Football's AI Revolution: How Data-Driven Players Like De Bruyne Exploit the $40M Salary Gap and Force a Legal Reckoning
🎧 Listen now on Spotify, Apple Podcasts, Amazon Music and other streaming services! Don’t forget to subscribe, share, and leave a ⭐⭐⭐⭐⭐ review to help more players and coaches discover the power of Techne Futbol and Data Technology in the beautiful game.
#TechneAfricaFutbol #TechneFutbol #TrainWithTechne #Playermaker #TraceFutbol #Hudl #FootballTech #FootballInnovation #PlayerDevelopment #CoachTools #SmartFootballTraining #AfricanFootball #CAFOnline #FIFA #FIFAYouth #CAFDevelopment #YouthFootballAfrica #NextGenFootball #FootballScouting #TalentIdentification #DigitalScouting #EliteYouthDevelopment #FootballJourney #FootballAnalytics #FIFAForward #FutebolAfricano #TrainTrackCompete #FootballExcellence #FootballAfrica #TheFutureOfFootball
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
(00:00):
Welcome to the deep dive. Today we're really locking in on
the high stakes world of professional football, soccer
for some of our listeners, wherethese massive multi €1,000,000
contracts are decided. And we're seeing, well, a
revolution, aren't we? The whole traditional power
structure seems to be getting, you know, challenged by data, by
advanced analytics, even AI. Oh.
(00:21):
Absolutely. Probably the biggest shake up in
sports labor management since, well, since the modern agent
system really emerged. The sheer amount of money flying
around. Now it just demands a level of
analytical rigor that frankly didn't exist 1015 years ago.
But the really core question, the one we need to wrestle with
today, is this. When a player sits down for that
(00:42):
negotiation, that life changing contract, what is the club
actually paying for? Is the stuff we can easily see
the historical stats, or are they paying for future
productivity? The stuff that's scientifically
proven to lead to wins. And the players themselves, I'm
not just sitting back and waiting for the clubs to figure
it out, are they? This power dynamic is shifting.
(01:03):
I mean, yeah, really shifting. Definitely.
We've seen reports, for instance, that a former
Manchester United star didn't just rely on his agent.
He reportedly used ChatGPT, you know, the AI language model, as
a sort of negotiating assistant for his new contract.
Wow. That just blurs the lines
completely, doesn't it? Between human advice and, well,
(01:23):
computational leverage. It really does.
And think about what that means.You've got immediate accessible
bait, informed council available204 seven.
It's incredible, yeah. But even beyond that cutting
edge AI stuff, we have these really solid examples of how
structured data is being used right now, kind of a blueprint
for the future. You mean like Kevin Debrina?
(01:44):
Exactly like Kevin Debrina, the Belgian midfielder, absolutely
brilliant player. When he was negotiating his big
extension with Manchester City, he made a very deliberate
choice. No agent, no traditional agent,
No. Instead, he hired his own
dedicated team data analysts, toarm him basically for that
negotiation. That is just that's the ultimate
(02:05):
power move, isn't it? It tells you that players at
that very top level, they see the traditional way of doing
contracts as well inefficient, an informational gap they can
personally exploit. They're commissioning their own
custom built data to fight back against the club's internal
valuations, which are often, youknow, a bit of a black box.
(02:26):
Totally. So OK, our mission today then is
to really unpack this whole conflict.
We're going to look at some ground breaking new research.
It tests whether these advanced analytics actually influence how
much a player gets paid, right. We'll dissect this trend of
data-driven self representation,like the Debruin example.
And then, crucially, we have to dive into the huge and frankly
(02:48):
urgent legal and ethical hurdlesthat AI needs to clear if it's
going to manage the careers of these $1,000,000 athletes.
There's a lot there. A huge amount.
It's a fascinating intersection of economics, technology and
law. OK, let's jump straight into
Part 1 then the Moneyball test. And this isn't just about, you
know, cool stats for fans. This really starts with cold
heart economics. Labor Economics One O 1.
(03:10):
Exactly. The basic theory is that your
pay should reflect how productive productive you are.
That's the foundation. And professional football, where
literally every pass, every run,every touch is tracked and
quantified. It should be the perfect place
for this theory to play out the perfect marketplace.
It should be, yes, in theory, because performance is so
measurable and the outcomes, wins, losses, goals scored,
(03:33):
they're crystal clear. But there's always a but.
There's always a but yeah, past research trying to link pay and
performance in football really struggled.
There was one major methodological flaw that kept
cropping up. It was it was the reliance on
multi year contracts. They just muddied the waters
completely. OK, explain that a bit more.
Why do those long contracts makethe research so messy?
(03:56):
Well, think about it, if a player signs say a six year deal
today in 2024, the salary they earn way out in 2028 that was
decided back in 2024, it reflects their perceived value
then, not how well they actuallyplayed in the season just before
in 2027. OK, so there's a time lag A.
Huge time lag and it fundamentally weakens any
(04:18):
correlation you try to draw between their current
performance and their current salary in those later years.
Because the salary was locked inyears ago.
Precisely that noise, that distortion caused by those
predetermined salaries from previous years.
It made it almost impossible to isolate the true effect of
performance metrics on what a player gets paid right now.
(04:38):
So, OK, how did this new Economic Research get around
that problem? What was their clever fix?
Their methodological breakthrough.
They used a really sharp, elegant constraint.
They decided to focus only on players who were in the very
first year of a new contract A. Clean slate.
Exactly. A clean slate for each player
observation. They built up a pretty
(04:58):
significant sample to 992 playerseason observations.
Wow. Yeah.
Across two of the biggest, most high profile leagues in the
world, the the English Premier League and Italy, Syria.
And the data covered the seasonsfrom 20/18/19 up to 20/21/22.
That feels crucial. By isolating just that first
year, they can make a much tighter assumption.
Right, a very reasonable assumption.
(05:20):
That the players performance in the season immediately before,
say 20/21/22 is what directly drove the specific salary offer
they just signed for the 20/22/23 season.
Exactly. The link between recent
performance and the new salary is much, much clearer.
Direct even. OK.
And what about the salary data itself?
How specific specific did they get?
(05:41):
Because contracts have bonuses, endorsements.
Good question. They were meticulous about this.
The salary data used was the verified guaranteed base salary
gross of tax. The before tax.
The before tax, yes. And crucially, it excluded
performance bonuses and any external endorsement deals.
So just the core guaranteed annual income, what the club is
(06:01):
fundamentally saying that players worth per year.
That's it, the core value proposition from the club's
perspective. OK, fantastic, robust
methodology in place. Let's start with the basics
then, the stuff fans and commentators talk about all the
time. Goals, assists, maybe simple
pass completion. Do these basic stats actually
predict team success? You know, winning.
And, critically, are they rewarded with bigger paychecks?
(06:22):
Broadly speaking, yes, on the basic level, there's alignment.
The research looked at these common metrics as predictors for
two key things, predicting team points, basically winning games,
and predicting player salary. And things like goals, assists,
simple pass completion percentage, the number of
touches a player has, even the distance they carry the ball.
(06:42):
All of these showed a positive correlation with the team
getting more points. They help you win.
OK, that makes sense. And does the money follow?
Do players get paid for doing those?
Things, yes. Strong alignment here too.
Players are overwhelmingly, significantly rewarded for goals
scored. That's the big one,
unsurprisingly. The glamour stat.
Absolutely, but they also see noticeable positive returns in
(07:06):
their salary for racking up assists, for having a high pass
completion rate, and even just for the sheer volume of touches
they get on the ball. So the most obvious, most easily
televised metrics. They definitely influence the
executives signing the checks. It confirms that yes, what's
easily seen and counted gets rewarded.
But even at this basic level, there was a really surprising
(07:28):
finding, wasn't there? Especially for defenders, it's
about the humble tackle. Ah yes, this one really jumps
out because it flies completely in the face of conventional
wisdom. You know, the idea of effort,
Defensive grit. Right, getting stuck in.
Exactly. The research found that the
number of successful tackles a player makes well in Turns out
(07:49):
it's not a significant predictorof team points.
It doesn't statistically help the team win more.
Wow, OK. And maybe even more
surprisingly, it's also not rewarded in their salary.
It was statistically insignificant in determining
pay. That's that's almost heresy in
football culture. You picture a midfielder making
10 crunching tackles in a game and you think the guy's earning
(08:12):
his money? How can the market just ignore
that? Well, it forces you to think a
bit more sophisticatedly about what actually constitutes good
defending. OK, the strong suspicion, the
conjecture coming out of this finding is that having to make a
lot of tackles might actually bea sign of failure earlier in the
defensive process. Meaning maybe the team's
positioning was poor or a teammate made a mistake further
(08:33):
up the pitch, forcing that player into a last ditch tackle
situation. So the tackle is fixing a
problem, not necessarily representing proactive dominant
defending. Potentially optimal, really high
value defensive work is often about anticipation.
It's about reading the game, cutting out passing lanes,
intercepting passes, preventing the need for a tackle in the
(08:55):
first place. And you don't really get a
flashy stat for intercepting a pass that would have led to a
dangerous situation 2 passes later.
Precisely that kind of invisibledefending is incredibly
valuable, but doesn't show up inbasic tackle counts.
So the data suggests executives are rewarding the tangible,
aggressive actions that touches the goals, but maybe missing the
(09:17):
value of that less visible but arguably more effective
defensive contribution. That's a really powerful
insight, and it perfectly sets the stage for this idea of the
analytics gap, doesn't it? If the basic stats are
potentially flawed or incomplete, then let's move into
the real Moneyball territory, the advanced analytics.
