Subscribe
Share
Share
Embed
Chasing the Game: Youth Soccer in America is a weekly podcast for soccer parents, coaches, and players who want to understand how youth soccer development really works in the United States.
Hosted by two dads, filmmaker Liron Unreich and investor Matt Tartaglia, the show covers everything from grassroots soccer to elite pathways like MLS NEXT and ECNL. Combining data, real experience, and expert insights from academy directors, college coaches, and former pros, each episode explains what families truly need to know.
Weekly episodes focus on the core aspects of youth soccer: player development, coaching culture, college recruiting, tryouts, travel costs, and the challenges of parenting in youth sports in today’s competitive environment.
For families navigating youth soccer’s complex system, Chasing the Game offers practical advice, credible voices, and relatable stories from two dads working to make sense of American player development, one episode at a time.
So before we start,
those of you on audio
don’t get the privilege,
but those on video get to see that Matt
and I are wearing the same shirt from
a team called Cambuur.
Why are we doing that,
Matt? Did we get hired by
the Cambuur coaching staff?
We unfortunately have not,
nor have we been approached.
But Ditmer de Jong, who works within
the academy system and is a scout
for the first team,
has long been a fan and
supporter of the pod.
He sent us some merch.
And Ditmer, we see you.
And appreciate you sending
us the merch. Love it.
Let’s get started with
the real meat and potatoes
of why you’re all here. Youth soccer.
We look at our kids and you go,
Okay, this kid has a great first touch.
This kid has a great first touch.
They’re doing the same thing in training,
but they’re not developing the same.
I want the coach to work on this kid’s
speed. I want the coach to work on this
kid’s spacing. Wouldn’t it be amazing
if there was some sort of bionic
brain that could help you decipher
what kid needs to work on at what time?
Yeah, it’s amazing. I would even add
to that IDPs for clubs that do them
consistently or do them well.
What else can we leverage that with,
context, to ultimately be able to measure
our kids’ development and
invest in it in the right way?
From time, resources, and
everything else that comes with it.
Our guest today on Chasing
the Game is Amir Donath,
co-founder and CEO of Game1.ai.
He grew up playing soccer in Switzerland
and began building the company while
earning his MBA at Stanford University.
He and his brother Yaniv are using player
data collected over ten years to help
coaches understand what a
child could work on next.
Yeah, and the magic word is AI, right?
Can it bring objectivity to an incredibly
subjective world? And if that’s the case,
no one’s going to want to hear my and
your opinion on some random kids running
on the field. What are we
going to do with our day, Matt?
Certainly not mine, I can promise you that.
But yours is still worth something.
Well, this interview sure helped me
change my mind on a lot of things.
And you’re going to hear why.
This is Chasing the Game,
prompting into the future.
Liron: Amir, welcome to
Chasing the Game, my friend.
It's really nice to have you here.
I'm gonna go straight to the chase, and I've
prepared this prompt all night.
So, here we go.
Look, we're parents, right?
Matt and I, and we wanna impress our
kids, we use AI for Roman history,
for weather, for movies, whatever.
But all of a sudden, when we hear that
it has to do with schools, we freak out,
and, God, now it has to do with soccer?
I'm not sure about this.
So why should a club or a family use
AI to forecast anything about their
kid, and where do we draw the line?
Amir: Yeah, I think it's
a very legitimate concern.
You know, we all have
big questions about AI.
In our case, AI just means it's a
technology that has learned over time,
you know, what works, what doesn't
work, and what helps kids the most.
So I think there's obviously different
ways we can talk about the term
AI, but essentially what AI is, in
our case, it's an algorithm that
has learned from thousands and thousands
of examples of young players that have
gone to the next level of the game.
So in the end of the day, you know, it's
AI for the purpose of developing players.
So there's nothing crazy
happening in the background
that, you know, is not
for the best of the kids.
I think that's where we try to educate a lot
and make sure that people
understand there's no evil AI
behind this or anything like that.
It's really just to help the kids.
Matt: You and Yaniv, who's
your partner in this, and your
brother grew up playing soccer.
Name one thing that Game1 measures or
deliberately doesn't measure that you
think is really important as part of this?
Amir: One thing it
measures is biological age.
So it's not about, you know, two
players could be 14 years old, on
their passport, but biologically
one could be 16, one could be 12.
And so we're comparing apples to apples.
So if you're a late developer, your stats,
the things that we test, the things that
we measure, are going to be compared to
other players that were at your stage,
and not just the same passport age.
So I think that's one really,
really important part.
There's a lot of things we don't measure.
I think there's a lot of things we
can't measure, but everything that we
measure has a purpose why we measure it.
There is nothing in there that we
measure for the sake of measuring it.
It's all because we have a lot of data
on exactly those things, and have seen
that these are an important part of,
you know, of young players' development.
Liron: But here's the thing.
