Manchester United’s Top Six Players

David de Gea
Without question, Manchester United’s best performing player of the 2017-18 season was David de Gea. He finished the season without any medals but claimed a couple of personal accolades. De Gea won the Golden Glove after picking up 18 clean sheets over the course of the season. He was also voted in the PFA Team of the Year and was named the Sir Matt Busby Player of the Season by the fans for a record-breaking fourth time.

He made 115 saves over the course of the season with an 80% success rate with the shots he faced. He made zero errors that lead to goals, compared to the 24 made by Petr Cech, 15 by Hugo Loris and three by Thibaut Courtois.

There are a few games that stand out, with his excellent saves winning United games and enabling them to finish second in the league, but none more so than the 3-1 over Arsenal at the Emirates.


De Gea doesn’t make it in to the top five using Twelve Football’s scoring system though, limited by his position in goal making it unlikely for him to gain any points for off ball action or shots. But he was head and shoulders above every player at the club this season.

Romelu Lukaku
The Belgian striker ended his first season at United as the club’s top scorer with 27 goals in all competitions.

When you consider that United created 70 fewer chances than any other top six team, Lukaku’s goal return is pretty impressive.

The striker wasn’t just a scorer though, but a creator of goals too, with Paul Pogba the only United player to assist more than him. Lukaku claimed more assists than the likes of Harry Kane, Sergio Aguero and Eden Hazard.

Having been criticised for failing to impress against the teams at the top, Lukaku was in fine form as United beat his former club Chelsea at Old Trafford. Having gone a goal down, Lukaku scored the equaliser before assisting Jesse Lingard’s winner. United will be left to rue his absence from the starting line-up in the FA Cup final against the same team through injury.


Lukaku is ranked the third best player by Twelve Football with over half his points coming from his contribution with shots, which should come as no surprise, given he’s largely been the only player Mourinho has used as a centre-forward this season.

Paul Pogba
The Frenchman has been strongly criticised this season, from pundits, the media and his own manager alike. That likely has more to do with how short of his potential he is currently performing, rather than his general contribution. For the most part, he did OK for United this season, with the odd poor performance.

When he plays well, United tend to follow suit, such is his influence on the team. Despite going off the boil for chunks of the season, only four players have more assists than him in the league this season, and all four of them had more playing time than him.

When played on the left of a midfield three against Everton at Goodison Park he stole the show. His stand out performance was against Manchester City at the Etihad, when he denied United’s hated rivals from being the team to win the Premier League after the fewest number of games.

After a disappointing first half, he came back in the second looking like a totally different player, scoring twice to draw the game level before Chris Smalling bagged the winner.

Per minute, Pogba is ranked the best player of the season by Twelve Football, with his influence on the attack his greatest strength. But he is still a long way from playing the way on a consistent basis the supporters believe he is capable of.

Jesse Lingard
Having joined the club as a seven-year-old, Lingard has plenty of United supporters who have been willing for him to do well. Before this season, he had claimed crucial goals in two Wembley cup finals, but had failed to play consistently well over the course of a season.

Under Jose Mourinho, Lingard has performed well, with the manager a clear fan of the forward. He claimed 13 goals and 5 assists in all competitions, a return of a goal or assist every 119 minutes he was on the pitch.

However, his most memorable performance of the season was in the aforementioned game against Arsenal when he scored two goals in United’s 3-1 win and Milly Rocked in celebration.


Lingard is ranked the 12th best player for United, with his attacking scores his best asset. Yet what comes as a surprise is how low his score is for off the ball action, given this is widely regarded as one of his greatest strengths. His pace and ability to run behind the defence drags players out of position and creates space for his teammates but this isn’t recognised by his off the ball score, where he is deemed less effective than the likes of Victor Lindelof and Phil Jones.

Nemanja Matic
The Serbian midfielder was reunited with his former manager Mourinho at United this season, after Chelsea made the mistake of letting him leave with the hopeless Tiémoué Bakayoko his replacement.

Matic started the season exceptionally well but went through periods where his lack of mobility was exposed and he failed to have the impact he made at the start of the season.

His campaign will be best remembered for his 91st minute winner against Crystal Palace after United had gone 2-0 down, where he blasted the ball in to the net from outside the box.

The best performance of the season from him came against Chelsea though when he totally bossed the midfield.


The midfielder was ranked United’s best player of the season by Twelve Football, with, surprisingly, his contribution to the attack his biggest strength.

