January 30, 2025 • 51 mins
In this episode, we dive into the rapidly evolving world of Name, Image, and Likeness (NIL) with Michael Schwimer, CEO of Big League Advantage and Alabama Men’s Basketball’s secret weapon to making it to the Final Four for the first time in school history.
First, we explore the origins of NIL, the NCAA’s initial resistance, and the mechanics of NIL deals, including how collectives operate. Schwimer shares insights on the opportunities and pitfalls of NIL and the long-term implications of NIL for athletes, especially female athletes.
Secondly, we dive into how data analytics can revolutionize roster-building in the NIL era. Schwimer explains how Big League Advantage is shaping data science in college athletics, and he offers a behind-the-scenes look at the thought process that was behind helping Alabama's Men’s Basketball team reach the Final Four with minimal NIL spending.
This episode is a must-listen for anyone curious about the intersection of sports, money, and data in the collegiate world.
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
(00:00):
Today on Data Nation, Michael Schwimer joins us to discuss the newfound role of Name, Image,
and Likeness and data in college athletics. We'll dive into the history of NIL, its opportunities
and its pitfalls and how to use data to build a strong college roster without breaking the bank.
(00:21):
I'm Liberty Vittert, a professor at Washington University in St. Louis and my co-host is
Munther Dahleh, William A. Coolidge professor in electrical engineering and computer science
at MIT. Together, we'll explore the development and function of NIL by following the money and
detailing all aspects of NIL deals. Michael Schwimer brings fascinating insights as the
(00:42):
CEO of Big League Advantage, a company that identifies star players based on analytics and
provides resources to these identified athletes to help make their dreams a reality. Ultimately,
there's so much going on behind the scenes in college athletics,
and there's so many unknowns, so today with Michael's help we'll bring clarity to the new
(01:02):
world of college athletics where NIL and data are unavoidable and revolutionary.
Liberty (01:09):
You know, this idea of paying college athletes is something that I feel like I've
been hearing about for years and now it's obviously happening through the power of NIL,
and initially though the NCAA seemed very reluctant at the idea of paying student
athletes or, I guess, college athletes and they even I think threaten to ban
(01:32):
schools from the NCAA if their states passed laws legalizing NIL. So can you
walk us through this - what is NIL? Why does the NCAA not like it? What's happening here?
Michael (01:45):
NIL came about because there's a large number of people that believe college athletes
should get be getting paid. These schools are making a tremendous amount of revenue through
their teams, their football teams, basketball teams are the big revenue generating sports but
other sports and different schools can generate revenue as well, and these are multi - this is a
multi-billion dollar industry where the talent aka athletes in this case weren't being compensated
(02:08):
at all. In fact, I think it's probably one of the only industries where that occurs,
and so there's a big uprising ‘hey these college athletes should get paid’,
but then the other side is we want to keep amateurism. Obviously, the point of the NCAA,
they're getting paid by free education via scholarship and ultimately that wasn't enough
and I've - as a former athlete myself I usually side with the athletes on this. I think that if
(02:32):
you are the product and you are what people are tuning into you probably should be getting paid.
Liberty (02:37):
What's like one of the biggest schools - let's say, I don't know,
University of Mississippi? Ole Miss, would that be fair? How much money do they make a
year? So it's from, what, TV deals and boosters or donors? Is that - just a ballpark about how
much money one of these schools makes a year off of these student athletes?
Michael (02:56):
Well it's a great question and it matters what conference you're in. So the SEC and the Big
10 make the most money, the ACC and the Big 12 are just that step below. That's what's called
the Power Four conferences, then you have the group of five underneath that which is
like Conference USA and all these other schools that make a lot less. So there's the media deal
(03:17):
that the schools negotiate if you're in the ACC in the Big 12 you're looking at 30 to 45-50 million,
you're in the SEC and the Big 10 it's double that. So there there's a lot of money at stake here and
that's where you see all these schools changing conferences, you see why did USC, UCLA, Oregon,
Washington go from the PAC-12 to the Big 10? Well, money, right? it's not because they want to fly on
(03:42):
a Tuesday to Rutgers and play a game and fly back from LA, right? It’s a big, big money grab here
and the reason these schools are making a lot more money is because of the media deals, right? So the
media deals - SEC football, look at the viewership on SEC football and Big 10 football right? This is
what drives the vast majority of these revenue streams to the schools. Now separately to your
(04:04):
point, the boosters and these collectives, so now you have a system in which boosters can pay
schools through a collective in order to help pay athletes through NIL deals. This was the situation
all through up until this coming July where it's all going to change. So back when NIL started, you
(04:26):
had multiple years of, really, a budgetless team you could create. As long as there was a donor
rich enough to give money through a collective, you could use that money to fund your roster.
Liberty (04:36):
Pay the students?
Michael (04:37):
Pay the students, correct. Pay the athletes,
exactly right. So this creates disparity. You have the Ohio States and Michigans of the world
with budgets in the twenty-plus million dollar range for football and then in the
same conference Maryland and Rutgers are under five million, because of the donors
and the fan base wanting to support those programs or not support those programs.
Munther (04:58):
I wanted to push back a little bit and even though I've actually traditionally
have argued against the NCAA rules, thinking that athletes ought to be paid,
I do find it a little bit of a - there's tension in that process, and I'll give you an example and
I think it would be great for you to draw on some of these analogies. So in universities,
(05:23):
a place like MIT, we don't have athletic programs like - but we do have -
Liberty (05:28):
Hey, I played field hockey at MIT,
thank you. *laughter* We had some athletic programs, thank you very much.
Michael (05:36):
No revenue-generating athletics. *laughter*
Munther (05:40):
*laughter* Yes, thank you, you saved me. Of course. But I would say we
have a parallel in terms of entrepreneurship and innovation. Where students can actually
have the ideas to create companies and make a lot of money, but we have similar rules to
the NCAA rules where it says for example if you're working with your advisor on an idea,
(06:04):
then you can no longer stay with that advisor and be a student if you want to develop it
into a company. Which then have you make a choice between staying at the university
and finishing your PhD say, for example, or moving on to a startup company which we know,
you know, the probability of success, and we put the students in that difficult place,
(06:25):
and I feel it's a similar situation of saying ‘oh we're not comfortable paying our student students
and we want them to finish their education this is the training ground and then they can go
later and and get their money’ the way we thought about collegiate athletes. So there's a tension,
I think, in converting that whole process into money and losing this ability of the
(06:46):
training ground that was happening in the universities. Any thoughts about this?
Michael (06:50):
So yeah, that's interesting and so you see this - a good comparison to what you
just referenced is especially in football where it's really hard to play as a true freshman,
right? So you go to a school and you're a true freshman, you're not going to play but that
school is going to develop you into a better player, right? It's going to help you succeed.
Munther (07:08):
Right.
Michael (07:09):
But the problem is, that school will help you right? Your freshman year,
they're turning you into a good player. Now you become a good player, well now sophomore year,
the next school over in your same conference is going to offer you double the amount of
money to switch schools and go over there. So now the school that has just trained you,
that has helped you fine-tune your craft now doesn't get the benefit of that because they're
not willing to to pay a high-enough price. You are seeing that in college athletics all the time.
Munther (07:35):
And that's the downside for the payment, right? If there was no payment,
maybe you would have committed and stayed with the school that trained you, with the potential
of you maybe even getting more playtime, because there's more playtime in one place versus more
money in another place, but less playtime because all the talents are concentrated,
right? So this equity division and the training for more future, better players
(08:00):
now is impacted by decisions of how much money I can make now.
Michael (08:03):
Correct, and the old rules - you couldn't transfer and leave, right? If you did,
you'd have to sit out for a year. So you had to stay with your original school. Again,
some people think that that's fair, other people - more of the free market
people - ‘hey the student should be able to, you should be able to make as much as you can
make wherever you're going to make it and if you want to bounce around’ - I mean, we are already
(08:25):
seeing multiple athletes that have started for four years at four different universities.
Munther (08:30):
Wow, really.
Michael (08:32):
I mean this is happening - this is not just one or two cases right now,
right? We’re in dozens if not a hundred cases throughout all of sports in the
NCAA world. I don't have the exact number there, but it's it's more than a dozen of
people that have played at more than four universities in their career.
Munther (08:48):
So this actually makes it very difficult for coaches and school
managers to manage a team that's so dynamic, people are coming in and out all the time,
there's no real history of people playing together.
Michael (09:01):
I mean, it's a very interesting dynamic and there are some coaches that can't handle it
and that have just quit. Some of the best coaches in the game have literally, like ‘I can't handle
this and I quit’ because they're not used to it. I am on the other side of this *laughs*, because
I'm just a big player guy. You know, in the NFL, you have a brand new team every year.
Munther (09:22):
Oh wow.
Michael (09:22):
Right? You get new free agents coming in every year, you have less time to train with them.
If you want to keep the players, you can keep your players, just pay them fairly. And look,
if you have a player that's a freshman, he's going to stay to sophomore, these
aren't always bottom-line guys. You can usually get twenty to fifty percent discount to stay,
generally speaking. So there is a component of that, it depends on the player, it depends
(09:44):
on the character. We work with you know UVA, Alabama, Texas Tech. Shoot, Alabama can get
massive discounts in football because they want to play at Alabama, right? So players do take a
longer-term view. There are - ‘Hey look, I'll come to Alabama and I'll get X dollars now,
which maybe is only twenty percent of my market value, but I know that I'll make so much more
(10:07):
money in the NFL because Alabama will train me and make me this good player because they have
this history of success.’ So this stuff does happen, this isn't - I don't want to make this
a black and white, everyone's going for bottom dollar - are there players that will always go
for the bottom dollar? Absolutely. Especially now because they all have agents. So this is another
world - all the agents are getting into this world and what agents want is maximum money,
(10:28):
right, because that's how they get their cut. So you're get you're seeing more and more of this.
Munther (10:32):
I guess we have to train all these players in probability theory, right,
because they all have to be calculating risk analysis and expected reward.
Michael (10:41):
I would argue that the whole world should be trained in probability a lot more
than they are *laughs* and this is inherently the issue with athletes, right? Athletes are the kings
and queens of being irrationally confident. I always joked, I was the captain of the irrational
confidence all-stars. When you're a pitcher - I was a pitcher - there's nobody in the world
(11:02):
that could get this out better than I could. Now realistically, looking back, were there?
Absolutely. I was not the best pitcher in the world, I wasn't even close, but when you have
the ball in your hand or the basketball in your hand taking a shot - if it's you or Steph Curry,
you're picking you. Even though it's an illogical decision. Athletes believe in themselves,
every athlete in college thinks they're going to be a pro athlete - less than five percent will.
(11:25):
But it's a really hard thing to do to explain this to them. So they, what they a lot of them will
often times think is - ‘well I'm going to make it anyways so I'll just get the most amount I can
right now, because it doesn't matter where I'll go.’ That's where you get the support systems,
the advisers, the parents that can sometimes rein in these kids in and be like ‘Hey, no,’ like talk
(11:46):
a little more sense ‘You're a lot more likely if you're going to this school or this school.’
Liberty (11:51):
So, I just want to make sure that I
know where we are right now. You said everything changes in July.
Michael (11:56):
Yes.
Liberty (11:57):
What happens? Where's all the money, what’s happening with the transfer forward,
the NIL, the collective. What is the game plan moving forward?
Michael (12:05):
So now it's pretty simple. Now they've made the revenue share of each
school and the schools can decide how they want to split it. There is no more collective,
it is not an all-you-can-eat buffet anymore.
