Players to sue for hundreds of millions over use of their statistics | Hacker News Reader

Players to sue for hundreds of millions over use of their statistics

theathletic.com

95 pointsTomte95 comments

I'm not a lawyer, and I'm especially not a British lawyer, but I do have an Athletic subscription. The article discusses a lawsuit on behalf of Primer League and mostly lower league soccer players in England around their "personal data."

> More than 400 current and former players have signed up to pursue gaming, betting and data-processing companies who utilise their personal statistics without consent or compensation.

As others have said, at least in the states, facts are not copyrightable. TA also states the lawsuit isn't going about this as part of image rights for the players, but doesn't say exactly what the argument will be. It makes it seem this is driven by lower league players who obviously don't have as lucrative careers.

Again IANAL but I don't see how if I go to a game and compile statistics myself how that's a breach of the players' data. Maybe if the club had a contact when I bought the ticket, but then I would think the players would need an agreement with the clubs to make that clear, because I'm not buying the ticket from the player I'm buying it from the club. Same with TV rights, that's league, club, and TV station rights, not player rights.

I think another argument is that the game is enjoyed in many forms, over radio, over television, streamed over the internet - and all of those rights must be purchased, etc. If the game statistics are another form of enjoyment, then it is another form, and should be subject to the same ownership issues as the other channels.

OTOH, I think the argument is very weak. First, precedent is strongly against the players here. Player stats have never been licensed to my knowledge. It's not clear if this is for lack of trying, or if the market for that data used to be superfans, and it was too small to matter.

That said, in any argument I tend to side with the underdog. I think it would be great if Athletic voluntarily shared some of their revenue with players! It would be a good move for them, because it would take wind out of the sails of the counter-parties, and it wouldn't acknowledge the players right to their data, except tacitly.

> and all of those rights must be purchased

this have been tried for Chess, and deemed not applicable. Others are allowed to do live coverage of the moves without purchasing any rights. The organizers control the live footage, can ban them from entering the premises to interview players etc., so most will adhere to some form of contract. But if you have no ties there's nothing stopping you from making your own content based on purely the moves being made.

Update: Specifically, the moves are not copyrightable https://chess24.com/en/read/news/us-judge-agrees-with-chess2...

It also mentions "NBA vs. Motorola" in which NBA didn't own statistics of NBA games and others were allowed to use them.

Interesting. Makes me wonder if you could train a machine learning model to take one or more football match video streams as input and turn them into a live 3d model. You then argue that the model simply documents the facts of the game, so does not infringe on the copyright of the original source streams. Then you sell live footage of the model (perhaps with slightly different camera angles)

Makes me wonder if you could train a machine learning model to take one or more football match video streams as input and turn them into a live 3d model.

It's been done.[1] Not for American football, for soccer.

[1] https://grail.cs.washington.edu/projects/soccer/soccer_on_yo...

> does not infringe on the copyright of the original source streams

Unless you're getting the raw video stream from all the cameras, it'll be transformed through the producer and director's creative control and that, I suspect, would block you from doing this - it's not longer just "facts", it's a particular interpretation of those "facts" (cf bare recounting of historical events vs someone's book covering the same, I suppose.

> it'll be transformed through the producer and director's creative control and that, I suspect, would block y

I don’t know about that. You’re using the directors production to establish the fact but once you have extruded those facts, it could be argued in principal of not legally that this is a new creative work based off the mere facts.

The "mere facts" would be a wide-angled camera showing the entire pitch and very few people would watch that, I think. Using your judgement as a director/producer to use different angles, cameras, positions, etc. to provide a more interesting spectacle would, for me (IANAL), be transformative enough.

Consider an analogy to classical music - Beethoven's 5th isn't copyrighted but a particular expression of it by an orchestra can be.