Ashley Young
Some United fans can hardly believe that Ashley Young is still at the club but he can owe his lengthier stay to his ability to reinvent himself as a full-back, having played most of his career on the wing.

While his defensive game does have flaws, with pacey and tricky wingers his downfall, he is always ready to put in a crunching tackle and crosses the ball better than anyone at the club (which, admittedly, doesn’t mean an awful lot).

Young claimed four assists this season, which is as many as Eden Hazard, Cesc Fabregas and Alexandre Lacazette, and more than Gabriel Jesus, Marcos Alonso and Harry Kane.

His best performance came against former club Watford when he scored two brilliant goals in United’s 4-2 win.


Young’s greatest strength, according to Twelve Football, is his work in attack, which paints an accurate picture. He is ranked third behind Matic and Paul Pogba.


Download Download

Nick Pope – A revelation for Burnley (and for England?)

10.09.2017 – Matchday Four of the 2017/18 Premier League season

Burnley v Crystal Palace (at Turf Moor)

3.36pm: Burnley’s captain and talisman, Tom Heaton, came to claim a cross, before falling awkwardly to the ground in the penalty area. Following a lengthy stoppage in play, the Clarets’ stopper was helped off the pitch in obvious discomfort, with what appeared to be a very serious shoulder injury.

Embed from Getty Images

To many of those watching, including several of Burnley’s own supporters, there was concern that this might have marked an end to the Clarets season, or at the very least, the start of a much more difficult survival battle.

But from the bench came the relatively unknown Nick Pope. Here was another goalkeeper who, like Heaton, had been plucked by Burnley from League One, following his team’s relegation from the Championship at the end of the previous season.

What happened for the remaining 54 minutes of the Crystal Palace match, and then for the other 34 matches of the Premier League season was remarkable. Firstly, a hard earned 1-0 victory for Burnley, and a clean sheet for Pope, even though official statistics do not credit it as such, as he did not play the entire ninety minutes.

There then followed a series of performances in goal that belied Nick Pope’s previous lack of top flight experience. It doesn’t necessarily require in-depth data analysis to know that Pope’s contribution to Burnley’s season was a vital one, but it can help to provide some important context.

A plethora of statistics, from various sources, suggests that Pope had become a vital replacement cog in Burnley’s defensive system – goals conceded, clean sheets, xG improvement, number of saves, number of high claims, all of which were impressive in their own right.

However, the data available from Twelve, which I have used and posted at various stages through the season, shows something else, in addition to information from elsewhere. Points are awarded for actions throughout each match, and although for goalkeepers there may otherwise be a tendency to focus on the activity in the penalty area, the system also credits other contributions, most notably in terms of attack.

Nick Pope topped the Twelve leaderboard when it came to attack, with the plot below highlighting his most important contributions.

Nick Pope went on to win all but two of the thirteen individual Player of the Year awards at Burnley FC’s supporters’ groups awards evening, and also earned the overall Player of the Season and Players’ Player of the season awards too. The subjective consensus was most definitely in Nick Pope’s favour, but it can also be helpful to consider other methods of classifying performance.

In fact, there are several other Burnley players who could also stake a claim to the overall Player of the Year award. Based on the Twelve data model, the table below shows the players who rank more highly, using points per minute (excluding Dwight McNeil, who only came on for the last few minutes of the final match):


Of the other highest scoring Burnley players, the three at the back who have shared central defensive duties, James Tarkowski, Kevin Long and Ben Mee, lead the pack. They are followed by two more offensive players, striker Chris Wood and midfielder Robbie Brady, who each only played less than half of the season. Each player displays their own particular strengths in each area to which points are allocated.

It helps to have data to support decision-making processes, rather than to use data blindly, but few would argue with Nick Pope’s status as Burnley’s Player of the Season. More recently, his call-up to the England squad for the World Cup in Russia is a well deserved reward for a very successful first season in the Premier League.


Download Download

Real Madrid vs Liverpool: Who have been their best players in Europe?

In my last article I took a look at who Liverpool’s player of the year has been when it comes to the stats in the Premier League. With the Champions League final coming up on Saturday, it’s now time to take a look at how the players of Liverpool and Real Madrid have fared in Europe in 2017/18.

It won’t surprise anyone to see Cristiano Ronaldo lead the way from Liverpool’s fabulous front three, as those four players are the top four scorers in the competition this season. However, it will probably confound people to see Dejan Lovren sneak into the top five, just as he did when looking for the Reds’ top players in domestic football.