Liberty (12:15):
So, this is gone?
Michael (12:17):
Yes, so this is gone.
Liberty (12:18):
The booster thing is gone-zo.
Michael (12:20):
It’s essentially a salary cap, okay? So it's $20.5 million that the school can allocate
however they want if they want to allocate $20.5 million to field hockey to help Liberty,
they can! *laughter* If they want - most schools are breaking it down the same way,
a somewhat similar way which is 13 to 15 million of that 20 is going to football, three to four is
(12:41):
going to basketball and then the other sports are splitting the rest of it. Which makes sense from
a revenue-generating standpoint, there's obviously a lot of Title Nine things that have come out that
I'm not going to, I don't think we need to get into on this podcast, but all the conventional
wisdom and thinking is all that stuff's going to be over - it's going to be a separate system,
and all the schools that we work with - we work with five schools - are all operating under the
premise of that 13 to 15 million football, three to four million in basketball. Now,
(13:06):
real NIL is still allowed, this is where it's going to get super dicey, and that is that players
are still allowed to sign deals with Joe Smith’s car dealership at Tuscaloosa for Alabama players.
Now the problem is, these must be fair market deals and the NCAA has hired Deloitte to audit
these deals. So this is going to be, I mean, this is going to be in court a zillion times right? ]
Liberty (13:30):
This is like the Wild West.
Michael (13:31):
Exactly, exactly, but what it does is it makes it so from a legal standpoint you can't
just - ‘hey, I have a booster, I have a lot of money here's a car dealership, I'm gonna pay
this basketball player two million dollars for a commercial’ because if they do that, it's going
to go to the NCAA, it's going to get in court but while it's in court that player can't play.
So it does no good, and then the car dealership is gonna have to explain why this player's worth
(13:53):
two million dollars and will likely lose that. Now again, a $100,000 you probably can get away with,
right? So it's this whole ground of fair market value, but what this does - I'm going to switch
gears now to the data science part of this, as this is a data science podcast, and that
is with - now that all schools have relatively even playing fields, more even than it's ever
(14:15):
been in the history of college sports, now how do you effectively allocate that capital? So
now you have athletic directors - I love athletic directors - they're not skilled in how much player
value is across multiple sports, right? Even head coaches in college, this is not their background,
they don't - if you look at NFL - a professional program - they have next to no say of who comes
(14:37):
on their team and how much a player's worth. So nobody knows how to do this, nobody knows how
to value this, yet they have to allocate twenty million dollars to do this. Now here comes this
beautiful data science puzzle. Now you can get the AD (athletic director)’s back to say okay,
this running back transfer from South Dakota State is really worth X to my organization,
(15:01):
and now they can have that backing because what can happen if you spend that money improperly
is going to be a disaster. We saw it at Florida State last year, they paid the wrong quarterback
a couple million dollars and they had one of the worst seasons in school history, and now you're
going to lose donor dollars. Now you're going to lose likelihood to move conferences because your
(15:21):
football team was garbage last year. Now, Florida State did a really good job in this year's portal
but that's a different story. But the idea is it gives you cover, because think about this,
the coaches I worked with beforehand all the theory was ‘here's good player,
I want good player, sign good player’ and they've been doing that for thirty years or however long
they've been coaching. So now they'll come to me like ‘oh, here's a running back I love like this
(15:46):
guy's a great player’ and schools I work with, they're not allowed to tell me that anymore. Is
he worth twenty million? Well, of course not, so okay he's good enough for how much? There's
now a ‘for what price’ component to this and it's really hard to get that mindset from coaches away
from ‘this player's good’ to ‘this player's worth X’ and this is where data science can really help.
Munther (16:07):
Yeah, I can imagine this being a really interesting problem to solve,
but I want to clarify one detail about this and that is the fundraising on
the side does not go into this spot of twenty million to spend.
Michael (16:20):
Correct. So the school can allocate - the school can do whatever they want. The school
can say - if you're the University of Virginia, you can say of the school's money, I’m giving the
$2.5 million of rev share. They can also raise the $20.5 million of rev share so it doesn't come out
of the school's pocket, but the cap is $20.5, cannot be more than that, period end of story.
Munther (16:43):
So okay, and as a data science - so this is great, so it's a fixed pot for everybody. Some
people may not have it but most people will try to figure out how to raise it and spend it, and
you're saying this is done across multiple sports including the managers and the directors or not?
Michael (16:59):
No, only players. Coaches and everybody else are all exempt,
so they still live in total free market world when it comes to this.
Munther (17:08):
Got it. Okay, and what you said is the real problem is that not so much that this is
a good player but how much would this player cost me, because there's also an opportunity
cost. But this is something NBA management had to do for many years with the salary cap,
you know, and sort of the swapping and saying ‘Okay I can get two for
this price of one person’ and so forth. So is is there any knowhow in the data
(17:31):
science world from the NBA or has the NBA never been a data science organization?
Michael (17:36):
No, the NBA does a great job in data science. It is a different mathematical
problem slightly because there's a lot of different rules in the NBA. For example,
there's you know max contracts you can sign. ln college, there are no rules with that money,
so like LeBron James can make only the maximum contract allowed by the CBA.
Munther (17:55):
I see
Michael (17:56):
So he can't make a 100 billion dollars if somebody wants to pay him 100 billion
dollars. In college, if you wanted to, you can give all $20.5 million to one person.
Munther (18:03):
Interesting
Michael (18:04):
There are no rules and there's no minimum. So the other really cool
thing about NBA - or cool or not cool depending on how you look at it - NBA,
NFL, MLB - is they all have rookie-scale contracts so in baseball you could be the
best player in the game for your two full years Tim Lincecum won two Cy
Youngs and he got paid the major league minimum, the lowest amount of anybody,
because it's more of like a tenure game of how it's set up. This is total free market in college.
Munther (18:28):
It’s really more like a a coalition game, it's not so much you know -‘what is the
optimal coalition that I can have for a fixed amount of budget?’ which also has
externalities in the sense that I'm also taking that player away from another team.
Michael (18:43):
Yes
Munzer (18:44):
So it's not like I just got them for me but I also took
it from another potential competitor which could be also a plus for me,
so I have to calculate the externality factor of my data science calculation.
Michael (18:56):
No question about it, and I'll draw back and give you a real life example. We’ve been
doing this for three years in basketball. So we partnered with the Alabama basketball program to
help their roster building. Our first year we took over, we helped with the recruiting in basketball,
we have every shot, every decision every players's made from fifteen years old and on,
all skill level adjustments, all opponent base, so we know how how they're going to perform to a
(19:21):
tighter degree than the eye test right. So we put together this roster and it was the
first time in school history, in over 100 years, they were number one in in the country going to
the postseason tournament. A lot went right there. Issue - seven of eight starters gone
the next year, because when you have a good team you leave for the NBA, you transfer out,
all these things. We have next to no money - under about about a million dollars, a little
(19:44):
less than a million dollar in NIL. All these other schools, a lot bigger donors for basketball,
because remember, Alabama donors are giving it to football *laughs* The basketball program was
a lot less at that time, we couldn't even get, like - talk about getting a recuit from Florida
to help that in the conference? Absolutely not. We couldn't even get a mid-major. Our
starting lineup was Hofstra University, Cal State Fullerton, North Dakota State,
(20:10):
Ohio University and Wofford, but our models looked at the coalition of how we can build this roster
and the talents of each of those players, not individually, but how those talents and those
skills are going to react amongst each other to create the most points on offense minus the
most points on defense, and for the first time in school history they got to the Final Four.
Munther (20:29):
Wow.
Michael (20:29):
With a roster that had the lowest salary or one of the lowest salaries in the entire SEC.
So you can do this type of stuff. Now, it takes a lot of work, I mean, again, we've been building
these models in basketball five, six years, have incredibly intelligent people - always hiring by
the way if anybody's listening, I'm throwing a plug in Liberty, but if anybody's listening and
(20:51):
like sports and data analytics, please apply for a job here at BLA. The company that works for the
schools it's called SAA, Sports Analytics Advantage, and you know it's really hard to
be a data scientist at SAA because it's a really different skill set than your normal data science
folks, and that is because we really put a premium on creativity, and it's really hard to get that
(21:17):
logical algorithm-building brain paired with a super creative outside-the-box-thinking-brain.
In academia, in general, most of these kids are taught to work at Meta and Uber and Facebook,
big money jobs - as they should be, by the way, that's where the money is,
which is a lot more execution type of data scientists. ‘Here's this list of data,
(21:41):
make all these executions, execute this stuff, find out these really cool predictive measures,
etc’ but in sports it's harder, it's tiny sample size, it's missing data,
you need to figure out creative constructions and how to build not only your data sets,
clean your data sets, but then build modeling frameworks around these in different types of
ways and so it's a really cool thing, and I think we've done a a pretty good job so far.
Munther (22:05):
I think the issue in these coalition structures that you're looking for is defining
what is a success, right, because I used to be an NBA junkie, I watched every single game,
basketball junkie in general, watch games and you think you become an expert - you're not,
but you think you become an expert and so forth, and it's often the case that it's
(22:28):
not really the obvious numbers that make a team the winning team, but rather a little bit more
intangibles about players and dedication and camaraderie and giving the ball an extra pass
and that kind of stuff that makes that one extra point at the end, and it's very difficult for
algorithms to capture this kind of dynamics. So I think it's really interesting as a data
(22:51):
science problem. You can be successful but can you really develop confidence in your methods?
Michael (22:56):
Yeah, in the last ten years since we've been operational,
2,500 data scientists have tried our assessment, under a hundred have passed it
Munther (23:03):
Wow.
Liberty (23:04):
I want to try to pass it, send it to me. *laughter*
Michael (23:11):
But we want there to be more, so I'd love to - you know, usually the people that pass
it aren't necessarily the highest GPA people, it's more of the people that - ‘On the side,
I made a fantasy football team’ or ‘I watched basketball all the time and I wondered why your
team's not fouling at this, so I built this model to determine this.’ It's like those types are,
(23:31):
generally speaking, the ones that are able to get through and pass our assessment.
Liberty (23:36):
So I just want to step back a little bit, because you're talking about building
these models and how much data you have on all these kids. How do you get all this play
around high school kids? If you're trying to recruit seniors in high school to come on as
freshman to the team, how do you do that, how do you account for them in your model?
Michael (23:54):
So it's different in all the sports. Basketball is what I was referring to with every
shot and every decision they made, 15 years old and on. The reason is because they have
this really cool system called the AAU system. So if you map out all the good players, play all the
good players, right, from fifteen years old and on. We also have about ten to fifteen of these
high schools that are athletic high schools - IMG Montverde, the Brewster Academies, like all these
(24:19):
types of schools and we will model those out, but we only count the games where we have opponent
adjustments. Obviously, we don't care if you score sixty points against somebody that we have no idea
about. So everything you do must have an opponent adjustment to get the correct skill levels,
and what we're looking for there is progression through the years. So we know different skill
(24:39):
sets develop with different players at different rates and different times so we can then project
what are you going to be at 18, 19, 20 years old. So one interesting example, if there's any - I
think we got a lot more data scientists than basketball players - but for ten seconds if you're
a basketball player out there and you're fifteen, finishing at the rim over elite rim protection
is a really good skill to have at an early age. Being crafty around the rim. Shooting doesn't
(25:03):
really matter as much at age fifteen. When you get to eighteen or nineteen, shooting is
the most important skill, right, like be able to shoot accurately from a distance, but each skill
develops again over time, and so when you look through and you can find these correlations you
can then predict what they're going to be. Brandon Miller was the number one player we thought in the
class coming into Alabama, he was a twenty-seven percent three-point shooter we projected him out
(25:27):
at like a thirty-seven percent shooter at Alabama and he ended up shooting close to forty percent
from three, was the first college player taken in the draft. Why? Because his free throw percentage,
his floater percentage, his two-point jump shot, all these other different percentages against
different type - we knew he had touch, he just hasn't developed a three-point shot quite yet.
