Football by numbers: How stats blur success at the elite level

Published 2 Comments

Football News

Arsenal’s dominant and shocking victory over Manchester City at the Etihad Stadium has apparently set the blueprint for how the Gunners should move forward. The uncharacteristic way that Arsene Wenger set up his side required defensive discipline and attacking efficiency in equal measure.

As Arsenal went on to win the game 2-0, the idea of criticising any individual player is something of a scandal to overzealous fans, who feel it unnecessary to read into something if their team comes away with maximum points.

But keen to expose one particularly bad performer, this writer sought evidence by heading to statistical websites. There was none to be found. How could this Arsenal star come away from the Etihad with his reputation enhanced despite a reckless display? Football is going through a drastic change, as this article will attempt to explain. If Arsenal are to move with it, there is a chance one star will become redundant to Wenger’s newfound style.

The basis for this thought process was born out of a harmless set of player ratings, published minutes after Mike Dean had blown the final whistle at the Etihad Stadium. The article in question awarded Aaron Ramsey a four out of ten–at least two points lower than any other Arsenal player.

This was not a judgement based on Ramsey's pass completion rate, number of interceptions or any other statistic you could throw up. Quite simply, it was based on an opinion inspired by an admission from Arsene Wenger before the game had even kicked off. The Arsenal boss admitted that Ramsey was under strict instructions to support Francis Coquelin and shield the back four from the dangerous passing combinations of David Silva.

Ramsey performed well in the first half, but let his side down in the second. He chose to take part in each and every one of Arsenal's counter-attacks as they attempted to double their one-goal lead. The counter attacks broke down and Ramsey's lung-busting 80-yard runs left him unable to regain position before Manchester City launched another offensive.

Stats struggle to tell you how many times a player was found out of position or calculate how many goal-scoring opportunities were created because the player in question wasn't quite where he should have been. Those indiscretions only ever get noticed when something bad happens, like conceding a goal. All is well that ends well, as the saying goes. awarded Ramsey a 7.5 for his performance against Manchester City, drawing up the question of how reliable statistics really are. Are important events going ignored in favour of non-important events just because they are perceived to be important? Is football even quantifiable?

Statistical white noise

If I told you Aaron Lennon had more Premier League assists than David Beckham, you would tell me that I was wrong. I probably am, but officially the Tottenham star has created more goals than the Manchester United legend. That's because statistics have only been collected by the Premier League since 2000/01.

In-depth analysis has grown in tandem with the dawn of the information age. People are no longer satisfied with just knowing what happened but why it happened and who is responsible for it happening. It started with basics like possession, shots on goal and shots on target, but now it goes much deeper than that. Via companies like Opta and Prozone, mountains of data are available on a subscription basis, but football is still waiting for the holy grail.

For John Coulson, head of professional football services at Opta, there will always be a part of the game that can't be broken down into numbers.

He told GiveMeSport: "I think clubs are changing what scouts are looking for now because our data doesn't capture the tactical awareness of a player.

"The timings of those runs, his movement, body shape and attitude; these are all qualities that our data can't capture so clubs still rely entirely on the intuition and experience of scouts and that will always be required.

"There is data that can tell us when a player has made a run and the speed of it but it is impossible to interpret whether he has been told by the coaching staff to be in that position or not."

The attempt to quantify a game as random and changeable as football is younger than any of the professionals currently playing it and, as of yet, no one is yet to find a winning formula. Even when successful tactics emerge, an antidote is soon found to nullify it.

Famously, Jose Mourinho countered Barcelona's high pressure, possession game in the 2010 Champions League semi-final by ordering his Inter Milan side to purposely give the ball away and invite attacks. Barcelona looked less than threatening and eventually went out 3-2 on aggregate. This was the anti-football of the time. Free-flowing possession football was the rebel alliance and rigid defensive tactics was the dark side.

Possession: The new enemy

Barcelona had 86 per cent possession in that game, but could only find one of the two goals they needed to progress. The ever-so-slight correlation between possession and winning has been consistent throughout the history of stat accumulation but now there is a strengthening argument for the advantages of not having the ball.

Mourinho seems to have always lived by the rule that possession is the enemy and, as one of the most successful managers in the history of football, it is surprising how few have followed his ethos. The Chelsea boss, according to Spanish journalist Diego Torres, has seven theories that suggests the team without the ball are more likely to win the game.

As the theory goes, the player with the ball is more nervous than the player without it. He has more to lose by giving the ball away than the challenging player does by not winning it. Those nerves lead to mistakes and the team that makes the most mistakes generally loses the game.

Everton boss Roberto Martinez has learned this fact the hard way. This season, the Toffees have conceded 11 goals in the Premier League directly through unforced errors. Their average possession of 57 per cent is the third highest in the Premier League.

