations about going elsewhere. And like every single time, it would be rewarded with another thing and continue to move up and continue to move up as fast as I did.

So I do think that the cliche of it all is definitely don't give up and just keep showing up. And PFF played a big part of that. I think Neil Hornsby played a big part of that. George Chihury played a big part of that. There were people that saw that I wasn't going to stop, dude. I'm showing up every single day. And eventually the, you know, the dam broke if you you if so to speak you know it's it is uh one of my absolute favorite stories to tell that i've ever told as a journalist is your life story in this book so i implore people to go get it and uh you know

just check it out football is a numbers game pff and how a data-driven approach shook up the sport go get amazon wherever you get your books you can find it uh i offered to send you one you're still waiting for it to be shipped there so uh hopefully if it doesn't arrive let me know but no it's really cool to see people starting to get them and especially you guys who i told your stories in the book reaching out and saying you know that it's cool or that, you know, it's fun to see their own life story laid out there in paper. So I can't thank you enough for your openness, your help with this project, and all the insight

that you gave me along the way to help me fully understand what was happening here and your journey now to The Ringer, which I'm very, very happy to see you there. So thanks so much, Austin. Um, follow your work where you write about Halloween movies now at the ringer. Yeah, yeah, yeah, yeah. I guess so. You can do that. You know, follow me on Twitter, Austin Gale underscore, but also just check out the ringer,

all the stuff that we're doing. And something I will, I will say too, I can't wait to read the book. I don't know what Amazon's doing. It's their fault. It's probably because they're just shipping out so many copies, but you know, they, I appreciate, you know,

I'm excited to see the story and stuff like that. Story's not done. I'll tell you that too. My story's not done. I got a lot more to do, so I'm excited for it. Absolutely. Well,

again, thanks so much, man. I really appreciate it. And this has been another episode of the self-promotion tour. And there will be more, there will be more amazon.comcom check it out football's a numbers game thanks again austin and uh we'll talk to you soon man sweet sounds good