Munther (25:44):
No it's interesting because let's say, let's imagine that BLA, I guess your company,
is the only company out there that does this - just bear with me for a second - and so then
what happens is that you would have these methodologies for selecting the players
and ranking them and so forth, and you effectively you are changing the game,
(26:08):
because your method assumes a certain type of a game that these people will be successful at,
you're selecting the players for that game, you would be changing the game.
And over time I think this impact will be felt by what we see as a top player.
Michael (26:25):
I appreciate that. I mean, tooting our own company horn here, we worked with four schools
this year - Alabama, Duke, Michigan and Texas Tech. Last year Michigan won six games, Texas Tech
nowhere close, right, and now we have all four in the top 12 in KenPom ratings right now, top 12 in
the country. I mean this is what can happen when - and it really just shows too how bad the market
(26:49):
is, how bad eye test can be, and it's hard it's - eye test is a little bit easier in the NBA,
there's thirty teams. In basketball, there's 363 teams. How am I supposed tom as a coach,
watch 363 teams play every game and and then try to figure out the starter at Eastern Kentucky
getting twenty a game how he correlates to the bench player at the University of Florida? How
(27:09):
do you do that? It’s just so hard with your eyes versus - again, NBA when everybody's professional
at the same level, I would still argue the data is far more important than the eyes,
but the eyes are a lot better because it's a lot of like-to-like comparisons.
Munther (27:23):
Sometimes in class I use this example for my students, from Red Auerbach,
from the old Celtics days, I think it was an interview with him and he presented this scenario,
he says ‘Your best players are playing and then they're losing they're not doing very well,
so you take them out and you put the second team in and the second team catches up and
(27:45):
now you're tied for the last shot. Do you stay with the second team or you bring back the the
first players?’ and I always present that as a statistical test for my students,
and that is the question of - how do you rationalize this problem from a statistical
point of view? Because when you have such a terrible day for these players and the
(28:07):
other team has brought you all the way to a tie game, and he was saying you bring your starters.
Michael (28:14):
I mean, I would - we have a lot of data now, we have heart rate monitors,
we know who's tired, we know how long everybody's been on the court. We can optimize for all these
things of who we think, there’s a lot more we have now than Red - I mean, you think,
Red Auerbach was just so far ahead of his time, a true legend of the game, and that's what you see
too, these coaches that were legend. We had the fortunate - we worked with Coach K, our
(28:37):
first two jobs were at Duke, were Coach K's last two years, and he was playing penetrate, kick -
Liberty (28:42):
Even I know who Coach K is.
Michael (28:43):
There you go. He was playing penetrate,
get in the paint, kick, shoot threes before anybody understood what that was.
Munther (28:51):
Yes.
Michael (28:51):
Right? And if you think about basketball it still annoys me,
it's - it took ’til the early 2000’s to realize three points was more than two.
Munther (28:59):
Yes, yeah yeah.
Michael (28:59):
Like, think about that. *laughs* It is a wild - you think
we're so advanced with all this stuff and then you realize - you think about
that it's - Sports is just behind on the analytical side than other aspects of life,
but when there's an opportunity there's something to be done about it.
Munther (29:20):
I think a lot of coaches and players didn't believe that three-point players can
actually hit with that percentage, so I think that's the other thing is that I think right now
I believe if you create a four-point line, we're going to have four-point players, and they'll
be hitting percentages. I mean, it's a skill, as you said, that when you're older you develop it,
(29:41):
and you play it and becomes a thing. So that's really interesting how I meant by changing the
game. The game changed with a three-point shot, it's a entirely different game because
of the percentages and the efficiency of a three-point shot versus two. So now big men
are not as as important as shooters, right, and the value - well they are, but it's a different -
Michael (30:03):
Yeah, I’m going to push back on that and this is where I think this is an
interesting point is everybody thinks the NBA is just running down shooting threes - there are so
many ways to build championship teams, so many ways. Especially in college,
because the paint's smaller, all the different rules. A good - analytically a great, elite big
man is actually better than a Steph Curry type player, because you're going to draw double teams,
(30:28):
and you're going to get open threes because of that. If you're a big man that can score
at 60 percent on one-on-one situations, you're the most lethal player in the country.
Munther (30:37):
Right, right.
Michael (30:38):
Because you're going to force double teams, and that's how you get open threes,
you don't have to dribble, drive, kick and do all that right?
Munther (30:45):
Yes.
Michael (30:45):
So, there's so many - and if you look at the four rosters we have, Michigan starts two
seven-footers and they play almost all inside. Alabama spaces it, runs shoots, more threes
than anybody in the country. I mean there's four completely different styles of play that we have,
and that's what I love about it. You get information from the coach. Gere's the team I want
(31:06):
to build and then you simply, you optimize the coach's brain. This is where I think a lot of data
science gets wrong, they say look I - you know, I think data science sometimes have too big of an
ego, I'm gonna say it. And I think that sometimes they go in and they'll say - here's the rules of
the game this is the most effective way to play, do this. Well, first of all, even if you're right,
coaches aren't going to listen to you because nobody has bigger egos than coaches, and second of
(31:31):
all, if you go to the coach and you say ‘this is what we do’, we go - ‘look, we know you know more
than we do, we know that. That's the premise we're going to work off of. You are smarter, you've been
around the game, how do you think about the game, how do you think about winning?’ and our job is to
build a model - because you have one fatal flaw as a coach, you got two eyes and one brain. Nothing
(31:52):
you can do about that, you can't watch 363 teams play every game. So we're going to build a model,
we're only going to be eighty percent as good as you - we hope! Like if you watched one game
and isolated every little thing, you're going to be better than our computer,
but if we can be eighty percent as as good as you, watching every game simultaneously,
grading every play and grading every player and grading every skill set, and now we know what
(32:13):
you're looking for, we're going to create you a roster that can be dynamic, and it's a roster
that you know how to coach because the players that you have. I mean Nate Oats at Alabama,
unbelievable coach, one of the best coaches in college basketball, You put Zack Edey on his
team who's a seven-foot-six monster post player, one of the best players in college basketball,
it's not going to work. He doesn't know how to - like, that's not how he coaches, that style.
Munther (32:35):
Yes, yes.
Michael (32:36):
So you have to match the players with the
coaching philosophy. That's a really cool part of the puzzle.
Munther (32:42):
That's really interesting, very interesting.
Liberty (32:45):
So I just want to make sure that I get this, So you're you're doing two things,
it sounds like. It’s not just building the roster, you also - it sounds like
you do realtime decision making too, when you said you have heart rate monitors,
you know how tired they are. So it's sort of like a two-part thing of - you build this
roster but then you're also deciding at any given moment in the game who's on the court.
Michael (33:07):
We do whatever the coaches want us to do. So there are some schools that want
us - in football to figure out what plays are going to work against what down-and-distances.
There other schools that are like ‘just build the roster, we're good.’ I mean it - just
however we can be of help to the university, we want to do that. We want to help win games.
Munther (33:25):
Are you allowed to make realtime decisions based on
computational capability. Is that - like an an AI assisted tool?
Michael (33:33):
Absolutely you're allowed to do that, there's different ways to do it. There are certain
things that you can and - there are different rules and things that you can and cannot do in
there. But you can, for example, build out every situation of a football game before the game
happens, and know that this defensive coordinator is going to be in cover two on third and four to
third and six 79.3 percent of the time, and the other time he's going to be in cover three shell
(33:58):
and now you can call a play that can beat both those coverages. So now you can go ahead and be,
like, when we're in these situations these are the four plays we can call.
Munther (34:07):
Got it
Michael (34:08):
Because we know the tendencies of everybody, so you know it in advance.
Munther (34:11):
You prepared the scenarios and then you react to the scenarios in real time.
Michael (34:15):
Correct, and there's also different real-time things that you can do. We have lineup
optimizers in there, so when a different team puts out a different lineup - what's our lineup
that's best to combat that, net expected points, there's all kinds of stuff that you're able to do.
Munzer (34:31):
And the biometrics are allowed to be used in the game,
like the NBA, I thought it's not possible?
Michael (34:36):
So biometrics are not. So biometrics - we get the data in and we know different players
playtime loads. So for example, one player gets - when he comes in the game does really
well until about six minutes and then after six minutes his play starts to go down. So we
knew that beforehand, so that's how we're using the hard data - so to your example,
to the Red Auerbach example, if those second teamers were right at their peak and about
(34:58):
to play great, I'd keep them in. If they're about to fall off a cliff,
then I'd bring in the starters. We would have that data and that information, but we would
know that ahead of time. This is not like a real time - he's at this level right now, correct.
Munther (35:10):
Yeah, yeah.
Michael (35:10):
I mean, we have the data you're just not allowed to use it. I should make that clear.
Liberty (35:14):
So how does it work, it sounds like what you're saying is that there's a huge
amount of - there's obviously a huge amount of data science that goes into this - but
when you're talking about matching up the data with the coach's ethos or how they know how to
coach or what they like to do or whatever, there's a lot of human decision-making in
here. So can you give me an example of one of the biggest mistakes, where the data is saying
(35:36):
one thing and it just blew up in your face, or somebody else if you guys haven't had one.
Michael. Oh yes, yeah I'll give you a big miss that we've had. So in basketball there was a
player, not ranked in top 100, we thought he was like a top 15 player in the class. Reason is,
he was a massively high IQ player on the computer. Super smart,
meaning when he would dribble in the lane, players would come and double team, he would always make
(35:59):
the right decision. This is exactly what you want to build - it's really hard to teach IQ,
and our computers have been unbelievable at generating player IQ, like we do - we build
NBA draft models for NBA teams. Tyrese Haliburton was like one of the best players in the class,
he waits late, like this was a great player, IQ index, all these things. Turns out we got this
(36:19):
guy completely wrong because of the play style, they basically ran one play. He was a left-handed
player - he'd get the ball on his right wing, he'd get the ball in his left hand. He knew
there was one read for him to make and he made that one read right every time. Well, when you're
playing a flow-style offense and there's 25 different reads to be making on a given play,
he was lost and was always making the wrong decision, so many turnovers. So here we thought
(36:44):
this was this great IQ player but we didn't - at the time, this was three or four years ago,
now obviously we've refined our model since then but - you learn from this stuff,
right? And so you learn - what are the types of decisions now is he making? It's not percentage
of correct decisions. It's the different types of decisions that are more relevant.
Munther (37:05):
So I want to get into the mentality of a player going to college and trying to
understand now all these factors that I have to be thinking about. So I'm a decent basketball player,
maybe division one player, maybe I'm good, I don't know. Maybe I had this
illusion that I'm going to be an NBA player, I wanted to go to a good program and train,
(37:26):
but now I'm thinking okay let's be reasonable if I go to a program that pays me and - maybe a program
that I get developed - but maybe if there's more opportunity for a deal that I signed with a shoe
store in the neighborhood or something. So now my decision is much more complicated. Before,
I wanted to be trained, maybe I'll get my best chance to be an NBA player. Now I'm going to
(37:49):
try to maximize my income in the meanwhile, in case things fail and I compromise that. From
the players perspective, who's helping them? I mean as a parent, I don't think I can help them.