Help us understand a little
bit better because I saw it on
your website. It says, and I'm gonna
quote from it, it says, "A forecast
informs and people decide." So you
have, just to help parents understand
the structure, you have a questionnaire
that the child fills, I guess, and maybe
you could talk a little bit about that.
But what are really the growth
estimates and forecasts?
How does that actually work?
And then if you present that to
a club, can a coach disagree with
your forecast or your analysis?
Amir: First of all, you know, always,
like a lot of these is to support any
decisions, to support any development.
It's never 100% this or 100% that, right?
It's always gonna be if
we speak about an actual
forecast, it's a probability.
And then one thing we always emphasize is
one test is good, two tests are better.
So one of the things that the
algorithm takes into consideration
very, very strongly is how you
develop from one test to the next one.
So nothing is taken as a given.
At a test day, any kid
can have a bad day, right?
So there's so many things
that come into consideration.
You had a, you know, rough day at
school, you were tired at the evening
that, you know, you did the testing.
So there's a lot of elements in there
where it's not like one test and then, you
know, the holy truth from there on.
It's really about development.
It's a lot of things about how you improved
over time, especially when we
talk about the biological age.
You know, suddenly you're in that
growth window, which means a lot
of times, you know, your feet aren't
really doing what you want anymore for
a few months because you grew a lot.
How are you expected to develop
now as a 14-year-old where
suddenly you're growing a lot?
You're probably gonna get faster, but
maybe ball control is gonna be a bit
more difficult the next few months.
That's all expected, so we just give
a way to benchmark that to get a sense
of, yes, for the next six months, maybe
your 10-meter sprint should improve
by 0.1 seconds, but you might have
a harder time with the ball for you.
So it's just about getting
insights on those things, right?
It's just we're opening up the black
box and say, "There's so many things
that are gonna happen in the
development over that time. Let's
take a look at what's expected, what's
realistic, how much could you improve
that, what's gonna be challenging."
Liron: But how long do you
follow, let's say, a child for
before you feel you have a good
analysis or forecast for that kid?
Amir: We generally recommend starting
testing at around 10 or 11 years old.
But then, we strongly recommend taking
at least two tests, meaning at least
three to six months apart before
taking really any sort of action on it.
So that's generally how we recommend it.
The very simple reason for that is, all
of the historic testing data that we
have gathered that is dating back up
to 10 years from a lot of many European
academies is for that age group.
And then on average, we have between
four and five tests per player.
So it's really about, you know, get
started with the testing and then
over time, more and more, you'll
get more and more insights on
any sort of future trajectory.
Liron: But when you say European teams,
the data set is something you guys
built, or this is something you've worked
with clubs that have given you a data
set to build your libraries?
Amir: Yeah, it's a fantastic
question because it's kind of like
the backbone of what we do.
The answer is both.
So there's a lot of European clubs
and academies that have been doing
standardized tests, you know, for up to
10 years back, some of them even more.
And we went from club to club,
worked together with them, and
asked them to share the data with
us in order to build that algorithm,
that backbone of what we do.
So it's a proprietary data set that took
us quite a bit of time to put together.
We also understood that for the clubs
themselves would be no incentive
to share that with each other.
So it needs a third party
like us to come in and say,
"We'll collect the data from all of them.
We're not gonna share it with anyone else.
We're just using that to build a
technology that could help everyone."
So this, it's a data set, again,
dating back up to 10 years, and,
you know, from young players, so a
lot of those we know nowadays, did
they go on to play professionally?
Did they stop playing?
So the technology's built on outcomes.
Matt: You mentioned in July that
Game1's top five usually overlaps with
the coaches on three or four players,
leaving one or two the coaches not picked.
Can you walk us through
one of those disagreements?
Amir: I think I would call it more, a
surprise sometimes than a disagreement.
You know, like it's, let's say
your typical team of, let's say 20,
25 players, the top five that the
algorithm flags as being the top
five, let's say the number one or two
might be the kid that just dribbles
past everyone, scores all the goals.
Like you don't need any sort of
technology to know, you know, that
kid probably has something.
I would say the only exception
to that would be if it's a very,
very early developer and they're just
scoring all the goals because they're
much, you know, much older
biologically than all the other kids.
But essentially how that looks like is,
you know, it's great that we're able to
confirm a lot of the hypotheses of the
coaches already so that, you know, that
builds trust, that creates buy-in.
They're seeing, okay, the
technology is not just, you know,
coming up with random things.
So it's a good balance then to
have those one or two other players
that suddenly, you know, it could
be maybe it's a number six or a
fullback, oftentimes positions
where it's more harder to pin down
what actually makes a good player.
And then maybe even on top of
that, they're a late developer.
It's always an interesting discussion
where they say, "Oh, yes, this guy, I
knew he's gonna be on the top of it.
Technology confirmed it.