We're talking expected goals, XG, progressive passes, shot
(09:39):
creating actions. These are the metrics close
supposedly use now to find valuethe market is missing.
Right, this is where the economic theory suggests
inefficiencies might lie, and wereally see that bear out in the
findings. And just a quick point on
expected goals XG. It's crucial because it moves
beyond just counting goals. It gives you a normative
appraisal. It calculates how many goals a
(10:00):
player should have scored given the quality and location of the
chances they had. It accounts for difficulty.
Right, a much smarter measure than just the raw goal tally.
So armed with these advanced metrics, what did the research
find? Are the things that really
predict winning being rewarded financially?
This is where the misalignment becomes stark.
There are actions scientificallyproven to contribute
(10:22):
significantly to winning, but the research shows they are not
being adequately rewarded in player salaries.
OK, give us the prime examples. What wins games but doesn't seem
to boost the paycheck? Two metrics really stood out
here. They were strong predictors of
team points, meaning they genuinely help teams win, but
they showed weak or statistically insignificant
(10:44):
evidence of actually being reflected in player salaries.
And they are. Successful presses and shot
creating actions. Wow, let's break those down.
Successful presses. That's winning the ball back
high up the pitch, right? Often leading to immediate
attacking chances. Exactly, it's fundamental to
modern high intensity football styles.
Think Liverpool under Clop or Guardiola City that pressing is
(11:06):
critical. And yet the player executing
that successful press, winning the ball in a dangerous area,
they don't see a significant financial return for that
specific action. The evidence suggests not nearly
as much as they should. Based on its contribution to
winning, it seems undervalued inthe pay structure.
Yeah, and the other one shot creating actions.
(11:27):
Yeah, this is another crucial 1Ashot.
Creating action is defined as one of the two offensive actions
directly preceding a shot. So it could be the pass before
the shot, or the dribble that beat a defender before the shot,
or drawing a foul that leads to a shot.
Basically the actions that directly generate scoring
opportunities. The engine room of the attack?
Really. Absolutely.
And again, while these actions strongly predict team success,
(11:50):
the research found much weaker evidence that they are being
rewarded appropriately in salaries.
So clubs are in effect systematically underpaying for
these incredibly high impact team success driving actions.
That's what the data points towards.
It highlights a clear inefficiency, an opportunity
that clubs, or at least their compensation models, seem to be
(12:10):
missing. Oh, and progressive passes to
moving the ball significantly closer to the opponent's goal.
They also predicted wins but weren't significantly linked to
higher pay. OK, now let's flip that.
What is being highly rewarded with a big contract, but doesn't
actually seem to predict wins all that well according to the
statistics? This is where that executive
(12:31):
bias you mentioned might become visible.
Yes, the primary example that emerged here was progressive
carries. OK, define that for us.
A progressive carry is when a player successfully moves the
ball forward towards the opponent's goal by dribbling or
carrying it covering a certain distance, usually at least 5
yards, or carrying it into the penalty area.
(12:52):
So running with the ball effectively up the pitch.
Exactly. And the analysis found that yes,
progressive carries are valued by teams.
Players who do this well tend toreceive higher salaries.
OK. That sounds reasonable.
But, and this is the key finding, yeah, the exact same
research found that progressive carries did not significantly
(13:12):
predict team wins. Wait, so running with the ball
up field gets you paid more, butit doesn't actually correlate
strongly with your team winning more games?
That's what this specific analysis found, yes.
It's the mirror image of the pressing and shot creation
findings. That's the exact opposite of
what pure data-driven decision making would dictate, isn't it?
Why would this action, which often looks exciting, involves
(13:35):
flair, gets fans off their seats, be rewarded with premium
pay if it doesn't reliably contribute to winning?
Well, this speaks volumes about what we're calling the executive
biases. Progressive carries are highly
visible. They're exciting to watch.
They often end up on highlight reels.
They contribute to a player's individual reputation for
dynamism and skill. They look good.
They look. Good.
(13:56):
And the suggestion is that club executives, maybe
subconsciously, maybe even consciously to some extent, are
rewarding that flair. They're rewarding the visible
output that appeals to the fan base, maybe helps sell jerseys,
rather than the more nuanced, perhaps less visually dramatic
but statistically more effectiveactions like successful pressing
or smart shot creating passes. So it's marketability and
(14:19):
perceived individual brilliance over the statistically proven
contribution to actually winningmatches.
That seems to be a strong possibility based on this
evidence, rewarding what looks impactful over what is impactful
according to the deeper analytics.
And this whole idea, this bias towards the visible and the
simple, it kind of culminates inthe most significant finding
(14:39):
regarding how executives might be making these decisions,
doesn't it? The finding around plus minus
goals. Ah yes, this is perhaps the
clearest indicator, the sort of smoking gun confirming this
potential bias towards simple outcomes.
OK. The researchers looked at a very
basic metric, plus minus goals. This is simply the team's goal
(14:59):
difference, goal scored minus goals conceded, but only during
the time that specific player was actually on the pitch.
Right, if I'm on the field for 70 minutes and my team scores 2
and concedes one while I'm playing, my plus minus for that
game is +1. Exactly, and unsurprisingly
perhaps, this simple plus minus metric was found to be
(15:19):
significant in predicting both player pay and team points.
OK, well, that seems reasonable on the surface, doesn't it?
If I'm on the pitch and the teamis generally outscored scoring
the opponent, surely I deserve ahigher salary.
It suggests I'm contributing to success.
It sounds logical, yes, but the implication is actually quite
profound when you dig deeper. How so?
It suggests that club executivesmight be taking a very
(15:42):
consequentialist approach. They're rewarding the player
simply for being present while the team is successful.
Rewarding success by proximity. Essentially, yes.
Rewarding presence during good outcomes, and this highlights a
potential fundamental inability or perhaps unwillingness to
correctly separate that individual players actual
(16:03):
productivity from the success generated by the high performing
teammates who might be around them.
So I might just be lucky to be playing alongside 3 superstars
who are carrying the team, but Istill get the pay bump because
the team wins when I happen to be on the field.
That's the risk inherent in relying on such a simple metric.
It doesn't isolate individual contribution effectively.
(16:23):
But how do we know this isn't just a really good simple metric
that magically captures everything important?
Maybe plus minus is just efficient.
That's a fair question, but we know it's likely not the full
picture because the researchers specifically tested a superior,
analytically much sounder version of the same basic idea.
Which was? Expected goals plus minus.
(16:43):
This metric does the same thing.Looks at the difference while
players on the pitch, but instead of raw goals it uses
expected goals. So it accounts for the quality
of chances created and conceited, opponent strength,
game state, all that complex stuff.
Exactly. It's far more informative about
a player's true net impact than just looking at the raw
(17:04):
scoreboard outcomes. And what did they find when they
tested that sophisticated metricagainst player salaries?
Crucially, the research found that this more sophisticated
metric, expected goals plus minus, was not significant in
determining player salaries. Wow, okay, let that sink in.
The decision makers, the executives, they have access to
the complex, more accurate data that isolates a players true net
(17:27):
impact on goal difference, but they seem to ignore it when
setting salaries. Instead they rely on the simple
scoreboard based metric, the rawplus minus.
That's the strong implication. It really confirms this
potential bias towards easily digestible results, even if
those results are frankly informationally quite shallow
compared to what's available. They're looking at the final
(17:49):
score rather than the sophisticated model that
explains how that score was likely achieved and who drove
it. Precisely, it suggests a
preference for outcome over process, analytically speaking.
OK, fascinating stuff. On the metrics, Before we move
on from salary determinants, what about the other
foundational factors? Things like age, experience,
maybe how a player moved clubs. Yeah, those baseline factors are
(18:12):
still very important. As you'd expect, age followed
the classic pattern we see in many professions, actually.
The inverted U curve. Exactly.
Age has a positive effect on salary, but it's a diminishing
effect. The returns get smaller as
players get older. So experience helps, but only up
to a point. Right.
And the data strongly suggested that the optimal time for a
(18:33):
player to sign that major careerdefining contract, the point of
highest earning potential, is typically reached around 26
years old. 26 That's the magic number for the big payday.
It appears to be the peak, yes. After that age, salary growth
tends to slow down, likely reflecting the market's
perception of the approaching physical decline, even if it
(18:54):
hasn't happened yet. So if you're a top player
hitting 28 and you haven't secured that massive long term
deal, you've potentially missed the absolute peak earning
window. That's what the data suggests,
yeah, timing is critical. What about experience and
reputation? Playing for your country?
Playing in top leagues. Clear premiums there, yeah.
Being a senior international player represent your national
(19:17):
team regularly provides a significant salary boost, which
confirms that clubs pay extra for that established reputation,
that pedigree and maybe also forunobserved qualities like
leadership, handling pressure, things associated with
international experience. Makes sense and playing in other
big leagues before. Also a positive factor,
experience accumulated in other top European leagues, say coming
(19:39):
from La Liga or the Bundesliga to the Premier League,
contributes positively to the salary offered.
It signals proven quality at a high level.
OK. And one last finding related to
player movement. What about players who move on a
free transfer when their contract expires and they're
free to sign anywhere? This was another compelling
finding, and it actually mirrorswhat we see in other sports
(20:01):
labor markets. Like baseball.
Players who moved clubs on a free transfer appear to receive
a measurable salary penalty. They seem to get paid slightly
less on average than similar players who moved for a transfer
fee. Really.
Why would that be? You'd think being free would
give them more bargaining power.You might think so, but the
common interpretation is about signaling.
(20:23):
Signaling. Yeah, the players former club,
by letting their contract expireand allowing them to leave for
free, is essentially signaling to the entire market that they
weren't desperate to keep that player.