Michael (38:01):
Well, you could as a parent, most parents can't *laugh* If you had the right date,
I promise you could help. But the idea - I get this question from parents all the time.
Munther (38:13):
Okay
Michael (38:13):
‘Here's my son, what do I do? He's getting offers from these four schools - this
school's offering us fifty grand, this school's offering us 200,000, this school's this,
this school's that, this school is guaranteeing me playing time, this school's not.’ I mean
there's so many factors as you say going in and by the way, you're seventeen years old. I mean,
I don't know what you were like at seventeen years old, but I probably wasn't making the most logical
(38:37):
decisions at the time. But the idea is, how I help parents is I model it all out for them. And so
what our company was based on before we got into this is projecting future earnings of athletes.
We lucked into this whole thing by the way, like this is a whole separate side story. We have a
whole company Big League Advantage which invests in future earnings of athletes. We project future
(38:59):
earnings and make investments based around that. So now, once the NIL stuff started to happen, we
were the only company around that had valuations for players. By luck, right? And so what I do is
I say - look, okay here's parent, here's the four schools, here's the head coach, here's the systems
(39:20):
that they run, here's how your kid is going to fit in these systems.’ We can build a simulation
model, we say ‘here's his general statistics if he goes to all these types of places’ and we say
the value for going to each of those schools, then you overlay the dollars. So you say, okay,
at this school I'm going to average 13.8 points, 14 rebound, whatever, whatever, whatever. For this
school I'm going to be played totally different and my future earnings at the end of that year is
(39:44):
XYZ, but now - so let's say your future earnings is eleven million dollars after year one in School
A, twelve million after School B, but School A is offering you two million, School B is offering
you $100,000, I'm going School A, even though it's not as good of a school for me technically,
the math still works. Now if School A was only five million versus twelve million,
(40:07):
I’m making a different decision. So we will do these analyses for players and their families.
Munther (40:13):
And these commitments that the universities make,
they are obligated to stick to it or can they change their minds?
Michael (40:19):
No, obligated but only one year. That's the problem. All these contracts
are one year deals, that's where you're seeing all this crazy action. I think this
is going to change. I think you're going to see contracts like coaching contracts,
where it's three year deals and you get a buyout in there. So it's like $100,000,
$300,000, $700,000, but after the first year if you're $100,000 now you're going
to $300,000 but you're the best player in the country and you're worth three million,
(40:42):
great you can transfer but you have to pay a million dollar buyout to the university.
Munther (40:46):
Wow.
Michael (40:47):
So now the school that wants you has to pay these buyouts, that's how I
think it should go, because this means that players will stay at the same school longer,
for the player they're locked in, a million dollars over three years,
and the school is now committed, they don't have to think about and by the way, if he leaves for
a million that's an extra million dollars of revenue, you can go get some more players.
Liberty (41:06):
Is anyone else making money off of this besides the players? As we said this,
is a little bit of the Wild West of what's going to be happening now,
in terms of the car dealership being able to give the kids money. Is there - it always feels
to me when something this new with this much money is starting that there's going to be
(41:27):
some funny business or some accidental funny business and things are going to be in court,
and I mean this is a lot of money, these are players lives, these are colleges that rely so
much on this money and this play. Where do you see the pitfalls of this coming?
Michael (41:42):
Oh there's a lot, but this is not like a new thing that - like the way you described it,
I would say like instead of the school making fifty million, now the school makes thirty and
the players make twenty. So it's not like oh, you know, it's a little bit, it's the definition of
revenue sharing, right? They're going to have to share this revenue with the players. Now of
(42:02):
course, all the schools are going to call broke on, poor on this, ‘now we have so much less money’
but the pitfall is going to happen with - the bag droppers are going to come back. When I say
bag droppers, I mean this how it used to be when those athletes were allowed to get paid - ‘Hey,
come here and we'll give you $50,000 in a brown paper bag.’ Now dentifying the bag droppers is
(42:26):
going be impossible. You used to be able to do it because all of a sudden the player driving
around a Ferrari you're like, well that doesn't make sense, right? Well now, if you're getting
a million dollars from the rev share and $500,000 under the table and now you're driving a Ferrari,
nobody's gonna know that. Now where I'm a little bit optimistic on the bag dropping is I think
(42:49):
donors are going to be a lot less, boosters a lot less likely to do that. Back in the day, you feel
bad for a player, you're the university, this guy's dominating, you're the star quarterback,
here's a few hundred thousand bucks. Morally, maybe that's more okay for a booster, trying
to get inside their head. ‘Now this player's making a million dollars, I gotta give them
an extra $300,000? Your guy's already making a million bucks, I’m not willing to do that’ - I
(43:15):
think is where you're gonna see it. Now will some people say, that are billionaires that ‘I don't
care here's money’? Probably, right? Take that risk, that's the big pitfall in this whole system.
Anytime you don't have a full, 100% free market, there's always going to be people getting paid
in different ways. You're going to see the NIL side of it, you're going to see here's a booster
(43:40):
in the hedge fund land that you can't give a true NIL deal to, so hey I'll go pay Microsoft
a million dollars, now Microsoft gives my player a million dollar NIL deal to go on a commercial. So
that indirect route of real NIL, you might start seeing some of that you know in there as well.
Munther (44:00):
The free markets - I am a big believer of free markets - but when they optimize the
long term, not the short term, right? So you can allow free markets in the short term and
we end up in monopolies and concentration of wealth and all of that stuff with free markets,
I mean it's tough, but with the long term and understanding it's putting on the right
(44:21):
objective. But maybe there was a different world, Michael, maybe we didn't think about a different
world. So let me propose a different world where there was no money in collegiate sports,
no money. The universities don't make money, the players don't make money. It's a training ground.
Nobody makes money on my courses, very small amount of money but we train these students in
(44:42):
statistics and and decision theory and all of that stuff and then they go to the real
world and then there is where there's the money. Why is that model not a good model?
Michael (44:52):
I love that model. That's what we were doing in 1930, when the NCAA started but now what
happened is that everyone started to watch and got addicted to these sports. So then,
people wanted to put it on TV. So then, people wanted to pay to put money on TV. So now the
(45:14):
schools are making a billion dollars. So now you could - this is the NCAA’s point - hey let's
just train them up, let's never pay them anything. This money - they never got paid anything in 1930.
Munther (45:23):
Yeah, that's not a good argument right, because someone is making money.
Michael (45:26):
Right, but I mean that's the issue with popularity,
and anytime someone's making money everybody wants a piece.
Munther (45:35):
Yeah, I agree.
Michael (45:36):
Now here's what's great about why it makes money. It’s an amazing thing,
because football is the golden goose. Why can UVA have a field hockey team,
because of football. This is a massive thing that's been incredible - the boom of women's
athletics can be directly attributed to the money going into football, and this
(45:58):
is amazing. As a father of a daughter right that now loves sports and goes to Caitlin Clark games,
that doesn't exist without football, if nobody got paid anything those sports wouldn't exist.
Munther (46:10):
Yeah.
Michael (46:11):
So, you know, that's how I would look at, I mean, I think it's a good thing and I - now
look, it doesn't mean it's an easy thing. This is incredibly complex and there's - every solution
is going to have detractors and downsides from economic theory to politics. By the way,
you're going to see Ted Cruz and Cory Booker duke this one out in terms of how this is coming,
(46:32):
be prepared for that. I would bet any amount of money within the next few months
Trump chimes in on this, this a real thing here. So, we're gonna see how it plays out.
Liberty (46:42):
I have one final question. So you guys basically built for Alabama this team that went
to the Final Four with the least amount of money, you named the roster. Is there ever a time where
one player is worth four million - a majority of the money - 25 to 50 percent of the money?
Michael (47:04):
25 percent, absolutely. 50 percent? It gets into externality - external stuff. So this
is what's happening with BYU. The number one player in the country AJ Dybantsa decided to
go to BYU for next year. Reportedly getting about eight million dollars. Now remember,
this is the last year that you can go crazy. But the idea for a school like BYU and pay one
(47:30):
player all the money essentially is that now if I can get a first player pick in the draft, I'll
for the next five years get a lot better recruits coming in. So the theory isn’t I'm getting him for
this year to build the best roster, the theory is longterm if I get one or two draft picks,
now the next best player in high school will be like look I can develop draft picks is the
(47:52):
concept. If you're looking purely mathematical - of all sports happen in just one year next year,
no player is worth half the portfolio in basketball. In football, the best
quarterback in the country will be worth about a third, which a third of $15 million is no joke.
Liberty (48:13):
Well Michael, thank you so much I thought this was fascinating and I know we're all going
to be so excited to see how your teams do next year, so we hope you're going to come back and
let us know the the successes and the pitfalls of your first year working in this new world.
Michael (48:31):
I appreciate it, thank you so much for having me
Munther (48:34):
Thank you so much, it is really informative. I wasn't really aware of all
of these things happening in the collegiate sports and this is - a lot is going on, right,
I mean payment decisions, fundraising, ability to do external contracts and
(48:57):
I don't think parents and kids are ready for all of this and I mean this
was an eye opener to say the least. I don't know what you felt Liberty.
Liberty (49:06):
You know, it's one thing when you think about basically kids going into the major leagues
for whatever sport, and they talk about how so many of these athletes lose all their money
because they don't know what they're doing. I mean, think about a seventeen year old, I mean
you give most seventeen year old boys $200,000, I don't know if they're going to survive it. I mean
(49:27):
you also have that like, just - these kids with all this money and a lot of times parents - I mean
I wouldn't know how to like advise a kid on this but you know, parents that just - I mean this is
just, it just feels like it's going to be the Wild West and I think it's all a little scary.
Munther (49:42):
You know, when my son was trying to - I mean, he was a division three basketball player
and we - like, there's no money in division three and yet we were struggling because the
balance there was more about playing basketball and getting a good education. And one thing to
make sure that we are in both, the good education is the is the formula for success in the future,
(50:07):
and so you have to make these decisions, but they were a lot simpler I would say than thinking about
these big contracts and so forth so yeah I think we have some people that do think about sports
analytics at MIT, and I'm going to throw at them this new world that I learned today and see where
(50:28):
maybe we can take it a little bit further in our students thinking about these issues.
Liberty (50:31):
Yeah, and I think especially given that you have these kids that might be making
two million a year for four years who will have no prospect making anything
like that afterwards. It’s like this is their only moment, these four years to
make this money or $800,000 which could be life-changing for one of these kids,
but that's it. When they get out of college they're going to be making $75,000 a year
(50:55):
doing something, so it's like how do you help them maximize that and not waste that.
Munther (51:01):
Very interesting.
Liberty (51:05):
Thank you for listening to this month's episode of Data Nation from the MIT
Institute for Data, Systems, and Society. You can learn more about IDSS and listen
to previous episodes at our website idss mit.edu or wherever you get your podcasts,
or @libertyvittert or @muntherdahleh. Thank you to our producers Tina Tobey Mack and assistant
(51:30):
producers Ben Stull and Maria Brooks. Don't forget to leave us a review and follow our
Twitter @mitidss to stay informed. Thank you again for listening to MIT’s Data Nation.