It would be a case to be heard in court no doubt. It could be argued that the footage is but one corroborating account of events in the dream theatre. All it would take would be to independently capture some off camera activity. Director has to assert that somehow your work is derivative. I’m not a lawyer of course and if you are I defer to your expertise. It’s an interesting thought experiment none the less.

Where they could get you is how they can protect their "likeness." If you reduced the simulation to only functional aspects of the game, then I don't see how it'd be a problem.

"in any argument I tend to side with the underdog. I think it would be great if Athletic voluntarily shared some of their revenue with players!"

The underdogs here are the consumers, who ultimately wind up paying all those hundreds of millions.

> Player stats have never been licensed to my knowledge.

Of course they are, that's the entire business model of data analytics companies - the customers are:

- clubs and national teams themselves (e.g. German national team coaches Klinsmann and Löw were famous for early adopting data-driven training)

- sports betting services, casinos and similar enterprises

- TV and radio stations so that the commenters can (at an instant) pull facts like "player xyz has a 80% successful pass rate over the last 30 games"

This stuff is called "soccer analytics", the (German) Wikipedia has a decent article: https://de.wikipedia.org/wiki/Soccer_Analytics

In those instances, nobody is licensing the statistics themselves, they are licensing access to the statistics databases that those organizations have compile and maintain. At least that's my understanding.

Facts are not subject to copyright in the U.S. However, there are other forms of intellectual property. In Illinois, where I am barred, there is a right of publicity - exclusive ownership over your identity when used for a commercial purpose. Could that extend to your name and unique statistical data when used as part of a for-pay entertainment endeavor? Maybe. I'm not aware of any case law one way or another, as I don't practice in IP anymore.

However, my point is that it doesn't seem completely crazy to me to think maybe the UK has some basis in law for this suit. It'll be interesting to keep an eye on it.

> facts are not copyrightable

This is the nail in the coffin. Imagine the lengths lawyers would go if they could monetize any abstract reference to something.

The results of my last STD test and the transaction history of my bank account are also facts. We have decided some facts deserve specific protections, at least in the US. So I don't think we can dismiss the players here with "facts are not copyrightable" without knowing what specific statistics are being recorded and to which the players reject. Sports technology has exploded over recent years and the data collection of athletes is now an industry of its own. There can be a range in the degree of ownership that a player deserves from the total number of goals scored in public matches to the heartrate of a player throughout a private practice session.

Those facts (STD test result and banking history) are protected by laws and private agreements which restrict your counter-parties; my understanding is that if some third party somehow learns those facts, they can do basically whatever they want with them (aside from blackmail).

are you saying that if i happen to stumble onto a data leak of those private facts, i can use them to create a non-blackmailing (non-profit?) business?

IANAL, but if you "stumble" onto data that is legally published, then you are free to use it in your business. It is on you to check that you are obtaining the data in a legal manner.

What if it was not intended to be legally published, but you still got access to it because they could not protect it well?

Someone I know actually got access to some stuff this way before; got access to a gold mine that was supposed to be private but someone misconfigured the webserver.

In the US, once a secret is out, it is no longer a secret. The person who originally disseminated the secret might be punished, but subsequent use wouldn't be. Facts aren't protected by copyright, they are typically just secrets. Your medical history is secret, and protected by law that would punish the person releasing them. But once released there isn't much you can do about it.

Using a somewhat relevant real world analogy: if I go into a house that should’ve been locked (but wasn’t), and I know I don’t have a right to be there, I can’t argue what I did was legal. It’s the same thing with computers: you can’t use[a] data you don’t have a right to use.

[a]: legally use

What if you see an open door with gold coins inside? Same answer.

The most you can do is inform the owner that their data isn't protected. And even that, unfairly IMO, is legally dicey.

if by legally published you mean that the publisher got a legally binding agreement with the patients to publish those 'facts', sure.

but the parent was not putting constraints on how i obtain it ('somehow learns those facts'). if the publisher makes a mistake and i hit the URL and the data is automatically downloaded, i didn't do anything wrong. doubt i can then just use that information though?