They were content to let them walk.
Exactly. It implicitly suggests that the
previous club felt they had or could find better alternatives.
(20:44):
That negative signal, that lack of strong interest from the
selling club, potentially dampens the salary offers the
player receives from their new club.
Interesting. It's like a subtle mark against
their perceived value. It seems that way, yes.
A small but statistically noticeable effect.
OK. So to wrap up Part 1, we've
established this clear, measurable inefficiency in the
(21:06):
football transfer market. Clubs often fail to pay
appropriately for the analyticalactions that genuinely predict
winning games. They seem to favor basic stats,
visibility, and simple proximityto success.
In this market failure, this gapbetween analytical value and
actual pay, it creates this massive vacuum, an opportunity
(21:26):
which, as we'll see in Part 2, the players themselves are now
starting to exploit in really interesting ways.
Let's move on to that, the new negotiating edge.
So yeah, if the clubs are being,let's say, analytically
inefficient in how they value players, that creates a
fantastic arbitrage opportunity for the players themselves.
Arbitrage, meaning they can profit from that inefficiency.
Precisely. And this seems to be a major
(21:47):
driver behind this growing trendwe're seeing towards self
representation, or at least moredirect involvement by players in
their own negotiations. Right.
Historically the model was pretty simple, wasn't it?
The player hires a single agent,and that agent handles basically
everything, finding clubs, negotiating contracts, maybe
even commercial deals. The player delegates.
(22:08):
That was the standard model, yes.
Delegate all the sporting, commercial and financial
interests to one person or agency.
So why is that model now being seen as potentially risky by
some top players? Well, I think the smartest
players and their advisors are realizing the complete
delegation means giving up a huge amount of informational
(22:28):
power. OK, there's a potential conflict
of interest too. Does a single agent always
prioritize maximizing this one specific contract for the
player, even if it means pushingthe club really hard?
Or might they sometimes prioritize maintaining their own
long term, smooth relationship with that club because they'll
have other players to place there in the future?
The repeat business incentive for the agent might conflict
(22:50):
with the players immediate best interest.
It's a potential concern, yes. So increasingly you see players
wanting more control. They're managing their careers
more actively themselves, maybe using specialized lawyers for
the legal aspects involving family members they trust.
And then hiring specific expertslike data analysts only when
they need that targeted expertise.
(23:11):
Exactly, it's a more modular, player controlled approach.
And the ultimate highest profileexample of this, the one that
really showed how data could be weaponized in a negotiation, is
Kevin De Bruyne. We need to dissect this case
properly. It really is the blueprint,
isn't it? A landmark case When Debruyn was
negotiating his big contract extension with Manchester City a
(23:32):
few years back, he completely bypassed the traditional agent
route. Just didn't use one.
Didn't use one. Instead, he commissioned his own
specialized, bespoke team of data analysts specifically for
this negotiation. So this wasn't just like looking
up his stats on a website. Oh no, absolutely not.
This is a dedicated analytical operation working directly for
(23:53):
him. What were the main tasks he gave
this analytical team? What did he want them to figure
out? The analysis they did was
brilliant really. It had two main prongs, 2 key
objectives. OK, what was the 1st?
The first was maybe obviously torigorously quantify his
individual performance in a pack.
Proving his own value. Exactly, using advanced metrics
(24:15):
linking his specific actions on the pitch, passes, chances
created, positioning everything directly to City's probability
of scoring, winning games, achieving success.
Basically proving with hard dataindispensable he was to the way
City played in one OK, makes sense use.
The club's own language data back at them to prove your
worth. But you said there were two
prongs. What was the second?
(24:36):
This is the really innovative part.
I think it really is the 2nd. Prong was a predictive analysis
of the team's continued chances of success moving forward, not
just about him. But about the whole team's
future? Yes, his analysts assess.
The Manchester City squad's age profile, the quality trajectory
of key players, the potential impact of planned signings and
even the likely strength of their main domestic and European
(24:58):
rivals over the next few years. Wow.
So he wasn't just negotiating his salary based on past
performance, he was negotiating based on the probability that
the club would remain competitive and successful
during the lifespan of his new contract.
That's exactly it. It was an incredibly strategic
move. He was essentially saying, OK,
you want me to commit my peak years to this club, Prove to me
(25:21):
with data that this club is structured to continue competing
for major trophies. Show me the numbers that justify
my commitment. He was demanding data.
Driven assurance about the club's future, not just his own
pay packet precisely he. Wanted to ensure that if he
signed, he wasn't just signing for money but for a continued
high probability of winning things and the outcome well.
(25:42):
The data, presumably. Backed up both his individual
value and the club's strong prospects, he secured a
significant pay rise and the contract extension he wanted.
And it was predicated not just on his brilliant past
performance, but on this data-driven forecast of the
club's trajectory. It's the perfect example of a
player actively commissioning specific analysis to completely
(26:04):
neutralize the informational advantages the club usually
holds in negotiations if the club tried to lowball.
Him using simplistic arguments or basic stats he could counter
with his own. Bespoke high end predictive
analysis proving exactly why he was worth more and why staying
was a good bet for success. It fundamentally flips the power
(26:25):
dynamic totally the club is usedto.
Being the only side in the room with the really sophisticated
spreadsheets and predictive models, now the player walks in
with their own custom built and highly persuasive valuation
model. It's a game changer.
And this transition from needinga bespoke team of human analysts
like Dabrina had to potentially using more scalable AI tools,
that's happening right now. Back to the ChatGPT.
(26:47):
Example exactly. That report about the former
Manchester United player using an AI like Chachi PT to assist
in contract negotiations. It confirms that the kind of
specialized analysis. De Bruyne Commission is already
merging with these readily accessible AI tools to the AI
can provide the. Speed.
The data synthesis? Maybe.
The analytical breadth that evena dedicated human team might
(27:10):
struggle to pull together instantly?
Potentially, yes. It can process vast amounts of
data, identify patterns, maybe even simulate negotiation
scenarios based on known variables.
It's becoming another tool in the players negotiating toolkit,
which means the traditional. Agent whose primary value
proposition was often based on relationships, access, and maybe
more subjective negotiation skills.
(27:31):
Their role is rapidly changing. It really is.
They're becoming perhaps more like strategic advisors who
manage a team of specialists, lawyers, financial advisors,
data analysts, maybe even AI prompts.
Rather than being the sole font of all knowledge and negotiation
power, they might become optional intermediaries,
supplemented or eventually even replaced by code and data in
(27:52):
some functions. But this explosion?
Of data use, players, commissioning, analysis, clubs
using AI for scouting and performance.
It leads us directly into this massive, complex minefield
regulation and ethics. Which brings us to Part 3, the
AI regulatory reality check. Because the tech is moving way
(28:12):
faster than the rules. OK, so AI is booming in sports
management. We're seeing it used everywhere
from talent scouting, identifying the next big star to
predicting injury risks, even automating aspects of training
schedules. The adoption rate is just
explosive. It really is.
But. As you say, the legal and
ethical frameworks needed to actually govern this powerful
(28:32):
technology, yeah, they are lagging dangerously behind.
How far behind do we have numbers?
On this adoption versus policy gap, we do and they're quite
stark. Frankly, a bit alarming.
A recent survey looked at about 150 major sports organizations
globally. OK, they found a staggering 78%
are already using AI tools in some capacity for talent
management. There's nearly four out of every
five organizations relying on predictive analytics, automated
(28:55):
scouting, or similar AI applications 78%.
Adoption. That's huge.
A massive technological shift ina very short time O the crucial
question, how many of those organizations actually have
comprehensive, well defined policies in place to govern how
they use that AI? That's where the gap is.
Shocking. Only 32 percent, 32, so almost.
(29:17):
80% are using it, but less than 1/3 have clear rules for it.
Exactly that. Massive gap, 78% usage versus
only 32% with established policies represents an enormous,
largely unmitigated legal and ethical exposure for these clubs
and organizations. Wow.
The urgency to adopt the new tech, you know, to in that
competitive edge on the pitch orin the transfer market, it has
(29:39):
completely outrun the organization's ability to manage
the risks that come with it. Risks around data privacy,
intellectual property, algorithmic bias.
It's a long list. OK, let's breakdown those risks.
What are the main legal and ethical challenges that are
really keeping the lawyers and executives at these sports
organizations awake at night right now?
What are the top concerns the survey identified?
(29:59):
Three primary areas of major concern.
The number one issue cited by 72% of the organization's is
intellectual property rights IP.OK.
How does that play out specifically in sports AI?
It sounds complex. It's incredibly nuanced.
Think about it, who actually owns the intellectual property
for the insights generated by anAI?
(30:19):
Who owns the algorithms themselves, especially if
they're constantly learning and adapting based on player data?
Who owns the predictive models derived from analyzing years of
player performance data? Right.
Give us a concrete example of. Where this gets sticky?
OK, let's say a premier. League Club hires an external AI
company to develop proprietary machine learning model.
(30:40):
This model predicts a young player's potential or their risk
of specific injuries based on tracking their movement patterns
and training using wearable sensors.
Now, if that player eventually leaves the club and signs for
arrival, who owns the IP relatedto the insights generated from
that specific player's unique biometric data?
Using that model, does the club retain rights over analysis
(31:02):
derived from data collected while he was their employee?
What if the player like De Bruyne commissions his own AI
driven valuation model using data from his matches?
Does he own the resulting IP or does the underlying performance
data technically belong to the league or the club since it was
generated during official competitions or training?
Yeah, I can see the problem. It's.
