• 3 days ago
Transcript
00:00Live from Radio Row, SportsGrid's coverage of Super Bowl 59 continues. Joe
00:05Lisi alongside Sam Schwartzstein, analytical expert for Thursday night,
00:10football. Sam, just a great atmosphere here today. It's unbelievable. It's my
00:15first time at Radio Row, first time getting to meet and greet a lot of the
00:18people I know from the internet through the analytics community, so it's been
00:21great being here. Analytics have changed the sport, not just in terms of the X's
00:26and O's, but obviously in regards to a number of different industries, fantasy
00:30and obviously the gambling industry. How did you get involved in the analytics
00:34end of it and more importantly with the NFL? Yeah, so I played football at
00:39Stanford. I was a center. I'm not the biggest, strongest guy or fastest guy in
00:43the world. Not many Schwartzsteins are, but I was able to find some value by
00:47understanding what the team was going to do, the defense was going to do before
00:50they do. Right. So I was able to then use analytics to help build up that kind of
00:54repertoire, and so analytics have been involved in football for years. It's
00:57probably the most analytical sport from the early days trying to identify
01:00tendencies. What is my opponent going to do? Then working in Silicon Valley after
01:04football and then ultimately getting a job at the XFL, I was tasked with
01:08reimagining the game from the on-field product to the technology that we're
01:11going to use, even the equipment we're going to use, and analytics helped drive
01:14my decision-making process when I was there to ultimately make rules that
01:19inspired the dynamic kickoff or coaching player communication systems or even our
01:22design of our football. So I got to know Amazon and they had this idea of
01:27making an analytics-focused broadcast, and that's where PrimeVision came to be.
01:30Let's focus on being able to give the people a unique view into the game,
01:35not over-inundate them with analytics, not just tell them a number for number's
01:38sake, but how do we really make the game more exciting by using analytics to show
01:42them new things? I'm always impressed when you see like a long touchdown run
01:45like Cheetah, right? Tyreek Hill, he's running 21 miles an hour, and then
01:50all of a sudden you have a college quarterback, Quinn Ewers, that was the
01:52discussion, he's running 22 miles an hour. I think if you put them head-to-head,
01:57Cheetah would win, but it's amazing in terms of how you break it
02:01down. We talk about, you know, yards after contact. Is there a specific category
02:07that you first started with and then evolved, or did you just dive into the
02:12whole broad base, and is it more so more offense than defense? I'm curious.
02:17Yeah, so it's easier to measure results in football. What we say in football is
02:21that we're a column-rich sport, but a row-poor sport, and what I mean by that
02:25is there's so many things you can measure on any given play. We have 11
02:29players on offense, 11 players on defense, down-distance location on field, time on
02:33clock, but there's only 65 offensive plays and a 55% pass rate. There's half
02:38of that, nearly half of that, for runs and passes, so just imagine it's not easy to
02:42measure everything outside of a vacuum, but what I really enjoy is analytics
02:47will take that hitting game of football. We'll show how many yards a quarterback
02:49has. Joe Burrow led the NFL on yards, but how important are those yards? Getting 10
02:54yards on first and 10 is really good. You get a new set of downs. 10 yards on
02:58third and 20, not as good, and so analytics will help evaluate those
03:02yards and provide context. So the first one for me was success rate, because that
03:05was the O-line stat. If we rush for four yards every play, we will be able to
03:10convert a first down, and that's what success rate is measuring. Did you get
03:13enough yards based on down and distance location on field to keep your team on
03:17schedule? And then EPA is then the measure of how much did you help your
03:21team's chance of scoring, so it really helps contextualize all the yards the
03:25team's getting. When I break down games, because I cover both college and the NFL,
03:29I feel personally the two biggest or maybe under-the-radar stats, and they are
03:34out there, are third down offense and third down defense, and for me I look at
03:39it as if you're sustaining drives, you have a better opportunity to
03:43score points, and obviously keep maybe a more explosive or potent offense on the
03:48sidelines, and when you have a defense, let's say, dominating 29%, 28%, especially
03:54in college football, but more importantly in the NFL, now you're taking games over
03:58and you're providing your offense a short field. Are those still two of the
04:02most important statistics offensively and defensively? Oh, absolutely, and those
04:07actually are reflected in both those stats, success rate and expected points
04:11added, because what you want to do is the thing you're not expected to do.