The law varies a lot around the world, but in the UK at least you need a lawful basis for collecting, storing and processing information like that (in the GDPR lingo, special category data). Unless you had very explicit permission from the individual I think you'd be hard pressed to demonstrate you had a lawful basis for collecting the information, as it would almost certainly not have been purposefully made available.

IANAL, but you can definitely gossip, and I think you can generally create a business based on selling the data (though you may be subject to some state laws).

There are patents on large portions of the human genome as it exists in nature

Where? Certainly not the US.

https://medlineplus.gov/genetics/understanding/testing/genep...

"On June 13, 2013, in the case of the Association for Molecular Pathology v. Myriad Genetics, Inc., the Supreme Court of the United States ruled that human genes cannot be patented in the U.S. because DNA is a "product of nature." The Court decided that because nothing new is created when discovering a gene, there is no intellectual property to protect, so patents cannot be granted. Prior to this ruling, more than 4,300 human genes were patented. The Supreme Court's decision invalidated those gene patents, making the genes accessible for research and for commercial genetic testing."

You're right, I was behind the times. Glad to see this was finally corrected!

Can't tell if serious or sarcastic in reference to copyrighted code

The legal argument appears to be:

1. There have been breaches of data protection law (GDPR as implemented in the Data Protection Act 2018), e.g. players did not consent to data transfer, the data isn't accurate, etc. and it's done on a commercial basis.

2. These breaches were injurious to the economic prospects of the affected players and therefore damages should be awarded.

I would imagine that the cause of action will be the tort of negligence against whoever sold the data on, and/or the gaming, betting, and data-processing companies. This is because they arguably had a duty of care to the players, the duty was breached, and the players suffered some harm -- based solely on the facts in this article.

Regarding personal compilation of statistics, that's fine - there's an exemption for activities of a purely personal nature in the GDPR - which is why you wouldn't get caught, but commercial exploitation of the data falls outside of that.

The idea that everything remotely involving data is forbidden under GDPR is giving it a bad name. I'm reminded of every Comcast sin being labeled as a "net neutrality" issue.

Here, specifically, GDPR prescribes exemptions for journalism: https://gdpr.eu/article-85-right-to-freedom-of-expression-an...

There are probably other exemptions that would apply for innocuous activity freely done in public, with the explicit understanding that it would be filmed.

I doubt statistics won't be consider personal data. For example "Someone shot 5 shots on goal" is not exactly personal data. "Simon Munster played 54 mintues" should not be personal data it's a fact of a public event.

A famous athlete wearing a brand sneaker is also just a fact of a public event. Yet if that brand uses it in advertisement they still need to pay the athlete for permission.

No they need to pay the athlete to wear the brand. Otherwise he wears a different brand that pays him. They also pay for the time the athlete spends making ads for the brand.

If the player just wears the brand through their own choice and they company advertises "As worn by Jordan in NBA all-star game!" then they would not need to pay.

Ads and stats are two completely different things.

GDPR personal data is anything relating to an identifiable person. It’s not an expectation of privacy standard. Data gathered by observing you in public is protected the same as data you share with the controller in confidence.

I think the challenge will be persuading courts that actions that users are paid to undertake in public and contractually agreed to reassign relevant image rights to are [i] 'personal data' under the intended meaning of GDPR and that [ii] something they didn't have informed consent about the possibility third parties might have access to when signing those contracts. Proving actual instances of harm suffered from the dissemination of unfavourable statistics is also going to be tricky (even though there are undoubtedly players that have lost out). It's plausible the net effect of increased statistics on player salaries is positive; certainly they haven't gone down on average.

Since most player contracts are frequently renegotiated and most clubs use third party databases and generate revenue from betting companies (so if necessary they'd all end up with clauses permitting this data use), the long term effects of a favourable stretch of the definition of 'personal data' are more likely to have chilling implications for people collating activity/performance metrics or compiling biographies of other types of public figure anyway...