Great." And then like, "Oh, interesting.
This guy, he wasn't really on my
radar." And then we have seen, you
know, behavioral changes sometimes. So
a coach was saying, you know, "I didn't
know this kid has high potential.
I'll give him more playing time 'cause
he needs that in order to develop,
obviously." But having that
different look on a player, can help
and also the other way around, right?
It's not like some players will
be flagged as high potential and
everyone else is left behind.
We're always showing where could this
player be in six months from now.
Matt: I was just gonna say, I
mean, we use the word development
like, you know, 400,000 times
a day, across youth soccer.
But then when it actually comes to
development and the investment in players,
I think you would be
hard-pressed to find parents
who have a great level of confidence
that development is really happening.
That's not universal, but I think
that that's a large swath, and what
you're talking about is actually
development, you know, as a part of
the development cycle for each kid.
Amir: It's exactly that.
It's, you know, if I take two or
three steps back, it is one of the
reasons we started this company.
When I was a young player, I remember
our coach would always say, "Okay, now
everyone does sprinting, or everyone
works on their first touch." And
I was always wondering, well, this
other kid has a great first touch.
Mine is not.
Why are we all doing the same thing?
Or on the flip side, when I was one of
the faster players, I was like, "Why are
we all working on speed where there's
so many other things I would want
to work on?" Would it be that difficult
to just have, you know, different
groups focus on different things?
And I get it.
It's hard for a coach, you know,
doing that for each individual
player is really, really difficult.
So one of the things we're highlighting
very simply is these are the five kids
that for the next three months should put
a bit more work on their first touch, and
these are the five kids that should
put a little bit more work onto their
agility or just their speed.
And, you know, that is one of
the very simple insights we say
that's why it's development focused.
It's very, very easy to act upon
it if a coach has, you know,
five different buckets of kids.
Yes, a lot of the trainings you
can do everything all together,
but take a little bit of time
to say, "You guys focus on this.
You guys focus on this, you guys focus on
this." Or even give it as homework, right?
Just on the weekend, there's no game.
At least as a player,
I know what to work on.
I know I should be focusing on this.
And the last thing I wanna add to this
is it's not always working on a weakness.
Sometimes, you know, the algorithm will
say, "I've seen players like you become
really, really good central midfielders.
You're never gonna be the
fastest player on the team.
You know, you should not go crazy or
trying to be the fastest, but make
sure you're the player that controls
the ball well, that moves on those
little pockets." So it's very, very
individualized, which could mean sometimes
working on a weakness or doubling
down on a strength, but it's all very,
very focused on the individual player.
Liron: No, it's amazing because
I told Matt before we started
today that, "I'm against this.
This is not gonna work," and obviously
I've already changed my mind because
I've seen this so many times with my son's
teams, is that kids who have a, let's
say, a super strength in one way but a
big weakness in another way, always
trying to catch up on their big weakness
because it's basically everybody's trying
to get to a certain equilibrium, and
it's very interesting that, obviously
there's a way to kind of fine-tune,
per kid what it is that they need.
But what's confusing to me is that
AI, I was telling Matt before, AI is
very much about personal preference.
It's very much about what you
feed it is what you get out.
So if you feed it a certain European
dataset, and I don't even know when
you say European 'cause Europe has
so many soccer cultures within it
and philosophies on what football
is and a good footballer is.
The profile of a German kid is not the
same profile as, let's say, a Spanish kid.
Northern Spain is not
the same as South Spain.
I don't know.
Now you're coming to the US trying to
get this algorithm or translator to help
out teams, but it's judging it on the
completely different set of cultures.
Isn't that a problem?
Amir: I'll add two things to this.
First of all, a lot of what
we do is there's a lot of
translation work that we do, right?
If a player played in this country
and then went on to play with
another player in another country,
and that player went on to play at a
separate country, we can use the data to
translate a lot in like in what regions,
what works better in which region.
So there's a lot of translation work
that we do there, which we can even boil
down to, you know, how does the, I don't
know, collegiate division two in the
U.S. compare to an under 19 in Germany?
There's a lot of things we can do there.
The other thing that I usually
like to highlight is, soccer is
very different, but it's also not.
It's a baseline is true across soccer
pretty much everywhere, that you
have to be able to control a ball.
You have to be able to change
directions at a specific speed.
And, you know, one of the things
that we also measure is game impact,
and essentially in the end of the day,
you know, it's about winning, it's
about what you're bringing into the team.
And so we might not be able
to say, "You're gonna be
the best player in northern Spain,"
but we're definitely able to create a
baseline of if you wanna play soccer
at a high level, these are the things
that are gonna be true everywhere.
Matt: Can I ask you, for
the data that you have for
the last ten plus years, is it from
all professional academies in Europe?