While the top four have clearly relied a lot on shots and goals (the blue bar) to reach their high scores, Liverpool’s Croatian centreback has understandably relied upon his defensive (red) points. But his off ball bar (yellow) is relevant too, and in fact Lovren leads the way on this particular attribute, either in total or on a points per minute basis.

Jürgen Klopp appears to value Lovren’s brand of front foot defending, even if it has Liverpool supporters tearing their hair out at times. Twelve value his contribution here too, as we can see on his map.

However, if anything is more surprising than Lovren’s inclusion in the top five players, it’s the fact Ronaldo has racked up far more defensive points than any of Liverpool’s front three, as that feels counter-intuitive when the Reds love to defend from the front.

The Portuguese legend is a useful asset when defending set plays though, thanks to his aerial prowess. Ronaldo has made one more clearance than the Reds’ attacking trident have between them, and here’s his defensive chart.

In terms of actual defenders, both Raphael Varane and Sergio Ramos have amassed more points for defending than Lovren has, and in fewer minutes on the pitch too. So how have Liverpool conceded two fewer goals in the competition this season?

An easier batch of opponents has obviously helped, but any Kopite will utter three words when asked why the Reds’ defence has improved recently: Virgil van Dijk. Sure enough, the Dutchman is top of the defensive pile when the scores are sorted on a points per minute basis.

That’s impressive work from the former Southampton man, and let’s not forget he has only featured in the knockout stages as he joined Liverpool in January. Van Dijk’s figures have not been boosted by facing lesser sides in the group stage.

Will his impact help in the final though? After all, Ronaldo is more inclined to drift to the left anyway (as his attack map testifies), and that is where the weaker side of Liverpool’s defence resides too. Then again, van Dijk should prove to be useful in marking Ronaldo on set pieces, even if they don’t face up too often in open play.

In the attack stats, it would be natural to assume James Milner would lead the way, as he has clocked up eight assists in the 2017/18 Champions League, yet he only takes the silver medal.

The fact the winner in this category has set up a single goal in the competition illustrates the random nature of assists nicely. Toni Kroos has created 24 goal scoring opportunities in the Champions League this season, so perhaps should’ve assisted more goals. Mohamed Salah is the only player on either side to create more chances, and he is behind Kroos on a per 90 minutes basis.

The German international has boosted his creativity figures with set plays, but he clearly contributes plenty to Madrid’s attack. While Liverpool will need to be wary of Kroos in Kiev, there’s clearly players on both sides who can cause damage. That makes the match both a mouthwatering prospect but also incredibly difficult to call. Don’t be surprised if one of the top four players in the Twelve rankings makes the key difference in the final.


Download Download

Spurs Player of the Season: Do Kids Know Best?

The Tottenham membership have ended the 2017/18 season divided among a generational line. Jan Vertonghen won the Members’ Player of the Season and Christian Eriksen the Junior Members’. Is much lauded experience the key to identifying Spurs’ Star or are the new generation of analytics awoken youth wise beyond their years?

Eriksen comes out top on Twelve’s ranking system with Vertonghen coming in 3rd to sandwich Harry Kane.

It is a testament to both his award-competing teammates and the standards set by the striker himself that he could record a 30 goal League season and still be deemed to have had a slightly off season.

Although the younger fans have opted for the more attacking player if they were making their selection based on goals and excitement then Kane would have been the obvious choice. Instead they collectively voted for the dainty Dane, a player of subtlety and guile.

Christian Eriksen

The Twelve widget captures him not only performing at various stages throughout the season but able to pick out a pass from anywhere on the pitch.


Eriksen has played wherever needed this season; left, right, central and even deep, sometimes all in the same game.

Whether this season’s consistency has come from within his own growth – mental, physical or technical – or whether this is simply a case of the team around him providing him with a more consistent platform of time on the ball and making the right movements around him is a matter of debate. The latter somewhat calls into question the very nature of individual awards.


Jan Vertonghen

The senior members’ pick too, has demonstrated a special consistency but this is less positional, and more about the various phases of the play. Vertonghen has trotted out an unwavering defensive solidity but combined this with line-breaking forwarded passes, mazy, one-footed, runs with the ball and aggressive switches of the play.