Today on Data Nation, Michael Schwimer joins us to discuss the newfound role of Name, Image,
and Likeness and data in college athletics. We'll dive into the history of NIL, its opportunities
and its pitfalls and how to use data to build a strong college roster without breaking the bank.
(00:21):
I'm Liberty Vittert, a professor at Washington University in St. Louis and my co-host is
Munther Dahleh, William A. Coolidge professor in electrical engineering and computer science
at MIT. Together, we'll explore the development and function of NIL by following the money and
detailing all aspects of NIL deals. Michael Schwimer brings fascinating insights as the
(00:42):
CEO of Big League Advantage, a company that identifies star players based on analytics and
provides resources to these identified athletes to help make their dreams a reality. Ultimately,
there's so much going on behind the scenes in college athletics,
and there's so many unknowns, so today with Michael's help we'll bring clarity to the new
(01:02):
world of college athletics where NIL and data are unavoidable and revolutionary.
Liberty (01:09):
You know, this idea of paying college athletes is something that I feel like I've
been hearing about for years and now it's obviously happening through the power of NIL,
and initially though the NCAA seemed very reluctant at the idea of paying student
athletes or, I guess, college athletes and they even I think threaten to ban
(01:32):
schools from the NCAA if their states passed laws legalizing NIL. So can you
walk us through this - what is NIL? Why does the NCAA not like it? What's happening here?
Michael (01:45):
NIL came about because there's a large number of people that believe college athletes
should get be getting paid. These schools are making a tremendous amount of revenue through
their teams, their football teams, basketball teams are the big revenue generating sports but
other sports and different schools can generate revenue as well, and these are multi - this is a
multi-billion dollar industry where the talent aka athletes in this case weren't being compensated
(02:08):
at all. In fact, I think it's probably one of the only industries where that occurs,
and so there's a big uprising ‘hey these college athletes should get paid’,
but then the other side is we want to keep amateurism. Obviously, the point of the NCAA,
they're getting paid by free education via scholarship and ultimately that wasn't enough
and I've - as a former athlete myself I usually side with the athletes on this. I think that if
(02:32):
you are the product and you are what people are tuning into you probably should be getting paid.
Liberty (02:37):
What's like one of the biggest schools - let's say, I don't know,
University of Mississippi? Ole Miss, would that be fair? How much money do they make a
year? So it's from, what, TV deals and boosters or donors? Is that - just a ballpark about how
much money one of these schools makes a year off of these student athletes?
Michael (02:56):
Well it's a great question and it matters what conference you're in. So the SEC and the Big
10 make the most money, the ACC and the Big 12 are just that step below. That's what's called
the Power Four conferences, then you have the group of five underneath that which is
like Conference USA and all these other schools that make a lot less. So there's the media deal
(03:17):
that the schools negotiate if you're in the ACC in the Big 12 you're looking at 30 to 45-50 million,
you're in the SEC and the Big 10 it's double that. So there there's a lot of money at stake here and
that's where you see all these schools changing conferences, you see why did USC, UCLA, Oregon,
Washington go from the PAC-12 to the Big 10? Well, money, right? it's not because they want to fly on
(03:42):
a Tuesday to Rutgers and play a game and fly back from LA, right? It’s a big, big money grab here
and the reason these schools are making a lot more money is because of the media deals, right? So the
media deals - SEC football, look at the viewership on SEC football and Big 10 football right? This is
what drives the vast majority of these revenue streams to the schools. Now separately to your
(04:04):
point, the boosters and these collectives, so now you have a system in which boosters can pay
schools through a collective in order to help pay athletes through NIL deals. This was the situation
all through up until this coming July where it's all going to change. So back when NIL started, you
(04:26):
had multiple years of, really, a budgetless team you could create. As long as there was a donor
rich enough to give money through a collective, you could use that money to fund your roster.
Liberty (04:36):
Pay the students?
Michael (04:37):
Pay the students, correct. Pay the athletes,
exactly right. So this creates disparity. You have the Ohio States and Michigans of the world
with budgets in the twenty-plus million dollar range for football and then in the
same conference Maryland and Rutgers are under five million, because of the donors
and the fan base wanting to support those programs or not support those programs.
Munther (04:58):
I wanted to push back a little bit and even though I've actually traditionally
have argued against the NCAA rules, thinking that athletes ought to be paid,
I do find it a little bit of a - there's tension in that process, and I'll give you an example and
I think it would be great for you to draw on some of these analogies. So in universities,
(05:23):
a place like MIT, we don't have athletic programs like - but we do have -
Liberty (05:28):
Hey, I played field hockey at MIT,
thank you. *laughter* We had some athletic programs, thank you very much.
Michael (05:36):
No revenue-generating athletics. *laughter*
Munther (05:40):
*laughter* Yes, thank you, you saved me. Of course. But I would say we
have a parallel in terms of entrepreneurship and innovation. Where students can actually
have the ideas to create companies and make a lot of money, but we have similar rules to
the NCAA rules where it says for example if you're working with your advisor on an idea,
(06:04):
then you can no longer stay with that advisor and be a student if you want to develop it
into a company. Which then have you make a choice between staying at the university
and finishing your PhD say, for example, or moving on to a startup company which we know,
you know, the probability of success, and we put the students in that difficult place,
(06:25):
and I feel it's a similar situation of saying ‘oh we're not comfortable paying our student students
and we want them to finish their education this is the training ground and then they can go
later and and get their money’ the way we thought about collegiate athletes. So there's a tension,
I think, in converting that whole process into money and losing this ability of the
(06:46):
training ground that was happening in the universities. Any thoughts about this?
Michael (06:50):
So yeah, that's interesting and so you see this - a good comparison to what you
just referenced is especially in football where it's really hard to play as a true freshman,
right? So you go to a school and you're a true freshman, you're not going to play but that
school is going to develop you into a better player, right? It's going to help you succeed.
Munther (07:08):
Right.
Michael (07:09):
But the problem is, that school will help you right? Your freshman year,
they're turning you into a good player. Now you become a good player, well now sophomore year,
the next school over in your same conference is going to offer you double the amount of
money to switch schools and go over there. So now the school that has just trained you,
that has helped you fine-tune your craft now doesn't get the benefit of that because they're
not willing to to pay a high-enough price. You are seeing that in college athletics all the time.
Munther (07:35):
And that's the downside for the payment, right? If there was no payment,
maybe you would have committed and stayed with the school that trained you, with the potential
of you maybe even getting more playtime, because there's more playtime in one place versus more
money in another place, but less playtime because all the talents are concentrated,
right? So this equity division and the training for more future, better players
(08:00):
now is impacted by decisions of how much money I can make now.
Michael (08:03):
Correct, and the old rules - you couldn't transfer and leave, right? If you did,
you'd have to sit out for a year. So you had to stay with your original school. Again,
some people think that that's fair, other people - more of the free market
people - ‘hey the student should be able to, you should be able to make as much as you can
make wherever you're going to make it and if you want to bounce around’ - I mean, we are already
(08:25):
seeing multiple athletes that have started for four years at four different universities.
Munther (08:30):
Wow, really.
Michael (08:32):
I mean this is happening - this is not just one or two cases right now,
right? We’re in dozens if not a hundred cases throughout all of sports in the
NCAA world. I don't have the exact number there, but it's it's more than a dozen of
people that have played at more than four universities in their career.
Munther (08:48):
So this actually makes it very difficult for coaches and school
managers to manage a team that's so dynamic, people are coming in and out all the time,
there's no real history of people playing together.
Michael (09:01):
I mean, it's a very interesting dynamic and there are some coaches that can't handle it
and that have just quit. Some of the best coaches in the game have literally, like ‘I can't handle
this and I quit’ because they're not used to it. I am on the other side of this *laughs*, because
I'm just a big player guy. You know, in the NFL, you have a brand new team every year.
Munther (09:22):
Oh wow.
Michael (09:22):
Right? You get new free agents coming in every year, you have less time to train with them.
If you want to keep the players, you can keep your players, just pay them fairly. And look,
if you have a player that's a freshman, he's going to stay to sophomore, these
aren't always bottom-line guys. You can usually get twenty to fifty percent discount to stay,
generally speaking. So there is a component of that, it depends on the player, it depends
(09:44):
on the character. We work with you know UVA, Alabama, Texas Tech. Shoot, Alabama can get
massive discounts in football because they want to play at Alabama, right? So players do take a
longer-term view. There are - ‘Hey look, I'll come to Alabama and I'll get X dollars now,
which maybe is only twenty percent of my market value, but I know that I'll make so much more
(10:07):
money in the NFL because Alabama will train me and make me this good player because they have
this history of success.’ So this stuff does happen, this isn't - I don't want to make this
a black and white, everyone's going for bottom dollar - are there players that will always go
for the bottom dollar? Absolutely. Especially now because they all have agents. So this is another
world - all the agents are getting into this world and what agents want is maximum money,
(10:28):
right, because that's how they get their cut. So you're get you're seeing more and more of this.
Munther (10:32):
I guess we have to train all these players in probability theory, right,
because they all have to be calculating risk analysis and expected reward.
Michael (10:41):
I would argue that the whole world should be trained in probability a lot more
than they are *laughs* and this is inherently the issue with athletes, right? Athletes are the kings
and queens of being irrationally confident. I always joked, I was the captain of the irrational
confidence all-stars. When you're a pitcher - I was a pitcher - there's nobody in the world
(11:02):
that could get this out better than I could. Now realistically, looking back, were there?
Absolutely. I was not the best pitcher in the world, I wasn't even close, but when you have
the ball in your hand or the basketball in your hand taking a shot - if it's you or Steph Curry,
you're picking you. Even though it's an illogical decision. Athletes believe in themselves,
every athlete in college thinks they're going to be a pro athlete - less than five percent will.
(11:25):
But it's a really hard thing to do to explain this to them. So they, what they a lot of them will
often times think is - ‘well I'm going to make it anyways so I'll just get the most amount I can
right now, because it doesn't matter where I'll go.’ That's where you get the support systems,
the advisers, the parents that can sometimes rein in these kids in and be like ‘Hey, no,’ like talk
(11:46):
a little more sense ‘You're a lot more likely if you're going to this school or this school.’
Liberty (11:51):
So, I just want to make sure that I
know where we are right now. You said everything changes in July.
Michael (11:56):
Yes.
Liberty (11:57):
What happens? Where's all the money, what’s happening with the transfer forward,
the NIL, the collective. What is the game plan moving forward?
Michael (12:05):
So now it's pretty simple. Now they've made the revenue share of each
school and the schools can decide how they want to split it. There is no more collective,
it is not an all-you-can-eat buffet anymore.
Liberty (12:15):
So, this is gone?
Michael (12:17):
Yes, so this is gone.
Liberty (12:18):
The booster thing is gone-zo.
Michael (12:20):
It’s essentially a salary cap, okay? So it's $20.5 million that the school can allocate
however they want if they want to allocate $20.5 million to field hockey to help Liberty,
they can! *laughter* If they want - most schools are breaking it down the same way,
a somewhat similar way which is 13 to 15 million of that 20 is going to football, three to four is
(12:41):
going to basketball and then the other sports are splitting the rest of it. Which makes sense from
a revenue-generating standpoint, there's obviously a lot of Title Nine things that have come out that
I'm not going to, I don't think we need to get into on this podcast, but all the conventional
wisdom and thinking is all that stuff's going to be over - it's going to be a separate system,
and all the schools that we work with - we work with five schools - are all operating under the
premise of that 13 to 15 million football, three to four million in basketball. Now,
(13:06):
real NIL is still allowed, this is where it's going to get super dicey, and that is that players
are still allowed to sign deals with Joe Smith’s car dealership at Tuscaloosa for Alabama players.