Amir: So the data comes from these
very elite academies, but I should
mention a lot of them, what they do is…
I'll give you an example in
Switzerland, you know, you have these
clubs that facilitate testing with
a ton of regional teams, and none of
these are in professional academies.
So it's a professional club that
facilitates that, and they also test
their own academy, and that's why we
have a lot of professional outcomes.
But there's so many kids in that
system, in that data set that come
from these non-professional academies.
And we, you know, we have seen
trajectories from a player that went
on from the little club in the village,
in the mountain village, to the bigger
city, to the professional club,
all the way to being professional.
But it also means while we have that top
player's trajectory, we also have that
player compared to the other players that
were in his or her team very early on,
that was not a professional environment.
So it's for any player
who wants to get better.
We have a ton of data that, you
know, we use to help players get
to a collegiate level much more
than a professional player.
So it's really, yeah,
really about development.
Matt: Who's investing?
Is it the club who's paying a licensing fee?
Is it an individual family or player who's
paying a fee to get access to the data
and then figuring out how to
record and do all of those things?
Like what does that part of it look like?
Amir: Yeah.
It is generally through the club
or through the academy, 'cause part
of the testing has a team element.
There's also a small set of games, that we
measure game impact with, so it's not that
as an individual you can do the full test.
You could do part of it.
What we generally like to do or
to say is, we partner up with a
club or an academy because we want
this to be part of the curriculum.
So it's not something you do
by yourself in your backyard.
It's really in that club environment, in
that academy environment, that we do this.
And so when the question
comes, who invests in this?
It's generally a conversation
that we have with the club.
You know, is this something you're
gonna offer out of your pocket
as a service to the parents?
Or is it something that
parents can opt into doing?
Which obviously, depending on the setup,
yes, if it's an MLS academy, they say,
"We're paying for this, no problem."
And then in other systems, in pay-to-play,
the club might not have the budgets
themselves for it, so they give
parents the options to opt into that.
If you chunk it up in, you know,
across enough parents, we can make
it very affordable for everyone.
Sometimes it's a mixture of the
club subsidizes part of it, and
then the parents take on the rest.
So it's really case by case.
Liron: On the website you say that
the parents don't get the same
report the team does if, again, if I
understand the literature correctly.
The team gets a full report plus what
you call a forecast, but the parent get a
edited version or they get something else.
Why did you make that kind of decision?
Amir: First of all, you know, we're
not gonna show one parent results of
another kid that is not their kid.
That's just to start with
data privacy. You're not
gonna see, oh, you know, my best friend
was better or anything like this.
The club has a full overview because,
you know, as I mentioned, with being
able to put players into different
groups that might need more
focus on different things, so
that's why they get everything.
And then also a lot of it is
education about, you know, making
sure the coaches are sensitive
to, you might not wanna share.
You know, we have seen like coaches say,
"Oh, yeah, this guy was faster than you,"
and like, just be a little bit conscious
about what you're gonna share with whom.
On the what we call, you know, projections
or predictions, that's something
we're also very careful with, right?
By default, playing professionally
is a very, very low percentage
for any kid anywhere.
But showing a kid when
they're 11 years old, your
chance is, let's say 5%, which actually
is a really high percentage when
you're 11 years old, 5%.
But probably me as an 11-year-old,
if I heard 5%, I would've
been like, "Oh my God, I can never
be professional at such a low number.
I thought it's gonna be 90%.
I'm so good." Right?
So having understanding of
what that means is obviously a
very tricky thing, and we wanna be
very cautious with what you're gonna
see at what age, making sure the
parents have the right information.
So it's also generally a
conversation with the club.
You know, depending on if the club is,
again, if they're MLS club that just,
you know, their number one goal is to
have players play professionally, and
the kids know they're in this very,
very tough system, they might just see,
you know, let's say full information of
this is your percentage to play there.
But it doesn't make sense for a club
where the main goal is to, let's say,
you know, develop kids for college,
where a lot of kids or a lot of parents
would just see a very low
percentage to play professionally.
So then it's much more about, "Hey, you
could improve your chances of
play collegiate by 10% this year.
Here are the steps how to get
there." That would be much more the
phrasing that we use in that context.
So it's really
Liron: I see.
So the information the parent
gets on the kid is relative
to certain target or the environment
that they're in, not… Okay.
Well, not bad.
Not bad at all.
All
right.
Matt: Your family page also says
that no child is told they're the
slowest or the weakest on the team.
I mean, this is great news if you
flash back to Liron's childhood.
And it offers a...
Liron: Always the dot dot dot.
Matt: And it offers a player
card that's similar to a pro's.
What does a 12-year-old actually see
after the test day, and who decides
what is shared with them beyond, you
just mentioned club, parent, like, so
I'm assuming there's some checkboxes
essentially of what they wanna see or what
they don't wanna see and who sees what.
Amir: Exactly.
So it's exactly this.