04:14If you take anything away from analytics, it tells you do the thing
04:17that's unexpected, and so the average conversion rate on about third and
04:22seven plus is around 35%. Well, Patrick Mahomes is
04:26converting nearly 50% of the time, and so that's showing their
04:30long sustained drives. Now what's interesting is they have not had an
04:33explosive play, which is a 10-yard run or a 15-yard pass. Don't ask me why we
04:37split it up, but we do, but also they are the lowest explosive play
04:42rate in the NFL, so they actually shorten games. Even though they're one of the
04:46best offense in the NFL, they have the fewest drives per game in the NFL, so
04:50they're making closer games by having one less drive. Now I think that the
04:53perception, per se, has led to them or pundits saying that they're not
04:59as dominant, where I disagree, and obviously as long as Pat Mahomes is on
05:05the center, you want him. I think if we surveyed 10,000 people here, all maybe
05:109,999 would say Pat Mahomes would be the best quarterback in
05:16that type of situation, right? Do you see that as well? Because everybody will say,
05:21well, they had three games this year where they won by two points, but I look
05:24at it as they won. It doesn't matter margin of victory. I'd rather have a
05:28two-point win than a two-point loss, and two is just as well as 40. Do you see it
05:34that way as well? Yeah, you know, I'm not one of those analytics people that's
05:37saying, go reach the singularity. We'll tell you, analytics will tell you, pass
05:41the ball every play. As offensive linemen, I know that if there's no
05:44threat of me going forward, they're gonna tee off on me if I'm moving
05:47backwards. So what I would tell a team, and what I used to consult with
05:51teams, is do whatever you're the best at. Be unique and have that unique
05:55flavor, and if their job is, hey, if we're in a close game, we have Patrick Mahomes,
05:58they don't, we're guaranteed to have that, I'll take that. Margin of victory,
06:03analytically, they might not be the best, but this clearly works for them. I would
06:06not say that this is the most talented Chiefs team they've had. You brought up
06:09Cheetah, right? They don't have the deep threat. Even Xavier Worthy, who they
06:12drafted to be there, he's more of a lateral guy, a screen guy for them this
06:15season. They don't have the deep threat. They might not be the best offense, but
06:18they're winning games, and they're winning it their way. If you're gonna ask
06:21a team to go try and be somebody else, like a lot of these teams, coaches will
06:25leave Sean McVay's system, and they'll try and be Sean McVay. You don't have
06:28Sean McVay, and so you need to be able to be the best version of yourself, and so I
06:33analytically, I could tell you they need to be better on short yardage, but no,
06:36they're doing just fine. Okay, let me ask you this, because this comes up a lot in
06:40terms of, especially what we cover, the gambling perspective, which fluctuates
06:44from an in-game perspective, going for it on fourth down and two-point
06:49conversions, and I've always brought up, and I want to get your take, yes, we talk
06:53about it from the offensive perspective, where your conversion rate, let's say, is
06:5755% or 60%, you would take that every time, or more than 50, right, or maybe in
07:03certain situations, fourth and one, fourth and two, inside the 20, right? Do we
07:07measure what it does emotionally, or from a points perspective, on the defensive
07:13side of the ball? So you go for it on fourth and two, you do not get it. That
07:18ensuing drive, or within the next five minutes, is a team apt to give up X
07:23amount of points. Do we cover it from that aspect as well? Yeah, so that's the
07:26beauty of expected points, Adam, is we'll see what your starting point was, and
07:30then we measure where you were at the other spot, and so on average, a turnover,
07:33an interception, let's say, I know this number off the top of my head, is worth negative
07:384.75 expected points, so nearly more than half of a
07:42touchdown, right, seven points is the end-all be-all. So think about it from that
07:46point of view, where it's actually huge, we do measure it, do we measure the
07:51emotion of it? No. You can't, right? Well, I don't want to say can't, Amazon
07:55probably can't. No, no, I'm just saying it's something that's not, it's a human element.
08:00Look, even if people have tried to measure momentum, and they say we can't
08:04measure it, just because there's an absence of data does not prove that
08:07something doesn't exist. I hope that made sense, but like, yes, as a player, I've
08:11known when a momentum is shifted, so I can feel it, I can't measure it right now,
08:15right now. So here's what I'll say about it, when I would consult with teams, I
08:20would tell them, you have a third down period, you have a red zone period, you
08:23have a two-minute drill, add a fourth down period. It starts in OTAs, you're
08:29preparing for fourth down. Look at Dan Campbell, they gained the most win
08:32probability, or second most win probability in the NFL this season on
08:35fourth downs, including the ones they did not get and the ones they got. He started
08:40that when they were 0-8 years ago. Hey, this is who we're going to be.
08:43So you can minimize the emotion by who you are as a team identity and
08:48culture, and culture will always matter more in those scenarios than the
08:52greatest play call or something like that. So the defense for the Eagles
08:57know, hey, every fourth and short, we're going for it. I don't care where we
09:00are, because they've seen that for the last four years, the tush push has been
09:0387% successful. Which is incredible. Right, and so you start there, you start
09:07early on in OTAs, we're gonna have a period for it, we're gonna have a play
09:10for it. At Stanford, we had 64 lead dive. It was run behind David DeCastro, all
09:14pro. That works, and so if you don't have that, you got to find the
09:17new play. And so get that mindset, then the momentum shift won't matter as
09:22much. Right. If your team's aware that this is who we are. Got it. Let me ask
09:26you this, too. When you look at coordinators, defensive coordinators, I
09:30feel like they're not, and this is just a personal opinion, I'll throw Eber
09:33Fluss and I'll throw Dennis Allen, who are no longer head coaches, but I feel
09:37like defensive minded head coaches typically aren't analytical guys. Is
09:41that true? Based off of just what you've seen and working with teams? You know,
09:47I think it's more of a mindset to build in, right? Because I played for Vic
09:53Stradio in college, and you know, the guy knew every single thing a team would do
09:57on third and seven plus. He would know what kind of, he would know all the
10:00different tendency measurables, but maybe he was not as focused on going
10:04for it. I think that newer guys come in the league who put, this is, sounds
10:07ridiculous, but if you played Madden as a little kid, you never punted. Right, right.