>Proving actual instances of harm suffered from the dissemination of unfavourable statistics is also going to be tricky (even though there are undoubtedly players that have lost out).

All you would need is one club saying they searched the data for say "any player over six foot" and one player who fit the search term but the data was inaccurate.

That would only count toward one person one club not a widespread pattern.

While I was being extreme, showing that similar searches happen often and errors are somewhat common makes the case.

I think betting is likely to have an unusually negative economic impact on the players. Loss avoidance means people are more likely to remember players ‘underperforming’ expectations than exceeding them. That’s going to have a negative impact on their marketability as paid sponsors independent of actual performance.

It’s questionable if they can win, but demonstrating damages may be the easiest part of this case.

I think proving a specific instance of lost sponsorship income to the satisfaction of a court is going to be extraordinarily difficult, especially if the argument is that consumers draw more adverse inferences from commercial datasets (which they generally don't have access to themselves; betting companies use them to help set the odds) than watching the matches.

Even more so when the context is that virtually any employer of professional footballers derives some of their revenue from betting, and betting companies are the primary sponsors of half the English Premier League teams and the English Football League organization

You can demonstrate harm without showing specific damages. Damaging someone’s reputation for example isn’t acceptable.

It’s the same basic principle as speeding or drunk driving being illegal even if nobody was actually harmed, putting people at significant risk of harm is not acceptable.

Isn't the harm to the players' reputations coming from their own quality of play in publicly viewed matches?

In my opinion a third party would only be liable for harming their reputation if the statistics being published were untrue.

It’s a team sport so they don’t optimize for fantasy leagues. If they are pulled to avoid risk of injury on a blowout their stats look worse. Essentially, a 2:0 win could look worse than a 2:3 loss.

In effect their a RNG that happens to make people dislike them.

Yes, I quite agree that the prospect of success appears remote. They would have an arguable case for negligence regarding data inaccuracy though, so I wouldn't be surprised if they submit that as an additional claim and abandon the GDPR claim if that seems likely to fail.

It seems there's a huge difference if your data is collected when you engage in a private activity (i.e. was not intended for public consumption)and participating in a public event where your behaviour is intended to be consumed by the public.

Does GDPR make this distinction?

The GDPR primarily focuses on the intent of the person processing or collecting the data rather than whether the activity you engaged in was private or public, although it does take into account reasonable expectations around privacy as well. I suppose it focuses on that because it's more flexible to enforce.

As matthewheath pointed out, this is around personal data use as opposed to copyright.

IANAL but I wouldn't be shocked however this falls.

> facts are not copyrightable

Yeah, but there are trade secrets, for example, or illegal numbers. I brought them up because thought they are related. There is also medical history or psychiatric history.

Do you need to license anything to do a radio broadcast of a game? All it’s doing is reporting the facts.

I suspect it’s less immediately dismissible than you suggest.

If you broadcast the game, that's not just facts but playing audio that someone records. Just telling the result or current state, sure, that's just data - but I never heard that there's a license for that.

Can we change the "Players"in the title to "Soccer Players"? Otherwise, I'd guess most people would think of videogame players.

Football players, to most of the world anyway ;^)

Not in Ireland ;o)

This is a fair point, because even if you are familiar with the source of the article (The Athletic), as I am, it's unclear which sports are discussed.

Football players for 90% of the people

Football players, but yes.

*footballers

It's no more "their" statistic than the sentence "Neil Armstrong was the first man on the moon" belongs to Neil Armstrong.

If I sit at a busy intersection and count the cars going by in a 5 minute span, do I owe something to each of those drivers.

IANAL, but this is just ridiculous.

Trademarked names... They'll want a license before anyone can talk about them at all; and then they'll wonder where their fame went.

You joke, but the name and likeness of these players is something EA have to license each year for the next FIFA release.