It starts the conversation with
the club, and then the parent
will get the information that, you
know, is agreed upon with the club.
What the parent shows the kid is
outside of our control, of course.
So we do rely on the parents,
you know, making sure, to
deliver that in the right way.
That being said, we also do a lot
of education in that regard, right?
Whether it's us first talking with
the club and then the club sharing the
information with the parents, or sometimes
also directly where we, you know, we
do a Zoom call with a lot of parents.
We explain to them how
to read certain things.
You know, what is a percentile?
If it says, you know, 99th
percentile, it doesn't mean you
were the 99th number in the club.
It actually means you're at the very top.
And the number one thing we also
highlight is, well, so what?
What do you actually do with it?
And in most cases, you can boil
it down to one sentence, right?
Busy parents, they're not gonna go maybe
now through all the different graphs and
start, you know, being data scientists.
They just wanna know, "Okay, so
what does it mean for me? Like, what
should I do?" And that we can boil
down to one sentence, which could
be, you know, A, maybe be patient.
Your kid might be a late developer.
That doesn't mean… Explain that to them.
Tell them, "You're gonna grow. Don't
worry. You're a little bit behind."
That could be one thing.
It could be another thing where we
say, "Work a lot with the ball."
That could be a very simple message.
So, you know, don't try to be the
fastest on the team right now.
Work on your first touch.
We also have a video library where
it can show you how to do that.
You know, professional players or
collegiate players have put it together.
This is how I work on my first touch.
Liron: Can it be as simple as me
saying, "Well, you know what?" Like, I
mean, that's the beauty of AI sometimes
when we try to overachieve learning things
we have no ability to do it very quickly.
Is there, I mean, would you say, "Okay,
well my goal is to get my kid into D1.
Where's the gap now? He's 11 years old."
Is that… I mean, obviously I
know we can't cheat nature, but
does it have matrix like that,
relative to that target?
Amir: Yeah, exactly the target is
always gonna be a probability, right?
So it's how much can you improve
the probability to play at that
level within the next few months.
And I should highlight, obviously,
focus only on the soccer part, right?
There's obviously other things
that come into consideration
when you apply for college.
Liron: No, there aren't.
No, there aren't.
I'm sorry, Amir, I would
have to beg to differ.
Talking to the Stanford
kid here, but look, Amir, I
got really bad news for you, man.
There are other companies that do this.
I don't know if you know this.
And, they're popping all over, and
Matt and I are bombarded by it.
And what is happening with the market?
Why is what you're doing
different, or is it not?
And is it the strong will survive?
What is the reality for
the sector you're in now?
Amir: Yeah, I think as you guys
get bombarded, or what I hear a lot
from coaches and academy directors,
they get bombarded with things,
you know, they call it AI, they
call it data, they call it stats.
My question is always, you know,
when someone asks me, "Should I use
Game1 or something else?" What's the
so what you get from there, right?
Is it just what does it tell
me if I know a kid ran a
10-meter sprint in 2.1 seconds?
So what?
We put it into context because we
have all of this historic data.
Data without context won't
help you much, right?
So I think that's the number one thing.
It's just because you're looking at data
doesn't mean you're working with data.
In fact, you could also flip it
and say you don't even need to
see the data we're showing you.
What you need to see is the so what,
is the takeaway, that actual takeaway.
So I haven't seen a company that does
what we do because the proprietary data
set we have, the algorithm that we have
is the most powerful out there because
no one else has that data, right?
Matt: Amir, a lot of these
teams, they're using Veo or Trace
to record the games, and there's
analytics that come with that.
Layer on top of that, for all MLS NEXT
academies and elite academies, there's
Taka, which the players get access to.
So more data, right?
With some analysis, and then
there's potentially what you have
to offer, and there's others.
I guess where I start to go with
this is what's the consistency?
So if you're a parent or even if
you're a club and you're a coach
and they're the experts, and we're
clearly not, we just play one on
podcasts, how do you start to…
How do you know what's correct?
How do you know what to actually
double-click into and really
base your analysis off of?
And then how do you not just get
deluged with data that it's like
paralysis by analysis, trying to
understand what to actually act on?
Amir: It's exactly this.
I would always ask as a parent, whether it's
data that I get from a platform or
from a coach directly is, "So what?
What should I do with this?" Right?
Like, one thing I love to highlight
when we get data from a game
is my kid has done 100 passes.
What the hell does that mean?
Is that a lot?
And then, okay, yes, it was
the most passes in the game.
Then I would also ask, "So what?
Did I just play back and forth
with the goalkeeper and that's
how he got 100 passes?" Like,
what does that actually mean?
Liron: To Matt's point,
I know from parents on
another team, another club, I'm not gonna
mention the name of the club, but they
use some analysis, AI algorithm, and then
they give the weekly report to the kids.
And the weekly reports is
exactly what I was asking you.