Now the problem is, these must be fair market deals and the NCAA has hired Deloitte to audit
these deals. So this is going to be, I mean, this is going to be in court a zillion times right? ]
Liberty (13:30):
This is like the Wild West.
Michael (13:31):
Exactly, exactly, but what it does is it makes it so from a legal standpoint you can't
just - ‘hey, I have a booster, I have a lot of money here's a car dealership, I'm gonna pay
this basketball player two million dollars for a commercial’ because if they do that, it's going
to go to the NCAA, it's going to get in court but while it's in court that player can't play.
So it does no good, and then the car dealership is gonna have to explain why this player's worth
(13:53):
two million dollars and will likely lose that. Now again, a $100,000 you probably can get away with,
right? So it's this whole ground of fair market value, but what this does - I'm going to switch
gears now to the data science part of this, as this is a data science podcast, and that
is with - now that all schools have relatively even playing fields, more even than it's ever
(14:15):
been in the history of college sports, now how do you effectively allocate that capital? So
now you have athletic directors - I love athletic directors - they're not skilled in how much player
value is across multiple sports, right? Even head coaches in college, this is not their background,
they don't - if you look at NFL - a professional program - they have next to no say of who comes
(14:37):
on their team and how much a player's worth. So nobody knows how to do this, nobody knows how
to value this, yet they have to allocate twenty million dollars to do this. Now here comes this
beautiful data science puzzle. Now you can get the AD (athletic director)’s back to say okay,
this running back transfer from South Dakota State is really worth X to my organization,
(15:01):
and now they can have that backing because what can happen if you spend that money improperly
is going to be a disaster. We saw it at Florida State last year, they paid the wrong quarterback
a couple million dollars and they had one of the worst seasons in school history, and now you're
going to lose donor dollars. Now you're going to lose likelihood to move conferences because your
(15:21):
football team was garbage last year. Now, Florida State did a really good job in this year's portal
but that's a different story. But the idea is it gives you cover, because think about this,
the coaches I worked with beforehand all the theory was ‘here's good player,
I want good player, sign good player’ and they've been doing that for thirty years or however long
they've been coaching. So now they'll come to me like ‘oh, here's a running back I love like this
(15:46):
guy's a great player’ and schools I work with, they're not allowed to tell me that anymore. Is
he worth twenty million? Well, of course not, so okay he's good enough for how much? There's
now a ‘for what price’ component to this and it's really hard to get that mindset from coaches away
from ‘this player's good’ to ‘this player's worth X’ and this is where data science can really help.
Munther (16:07):
Yeah, I can imagine this being a really interesting problem to solve,
but I want to clarify one detail about this and that is the fundraising on
the side does not go into this spot of twenty million to spend.
Michael (16:20):
Correct. So the school can allocate - the school can do whatever they want. The school
can say - if you're the University of Virginia, you can say of the school's money, I’m giving the
$2.5 million of rev share. They can also raise the $20.5 million of rev share so it doesn't come out
of the school's pocket, but the cap is $20.5, cannot be more than that, period end of story.
Munther (16:43):
So okay, and as a data science - so this is great, so it's a fixed pot for everybody. Some
people may not have it but most people will try to figure out how to raise it and spend it, and
you're saying this is done across multiple sports including the managers and the directors or not?
Michael (16:59):
No, only players. Coaches and everybody else are all exempt,
so they still live in total free market world when it comes to this.
Munther (17:08):
Got it. Okay, and what you said is the real problem is that not so much that this is
a good player but how much would this player cost me, because there's also an opportunity
cost. But this is something NBA management had to do for many years with the salary cap,
you know, and sort of the swapping and saying ‘Okay I can get two for
this price of one person’ and so forth. So is is there any knowhow in the data
(17:31):
science world from the NBA or has the NBA never been a data science organization?
Michael (17:36):
No, the NBA does a great job in data science. It is a different mathematical
problem slightly because there's a lot of different rules in the NBA. For example,
there's you know max contracts you can sign. ln college, there are no rules with that money,
so like LeBron James can make only the maximum contract allowed by the CBA.
Munther (17:55):
I see
Michael (17:56):
So he can't make a 100 billion dollars if somebody wants to pay him 100 billion
dollars. In college, if you wanted to, you can give all $20.5 million to one person.
Munther (18:03):
Interesting
Michael (18:04):
There are no rules and there's no minimum. So the other really cool
thing about NBA - or cool or not cool depending on how you look at it - NBA,
NFL, MLB - is they all have rookie-scale contracts so in baseball you could be the
best player in the game for your two full years Tim Lincecum won two Cy
Youngs and he got paid the major league minimum, the lowest amount of anybody,
because it's more of like a tenure game of how it's set up. This is total free market in college.
Munther (18:28):
It’s really more like a a coalition game, it's not so much you know -‘what is the
optimal coalition that I can have for a fixed amount of budget?’ which also has
externalities in the sense that I'm also taking that player away from another team.
Michael (18:43):
Yes
Munzer (18:44):
So it's not like I just got them for me but I also took
it from another potential competitor which could be also a plus for me,
so I have to calculate the externality factor of my data science calculation.
Michael (18:56):
No question about it, and I'll draw back and give you a real life example. We’ve been
doing this for three years in basketball. So we partnered with the Alabama basketball program to
help their roster building. Our first year we took over, we helped with the recruiting in basketball,
we have every shot, every decision every players's made from fifteen years old and on,
all skill level adjustments, all opponent base, so we know how how they're going to perform to a
(19:21):
tighter degree than the eye test right. So we put together this roster and it was the
first time in school history, in over 100 years, they were number one in in the country going to
the postseason tournament. A lot went right there. Issue - seven of eight starters gone
the next year, because when you have a good team you leave for the NBA, you transfer out,
all these things. We have next to no money - under about about a million dollars, a little
(19:44):
less than a million dollar in NIL. All these other schools, a lot bigger donors for basketball,
because remember, Alabama donors are giving it to football *laughs* The basketball program was
a lot less at that time, we couldn't even get, like - talk about getting a recuit from Florida
to help that in the conference? Absolutely not. We couldn't even get a mid-major. Our
starting lineup was Hofstra University, Cal State Fullerton, North Dakota State,
(20:10):
Ohio University and Wofford, but our models looked at the coalition of how we can build this roster
and the talents of each of those players, not individually, but how those talents and those
skills are going to react amongst each other to create the most points on offense minus the
most points on defense, and for the first time in school history they got to the Final Four.
Munther (20:29):
Wow.
Michael (20:29):
With a roster that had the lowest salary or one of the lowest salaries in the entire SEC.
So you can do this type of stuff. Now, it takes a lot of work, I mean, again, we've been building
these models in basketball five, six years, have incredibly intelligent people - always hiring by
the way if anybody's listening, I'm throwing a plug in Liberty, but if anybody's listening and
(20:51):
like sports and data analytics, please apply for a job here at BLA. The company that works for the
schools it's called SAA, Sports Analytics Advantage, and you know it's really hard to
be a data scientist at SAA because it's a really different skill set than your normal data science
folks, and that is because we really put a premium on creativity, and it's really hard to get that
(21:17):
logical algorithm-building brain paired with a super creative outside-the-box-thinking-brain.
In academia, in general, most of these kids are taught to work at Meta and Uber and Facebook,
big money jobs - as they should be, by the way, that's where the money is,
which is a lot more execution type of data scientists. ‘Here's this list of data,
(21:41):
make all these executions, execute this stuff, find out these really cool predictive measures,
etc’ but in sports it's harder, it's tiny sample size, it's missing data,
you need to figure out creative constructions and how to build not only your data sets,
clean your data sets, but then build modeling frameworks around these in different types of
ways and so it's a really cool thing, and I think we've done a a pretty good job so far.
Munther (22:05):
I think the issue in these coalition structures that you're looking for is defining
what is a success, right, because I used to be an NBA junkie, I watched every single game,
basketball junkie in general, watch games and you think you become an expert - you're not,
but you think you become an expert and so forth, and it's often the case that it's
(22:28):
not really the obvious numbers that make a team the winning team, but rather a little bit more
intangibles about players and dedication and camaraderie and giving the ball an extra pass
and that kind of stuff that makes that one extra point at the end, and it's very difficult for
algorithms to capture this kind of dynamics. So I think it's really interesting as a data
(22:51):
science problem. You can be successful but can you really develop confidence in your methods?
Michael (22:56):
Yeah, in the last ten years since we've been operational,
2,500 data scientists have tried our assessment, under a hundred have passed it
Munther (23:03):
Wow.
Liberty (23:04):
I want to try to pass it, send it to me. *laughter*
Michael (23:11):
But we want there to be more, so I'd love to - you know, usually the people that pass
it aren't necessarily the highest GPA people, it's more of the people that - ‘On the side,
I made a fantasy football team’ or ‘I watched basketball all the time and I wondered why your
team's not fouling at this, so I built this model to determine this.’ It's like those types are,
(23:31):
generally speaking, the ones that are able to get through and pass our assessment.
Liberty (23:36):
So I just want to step back a little bit, because you're talking about building
these models and how much data you have on all these kids. How do you get all this play
around high school kids? If you're trying to recruit seniors in high school to come on as
freshman to the team, how do you do that, how do you account for them in your model?
Michael (23:54):
So it's different in all the sports. Basketball is what I was referring to with every
shot and every decision they made, 15 years old and on. The reason is because they have
this really cool system called the AAU system. So if you map out all the good players, play all the
good players, right, from fifteen years old and on. We also have about ten to fifteen of these
high schools that are athletic high schools - IMG Montverde, the Brewster Academies, like all these
(24:19):
types of schools and we will model those out, but we only count the games where we have opponent
adjustments. Obviously, we don't care if you score sixty points against somebody that we have no idea
about. So everything you do must have an opponent adjustment to get the correct skill levels,
and what we're looking for there is progression through the years. So we know different skill
(24:39):
sets develop with different players at different rates and different times so we can then project
what are you going to be at 18, 19, 20 years old. So one interesting example, if there's any - I
think we got a lot more data scientists than basketball players - but for ten seconds if you're
a basketball player out there and you're fifteen, finishing at the rim over elite rim protection
is a really good skill to have at an early age. Being crafty around the rim. Shooting doesn't
(25:03):
really matter as much at age fifteen. When you get to eighteen or nineteen, shooting is
the most important skill, right, like be able to shoot accurately from a distance, but each skill
develops again over time, and so when you look through and you can find these correlations you
can then predict what they're going to be. Brandon Miller was the number one player we thought in the
class coming into Alabama, he was a twenty-seven percent three-point shooter we projected him out
(25:27):
at like a thirty-seven percent shooter at Alabama and he ended up shooting close to forty percent
from three, was the first college player taken in the draft. Why? Because his free throw percentage,
his floater percentage, his two-point jump shot, all these other different percentages against
different type - we knew he had touch, he just hasn't developed a three-point shot quite yet.
Munther (25:44):
No it's interesting because let's say, let's imagine that BLA, I guess your company,
is the only company out there that does this - just bear with me for a second - and so then
what happens is that you would have these methodologies for selecting the players
and ranking them and so forth, and you effectively you are changing the game,
(26:08):
because your method assumes a certain type of a game that these people will be successful at,
you're selecting the players for that game, you would be changing the game.