10:11And so what's ingrained in you? Well, you know, at East Stroudsburg, where Vic
10:15Fangio went to school, you know, which is a beautiful campus. It's run it with three yards in a cloud of dust. I always say, like,
10:21defensive coordinators that become head coaches sometimes, they want to win
10:24games 17-13. Losing games, they want to win it 17-13, but they
10:30refuse to win it, to try and win it 40-20, and they wind up losing those games
10:35like 16-14. Which is ironic, because you'd think, oh, if I'm Vic Fangio defense, I can
10:39handle them getting the ball anywhere on the field, so I might as well build in.
10:43And then look at the Eagles. Right now, they have under two and a
10:48half yards average as it's a go on fourth down. That's incredible. Because they know on
10:51third and eight, the hardest play in football, they can get five, they can get
10:56six yards, they don't need to get all eight. And so that line on the field,
10:59those chains, there's an imaginary line two yards. On prime vision and
11:02addiction stats, we tell you what yard you need to get to on third down for
11:07it to be a go on fourth down, which is how the coaches are communicated within
11:10their headset, saying, hey, we only need six here. If you only need six all
11:14those times, you can have so much a more higher advantage on those fourth down
11:18plays. Absolutely. I do want to break down the game itself from an analytical
11:23standpoint. I feel you have to make Kansas City a favorite, right? But just
11:27from what you see, I do feel the more complete team, assuming they both
11:32play their A game, would be the Eagles. They're more balanced, they can run the
11:36football, not just with Barkley, but obviously Jalen Hurts with the RPO
11:40game. Do you see that as well? Or do you give a significant advantage or
11:44disadvantage to either team? Look, I'm not going against Patrick Mahomes. I
11:47think we all know that no one's gonna get burned three times in a row. But I
11:51don't doubt you if you go with the Eagles. What a roster Harry Roseman's
11:54built, right? What a great team that they have, especially starting up front
11:58in the run game. Now, as teams have gone away from base defense, right, right,
12:03at taking that third line back off the field and bringing a defensive back on,
12:05it's really stifled the pass game. Passing game's gotten worse, right? It's
12:09let run games go up. And what's been great about the Eagles is they make it
12:13very easy. When you talk to coaches around the league, it says this is just
12:16day one install stuff, meaning they're gonna run the simple plays, the RPOs
12:19you brought up exactly. They run to the weak side, the sixth most in the NFL,
12:23say from Barkley, because Jalen Hurts is reading the backside for that RPO. And
12:28so they telegraph what they're doing. But at some point, you've got to be a
12:31man and take on that block and do it. And so you're 100% right. Here's the
12:35key. A lot of people have talked about it. Chris Jones will find the fish. He's
12:39gonna play against the worst offensive lineman. Who's the fish on this Eagles
12:42offensive line? There really isn't one. So you can focus on, hey, if we can take
12:46Chris Jones out of the game, get Nick Bolton, who I think is one of the best
12:50inside linebackers, he's not gonna be able to run with Saquon Barkley. Get
12:54Saquon to the edge, let us move side with side, and then let's see. Now,
12:58Saquon Barkley's such a unique player. He's almost like a wide receiver in
13:01terms of production. He's got the fourth lowest success rate, meaning he's
13:05one of the worst at consistently getting yards needed. But he's the
13:08fourth best at EPA per carry. And you know, there's a lot of guys who have
13:12not played as many playoff games in this playoffs because his explosive
13:15plays are so big. So part of their explosive plays come from the
13:19running back, which is so rare in this game. I think the running back will be
13:22a factor. Barkley as well, but not on the ground through the air. Because
13:25you look at former Stanford Cardinal, you know him very well. Christian
13:28McCaffrey. He did it on the ground, but he had over 60 yards receiving last
13:32year. And I think that's the mismatch that they could utilize, especially if
13:36they break tendency. That's the biggest thing, right? You have to break
13:39tendency in big games. If the analytics or the tendency show we're going to
13:43run on first and second down, why not play action pass? Absolutely. I think
13:46you're 100% right there. One of the best screen teams in the postseason
13:50this year. They're 13 for 13 completion, 70% success rate. The Eagles are
13:54great inside and information. We'll be right back.

Recommended