It's not just for the kid itself, it's
also, A, it compares it to the other kids.
So you're number four in
passing, number three in running.
And a couple times the dad
would say, "Well, you know,
Duke isn't this great," but deep
inside, in a cynical, I'm like,
"Relative to what?" What is the
information actually telling you?
How are you deciphering
what it is it's saying?
And it's an excellent point,
and, I don't have an answer for
Amir: What I would argue here, I wanna
double-click on that point really quickly.
If you had stats on Messi's game, but you
wouldn't see the goals and the assist,
you'd probably say, "Doesn't run
much, probably doesn't pass too much."
He wins most of the games
for his team because
he scores the goals, he has the assists.
But I'm pretty sure if you would
cover up the goals, the assists, and
maybe the name Messi, and we just showed
the stats to a lot of people, they
would say, "Yeah, he
was the last in his team
on how much he runs. You know,
he's like, who is this guy?" Right?
So if you don't have
context, you're just gonna,
Matt: He's never in a duel.
He just floats around out there.
Who is this guy?
Yeah.
Liron: Context, context, context,
Matt, context
Matt: Is king around all of this.
What are the concerns that you
have with technology coming in,
you know, incredibly strongly into this
space that can either get misinterpreted,
not interpreted at all, and actually
change the dynamics in a negative way?
Amir: My concern is I think
what we touched on before. You
know, everyone thinks they need to
use AI, and so just by using the
word AI and just dumping a lot of
things on coaches or on parents
is obviously a big risk, right?
And then I think from the
parent's perspective, what I hear
a lot and what we get asked a lot
is, "Well, what's happening with my
kid's data?" And I think that is a
really, really important question.
You know, we take the data
privacy very, very seriously.
I think it is questions that you
should ask as parents, you know,
like what happens with all of the
data that gets recorded during games
or during tests or anything there.
And I think, you know, it's
obviously your right to ask like,
"Yeah, I wanna see your data
policy" or things like this.
So I think these are the things that we
hear as a concern and that you should
address as a parent, proactively.
Liron: I mean, this is a
complicated issue, right?
I was at a presentation.
The club organized a
nutritionist from a large company to
come explain nutrition for the kids
and then sign up for their program.
And the presentation was fine, it
was pretty generic, but I mean,
there were a lot of parents there.
And at the end, the presenter was
kind of, "Well, you know, and then we
take your kids' sleep patterns and
food and questions, and we feed
this algorithm we built and AI."
And then I was just like, "Wait."
And I could see kind of parents, some
pushing forward, "Oh, tell me about it.
I want somebody to tell me
how to run my kid's life."
The others pushing back.
And I raised my hand because it's
what I do, just no mic, and I
said, "Well, how do you deal with
privacy for the kids?" And the
answer was, "Well, we just use industry
standards." And that was pretty much
the death of that conversation.
It's a real, real concern,
especially on your business model,
which what happens if a family
is not comfortable while the team
does sign up for your service?
Amir: I mean, obviously no one is,
you know, in the end of the day,
no one is forced to do it, right?
Even if a full club signs up,
all the teams sign up, you
obviously don't have to participate.
Or you can participate and say,
"Look, I don't like what you guys
did. Please delete the data."
AI is so new. I think talking
about an industry standard is a
sneaky way around the actual answer.
So, the industry standard might not
be more than a few weeks old, right?
So, I don't think there
is really that standard,
especially in this space.
Maybe there's an industry
standard for, I don't know, how to
use AI in manufacturing, but you
know, we're not in manufacturing,
we're in kid soccer here.
Ask, "Okay, what does it
mean, like industry standard?
What's happening with the data?
Like what is it?
Who else can see it?
Can someone else see it?
What's happening with it?" So I
think I would really just ask those
questions and then, you know, it's on
us as the platforms to answer that.
And if you don't like the answer, you
should ask for the data to be deleted.
Matt: I would argue that the MLS
teams and academies could eventually
decide to subsidize, you know, the
10 strongest elite academies in their
regions, because the more data they
have, the more the elite academies can
focus on areas of strength and areas
of improvement, the better likelihood
of a player developing into one
that moves on to an elite academy.
Is that part of the
conversation at this point too?
Amir: Definitely.
I mean, there's three different ways
that, you know, these MLS clubs usually
use the technology, and I'll
use that to answer the question.
One is, you know, for their own
academy, one is for trialists, and then
one is, let's call it scouting days.
And, you know, when we talk about a
scouting day or tryouts, they invite
obviously kids from the local clubs,
which means let's say they invite 50
players, they're already pre-selected.
And then what we usually tell them
is, well, you wouldn't even need to...
Like how do you figure
out from those 20 clubs
or 10 clubs which 50 to invite?
You know, it's based on recommendations.
So if those clubs would already
do the testing and that helped
them filter down more already who
to invite, everyone would win.