And over time I think this impact will be felt by what we see as a top player.
Michael (26:25):
I appreciate that. I mean, tooting our own company horn here, we worked with four schools
this year - Alabama, Duke, Michigan and Texas Tech. Last year Michigan won six games, Texas Tech
nowhere close, right, and now we have all four in the top 12 in KenPom ratings right now, top 12 in
the country. I mean this is what can happen when - and it really just shows too how bad the market
(26:49):
is, how bad eye test can be, and it's hard it's - eye test is a little bit easier in the NBA,
there's thirty teams. In basketball, there's 363 teams. How am I supposed tom as a coach,
watch 363 teams play every game and and then try to figure out the starter at Eastern Kentucky
getting twenty a game how he correlates to the bench player at the University of Florida? How
(27:09):
do you do that? It’s just so hard with your eyes versus - again, NBA when everybody's professional
at the same level, I would still argue the data is far more important than the eyes,
but the eyes are a lot better because it's a lot of like-to-like comparisons.
Munther (27:23):
Sometimes in class I use this example for my students, from Red Auerbach,
from the old Celtics days, I think it was an interview with him and he presented this scenario,
he says ‘Your best players are playing and then they're losing they're not doing very well,
so you take them out and you put the second team in and the second team catches up and
(27:45):
now you're tied for the last shot. Do you stay with the second team or you bring back the the
first players?’ and I always present that as a statistical test for my students,
and that is the question of - how do you rationalize this problem from a statistical
point of view? Because when you have such a terrible day for these players and the
(28:07):
other team has brought you all the way to a tie game, and he was saying you bring your starters.
Michael (28:14):
I mean, I would - we have a lot of data now, we have heart rate monitors,
we know who's tired, we know how long everybody's been on the court. We can optimize for all these
things of who we think, there’s a lot more we have now than Red - I mean, you think,
Red Auerbach was just so far ahead of his time, a true legend of the game, and that's what you see
too, these coaches that were legend. We had the fortunate - we worked with Coach K, our
(28:37):
first two jobs were at Duke, were Coach K's last two years, and he was playing penetrate, kick -
Liberty (28:42):
Even I know who Coach K is.
Michael (28:43):
There you go. He was playing penetrate,
get in the paint, kick, shoot threes before anybody understood what that was.
Munther (28:51):
Yes.
Michael (28:51):
Right? And if you think about basketball it still annoys me,
it's - it took ’til the early 2000’s to realize three points was more than two.
Munther (28:59):
Yes, yeah yeah.
Michael (28:59):
Like, think about that. *laughs* It is a wild - you think
we're so advanced with all this stuff and then you realize - you think about
that it's - Sports is just behind on the analytical side than other aspects of life,
but when there's an opportunity there's something to be done about it.
Munther (29:20):
I think a lot of coaches and players didn't believe that three-point players can
actually hit with that percentage, so I think that's the other thing is that I think right now
I believe if you create a four-point line, we're going to have four-point players, and they'll
be hitting percentages. I mean, it's a skill, as you said, that when you're older you develop it,
(29:41):
and you play it and becomes a thing. So that's really interesting how I meant by changing the
game. The game changed with a three-point shot, it's a entirely different game because
of the percentages and the efficiency of a three-point shot versus two. So now big men
are not as as important as shooters, right, and the value - well they are, but it's a different -
Michael (30:03):
Yeah, I’m going to push back on that and this is where I think this is an
interesting point is everybody thinks the NBA is just running down shooting threes - there are so
many ways to build championship teams, so many ways. Especially in college,
because the paint's smaller, all the different rules. A good - analytically a great, elite big
man is actually better than a Steph Curry type player, because you're going to draw double teams,
(30:28):
and you're going to get open threes because of that. If you're a big man that can score
at 60 percent on one-on-one situations, you're the most lethal player in the country.
Munther (30:37):
Right, right.
Michael (30:38):
Because you're going to force double teams, and that's how you get open threes,
you don't have to dribble, drive, kick and do all that right?
Munther (30:45):
Yes.
Michael (30:45):
So, there's so many - and if you look at the four rosters we have, Michigan starts two
seven-footers and they play almost all inside. Alabama spaces it, runs shoots, more threes
than anybody in the country. I mean there's four completely different styles of play that we have,
and that's what I love about it. You get information from the coach. Gere's the team I want
(31:06):
to build and then you simply, you optimize the coach's brain. This is where I think a lot of data
science gets wrong, they say look I - you know, I think data science sometimes have too big of an
ego, I'm gonna say it. And I think that sometimes they go in and they'll say - here's the rules of
the game this is the most effective way to play, do this. Well, first of all, even if you're right,
coaches aren't going to listen to you because nobody has bigger egos than coaches, and second of
(31:31):
all, if you go to the coach and you say ‘this is what we do’, we go - ‘look, we know you know more
than we do, we know that. That's the premise we're going to work off of. You are smarter, you've been
around the game, how do you think about the game, how do you think about winning?’ and our job is to
build a model - because you have one fatal flaw as a coach, you got two eyes and one brain. Nothing
(31:52):
you can do about that, you can't watch 363 teams play every game. So we're going to build a model,
we're only going to be eighty percent as good as you - we hope! Like if you watched one game
and isolated every little thing, you're going to be better than our computer,
but if we can be eighty percent as as good as you, watching every game simultaneously,
grading every play and grading every player and grading every skill set, and now we know what
(32:13):
you're looking for, we're going to create you a roster that can be dynamic, and it's a roster
that you know how to coach because the players that you have. I mean Nate Oats at Alabama,
unbelievable coach, one of the best coaches in college basketball, You put Zack Edey on his
team who's a seven-foot-six monster post player, one of the best players in college basketball,
it's not going to work. He doesn't know how to - like, that's not how he coaches, that style.
Munther (32:35):
Yes, yes.
Michael (32:36):
So you have to match the players with the
coaching philosophy. That's a really cool part of the puzzle.
Munther (32:42):
That's really interesting, very interesting.
Liberty (32:45):
So I just want to make sure that I get this, So you're you're doing two things,
it sounds like. It’s not just building the roster, you also - it sounds like
you do realtime decision making too, when you said you have heart rate monitors,
you know how tired they are. So it's sort of like a two-part thing of - you build this
roster but then you're also deciding at any given moment in the game who's on the court.
Michael (33:07):
We do whatever the coaches want us to do. So there are some schools that want
us - in football to figure out what plays are going to work against what down-and-distances.
There other schools that are like ‘just build the roster, we're good.’ I mean it - just
however we can be of help to the university, we want to do that. We want to help win games.
Munther (33:25):
Are you allowed to make realtime decisions based on
computational capability. Is that - like an an AI assisted tool?
Michael (33:33):
Absolutely you're allowed to do that, there's different ways to do it. There are certain
things that you can and - there are different rules and things that you can and cannot do in
there. But you can, for example, build out every situation of a football game before the game
happens, and know that this defensive coordinator is going to be in cover two on third and four to
third and six 79.3 percent of the time, and the other time he's going to be in cover three shell
(33:58):
and now you can call a play that can beat both those coverages. So now you can go ahead and be,
like, when we're in these situations these are the four plays we can call.
Munther (34:07):
Got it
Michael (34:08):
Because we know the tendencies of everybody, so you know it in advance.
Munther (34:11):
You prepared the scenarios and then you react to the scenarios in real time.
Michael (34:15):
Correct, and there's also different real-time things that you can do. We have lineup
optimizers in there, so when a different team puts out a different lineup - what's our lineup
that's best to combat that, net expected points, there's all kinds of stuff that you're able to do.
Munzer (34:31):
And the biometrics are allowed to be used in the game,
like the NBA, I thought it's not possible?
Michael (34:36):
So biometrics are not. So biometrics - we get the data in and we know different players
playtime loads. So for example, one player gets - when he comes in the game does really
well until about six minutes and then after six minutes his play starts to go down. So we
knew that beforehand, so that's how we're using the hard data - so to your example,
to the Red Auerbach example, if those second teamers were right at their peak and about
(34:58):
to play great, I'd keep them in. If they're about to fall off a cliff,
then I'd bring in the starters. We would have that data and that information, but we would
know that ahead of time. This is not like a real time - he's at this level right now, correct.
Munther (35:10):
Yeah, yeah.
Michael (35:10):
I mean, we have the data you're just not allowed to use it. I should make that clear.
Liberty (35:14):
So how does it work, it sounds like what you're saying is that there's a huge
amount of - there's obviously a huge amount of data science that goes into this - but
when you're talking about matching up the data with the coach's ethos or how they know how to
coach or what they like to do or whatever, there's a lot of human decision-making in
here. So can you give me an example of one of the biggest mistakes, where the data is saying
(35:36):
one thing and it just blew up in your face, or somebody else if you guys haven't had one.
Michael. Oh yes, yeah I'll give you a big miss that we've had. So in basketball there was a
player, not ranked in top 100, we thought he was like a top 15 player in the class. Reason is,
he was a massively high IQ player on the computer. Super smart,
meaning when he would dribble in the lane, players would come and double team, he would always make
(35:59):
the right decision. This is exactly what you want to build - it's really hard to teach IQ,
and our computers have been unbelievable at generating player IQ, like we do - we build
NBA draft models for NBA teams. Tyrese Haliburton was like one of the best players in the class,
he waits late, like this was a great player, IQ index, all these things. Turns out we got this
(36:19):
guy completely wrong because of the play style, they basically ran one play. He was a left-handed
player - he'd get the ball on his right wing, he'd get the ball in his left hand. He knew
there was one read for him to make and he made that one read right every time. Well, when you're
playing a flow-style offense and there's 25 different reads to be making on a given play,
he was lost and was always making the wrong decision, so many turnovers. So here we thought
(36:44):
this was this great IQ player but we didn't - at the time, this was three or four years ago,
now obviously we've refined our model since then but - you learn from this stuff,
right? And so you learn - what are the types of decisions now is he making? It's not percentage
of correct decisions. It's the different types of decisions that are more relevant.
Munther (37:05):
So I want to get into the mentality of a player going to college and trying to
understand now all these factors that I have to be thinking about. So I'm a decent basketball player,
maybe division one player, maybe I'm good, I don't know. Maybe I had this
illusion that I'm going to be an NBA player, I wanted to go to a good program and train,
(37:26):
but now I'm thinking okay let's be reasonable if I go to a program that pays me and - maybe a program
that I get developed - but maybe if there's more opportunity for a deal that I signed with a shoe
store in the neighborhood or something. So now my decision is much more complicated. Before,
I wanted to be trained, maybe I'll get my best chance to be an NBA player. Now I'm going to
(37:49):
try to maximize my income in the meanwhile, in case things fail and I compromise that. From
the players perspective, who's helping them? I mean as a parent, I don't think I can help them.
Michael (38:01):
Well, you could as a parent, most parents can't *laugh* If you had the right date,
I promise you could help. But the idea - I get this question from parents all the time.
Munther (38:13):
Okay
Michael (38:13):
‘Here's my son, what do I do? He's getting offers from these four schools - this
school's offering us fifty grand, this school's offering us 200,000, this school's this,
this school's that, this school is guaranteeing me playing time, this school's not.’ I mean
there's so many factors as you say going in and by the way, you're seventeen years old. I mean,
I don't know what you were like at seventeen years old, but I probably wasn't making the most logical
(38:37):
decisions at the time. But the idea is, how I help parents is I model it all out for them. And so
what our company was based on before we got into this is projecting future earnings of athletes.