While it is easy to pass it off as 'parking the bus', Mourinho's way of winning, or not losing as it were, is one of the most efficient and effective tactics in the modern game. Chelsea went to Old Trafford, Anfield and the Etihad Stadium in the first half of this season, coming back with a total of five points and undefeated despite having possession of the 43 per cent of the time on average.

Arsenal's possession against Manchester City was the lowest Opta has ever recorded for the Gunners at 35 per cent, yet they won in convincing fashion. Rather astonishingly, of the eight times Arsenal have had the ball less than 45 per cent of the time they have won seven, drawing the other.

Three of those eight games (West Ham, Liverpool & Manchester City) have come this season hinting that even the loudest advocate of possession football, Wenger, is ready to call time on an ageing tactic. While possession can be a successful way to dominate and, therefore, win football matches, there is an equal opportunity to do exactly the opposite with similar results.

The sticking point is that the tactics Arsenal displayed against Manchester City and Liverpool earlier this season are designed as a direct repellent to free-flowing, possession football: organisation v disorganisation–intuitive logic tells us the former will always beat the latter and they're not the only ones achieving results this way.

Southampton are a team on the verge of achieving something very special. They sold over £100 million worth of talent in the summer but are currently sitting above every single one of the players they sold in third place. Should they qualify for a Champions League spot, beating Manchester United, Arsenal and/or Liverpool to the punch, it will be remarkable achievement.

Ronald Koeman came in for Mauricio Pochettino and seemed to carry on where he left off. But there is a major difference to the way Southampton are playing. Last season they were the most possessive team in the league with an average of 58.4 per cent, but this season has seen that drop significantly to 51.9 per cent. The result has been just two goals conceded from direct mistakes–the second lowest in the league.

Eibar, the small La Liga club, are also defying apparent rules of possession football, currently sitting eighth in the division despite having the second lowest average possession in the entire league (42.2 per cent). The team below them in that statistic, Levante, and the team just above them, Almeria, are both in the relegation zone.

A new dawn

So less possession can be detrimental if you don't have the tactical discipline required to keep the flow of attacks at bay, but a blueprint is steadily emerging from the likes of Arsenal, Southampton and Eibar. It's a zeitgeist that would leave the statistic of possession completely defunct.

Ramsey ignored this blueprint, designed purely around defensive discipline and it could have ultimately let his team-mates down. He may have won five tackles and completed over 80 per cent of his passes, but his disregard for defensive responsibility did not comply with a team trying to keep a clean sheet with under 40 per cent of possession.

The same performance drew wide criticism after Arsenal threw a three-goal lead against Anderlecht away. Despite being 3-0 and then 3-1 up, Ramsey shot from distance several times and went forward with a certain disregard at every opportunity. It left gaping holes at the back and would later allow the Champions League minnows an unlikely point.

As Manchester City failed to take advantage of Ramsey's disobedience, there was no need for an enquiry. Had Arsenal fallen to a 2-1 defeat after conceding two second-half goals, a responsible party would be hunted down and exposed.

Unquantifiable sport

Modern society wants explanations for everything. It is why stats were invented in the first place. But anything multiplied by zero is equal to zero. Zero goals conceded equals zero mistakes. Stats can't tell us anything about what did not happen in a game – that is why, despite the wonders of technology, the best clubs in the world still send a group of old men to football games to scout players they may want to sign, looking particularly at what they do when the ball is nowhere near them.

There will come a day when a club can sign a player based purely on statistics, just like Billy Beane has done with MLB's Oakland A's so successfully, but football analysis has not reached that point yet. There are too many scenarios that don't happen in football to say what and what is not a key point in any given match.

A missed tackle could lead to winning goal, in an attack that was only possible because a player was in the wrong position to intercept an earlier pass. But winning that tackle saves the goal and spares the man who made the original mistake from a torrent of criticism. That is where the mystique of football will always lie and why player ratings should always be based on one's own perception of the game, not statistics.

Do YOU want to write for GiveMeSport? Get started today by signing-up and submitting an article HERE:

Premier League

Article Comments

Read more

Report author of article

Please let us know if you believe this article is in violation of our editorial policy, please only report articles for one of the following reasons.

Report author


This article has been written by a member of the GiveMeSport Writing Academy and does not represent the views of or SportsNewMedia. The views and opinions expressed are solely that of the author credited at the top of this article. and SportsNewMedia do not take any responsibility for the content of its contributors.

Want more content like this?

Like our GiveMeSport Facebook Page and you will get this directly to you.

Already Subscribed to Facebook, don't ask me again

Follow GiveMeSport on Twitter and you will get this directly to you.

Already Following, don't ask me again

Like our GiveMeSport Page and you will get this directly to you.

Already Subscribed to G+, don't ask me again