If you record the license plates or images of the drivers, 100%.

If you record something like make and model, and you catch a car that’s rare enough in the area that there is only one person it could be, then also yes.

It’s absolutely wild to me that someone’s own observations of a public place can be legally restricted, but that’s GDPR.

If you write down the plate, the name of the intersection, and when the car passed by, then yeah that's personal data and GDPR regulated. You can't just compile that DB and sell it, dude. That's literally what GDPR is for. Not even slightly different from online advertising.

Hard for me to believe that personal statistics aren't like that. Meet all the GDPR conditions.

I think it's the resale factor that alot of people seem to be missing, The data used would appear in scouting databases that are traded and sold, this could have a rather large impact on your career trajectory behind the scenes. Thats before you factor in the videogame market. While i understand this is just about getting more money I would find it weird if the number of points I achieved every sprint was tracked and shared but hey I guess thats just the industry.

FYI this "reveal" is 3 months old, from July 2020

Here's a related article not behind a paywall

https://www.bbc.com/sport/football/53557706

It seems like the players should want their stats out there. Erasing a player from a dataset makes it like they didn’t even play which would probably hurt their career.

This seems like publicly available info. Should I receive a royalty payment every time somebody views the value of my home on Zillow?

IANAL. Statistics sound like a compilation of facts. Facts are not copyrightable by themselves. Compilations are, but compilation were not authored by players, but by someone else. I really hope they will lose. Also, paywall.

Where you were last Friday night is also a fact, and even if you were somewhere in public most people would consider that personal information.

Same with your age, your gender, your height and weight, your address, your phone number, where you work, etc.

All could be either measured in public or are likely to already be in a public database somewhere.

Yet most would consider those to be personal information.

That information might not be private, but that doesn't make it any less personal.

I don't have access to the article but they'll surely be suing on GDPR grounds, for unpermissioned use of personal data, not the copyright (which, I agree, they probably won't hold).

It's akin to an employer now tracking 24/7 whether you are sat at your home office desk (in my view: this shouldn't be allowed, with or without permission).

These are football players playing matches in public and the statistics are trivial I imagine, like how many passes they make and stuff like that, no? This seems like they just want free money and I hope they lose

I agree I also hope they lose

Actually this sort of case could have quite far reaching implications. What other data would fall under similar protection? I can think of quite innocent use cases, like IMDB or other databases of who worked on artistic product like a book or game. If your name is in one, should you be paid? Where would this end?

Sure and when you drive your car in public, it's trivial for me to place cameras around the city and track your plate for movement data. You think I should be able to sell that?

Personal data doesn't just stop being personal data because it's someone else's.

IANAL, but I recall the phrase “reasonable expectation of privacy” coming up in these kinds of situations. I’m really not sure how performing public entertainment exhibitions overlaps with that, whereas driving to work and such is an activity that merely takes place in public but is not performed for the public.

> the statistics are trivial I imagine, like how many passes they make

If it's Opta, they record a whole bunch of stats for every player in every game in the top 4 (I think) leagues - distance run, passes attempted, passes completed, whether shots were on target, etc.

Even providing "how good was the shot?" metrics - https://www.optasports.com/services/analytics/advanced-metri...

Huge investment in tech and people. Definitely not "trivial".

The articles don't make it very clear what sort of statistics are involved. Are they talking about simple things anyone could scribble down while watching TV, like it's 1980? I thought that, for many sports, they now collected much more. Like where every player's feet were, to the cm, in every frame, that sort of level. Would this change any arguments?

If it is statistics to a degree that can be used for hiring decisions or sold to competitors (or: for other people to make hundreds of millions from your work), is it trivial?

Like in moneyball. It could have a bearing on players wages, sponsorship etc. I hope too hope they loose though.