We lucked into this whole thing by the way, like this is a whole separate side story. We have a
whole company Big League Advantage which invests in future earnings of athletes. We project future
(38:59):
earnings and make investments based around that. So now, once the NIL stuff started to happen, we
were the only company around that had valuations for players. By luck, right? And so what I do is
I say - look, okay here's parent, here's the four schools, here's the head coach, here's the systems
(39:20):
that they run, here's how your kid is going to fit in these systems.’ We can build a simulation
model, we say ‘here's his general statistics if he goes to all these types of places’ and we say
the value for going to each of those schools, then you overlay the dollars. So you say, okay,
at this school I'm going to average 13.8 points, 14 rebound, whatever, whatever, whatever. For this
school I'm going to be played totally different and my future earnings at the end of that year is
(39:44):
XYZ, but now - so let's say your future earnings is eleven million dollars after year one in School
A, twelve million after School B, but School A is offering you two million, School B is offering
you $100,000, I'm going School A, even though it's not as good of a school for me technically,
the math still works. Now if School A was only five million versus twelve million,
(40:07):
I’m making a different decision. So we will do these analyses for players and their families.
Munther (40:13):
And these commitments that the universities make,
they are obligated to stick to it or can they change their minds?
Michael (40:19):
No, obligated but only one year. That's the problem. All these contracts
are one year deals, that's where you're seeing all this crazy action. I think this
is going to change. I think you're going to see contracts like coaching contracts,
where it's three year deals and you get a buyout in there. So it's like $100,000,
$300,000, $700,000, but after the first year if you're $100,000 now you're going
to $300,000 but you're the best player in the country and you're worth three million,
(40:42):
great you can transfer but you have to pay a million dollar buyout to the university.
Munther (40:46):
Wow.
Michael (40:47):
So now the school that wants you has to pay these buyouts, that's how I
think it should go, because this means that players will stay at the same school longer,
for the player they're locked in, a million dollars over three years,
and the school is now committed, they don't have to think about and by the way, if he leaves for
a million that's an extra million dollars of revenue, you can go get some more players.
Liberty (41:06):
Is anyone else making money off of this besides the players? As we said this,
is a little bit of the Wild West of what's going to be happening now,
in terms of the car dealership being able to give the kids money. Is there - it always feels
to me when something this new with this much money is starting that there's going to be
(41:27):
some funny business or some accidental funny business and things are going to be in court,
and I mean this is a lot of money, these are players lives, these are colleges that rely so
much on this money and this play. Where do you see the pitfalls of this coming?
Michael (41:42):
Oh there's a lot, but this is not like a new thing that - like the way you described it,
I would say like instead of the school making fifty million, now the school makes thirty and
the players make twenty. So it's not like oh, you know, it's a little bit, it's the definition of
revenue sharing, right? They're going to have to share this revenue with the players. Now of
(42:02):
course, all the schools are going to call broke on, poor on this, ‘now we have so much less money’
but the pitfall is going to happen with - the bag droppers are going to come back. When I say
bag droppers, I mean this how it used to be when those athletes were allowed to get paid - ‘Hey,
come here and we'll give you $50,000 in a brown paper bag.’ Now dentifying the bag droppers is
(42:26):
going be impossible. You used to be able to do it because all of a sudden the player driving
around a Ferrari you're like, well that doesn't make sense, right? Well now, if you're getting
a million dollars from the rev share and $500,000 under the table and now you're driving a Ferrari,
nobody's gonna know that. Now where I'm a little bit optimistic on the bag dropping is I think
(42:49):
donors are going to be a lot less, boosters a lot less likely to do that. Back in the day, you feel
bad for a player, you're the university, this guy's dominating, you're the star quarterback,
here's a few hundred thousand bucks. Morally, maybe that's more okay for a booster, trying
to get inside their head. ‘Now this player's making a million dollars, I gotta give them
an extra $300,000? Your guy's already making a million bucks, I’m not willing to do that’ - I
(43:15):
think is where you're gonna see it. Now will some people say, that are billionaires that ‘I don't
care here's money’? Probably, right? Take that risk, that's the big pitfall in this whole system.
Anytime you don't have a full, 100% free market, there's always going to be people getting paid
in different ways. You're going to see the NIL side of it, you're going to see here's a booster
(43:40):
in the hedge fund land that you can't give a true NIL deal to, so hey I'll go pay Microsoft
a million dollars, now Microsoft gives my player a million dollar NIL deal to go on a commercial. So
that indirect route of real NIL, you might start seeing some of that you know in there as well.
Munther (44:00):
The free markets - I am a big believer of free markets - but when they optimize the
long term, not the short term, right? So you can allow free markets in the short term and
we end up in monopolies and concentration of wealth and all of that stuff with free markets,
I mean it's tough, but with the long term and understanding it's putting on the right
(44:21):
objective. But maybe there was a different world, Michael, maybe we didn't think about a different
world. So let me propose a different world where there was no money in collegiate sports,
no money. The universities don't make money, the players don't make money. It's a training ground.
Nobody makes money on my courses, very small amount of money but we train these students in
(44:42):
statistics and and decision theory and all of that stuff and then they go to the real
world and then there is where there's the money. Why is that model not a good model?
Michael (44:52):
I love that model. That's what we were doing in 1930, when the NCAA started but now what
happened is that everyone started to watch and got addicted to these sports. So then,
people wanted to put it on TV. So then, people wanted to pay to put money on TV. So now the
(45:14):
schools are making a billion dollars. So now you could - this is the NCAA’s point - hey let's
just train them up, let's never pay them anything. This money - they never got paid anything in 1930.
Munther (45:23):
Yeah, that's not a good argument right, because someone is making money.
Michael (45:26):
Right, but I mean that's the issue with popularity,
and anytime someone's making money everybody wants a piece.
Munther (45:35):
Yeah, I agree.
Michael (45:36):
Now here's what's great about why it makes money. It’s an amazing thing,
because football is the golden goose. Why can UVA have a field hockey team,
because of football. This is a massive thing that's been incredible - the boom of women's
athletics can be directly attributed to the money going into football, and this
(45:58):
is amazing. As a father of a daughter right that now loves sports and goes to Caitlin Clark games,
that doesn't exist without football, if nobody got paid anything those sports wouldn't exist.
Munther (46:10):
Yeah.
Michael (46:11):
So, you know, that's how I would look at, I mean, I think it's a good thing and I - now
look, it doesn't mean it's an easy thing. This is incredibly complex and there's - every solution
is going to have detractors and downsides from economic theory to politics. By the way,
you're going to see Ted Cruz and Cory Booker duke this one out in terms of how this is coming,
(46:32):
be prepared for that. I would bet any amount of money within the next few months
Trump chimes in on this, this a real thing here. So, we're gonna see how it plays out.
Liberty (46:42):
I have one final question. So you guys basically built for Alabama this team that went
to the Final Four with the least amount of money, you named the roster. Is there ever a time where
one player is worth four million - a majority of the money - 25 to 50 percent of the money?
Michael (47:04):
25 percent, absolutely. 50 percent? It gets into externality - external stuff. So this
is what's happening with BYU. The number one player in the country AJ Dybantsa decided to
go to BYU for next year. Reportedly getting about eight million dollars. Now remember,
this is the last year that you can go crazy. But the idea for a school like BYU and pay one
(47:30):
player all the money essentially is that now if I can get a first player pick in the draft, I'll
for the next five years get a lot better recruits coming in. So the theory isn’t I'm getting him for
this year to build the best roster, the theory is longterm if I get one or two draft picks,
now the next best player in high school will be like look I can develop draft picks is the
(47:52):
concept. If you're looking purely mathematical - of all sports happen in just one year next year,
no player is worth half the portfolio in basketball. In football, the best
quarterback in the country will be worth about a third, which a third of $15 million is no joke.
Liberty (48:13):
Well Michael, thank you so much I thought this was fascinating and I know we're all going
to be so excited to see how your teams do next year, so we hope you're going to come back and
let us know the the successes and the pitfalls of your first year working in this new world.
Michael (48:31):
I appreciate it, thank you so much for having me
Munther (48:34):
Thank you so much, it is really informative. I wasn't really aware of all
of these things happening in the collegiate sports and this is - a lot is going on, right,
I mean payment decisions, fundraising, ability to do external contracts and
(48:57):
I don't think parents and kids are ready for all of this and I mean this
was an eye opener to say the least. I don't know what you felt Liberty.
Liberty (49:06):
You know, it's one thing when you think about basically kids going into the major leagues
for whatever sport, and they talk about how so many of these athletes lose all their money
because they don't know what they're doing. I mean, think about a seventeen year old, I mean
you give most seventeen year old boys $200,000, I don't know if they're going to survive it. I mean
(49:27):
you also have that like, just - these kids with all this money and a lot of times parents - I mean
I wouldn't know how to like advise a kid on this but you know, parents that just - I mean this is
just, it just feels like it's going to be the Wild West and I think it's all a little scary.
Munther (49:42):
You know, when my son was trying to - I mean, he was a division three basketball player
and we - like, there's no money in division three and yet we were struggling because the
balance there was more about playing basketball and getting a good education. And one thing to
make sure that we are in both, the good education is the is the formula for success in the future,
(50:07):
and so you have to make these decisions, but they were a lot simpler I would say than thinking about
these big contracts and so forth so yeah I think we have some people that do think about sports
analytics at MIT, and I'm going to throw at them this new world that I learned today and see where
(50:28):
maybe we can take it a little bit further in our students thinking about these issues.
Liberty (50:31):
Yeah, and I think especially given that you have these kids that might be making
two million a year for four years who will have no prospect making anything
like that afterwards. It’s like this is their only moment, these four years to
make this money or $800,000 which could be life-changing for one of these kids,
but that's it. When they get out of college they're going to be making $75,000 a year
(50:55):
doing something, so it's like how do you help them maximize that and not waste that.
Munther (51:01):
Very interesting.
Liberty (51:05):
Thank you for listening to this month's episode of Data Nation from the MIT
Institute for Data, Systems, and Society. You can learn more about IDSS and listen
to previous episodes at our website idss mit.edu or wherever you get your podcasts,
or @libertyvittert or @muntherdahleh. Thank you to our producers Tina Tobey Mack and assistant
(51:30):
producers Ben Stull and Maria Brooks. Don't forget to leave us a review and follow our
Twitter @mitidss to stay informed. Thank you again for listening to MIT’s Data Nation.
Popular Podcasts
On Purpose with Jay Shetty
I’m Jay Shetty host of On Purpose the worlds #1 Mental Health podcast and I’m so grateful you found us. I started this podcast 5 years ago to invite you into conversations and workshops that are designed to help make you happier, healthier and more healed. I believe that when you (yes you) feel seen, heard and understood you’re able to deal with relationship struggles, work challenges and life’s ups and downs with more ease and grace. I interview experts, celebrities, thought leaders and athletes so that we can grow our mindset, build better habits and uncover a side of them we’ve never seen before. New episodes every Monday and Friday. Your support means the world to me and I don’t take it for granted — click the follow button and leave a review to help us spread the love with On Purpose. I can’t wait for you to listen to your first or 500th episode!
The Breakfast Club
The World's Most Dangerous Morning Show, The Breakfast Club, With DJ Envy And Charlamagne Tha God!
The Joe Rogan Experience
The official podcast of comedian Joe Rogan.