(31:23):
Incredibly ambiguous. Who owns the analysis of the
performance exactly and? This ambiguity creates a
potential legal time bomb. You can easily foresee future
disputes, arbitrations, maybe even lawsuits over the ownership
and use of these valuable AI generated insights and the
underlying data. OK, IP is number.
One, what was the second major concern, the second sided by?
68% revolves around data privacyconcerns.
(31:46):
This is huge, especially Europe with GDPR but relevant globally.
General data protection. Regulation, right we're talking
about. Managing incredibly sensitive,
often hyper specific athlete data.
Performance metrics are one thing, but AI often relies on
much deeper data. This is where the wearable.
Tech and the advanced tracking systems really complicate
(32:06):
things. Isn't it the stuff players wear
in training now? The cameras tracking every
movement precisely. We already have some precedent
in North American sports like the NBA, where there are
specific rules prohibiting the use of data from wearables in
contract negotiations, largely to protect players.
But in top level football, theseadvanced tracking systems are
(32:26):
becoming ubiquitous. They capture biometric data,
heart rate, acceleration, deceleration, fatigue markers,
incredibly complex movement profiles.
Clubs need explicit informed consent for collecting this, and
more importantly, they need extremely careful governance
protocols dictating how this sensitive data can be collected,
how long it could be stored, whocan access it, and specifically
(32:47):
how it could be used, especiallyif it might lead to decisions
affecting a player's contract, they're playing time, their
financial future, or even their employment status.
You can imagine scenarios. Where AI flags a potential long
term injury risk based on subtlemovement changes and a club
might hesitate to offer a long contract.
That's a minefield, a huge minefield.
The need for advanced data anonymization techniques where
(33:09):
possible and robust security protocols is absolutely
paramount to comply with privacylaws and maintain player trust.
OK, so IP. And data privacy.
What was the third major legal and ethical challenge
identified? The third one cited by. 61% is
the potential for bias in decision making.
Algorithmic bias. The AI.
(33:30):
Inheriting human biases exactly this speaks.
Directly to the legal and ethical imperative for these AI
systems to ensure fairness and avoid discrimination.
How might that bias creep? In, well, imagine an AI.
Scouting tool that's trained primarily on historical scouting
reports and performance data from, say, the last 20 years.
If the human scouts and managerswho generated that historical
(33:50):
data had unconscious biases, maybe they tended to overvalue
players from certain countries or leagues, or players with a
particular physical profile while undervaluing others.
The AI learning from that data will simply learn, replicate,
and potentially amplify those historical biases.
So the AI wouldn't be. Objective.
It would just automate the old prejudices at scale.
(34:12):
That's the risk. It could lead to unfairly
filtering out perfectly capable players from underrepresented
backgrounds or those who don't fit the historical mold simply
because the training data was skewed.
And legally, that's obviously. Problematic, but practically
too. You miss out on Absolutely.
And this demands transparency. It underlines the need for
what's called explainable AI, orXAI.
(34:35):
Clubs and organizations need to be able to understand why the AI
made a particular recommendation, whether it was
flagging a player to sign, suggesting a transfer value, or
even monitoring training intensity.
They can't just hide behind, thealgorithm told us so they need
to be able to audit the decisionmaking process to ensure it's
fair and justifiable. So faced with these.
Huge challenges, IP privacy bias, putting actual policies in
(35:00):
place moves from being just, youknow, good governance or nice to
have to being an absolutely. Critical risk mitigation
strategy. It's essential.
Do we have concrete proof? That having these policies
actually works. Does it tangibly reduce the
risks? Yes, we do.
The survey provided some really compelling quantitative evidence
on this point. Organizations that had already
(35:20):
implemented robust proactive AI policies, the 32% who are head
of the curve, they reported a demonstrable 40% reduction in
legal disputes related to talentmanagement compared to those
without such policies. 40% reduction.
That's that's not a small number, that's significant.
It's a massive difference. It's clear evidence that taking
the time to establish pre emptive legal and ethical
(35:43):
governance frameworks drastically reduces the
likelihood of ending up in costly litigation, arbitration
or regulatory trouble down the line.
That 40% figure. Makes a very strong financial
case for investing time and resources into developing these
policies, doesn't it? It's not just about doing the
right thing. It saves money and hassle,
absolutely. And.
Interestingly, the ethical governance side seems to tie
(36:04):
directly to performance outcomestoo, not just risk reduction.
How so? The research.
Also found. A strong positive correlation.
They quantified it with an R value of .71 between
organizations establishing dedicated AI ethics committees
or review boards and their success and talent acquisition R
value of point. 7/1 for listeners may be less familiar
(36:25):
with statistics that indicates avery strong positive
relationship, right? Almost undeniable cause and
effect. It's a very strong.
Correlation. Yes, it suggests that
organizations actively grapplingwith the ethical implications of
AI setting up oversight structures, they're also the
ones having more success in actually identifying, valuing,
and securing the talent they need so ethically governed.
AI isn't just safer legally, it might actually be more effective
(36:48):
at its job, perhaps because the internal systems are less likely
to be thrown off by hidden biases or flawed data practices.
That's a very plausible. Interpretation Good ethics leads
to better, fairer and ultimatelymore successful AI
implementation in this context. So what's the?
Way forward, then. How do sports organizations
navigate this? It requires a proactive.
(37:10):
Approach implementing adaptive regulatory frameworks,
internally enhancing data governance protocols, constantly
investing in advanced data anonymization and security, and
crucially, providing continuous legal and ethical education for
staff involved in using these AItools.
It's not a one off policy. Document.
It's an ongoing process, exactlythe.
(37:32):
Organizations that are going to thrive in this new data-driven,
AI influenced era of sport will be the ones that view data
security, pretty ethical compliance and robust
governance. Not as inconvenient burdens, but
as core strategic assets. They are fundamental to
sustainable success. OK, let's try and synthesize.
This whole deep dive then, it's been quite a journey.
We started with that fundamentaleconomic theory, pay should
(37:54):
reflect productivity. And we looked at brand new
research using that clever methodology, focusing only on
players in the first year of newcontracts in the Premier League
in Syria to really test that theory in football.
And what we found was. Well, a pretty profound
disconnect, wasn't there significant misalignment?
Yeah, while. The basic highly visible metrics
like goal scored and maybe touches on the ball are
(38:17):
definitely rewarded financially.The market compensation.
Structure often seems to overlook or undervalue these
high impact, analytically superior actions like successful
presses and shot creating actions.
The very things the statistical models show are strongly
predictive of team victories, which signals the systematic.
Issue perhaps this executive bias.
(38:38):
We discussed a tendency for decision makers to reward the
simple, easily observed scoreboard outcomes like raw
plus minus scores, rather than digging into the more complex
analytics that reveal true isolated individual
contribution. Exactly.
And that. Inefficiency, that failure of
the market to correctly price analytical value, has directly
(39:00):
fueled this counter movement, the shift towards data.
Driven self representation by the players themselves, right?
Players like Kevin De Bruyne. Literally weaponizing custom
built data analysis to quantify their own indispensable worth
and to get predictive assurancesabout the team's future
trajectory. They are actively stepping in to
correct the market's failures and capture that value for
themselves. And this trend is?
(39:21):
Already evolving, moving from bespoke human analyst teams
towards more accessible, scalable tools like AI agents
assisting in negotiations, It's fundamentally reshaping the role
and necessity of the traditionalfootball agent.
But. And it's a huge.
But this rapid rush towards technology adoption.
Remember that 78% of sports organizations are already using
(39:42):
AI and talent management has created this massive governance
vacuum with only 32. Percent having comprehensive
policies in place, which leaves clubs and leagues.
Dangerously exposed to those bigThree legal and ethical risks,
we impact the thorny ambiguitiesaround intellectual property
ownership for AI models and dataderived insights, the serious
data privacy concerns, especially with advanced
(40:04):
biometric tracking, and the absolute imperative to identify
and mitigate algorithmic bias toensure fairness.
But crucially, there's a clear. Solution path.
Those organizations that are proactively implementing
policies and tablishing ethics committees, they're seeing
tangible benefits like that massive 40% reduction in related
legal disputes. Good governance isn't just.
Risk management. It's enabling better
(40:25):
performance, so the. Key knowledge application here
for you, the listener, whether you're a fan, work in the
industry, maybe even an aspiringplayer or agent, is crystal
clear. Success in this new era hinges
on understanding not just which statistics are being measured,
because everything is measured now, but understanding which of
those statistics are actually being valued and rewarded in the
(40:47):
marketplace. Because right now those two sets
of data, what wins games versus what gets paid are often
significantly misaligned. Mastering that analytical gap,
understanding that inefficiency,that's the difference between a
good contract and a potentially game changing one.
Absolutely. Which?
Leaves us with one final, perhaps slightly provocative
(41:07):
thought for everyone to Mull over after this deep dive.
OK, let's hear it. If the.
Data analysis. Like the research we've
discussed repeatedly shows that player salaries often don't
reflect the analytical actions that contribute most effectively
to winning games. Things like those crucial but
undervalued presses and shot creating actions.
Then, does the current misalignment and how pay is
determined actually inadvertently incentivize
(41:30):
players to optimize their own individual visible statistics?
Things like goals, or maybe those progressive carries that
look good but don't strongly predict wins?
Are players being subtly encouraged to play for a bigger
contract based on visible stats,rather than always prioritizing
the most effective but perhaps statistically unrewarded action
that truly benefits the team's chances of winning?
(41:52):
Wow, that's a. Profound question isn't it?
Does the way we pay players actually undermine team optimal
play because the incentives are skewed towards individual,
measurable but not always most effective actions?