Again, stylistically, this is nothing new from the Belgian but taken to new levels this year. Perhaps this was a vote of relief for the One Hotspur members who have seen it all. This season they witnessed Jan shining in the absence of his countryman Toby Aldweireld.

Alderweireld asserted himself as the best defender in the Premier League over the last two season before missing the vast majority of 2017/18, at first through injury, and what now seems fairly clear to have been extended due to contractual disputes.

To have Vertonghen immediately step up and fill his place, and in turn new boy Davinson Sanchez turn in a highly impressive debut season, has provided an immediate and easy solution to what could have been a quite catastrophic problem.

So who wins?

For my two cents, I think oldies’ favourite Jan Vertonghen just edges it, but to leave Christian Eriksen un-celebrated would be an injustice of it’s own. On balance an even divide of individual awards between the two ex-Ajax stars probably captures the mood on a very close call.


Download Download

Salah, Firmino & Mane all make Twelve’s top 5 Liverpool players, but should Lovren and Matip be there too?

Using algorithms and points to assess the performance of footballers is nothing new. But how do Twelve shape up against the likes of WhoScored and Squakwa? I’ve taken a look at Twelve’s stats for Liverpool’s players to see if their ranking for the Reds’ player of the year matches what Kopites would believe.

However cynical you might be about football statistics, and whether you don’t really care about how many goals and assists he’s racked up, you’re going to say that Mohamed Salah is Liverpool’s player of the season, right?

It’s a relief for all concerned to see Salah atop this chart. The runner up aligns with public perception too; I ran a quick Twitter poll to see who Liverpool’s non-Egyptian player of the year is, and Roberto Firmino was the runaway winner with over 90 percent of the vote.

Few Reds fans would disagree with the notion that Sadio Mané would be in the top five either, but Dejan Lovren and Joel Matip? Careful now, Twelve, this is how football stats pick up a bad reputation.

They have an advantage over some of their colleagues in that they are ranked sixth and eighth respectively for minutes played in the Premier League, but amending the chart to ‘Points Per Minute’ does not dislodge them from the top five.

Defensive contribution is highlighted in red, so of course the players in third to fifth here would perform well as it’s what they’re paid to do. But do they really contribute to attack (which is the green bar, and excludes shots) more frequently than Salah? And not just them either; there are 15 players who contributed more points per minute to attack than the newly crowned Golden Boot winner.

It seems Twelve rates contribution from further back on the field more highly, as the likes of Firmino, Ings, Solanke and Sturridge are all outside the top fifteen too. While my scepticism surrounding these figures remains, I am at least heartened by seeing Jordan Henderson at the summit of the attack rankings.

The skipper is a very divisive figure among Liverpool fans. I’m of the opinion the Reds have looked better with him in the side, and while it’s easy to quibble over Twelve’s scoring system, it’s clear that Henderson is definitely contributing to the attack and not just playing safe, sideways passes as his detractors will tell you.

When it comes to off ball work, my player of the season for Liverpool was undoubtedly Roberto Firmino. What he provides when the Reds don’t have the ball is a sight to behold, so why is he so low down in the Twelve rankings?

As with the attack figures, it seems the system rates these actions more highly at the back end of the pitch. Jürgen Klopp would beg to differ though, as he famously stated that “no playmaker in the world can be as good as a good counter pressing situation”, and it’s for this reason that Bobby Firmino will always be one of the first names on his team sheet.

Even if Twelve doesn’t value Firmino as highly as Klopp does, the level of off ball work done by the Brazilian is abundantly clear from their interactive widget. It will initially appear blank, so make sure you select the ‘Only important actions’ button.

Look at how many actions there are, and note how they’re across the whole width of the pitch, even with the final third excluded. Click on any of those bubbles, and it will say ‘Pressure on Pass’, illustrating how the former Hoffenheim man is a master of closing down opponents with the aim of springing Liverpool into attack.

There aren’t any surprises with regards to the top points scorers for shots or errors, only to add that the latter emphasises why Klopp was right to abandon Simon Mignolet and opt for Loris Karius between the sticks for the run in.

No model will ever be perfect, and Twelve don’t claim that theirs is. While I don’t agree with their scoring system entirely, with Salah and Firmino in first and second place for Liverpool’s player of the season, Twelve are on the right track here in many ways.


Download Download

The best attacking players outside of the Premier League’s big six.

The top of the Twelve rankings are dominated by players from the big six clubs. But if we exclude these players who tops our statistics? This is certainly an interesting question for fans of teams lower down in the league, wondering whether their favourite star will be whisked away by one of the richer clubs during the summer transfers.