If they win, instead the firms could simply give them the rights to remove their data, which honestly none of them would exercise. Not being in the professional sports statistics databases is worse, even for poorly performing players, as it implies that aren't famous or noteworthy enough to be included. Under GDPR a person doesn't get to cherry pick which data to delete right? Either you request all your data to be deleted by a firm or you don't, right?

GDPR clearly needs an exemption for anything that is already publicly available data.

Article seems to be from 4 months ago and there doesn't seem to be much follow up in the mainstream press.

In the computer weekly article linked in another comment it states the players are not claiming they own the data - but that they have data rights and are seeking to address the issue of people using that data without consent.

That sounds a little less clear cut, I'm not a legal expert but I guess a case might be able to be made about privacy and GDPR conformance. But probably not that they can be paid for the data.

My understanding is that they should have some rights, in this case namely they could be to request their own data, to correct their own data and to ask their data to be removed(right to be forgotten)...

some of those "stats" are the opinions of scouts(good at tackling?) right? other stats are historical data that is publicly observable(tackles per game, etc).

i can't see their logic...

Related, non-paywall

https://www.computerweekly.com/news/252487262/Footballers-to...

Paywalled. Couldn't (quickly) find a readable version in the usual archives.

The belief that information and ideas should be "owned" as if they are physical property is insidious, was always going to lead to absurd lawsuits like this, and will lead to worse in the future.

Information is non-rivalrous. It cannot be "stolen" in the same way that, say, an apple can. If I take your apple, you no longer have it. If someone repeats your information, you haven't been deprived of it. These athletes are as free as ever to collect and share statistics about their performances, regardless of who else is also doing it.

When you move through the world, you emit information, but that doesn't mean you "own" it.

We live in a knowledge economy. If I spend $200M making a movie and you start showing it in your theaters without paying me, saying that you aren’t depriving me of my copy is irrelevant.

The property you’re talking about is orthogonal to questions about who has the right to control what information.

The economy we've created that transacts in information is the direct result of laws that support it. If, for example, anybody could copy your movie, then spending $200M to make it would be a very bad business decision, so people would prioritize other business models and endeavors. Hollywood would be weaker, and Broadway would be stronger. There'd be a much larger emphasis on live, interactive, and/or customizable entertainment experiences.

I fail to see a problem here. "But that's not how it works today," isn't a problem imo.

Let's say we made recipes broadly patentable. I'm sure that would change the restaurant industry into something unrecognizable over time. 100 years from now we'd see all sorts of massive businesses profiting in ways that are unimaginable today. Those businesses' existence wouldn't imply the laws were good. However I'm sure somebody might say, "But how is MegaPizzaWorld ever going to justify its $10 billion investment into monopolizing pizza if anybody can just copy it?"

GDPR does not use ownership semantics. There are "data subjects", "processors", "controllers" and they each interact with "personal data" in some way.

Then certain restrictions are enforced on processors and controllers with data subjects having certain rights recognized.

The rights and restrictions differ from ownership semantics sufficiently that it's not useful to talk about as owning.

Yes, this is why I'm careful to limit my criticism specifically to the belief that information can be "owned." That's subtly different from the belief that we can make contracts around information.

But if you ask the average layperson or watch popular media about data, they speak and think in terms of ownership. That's the wrong model, and it leads to the wrong assumptions and conclusions.

From what I understand, this wasn't always the case. The belief rose alongside the term "intellectual property" in recent decades, which was itself popularized by media companies who have the rights to lots of information, and who stand to gain financially the more we shift toward conceiving of these rights as permanent ownership.

Ah, I see. Okay, yeah, I think we're in agreement then. OP article is about GDPR-based suit.

Apple also has a nice statistics of >>all used apps on Macs<< since 2017 - every app start is reported back to apple. One could see from such stat what is being used more, what is being used less and then make business decisions... Maybe somebody will sue in the future...

It actually depends on if this data is maintained at an individual level and is personally identifiable. This data is almost certainly deidentified in Apple's case, which would make GDPR not applicable.

Reply on news.ycombinator.com