Something to definitely think about.
Indeed. A huge.
Thank you for joining. Us and sharing your insights on
this fascinating topic today. And thank you all for listening
to the deep dive. We'll see you next time.
Welcome to the deep dive. Today we're really locking in on
the high stakes world of professional football, soccer
for some of our listeners, wherethese massive multi €1,000,000
contracts are decided. And we're seeing, well, a
revolution, aren't we? The whole traditional power
structure seems to be getting, you know, challenged by data, by
advanced analytics, even AI. Oh.
(00:21):
Absolutely. Probably the biggest shake up in
sports labor management since, well, since the modern agent
system really emerged. The sheer amount of money flying
around. Now it just demands a level of
analytical rigor that frankly didn't exist 1015 years ago.
But the really core question, the one we need to wrestle with
today, is this. When a player sits down for that
(00:42):
negotiation, that life changing contract, what is the club
actually paying for? Is the stuff we can easily see
the historical stats, or are they paying for future
productivity? The stuff that's scientifically
proven to lead to wins. And the players themselves, I'm
not just sitting back and waiting for the clubs to figure
it out, are they? This power dynamic is shifting.
(01:03):
I mean, yeah, really shifting. Definitely.
We've seen reports, for instance, that a former
Manchester United star didn't just rely on his agent.
He reportedly used ChatGPT, you know, the AI language model, as
a sort of negotiating assistant for his new contract.
Wow. That just blurs the lines
completely, doesn't it? Between human advice and, well,
(01:23):
computational leverage. It really does.
And think about what that means.You've got immediate accessible
bait, informed council available204 seven.
It's incredible, yeah. But even beyond that cutting
edge AI stuff, we have these really solid examples of how
structured data is being used right now, kind of a blueprint
for the future. You mean like Kevin Debrina?
(01:44):
Exactly like Kevin Debrina, the Belgian midfielder, absolutely
brilliant player. When he was negotiating his big
extension with Manchester City, he made a very deliberate
choice. No agent, no traditional agent,
No. Instead, he hired his own
dedicated team data analysts, toarm him basically for that
negotiation. That is just that's the ultimate
(02:05):
power move, isn't it? It tells you that players at
that very top level, they see the traditional way of doing
contracts as well inefficient, an informational gap they can
personally exploit. They're commissioning their own
custom built data to fight back against the club's internal
valuations, which are often, youknow, a bit of a black box.
(02:26):
Totally. So OK, our mission today then is
to really unpack this whole conflict.
We're going to look at some ground breaking new research.
It tests whether these advanced analytics actually influence how
much a player gets paid, right. We'll dissect this trend of
data-driven self representation,like the Debruin example.
And then, crucially, we have to dive into the huge and frankly
(02:48):
urgent legal and ethical hurdlesthat AI needs to clear if it's
going to manage the careers of these $1,000,000 athletes.
There's a lot there. A huge amount.
It's a fascinating intersection of economics, technology and
law. OK, let's jump straight into
Part 1 then the Moneyball test. And this isn't just about, you
know, cool stats for fans. This really starts with cold
heart economics. Labor Economics One O 1.
(03:10):
Exactly. The basic theory is that your
pay should reflect how productive productive you are.
That's the foundation. And professional football, where
literally every pass, every run,every touch is tracked and
quantified. It should be the perfect place
for this theory to play out the perfect marketplace.
It should be, yes, in theory, because performance is so
measurable and the outcomes, wins, losses, goals scored,
(03:33):
they're crystal clear. But there's always a but.
There's always a but yeah, past research trying to link pay and
performance in football really struggled.
There was one major methodological flaw that kept
cropping up. It was it was the reliance on
multi year contracts. They just muddied the waters
completely. OK, explain that a bit more.
Why do those long contracts makethe research so messy?
(03:56):
Well, think about it, if a player signs say a six year deal
today in 2024, the salary they earn way out in 2028 that was
decided back in 2024, it reflects their perceived value
then, not how well they actuallyplayed in the season just before
in 2027. OK, so there's a time lag A.
Huge time lag and it fundamentally weakens any
(04:18):
correlation you try to draw between their current
performance and their current salary in those later years.
Because the salary was locked inyears ago.
Precisely that noise, that distortion caused by those
predetermined salaries from previous years.
It made it almost impossible to isolate the true effect of
performance metrics on what a player gets paid right now.
(04:38):
So, OK, how did this new Economic Research get around
that problem? What was their clever fix?
Their methodological breakthrough.
They used a really sharp, elegant constraint.
They decided to focus only on players who were in the very
first year of a new contract A. Clean slate.
Exactly. A clean slate for each player
observation. They built up a pretty
(04:58):
significant sample to 992 playerseason observations.
Wow. Yeah.
Across two of the biggest, most high profile leagues in the
world, the the English Premier League and Italy, Syria.
And the data covered the seasonsfrom 20/18/19 up to 20/21/22.
That feels crucial. By isolating just that first
year, they can make a much tighter assumption.
Right, a very reasonable assumption.
(05:20):
That the players performance in the season immediately before,
say 20/21/22 is what directly drove the specific salary offer
they just signed for the 20/22/23 season.
Exactly. The link between recent
performance and the new salary is much, much clearer.
Direct even. OK.
And what about the salary data itself?
How specific specific did they get?
(05:41):
Because contracts have bonuses, endorsements.
Good question. They were meticulous about this.
The salary data used was the verified guaranteed base salary
gross of tax. The before tax.
The before tax, yes. And crucially, it excluded
performance bonuses and any external endorsement deals.
So just the core guaranteed annual income, what the club is
(06:01):
fundamentally saying that players worth per year.
That's it, the core value proposition from the club's
perspective. OK, fantastic, robust
methodology in place. Let's start with the basics
then, the stuff fans and commentators talk about all the
time. Goals, assists, maybe simple
pass completion. Do these basic stats actually
predict team success? You know, winning.
And, critically, are they rewarded with bigger paychecks?
(06:22):
Broadly speaking, yes, on the basic level, there's alignment.
The research looked at these common metrics as predictors for
two key things, predicting team points, basically winning games,
and predicting player salary. And things like goals, assists,
simple pass completion percentage, the number of
touches a player has, even the distance they carry the ball.
(06:42):
All of these showed a positive correlation with the team
getting more points. They help you win.
OK, that makes sense. And does the money follow?
Do players get paid for doing those?
Things, yes. Strong alignment here too.
Players are overwhelmingly, significantly rewarded for goals
scored. That's the big one,
unsurprisingly. The glamour stat.
Absolutely, but they also see noticeable positive returns in
(07:06):
their salary for racking up assists, for having a high pass
completion rate, and even just for the sheer volume of touches
they get on the ball. So the most obvious, most easily
televised metrics. They definitely influence the
executives signing the checks. It confirms that yes, what's
easily seen and counted gets rewarded.
But even at this basic level, there was a really surprising
(07:28):
finding, wasn't there? Especially for defenders, it's
about the humble tackle. Ah yes, this one really jumps
out because it flies completely in the face of conventional
wisdom. You know, the idea of effort,
Defensive grit. Right, getting stuck in.
Exactly. The research found that the
number of successful tackles a player makes well in Turns out
(07:49):
it's not a significant predictorof team points.
It doesn't statistically help the team win more.
Wow, OK. And maybe even more
surprisingly, it's also not rewarded in their salary.
It was statistically insignificant in determining
pay. That's that's almost heresy in
football culture. You picture a midfielder making
10 crunching tackles in a game and you think the guy's earning
(08:12):
his money? How can the market just ignore
that? Well, it forces you to think a
bit more sophisticatedly about what actually constitutes good
defending. OK, the strong suspicion, the
conjecture coming out of this finding is that having to make a
lot of tackles might actually bea sign of failure earlier in the
defensive process. Meaning maybe the team's
positioning was poor or a teammate made a mistake further
(08:33):
up the pitch, forcing that player into a last ditch tackle
situation. So the tackle is fixing a
problem, not necessarily representing proactive dominant
defending. Potentially optimal, really high
value defensive work is often about anticipation.
It's about reading the game, cutting out passing lanes,
intercepting passes, preventing the need for a tackle in the
(08:55):
first place. And you don't really get a
flashy stat for intercepting a pass that would have led to a
dangerous situation 2 passes later.
Precisely that kind of invisibledefending is incredibly
valuable, but doesn't show up inbasic tackle counts.
So the data suggests executives are rewarding the tangible,
aggressive actions that touches the goals, but maybe missing the
(09:17):
value of that less visible but arguably more effective
defensive contribution. That's a really powerful
insight, and it perfectly sets the stage for this idea of the
analytics gap, doesn't it? If the basic stats are
potentially flawed or incomplete, then let's move into
the real Moneyball territory, the advanced analytics.
We're talking expected goals, XG, progressive passes, shot
(09:39):
creating actions. These are the metrics close
supposedly use now to find valuethe market is missing.
Right, this is where the economic theory suggests
inefficiencies might lie, and wereally see that bear out in the
findings. And just a quick point on
expected goals XG. It's crucial because it moves
beyond just counting goals. It gives you a normative
appraisal. It calculates how many goals a
(10:00):
player should have scored given the quality and location of the
chances they had. It accounts for difficulty.
Right, a much smarter measure than just the raw goal tally.
So armed with these advanced metrics, what did the research
find? Are the things that really
predict winning being rewarded financially?
This is where the misalignment becomes stark.
There are actions scientificallyproven to contribute
(10:22):
significantly to winning, but the research shows they are not
being adequately rewarded in player salaries.