And so, it's exactly that.
Liron: So, now that you've done this long
enough, maybe you can tell us what you
think you got wrong as you were growing.
What surprised you between sitting
on your computers and seeing
what data and analysis is all about,
but very different on the field?
Amir: One of the funniest
things I think we got wrong
initially is we were very focused
on the analysis, the algorithm,
everything that comes out of it.
What we underestimated is the testing
itself needs to be a good experience.
And that is one of the key things
that I think from when I think back
about like the very first tests
that we have facilitated with a
professional club in Europe, and it
was a horrible experience for them.
I think it took them like three hours to
test 20 kids, and then what we learned is,
well, we need to help them with, you know,
if you have a group of players, what's
the most efficient way to test them?
You know, are you gonna use a stopwatch?
Are you gonna use a laser timing gate?
Like how do you actually collect the data?
And so what we have doubled down on
was the actual testing technology.
So make it super easy, right?
Matt: I think it's a great idea.
I know my sons through the years
have gone through different testing
situations, in different environments,
and it seems like it's like
two or three hours after training.
And by the way, then you don't even see--
or if you see the results of something,
which is seemingly 20 to 25% of
the time, there's nothing to do.
There's no context around
it whatsoever, right?
And then it sours everybody a bit
on the wasted time and what exactly
we're trying to achieve through it.
Amir: 100%.
We have doubled down so much on
that, that we have gotten requests
from other sports about just our
testing technology because, you know,
all you do is you put up phones, it's
automatic, you can upload, you know,
several hundred players into the system.
It's like a combine.
It's fun.
You're in small groups
of players all the time.
It's very easy.
It's very simple.
You don't need an army of coaches to do it.
One of the most extreme things we have
done is, you know, we helped a club test
all their 800 players in one weekend.
And it was quick, it was fun, it was fast.
And that is because we have
developed that really, really, really
good testing technology around it.
Liron: Matt, am I hearing a
dystopian, I can't tell if it's
rosy or not, future where basically,
and Amir, are showcases done?
Like, if you look at it five years
down the road, let's say this thing
grows where you want it to grow.
I know you have plans.
Is what we think is these send 1,000
emails, pay hundreds of dollars
to showcase your kid in front
of a college or whoever, calling
scouts, is this what this is gonna replace?
Amir: I really hope it gives players,
I wanna say, a better chance.
Every single player that I've spoken to
that either went on to play collegiate
or professionally said they were lucky
to be discovered on that one day.
So I really don't want a player to be
stuck in, you know, an unread inbox
of another college coach, like.
So ideally, that would exactly be that.
We want to replace that you're just
sending, you know, hopeless emails with
video highlights that you've paid
so much money for and then no one's
ever gonna watch, but rather, you
know, being able to help the collegiate
coaches to filter through much faster.
You know, "Oh, these are all the
players that could be relevant for me.
Great.
I can actually watch all of them,"
rather than being overloaded
with all of those emails.
Matt: You were talking about the data
privacy, which I totally understand.
If you think about Taka today with MLS
clubs, college coaches have access to
Taka for individual players, right?
So they have that data already, which
presumably that's used for good.
I'm sure it is.
But Taka isn't producing the type
of analytics that what you're
talking about specifically.
Does this turn into eventually a massive
portal where you can go in and search
by right backs and then look across
the board or certain measurements?
If you think about how baseball and
Moneyball happened, right, where
the entire game scouting changed
drastically, how they paid people,
evaluated people, is that ultimately
where you think we're going here?
And this could be a portal that's
accessible by professional scouts, college
coaches, you name it, to be able to better
identify players from that perspective.
Amir: I wanna say yes and no at the
same time, because on the one hand, it
has the capability to do that, right?
But my goal is not to
make sure we can always
just pick the very, very top ones.
What this technology can do is it
can help every player get better,
and I think that's really, really the
point where, you know, when you talk
about, you know, collegiate coaches
having access to certain platforms,
there's also a risk there, right?
There's a risk that you have a lot
of information without context, and
those decisions might not
be the best decisions.
You just get more information, doesn't
have to be necessarily better information.
So I think while a Moneyball sort
of picking is possible, especially
if you're a professional club that
is looking to optimize for that, my
goal is much more to serve, I wanna
say, far more down the pyramid.
It's really not just about like,
"Let's pick the top ones and leave
everyone else behind." It's much
more how can we raise the floor?
How can we give everyone access to
a technology that is used by, you
know, most elite clubs, in Europe?
But then again, why should
it not be accessible to you?
You know, you're a young player in maybe
a underdeveloped, area of the world.
That's the great thing
about technology and AI.
It's so easy to make it
accessible to everyone.
So I wanna say it's much more about
the insights that you get playing at an
elite academy can be the same insights
that you get as a 12-year-old in rural
somewhere, 'cause you still know
what to work on next, right?
