Hundreds of FIFA 17 players may have inaccurate ratings because of some of the methods EA uses to score them, according to a former EA data collector.
The source, who worked as part of EA’s 9,000 strong FIFA data collection team, claims:
- Some players could be given ratings by people who may not even have seen them play
- There is no official training or screening process for the work
- Senior data collectors sometimes even overrule rating suggestions without any evidence
- And they can get paid just £30 a month for “hours of work”
Rather than relying solely on statistics, EA apparently uses voluntary data collectors to determine their prestigious FIFA ratings.
The collectors watch games and upload subjective feedback on every player – all 18,000 of them – to a secure EA website.
This data is then “reviewed” by senior collectors, known as ‘editors’, before it is collated into 300 different data fields and 35 specific attribute categories, which come together to determine a player’s overall rating.
Last year, EA programme lead Michael Mueller-Moehring told ESPN that “we guess a little bit” regarding some player ratings.
But the source claims the scale of this guesswork is far greater than previously thought – which echoes sentiment expressed by some FIFA fans.
According to the source, the data collection team was “fairly short on staff at times”.
As a result, smaller leagues were often passed to people in the programme who didn’t or couldn’t watch the games.
They were sometimes asked to “fill in ratings as best they could” with the understanding that there was going to be “a fair bit of guesswork”.
Consequently, the former data collector claims hundreds of player attributes may have been effectively plucked from thin air and submitted with “no time or way to check them”.
The source first got involved with the programme after replying to an advert posted on social media.
He was a “massive FIFA fan” and it “seemed a nice way to combine a love for football and the game”.
“Lots of ratings I saw didn’t seem fair. I was sure I could do a better job,” the source said.
But alarm bells started ringing when, after an initial online application, he claims he was accepted on to the programme without any interview.
Training consisted of “a series of emails”, relaying instructions with minimal guidance, before he was sent out to work.
The source was given full access to the database, which contains statistics from the entire roster of players in EA’s catalogue.
He was then assigned a club and asked to score players across different attributes.
It wasn’t long before the lack of training came home to roost as, he admits, he struggled to score players accurately.
For example, if Cristiano Ronaldo beat a player in a footrace, it was not made clear if Ronaldo’s speed rating should be boosted or his opponent’s pace decreased.
As a result he claims he had no choice but to “make things up as he went along”.
To protect against subjectivity, the programme has a team of Editors, who are said to make just £30 a month, who review all stats from the data collectors. As the source puts it, “they are the last line of defence”.
But worryingly, according to the source, their lack of transparency can sometimes feel like it’s compounding the problem, as they often override data reports from collectors, seemingly “without evidence or explanation”.
The source thinks that while EA provided guidelines on using evidence, they were often not adhered to.
“When they didn’t agree with the data they just sent you a message saying that they disagreed without any explanation – even though I’m pretty sure they weren’t at the games,” he said.
The source was eventually promoted to an Editor role – but could do very little to guard against bogus data as there was simply no way of verifying the opinions and claims of data collectors scattered around the country.
He spoke of collectors being assigned to far-flung foreign leagues, despite superiors allegedly knowing they had no means by which to watch the games.
EA has been open about the failings in the system.
Mueller-Moehring told ESPN that, due to the volume of player data required, stats alone are not sufficient.
“We have many leagues in the game; no stats provider could offer us data for all these leagues, teams and players,” Mueller-Moehring said last year.
“This is also the reason why we use this online database, because it’s not possible to buy this data some way — it just doesn’t exist.”
Mueller-Moehring also revealed that German striker Thomas Muller had to have his rating boosted because the data did not accurately reflect the striker’s ability.
Now it seems that Muller’s inaccurate rating may just be the tip of the iceberg when it comes to how players are classified in the world’s most popular football game.
Dream Team contacted EA regarding these claims but they respectfully refused to comment