OK, give us the prime examples. What wins games but doesn't seem
to boost the paycheck? Two metrics really stood out
here. They were strong predictors of
team points, meaning they genuinely help teams win, but
they showed weak or statistically insignificant
(10:44):
evidence of actually being reflected in player salaries.
And they are. Successful presses and shot
creating actions. Wow, let's break those down.
Successful presses. That's winning the ball back
high up the pitch, right? Often leading to immediate
attacking chances. Exactly, it's fundamental to
modern high intensity football styles.
Think Liverpool under Clop or Guardiola City that pressing is
(11:06):
critical. And yet the player executing
that successful press, winning the ball in a dangerous area,
they don't see a significant financial return for that
specific action. The evidence suggests not nearly
as much as they should. Based on its contribution to
winning, it seems undervalued inthe pay structure.
Yeah, and the other one shot creating actions.
(11:27):
Yeah, this is another crucial 1Ashot.
Creating action is defined as one of the two offensive actions
directly preceding a shot. So it could be the pass before
the shot, or the dribble that beat a defender before the shot,
or drawing a foul that leads to a shot.
Basically the actions that directly generate scoring
opportunities. The engine room of the attack?
Really. Absolutely.
And again, while these actions strongly predict team success,
(11:50):
the research found much weaker evidence that they are being
rewarded appropriately in salaries.
So clubs are in effect systematically underpaying for
these incredibly high impact team success driving actions.
That's what the data points towards.
It highlights a clear inefficiency, an opportunity
that clubs, or at least their compensation models, seem to be
(12:10):
missing. Oh, and progressive passes to
moving the ball significantly closer to the opponent's goal.
They also predicted wins but weren't significantly linked to
higher pay. OK, now let's flip that.
What is being highly rewarded with a big contract, but doesn't
actually seem to predict wins all that well according to the
statistics? This is where that executive
(12:31):
bias you mentioned might become visible.
Yes, the primary example that emerged here was progressive
carries. OK, define that for us.
A progressive carry is when a player successfully moves the
ball forward towards the opponent's goal by dribbling or
carrying it covering a certain distance, usually at least 5
yards, or carrying it into the penalty area.
(12:52):
So running with the ball effectively up the pitch.
Exactly. And the analysis found that yes,
progressive carries are valued by teams.
Players who do this well tend toreceive higher salaries.
OK. That sounds reasonable.
But, and this is the key finding, yeah, the exact same
research found that progressive carries did not significantly
(13:12):
predict team wins. Wait, so running with the ball
up field gets you paid more, butit doesn't actually correlate
strongly with your team winning more games?
That's what this specific analysis found, yes.
It's the mirror image of the pressing and shot creation
findings. That's the exact opposite of
what pure data-driven decision making would dictate, isn't it?
Why would this action, which often looks exciting, involves
(13:35):
flair, gets fans off their seats, be rewarded with premium
pay if it doesn't reliably contribute to winning?
Well, this speaks volumes about what we're calling the executive
biases. Progressive carries are highly
visible. They're exciting to watch.
They often end up on highlight reels.
They contribute to a player's individual reputation for
dynamism and skill. They look good.
They look. Good.
(13:56):
And the suggestion is that club executives, maybe
subconsciously, maybe even consciously to some extent, are
rewarding that flair. They're rewarding the visible
output that appeals to the fan base, maybe helps sell jerseys,
rather than the more nuanced, perhaps less visually dramatic
but statistically more effectiveactions like successful pressing
or smart shot creating passes. So it's marketability and
(14:19):
perceived individual brilliance over the statistically proven
contribution to actually winningmatches.
That seems to be a strong possibility based on this
evidence, rewarding what looks impactful over what is impactful
according to the deeper analytics.
And this whole idea, this bias towards the visible and the
simple, it kind of culminates inthe most significant finding
(14:39):
regarding how executives might be making these decisions,
doesn't it? The finding around plus minus
goals. Ah yes, this is perhaps the
clearest indicator, the sort of smoking gun confirming this
potential bias towards simple outcomes.
OK. The researchers looked at a very
basic metric, plus minus goals. This is simply the team's goal
(14:59):
difference, goal scored minus goals conceded, but only during
the time that specific player was actually on the pitch.
Right, if I'm on the field for 70 minutes and my team scores 2
and concedes one while I'm playing, my plus minus for that
game is +1. Exactly, and unsurprisingly
perhaps, this simple plus minus metric was found to be
(15:19):
significant in predicting both player pay and team points.
OK, well, that seems reasonable on the surface, doesn't it?
If I'm on the pitch and the teamis generally outscored scoring
the opponent, surely I deserve ahigher salary.
It suggests I'm contributing to success.
It sounds logical, yes, but the implication is actually quite
profound when you dig deeper. How so?
It suggests that club executivesmight be taking a very
(15:42):
consequentialist approach. They're rewarding the player
simply for being present while the team is successful.
Rewarding success by proximity. Essentially, yes.
Rewarding presence during good outcomes, and this highlights a
potential fundamental inability or perhaps unwillingness to
correctly separate that individual players actual
(16:03):
productivity from the success generated by the high performing
teammates who might be around them.
So I might just be lucky to be playing alongside 3 superstars
who are carrying the team, but Istill get the pay bump because
the team wins when I happen to be on the field.
That's the risk inherent in relying on such a simple metric.
It doesn't isolate individual contribution effectively.
(16:23):
But how do we know this isn't just a really good simple metric
that magically captures everything important?
Maybe plus minus is just efficient.
That's a fair question, but we know it's likely not the full
picture because the researchers specifically tested a superior,
analytically much sounder version of the same basic idea.
Which was? Expected goals plus minus.
(16:43):
This metric does the same thing.Looks at the difference while
players on the pitch, but instead of raw goals it uses
expected goals. So it accounts for the quality
of chances created and conceited, opponent strength,
game state, all that complex stuff.
Exactly. It's far more informative about
a player's true net impact than just looking at the raw
(17:04):
scoreboard outcomes. And what did they find when they
tested that sophisticated metricagainst player salaries?
Crucially, the research found that this more sophisticated
metric, expected goals plus minus, was not significant in
determining player salaries. Wow, okay, let that sink in.
The decision makers, the executives, they have access to
the complex, more accurate data that isolates a players true net
(17:27):
impact on goal difference, but they seem to ignore it when
setting salaries. Instead they rely on the simple
scoreboard based metric, the rawplus minus.
That's the strong implication. It really confirms this
potential bias towards easily digestible results, even if
those results are frankly informationally quite shallow
compared to what's available. They're looking at the final
(17:49):
score rather than the sophisticated model that
explains how that score was likely achieved and who drove
it. Precisely, it suggests a
preference for outcome over process, analytically speaking.
OK, fascinating stuff. On the metrics, Before we move
on from salary determinants, what about the other
foundational factors? Things like age, experience,
maybe how a player moved clubs. Yeah, those baseline factors are
(18:12):
still very important. As you'd expect, age followed
the classic pattern we see in many professions, actually.
The inverted U curve. Exactly.
Age has a positive effect on salary, but it's a diminishing
effect. The returns get smaller as
players get older. So experience helps, but only up
to a point. Right.
And the data strongly suggested that the optimal time for a
(18:33):
player to sign that major careerdefining contract, the point of
highest earning potential, is typically reached around 26
years old. 26 That's the magic number for the big payday.
It appears to be the peak, yes. After that age, salary growth
tends to slow down, likely reflecting the market's
perception of the approaching physical decline, even if it
(18:54):
hasn't happened yet. So if you're a top player
hitting 28 and you haven't secured that massive long term
deal, you've potentially missed the absolute peak earning
window. That's what the data suggests,
yeah, timing is critical. What about experience and
reputation? Playing for your country?
Playing in top leagues. Clear premiums there, yeah.
Being a senior international player represent your national
(19:17):
team regularly provides a significant salary boost, which
confirms that clubs pay extra for that established reputation,
that pedigree and maybe also forunobserved qualities like
leadership, handling pressure, things associated with
international experience. Makes sense and playing in other
big leagues before. Also a positive factor,
experience accumulated in other top European leagues, say coming
(19:39):
from La Liga or the Bundesliga to the Premier League,
contributes positively to the salary offered.
It signals proven quality at a high level.
OK. And one last finding related to
player movement. What about players who move on a
free transfer when their contract expires and they're
free to sign anywhere? This was another compelling
finding, and it actually mirrorswhat we see in other sports
(20:01):
labor markets. Like baseball.
Players who moved clubs on a free transfer appear to receive
a measurable salary penalty. They seem to get paid slightly
less on average than similar players who moved for a transfer
fee. Really.
Why would that be? You'd think being free would
give them more bargaining power.You might think so, but the
common interpretation is about signaling.
(20:23):
Signaling. Yeah, the players former club,
by letting their contract expireand allowing them to leave for
free, is essentially signaling to the entire market that they
weren't desperate to keep that player.
They were content to let them walk.
Exactly. It implicitly suggests that the
previous club felt they had or could find better alternatives.
(20:44):
That negative signal, that lack of strong interest from the
selling club, potentially dampens the salary offers the
player receives from their new club.
Interesting. It's like a subtle mark against
their perceived value. It seems that way, yes.
A small but statistically noticeable effect.
OK. So to wrap up Part 1, we've
established this clear, measurable inefficiency in the
(21:06):
football transfer market. Clubs often fail to pay
appropriately for the analyticalactions that genuinely predict
winning games. They seem to favor basic stats,
visibility, and simple proximityto success.
In this market failure, this gapbetween analytical value and
actual pay, it creates this massive vacuum, an opportunity
(21:26):
which, as we'll see in Part 2, the players themselves are now
starting to exploit in really interesting ways.