So that's,
Matt: You're democratizing it.
Yeah, I love that.
I mean, just yesterday we heard
from a parent who's not in an urban
area, and we tend to cover urban areas,
professional academies, and the pathways.
And I thought it was a great
question, which is, if you're not even
close to one of these professional
academies, you know, where do you
go to get that type of development?
How do you measure that development?
Like, where are the opportunities?
And I think this provides, you
know, it doesn't solve that puzzle
entirely, but that's, you know, that's
a very difficult puzzle to solve.
But this helps democratize it a great deal.
Amir: Yeah.
I think for that, you know, to that
example, what I would say is if the local
club at, you know, the bigger urban club
is already using, our technology, you
as a parent could send your score to
the club and say, "Hey, we have done the
testing here. My kid came out really,
really high. You should look at this."
And it's not as long as a video highlight.
It's not as, I wanna say,
intransparent as a video highlight.
It's the stats they can compare it to
what they already have within their
team, which makes it a no-brainer to
say, "Yes, of course, we're gonna, you
know, invite this kid for a tryout.
It outperformed all of our kids.
Like, why should we not
just invite that kid?"
Liron: Wow.
Oof, if only they had, one of these
for podcasts, we would be, top ranked.
Amir, you managed to answer all of
the questions I thought you wouldn't
be able to, and I gotta tell you,
this is incredibly fascinating.
You said that you have a couple of
partners in MLS as well, so I don't
know if you could elaborate on that.
Amir: Yeah, happy to mention some
of the clubs that we work with.
I should also mention
some we can't disclose.
There's clubs that, you know,
on purpose want us to not to.
But you know, happy to.
I spent this Friday in Houston
with the Houston Dynamo.
We did a really cool project
together with them where they
invited players that are actively
part of their Futures program.
And so these are, you know, players
from 20 different clubs, local clubs
that they're keeping an eye on.
And, as part of keeping an
eye on them, you know, we
facilitated testing with them.
We're gonna retest them
again in a few months.
So it's just a very good way for them
to keep track, to highlight some…
Again, I mean, there are players there
where like one kid was basically like
double the size of the other kid.
Both of them were amazing players,
but then getting insights on like,
you know, who's an early developer,
who's a late bloomer, things like this.
How can we give, as you
mentioned, democratizing the
access to a developmental
technology in the region?
Matt: Amazing.
Why don't we end with this?
So if an elite academy, so not a
professional academy, but one level
below that, whether that be MLS NEXT or
GA for girls or ECNL for boys or girls, if
they wanna work with you, what does
that process actually look like?
Amir: We'll always have a team member
coming to a first test, showing how to set
up the test, onboarding all the
coaches on the ground, also doing
the educational part of what happens
afterwards, what happens with the results.
So everyone is familiar with, you know,
the testing technology, the platform,
and then also the parent education part.
Liron: Okay.
Well, good luck to you and
good luck to all of us.
Matt: Yeah.
No, it's amazing, man.
Getting the opportunity to hear
about this in the earlier stages of
it and already seeing the progress that
you guys have made in your ability to
go get 10 years' worth of data to make
this differentiated is pretty impressive.
So, we're rooting for you and look
forward to following the progress.
Amir: I appreciate that.
Maybe one last thing I wanted to throw
in was, you know, you mentioned in your
very last question, boys and girls.
I think we haven't spoken enough about this.
The technology works for boys and girls.
So just very important to highlight, you
know, we're not just focused on boys.
It is really for every young
kid that wants to develop.
Liron: For the record, Matt had
that question on the script, and
I took it out during the interview
because we were running out of time.
So I'm the jerk on this one.
Amir: No worries.
No worries at all.
No, no, don't worry.
Before you pay for another
report for your kid,
you’ve got to ask, what are
you supposed to do with it?
So Amir said the most important
word in this whole conversation
out of the thousands spoken by me.
He said the word “context.”
And without context,
none of this evaluation
is going to be worthwhile.
Yeah, it is a good reminder we can
get inundated with information,
but without that context and without
an understanding of what the overall
objective is and where you can invest and
where you can move the needle and where
you need to focus, or the club needs to
focus and the coach needs to focus with
your child and vice versa, it’s
really difficult to feel like you’re
making ground or you’re gaining ground.
Yeah, I mean, look, AI is
democratizing a lot of things,
and they’re trying to
democratize development,
which, as we know, is
incredibly difficult to do,
and not for a lack of interest
or trying or capable people.
And I’m going to quote Amir.
He says, “I really hope it gives
players a better chance to succeed.”
I mean, that’s really what
it’s all about, right?
Yeah,
well, thank you, Amir Donath,
for giving us your time.
And Matt, this is Chasing the Game.
Not avatars quite yet.
Not quite.
- See you, man. Bye.
- See ya.