Let's move on to that, the new negotiating edge.
So yeah, if the clubs are being,let's say, analytically
inefficient in how they value players, that creates a
fantastic arbitrage opportunity for the players themselves.
Arbitrage, meaning they can profit from that inefficiency.
Precisely. And this seems to be a major
(21:47):
driver behind this growing trendwe're seeing towards self
representation, or at least moredirect involvement by players in
their own negotiations. Right.
Historically the model was pretty simple, wasn't it?
The player hires a single agent,and that agent handles basically
everything, finding clubs, negotiating contracts, maybe
even commercial deals. The player delegates.
(22:08):
That was the standard model, yes.
Delegate all the sporting, commercial and financial
interests to one person or agency.
So why is that model now being seen as potentially risky by
some top players? Well, I think the smartest
players and their advisors are realizing the complete
delegation means giving up a huge amount of informational
(22:28):
power. OK, there's a potential conflict
of interest too. Does a single agent always
prioritize maximizing this one specific contract for the
player, even if it means pushingthe club really hard?
Or might they sometimes prioritize maintaining their own
long term, smooth relationship with that club because they'll
have other players to place there in the future?
The repeat business incentive for the agent might conflict
(22:50):
with the players immediate best interest.
It's a potential concern, yes. So increasingly you see players
wanting more control. They're managing their careers
more actively themselves, maybe using specialized lawyers for
the legal aspects involving family members they trust.
And then hiring specific expertslike data analysts only when
they need that targeted expertise.
(23:11):
Exactly, it's a more modular, player controlled approach.
And the ultimate highest profileexample of this, the one that
really showed how data could be weaponized in a negotiation, is
Kevin De Bruyne. We need to dissect this case
properly. It really is the blueprint,
isn't it? A landmark case When Debruyn was
negotiating his big contract extension with Manchester City a
(23:32):
few years back, he completely bypassed the traditional agent
route. Just didn't use one.
Didn't use one. Instead, he commissioned his own
specialized, bespoke team of data analysts specifically for
this negotiation. So this wasn't just like looking
up his stats on a website. Oh no, absolutely not.
This is a dedicated analytical operation working directly for
(23:53):
him. What were the main tasks he gave
this analytical team? What did he want them to figure
out? The analysis they did was
brilliant really. It had two main prongs, 2 key
objectives. OK, what was the 1st?
The first was maybe obviously torigorously quantify his
individual performance in a pack.
Proving his own value. Exactly, using advanced metrics
(24:15):
linking his specific actions on the pitch, passes, chances
created, positioning everything directly to City's probability
of scoring, winning games, achieving success.
Basically proving with hard dataindispensable he was to the way
City played in one OK, makes sense use.
The club's own language data back at them to prove your
worth. But you said there were two
prongs. What was the second?
(24:36):
This is the really innovative part.
I think it really is the 2nd. Prong was a predictive analysis
of the team's continued chances of success moving forward, not
just about him. But about the whole team's
future? Yes, his analysts assess.
The Manchester City squad's age profile, the quality trajectory
of key players, the potential impact of planned signings and
even the likely strength of their main domestic and European
(24:58):
rivals over the next few years. Wow.
So he wasn't just negotiating his salary based on past
performance, he was negotiating based on the probability that
the club would remain competitive and successful
during the lifespan of his new contract.
That's exactly it. It was an incredibly strategic
move. He was essentially saying, OK,
you want me to commit my peak years to this club, Prove to me
(25:21):
with data that this club is structured to continue competing
for major trophies. Show me the numbers that justify
my commitment. He was demanding data.
Driven assurance about the club's future, not just his own
pay packet precisely he. Wanted to ensure that if he
signed, he wasn't just signing for money but for a continued
high probability of winning things and the outcome well.
(25:42):
The data, presumably. Backed up both his individual
value and the club's strong prospects, he secured a
significant pay rise and the contract extension he wanted.
And it was predicated not just on his brilliant past
performance, but on this data-driven forecast of the
club's trajectory. It's the perfect example of a
player actively commissioning specific analysis to completely
(26:04):
neutralize the informational advantages the club usually
holds in negotiations if the club tried to lowball.
Him using simplistic arguments or basic stats he could counter
with his own. Bespoke high end predictive
analysis proving exactly why he was worth more and why staying
was a good bet for success. It fundamentally flips the power
(26:25):
dynamic totally the club is usedto.
Being the only side in the room with the really sophisticated
spreadsheets and predictive models, now the player walks in
with their own custom built and highly persuasive valuation
model. It's a game changer.
And this transition from needinga bespoke team of human analysts
like Dabrina had to potentially using more scalable AI tools,
that's happening right now. Back to the ChatGPT.
(26:47):
Example exactly. That report about the former
Manchester United player using an AI like Chachi PT to assist
in contract negotiations. It confirms that the kind of
specialized analysis. De Bruyne Commission is already
merging with these readily accessible AI tools to the AI
can provide the. Speed.
The data synthesis? Maybe.
The analytical breadth that evena dedicated human team might
(27:10):
struggle to pull together instantly?
Potentially, yes. It can process vast amounts of
data, identify patterns, maybe even simulate negotiation
scenarios based on known variables.
It's becoming another tool in the players negotiating toolkit,
which means the traditional. Agent whose primary value
proposition was often based on relationships, access, and maybe
more subjective negotiation skills.
(27:31):
Their role is rapidly changing. It really is.
They're becoming perhaps more like strategic advisors who
manage a team of specialists, lawyers, financial advisors,
data analysts, maybe even AI prompts.
Rather than being the sole font of all knowledge and negotiation
power, they might become optional intermediaries,
supplemented or eventually even replaced by code and data in
(27:52):
some functions. But this explosion?
Of data use, players, commissioning, analysis, clubs
using AI for scouting and performance.
It leads us directly into this massive, complex minefield
regulation and ethics. Which brings us to Part 3, the
AI regulatory reality check. Because the tech is moving way
(28:12):
faster than the rules. OK, so AI is booming in sports
management. We're seeing it used everywhere
from talent scouting, identifying the next big star to
predicting injury risks, even automating aspects of training
schedules. The adoption rate is just
explosive. It really is.
But. As you say, the legal and
ethical frameworks needed to actually govern this powerful
(28:32):
technology, yeah, they are lagging dangerously behind.
How far behind do we have numbers?
On this adoption versus policy gap, we do and they're quite
stark. Frankly, a bit alarming.
A recent survey looked at about 150 major sports organizations
globally. OK, they found a staggering 78%
are already using AI tools in some capacity for talent
management. There's nearly four out of every
five organizations relying on predictive analytics, automated
(28:55):
scouting, or similar AI applications 78%.
Adoption. That's huge.
A massive technological shift ina very short time O the crucial
question, how many of those organizations actually have
comprehensive, well defined policies in place to govern how
they use that AI? That's where the gap is.
Shocking. Only 32 percent, 32, so almost.
(29:17):
80% are using it, but less than 1/3 have clear rules for it.
Exactly that. Massive gap, 78% usage versus
only 32% with established policies represents an enormous,
largely unmitigated legal and ethical exposure for these clubs
and organizations. Wow.
The urgency to adopt the new tech, you know, to in that
competitive edge on the pitch orin the transfer market, it has
(29:39):
completely outrun the organization's ability to manage
the risks that come with it. Risks around data privacy,
intellectual property, algorithmic bias.
It's a long list. OK, let's breakdown those risks.
What are the main legal and ethical challenges that are
really keeping the lawyers and executives at these sports
organizations awake at night right now?
What are the top concerns the survey identified?
(29:59):
Three primary areas of major concern.
The number one issue cited by 72% of the organization's is
intellectual property rights IP.OK.
How does that play out specifically in sports AI?
It sounds complex. It's incredibly nuanced.
Think about it, who actually owns the intellectual property
for the insights generated by anAI?
(30:19):
Who owns the algorithms themselves, especially if
they're constantly learning and adapting based on player data?
Who owns the predictive models derived from analyzing years of
player performance data? Right.
Give us a concrete example of. Where this gets sticky?
OK, let's say a premier. League Club hires an external AI
company to develop proprietary machine learning model.
(30:40):
This model predicts a young player's potential or their risk
of specific injuries based on tracking their movement patterns
and training using wearable sensors.
Now, if that player eventually leaves the club and signs for
arrival, who owns the IP relatedto the insights generated from
that specific player's unique biometric data?
Using that model, does the club retain rights over analysis
(31:02):
derived from data collected while he was their employee?
What if the player like De Bruyne commissions his own AI
driven valuation model using data from his matches?
Does he own the resulting IP or does the underlying performance
data technically belong to the league or the club since it was
generated during official competitions or training?
Yeah, I can see the problem. It's.
(31:23):
Incredibly ambiguous. Who owns the analysis of the
performance exactly and? This ambiguity creates a
potential legal time bomb. You can easily foresee future
disputes, arbitrations, maybe even lawsuits over the ownership
and use of these valuable AI generated insights and the
underlying data. OK, IP is number.
One, what was the second major concern, the second sided by?
68% revolves around data privacyconcerns.
(31:46):
This is huge, especially Europe with GDPR but relevant globally.
General data protection. Regulation, right we're talking
about. Managing incredibly sensitive,
often hyper specific athlete data.
Performance metrics are one thing, but AI often relies on
much deeper data. This is where the wearable.
Tech and the advanced tracking systems really complicate
(32:06):
things. Isn't it the stuff players wear
in training now? The cameras tracking every
movement precisely. We already have some precedent
in North American sports like the NBA, where there are
specific rules prohibiting the use of data from wearables in
contract negotiations, largely to protect players.
But in top level football, theseadvanced tracking systems are
(32:26):
becoming ubiquitous. They capture biometric data,
heart rate, acceleration, deceleration, fatigue markers,
incredibly complex movement profiles.
Clubs need explicit informed consent for collecting this, and
more importantly, they need extremely careful governance
protocols dictating how this sensitive data can be collected,
how long it could be stored, whocan access it, and specifically
(32:47):
how it could be used, especiallyif it might lead to decisions
affecting a player's contract, they're playing time, their
financial future, or even their employment status.
You can imagine scenarios. Where AI flags a potential long
term injury risk based on subtlemovement changes and a club
might hesitate to offer a long contract.
That's a minefield, a huge minefield.
The need for advanced data anonymization techniques where
(33:09):
possible and robust security protocols is absolutely
paramount to comply with privacylaws and maintain player trust.
OK, so IP. And data privacy.
What was the third major legal and ethical challenge
identified? The third one cited by. 61% is
the potential for bias in decision making.
Algorithmic bias. The AI.
(33:30):
Inheriting human biases exactly this speaks.
Directly to the legal and ethical imperative for these AI
systems to ensure fairness and avoid discrimination.
How might that bias creep? In, well, imagine an AI.
Scouting tool that's trained primarily on historical scouting
reports and performance data from, say, the last 20 years.
If the human scouts and managerswho generated that historical
(33:50):
data had unconscious biases, maybe they tended to overvalue
players from certain countries or leagues, or players with a
particular physical profile while undervaluing others.
The AI learning from that data will simply learn, replicate,
and potentially amplify those historical biases.
So the AI wouldn't be. Objective.
It would just automate the old prejudices at scale.
(34:12):
That's the risk. It could lead to unfairly
filtering out perfectly capable players from underrepresented
backgrounds or those who don't fit the historical mold simply
because the training data was skewed.
And legally, that's obviously. Problematic, but practically
too. You miss out on Absolutely.
And this demands transparency. It underlines the need for
what's called explainable AI, orXAI.
(34:35):
Clubs and organizations need to be able to understand why the AI
made a particular recommendation, whether it was
flagging a player to sign, suggesting a transfer value, or
even monitoring training intensity.
They can't just hide behind, thealgorithm told us so they need
to be able to audit the decisionmaking process to ensure it's
fair and justifiable. So faced with these.
Huge challenges, IP privacy bias, putting actual policies in
(35:00):
place moves from being just, youknow, good governance or nice to
have to being an absolutely. Critical risk mitigation
strategy. It's essential.
Do we have concrete proof? That having these policies
actually works. Does it tangibly reduce the
risks? Yes, we do.
The survey provided some really compelling quantitative evidence
on this point. Organizations that had already
(35:20):
implemented robust proactive AI policies, the 32% who are head
of the curve, they reported a demonstrable 40% reduction in
legal disputes related to talentmanagement compared to those
without such policies. 40% reduction.
That's that's not a small number, that's significant.
It's a massive difference. It's clear evidence that taking
the time to establish pre emptive legal and ethical
(35:43):
governance frameworks drastically reduces the
likelihood of ending up in costly litigation, arbitration
or regulatory trouble down the line.
That 40% figure. Makes a very strong financial
case for investing time and resources into developing these
policies, doesn't it? It's not just about doing the
right thing. It saves money and hassle,
absolutely. And.
Interestingly, the ethical governance side seems to tie
(36:04):
directly to performance outcomestoo, not just risk reduction.
How so? The research.
Also found. A strong positive correlation.
They quantified it with an R value of .71 between
organizations establishing dedicated AI ethics committees
or review boards and their success and talent acquisition R
value of point. 7/1 for listeners may be less familiar
(36:25):
with statistics that indicates avery strong positive
relationship, right? Almost undeniable cause and
effect. It's a very strong.
Correlation. Yes, it suggests that
organizations actively grapplingwith the ethical implications of
AI setting up oversight structures, they're also the
ones having more success in actually identifying, valuing,
and securing the talent they need so ethically governed.
AI isn't just safer legally, it might actually be more effective
(36:48):
at its job, perhaps because the internal systems are less likely
to be thrown off by hidden biases or flawed data practices.
That's a very plausible. Interpretation Good ethics leads
to better, fairer and ultimatelymore successful AI
implementation in this context. So what's the?
Way forward, then. How do sports organizations
navigate this? It requires a proactive.
(37:10):
Approach implementing adaptive regulatory frameworks,
internally enhancing data governance protocols, constantly
investing in advanced data anonymization and security, and
crucially, providing continuous legal and ethical education for
staff involved in using these AItools.
It's not a one off policy. Document.
It's an ongoing process, exactlythe.
(37:32):
Organizations that are going to thrive in this new data-driven,
AI influenced era of sport will be the ones that view data
security, pretty ethical compliance and robust
governance. Not as inconvenient burdens, but
as core strategic assets. They are fundamental to
sustainable success. OK, let's try and synthesize.
This whole deep dive then, it's been quite a journey.
We started with that fundamentaleconomic theory, pay should
(37:54):
reflect productivity. And we looked at brand new
research using that clever methodology, focusing only on
players in the first year of newcontracts in the Premier League
in Syria to really test that theory in football.
And what we found was. Well, a pretty profound
disconnect, wasn't there significant misalignment?
Yeah, while. The basic highly visible metrics
like goal scored and maybe touches on the ball are
(38:17):
definitely rewarded financially.The market compensation.
Structure often seems to overlook or undervalue these
high impact, analytically superior actions like successful
presses and shot creating actions.
The very things the statistical models show are strongly
predictive of team victories, which signals the systematic.
Issue perhaps this executive bias.
(38:38):
We discussed a tendency for decision makers to reward the
simple, easily observed scoreboard outcomes like raw
plus minus scores, rather than digging into the more complex
analytics that reveal true isolated individual
contribution. Exactly.
And that. Inefficiency, that failure of
the market to correctly price analytical value, has directly
(39:00):
fueled this counter movement, the shift towards data.
Driven self representation by the players themselves, right?
Players like Kevin De Bruyne. Literally weaponizing custom
built data analysis to quantify their own indispensable worth
and to get predictive assurancesabout the team's future
trajectory. They are actively stepping in to
correct the market's failures and capture that value for
themselves. And this trend is?
(39:21):
Already evolving, moving from bespoke human analyst teams
towards more accessible, scalable tools like AI agents
assisting in negotiations, It's fundamentally reshaping the role
and necessity of the traditionalfootball agent.
But. And it's a huge.
But this rapid rush towards technology adoption.
Remember that 78% of sports organizations are already using
(39:42):
AI and talent management has created this massive governance
vacuum with only 32. Percent having comprehensive
policies in place, which leaves clubs and leagues.
Dangerously exposed to those bigThree legal and ethical risks,
we impact the thorny ambiguitiesaround intellectual property
ownership for AI models and dataderived insights, the serious
data privacy concerns, especially with advanced
(40:04):
biometric tracking, and the absolute imperative to identify
and mitigate algorithmic bias toensure fairness.
But crucially, there's a clear. Solution path.
Those organizations that are proactively implementing
policies and tablishing ethics committees, they're seeing
tangible benefits like that massive 40% reduction in related
legal disputes. Good governance isn't just.
Risk management. It's enabling better
(40:25):
performance, so the. Key knowledge application here
for you, the listener, whether you're a fan, work in the
industry, maybe even an aspiringplayer or agent, is crystal
clear. Success in this new era hinges
on understanding not just which statistics are being measured,
because everything is measured now, but understanding which of
those statistics are actually being valued and rewarded in the
(40:47):
marketplace. Because right now those two sets
of data, what wins games versus what gets paid are often
significantly misaligned. Mastering that analytical gap,
understanding that inefficiency,that's the difference between a
good contract and a potentially game changing one.
Absolutely. Which?
Leaves us with one final, perhaps slightly provocative
(41:07):
thought for everyone to Mull over after this deep dive.
OK, let's hear it. If the.
Data analysis. Like the research we've
discussed repeatedly shows that player salaries often don't
reflect the analytical actions that contribute most effectively
to winning games. Things like those crucial but
undervalued presses and shot creating actions.
Then, does the current misalignment and how pay is
determined actually inadvertently incentivize
(41:30):
players to optimize their own individual visible statistics?
Things like goals, or maybe those progressive carries that
look good but don't strongly predict wins?
Are players being subtly encouraged to play for a bigger
contract based on visible stats,rather than always prioritizing
the most effective but perhaps statistically unrewarded action
that truly benefits the team's chances of winning?
(41:52):
Wow, that's a. Profound question isn't it?
Does the way we pay players actually undermine team optimal
play because the incentives are skewed towards individual,
measurable but not always most effective actions?
Something to definitely think about.
Indeed. A huge.
Thank you for joining. Us and sharing your insights on
this fascinating topic today. And thank you all for listening
to the deep dive. We'll see you next time.
Popular Podcasts
Stuff You Should Know
If you've ever wanted to know about champagne, satanism, the Stonewall Uprising, chaos theory, LSD, El Nino, true crime and Rosa Parks, then look no further. Josh and Chuck have you covered.
Dateline NBC
Current and classic episodes, featuring compelling true-crime mysteries, powerful documentaries and in-depth investigations. Follow now to get the latest episodes of Dateline NBC completely free, or subscribe to Dateline Premium for ad-free listening and exclusive bonus content: DatelinePremium.com
The Breakfast Club
The World's Most Dangerous Morning Show, The Breakfast Club, With DJ Envy, Jess Hilarious, And Charlamagne Tha God!