The biggest within Premier League transfer-that-never-was in the January window was Riyad Mahrez (not) to Manchester City. And the statistics show us exactly why. When we look at attack, Mahrez tops the table outside (click here to see the whole list).

Looking in detail at his contributions over his last ten games we see that Mahrez’s attacking points come primarily from delivering longer balls that change the direction of play and from his dribbles in and around the box.

Add his goal-scoring to that and City would have a flexible player that would fit in to several roles in their attacking system.

Both Aaron Mooy and Pascal Gross, in second and third place respectively, have helped their teams survive to fight another Premier League campaign, contributing a lot to attack. However, neither of these players’ styles is particular similar to what the top teams are looking for. If we look at Gross’ most important contributions, they are primarily very long passes made as Brighton counter-attack.

This isn’t to say that Gross couldn’t play a different role, but he isn’t doing the things that most big-six scouts are looking for.

Probably the most interesting player from a big-six-signing perspective is Abdoulaye Doucoure. He is dribbling and making passes in that very difficult area in front of the box.

This is typically the type of player Arsenal like a lot and he could be of interest to a post-Coutinho Liverpool too.

Lastly, Xerdhan Shaqiri marks himself out not only in his creative midfield play, but also his goal scoring. Here are his shots and goals this season.

It seems very unlikely that Shaqiri will be playing in the Championship next season.


Download Download

Who is the best player in the Premier League? Here is what the numbers say.

The Premier League is drawing to a close and its time to look at who the best players have been. We at Twelve don’t think that numbers alone can’t answer the question, but we can make the best possible answer based on stats. And that is where Twelve comes in.

This is the first full season over which we have applied the Twelve model. Our aim is to be able to give a value to every action that occurs during a match. If you play Premier League Fantasy Football, you will know that you get points for goals, assists, minutes played etc. But you will also know that these points feel rather arbitrary. While the strikers like Mohamed Salah and Harry Kane who get the most points are undoubtably good, other players—in particular defensive midfielders such as Fernandinho and defenders such as Azpilicueta—are not given the appreciation they deserve in fantasy football.

Our model puts this right, because it is based on an underlying statistical model of how each action increases a team’s chance of scoring or decreases the opposition’s chance of scoring. We assign a goal 1000 points and every other action can be evaluated relative to that goal. So if a pass increases the chance of a team scoring by 5%, we give the pass 50 points. If a defender recovers the ball in a position where a goal is scored 23% of the time, he get 230 points. Its as simple and as powerful as that.

If you have heard of expected goals, then Twelve takes expected goals and extends it to every action in every game.

OK. Enough of the technical stuff. Here are Twelve’s top ten players.

Green is attack, red is defence, yellow is off-the-ball and blue is shots (and goals). If you click on the picture you’ll go in to an interactive version.

It is Kevin De Bruyne who takes it this season. Salah scored more goals, of course, but De Bruyne’s contribution to attack edged it for him.

One of the most powerful features of Twelve is you can see exactly why he is best. The widget below shows his attacking contributions, making passes from everywhere in front of the box.

That De Bruyne and Salah top the charts is hardly surprising (although it does also provide evidence that our model works!), but a bigger surprise is Granit Xhaka in third place.

Xhaka has only scored once this season, but has collected points both for his defensive work and his long distance attacking balls. This solves the problem seen in other fantasy football games, despite playing in defensive midfield Xhaka is properly rewarded for his actions.

Actually, Xhaka ranks highly for another reason. He had alot of playing time. If we look at points per minute, he drops to 10th place. De Bruyne and Salah also drop and David Silva jumps to the top.

The Twelve system likes creative midfielders, like Silva, Fabregas, Pogba and Hazard a lot, they make the passes that create better positions for the team. It also ranks defenders, like Otamendi (7th overall) and Azpilicueta (13th overall), highly for both getting the ball back and creating the next opportunity for their teammates.

But Twelve doesn’t forget defenders, for the teams lower down the league table, who have been working hard to keep the ball out their teams net. Best defender this season (in 6th place overall) is Shane Duffy.

Lots of clearances, won arial duels and blocks have put him up there with the stars of the big-six.

When it comes to choosing our season’s favourites there is always room for personal opinions and discussion. But if you just want to know which players contributed most to their teams, from a statistically rigorous model of football, then Twelve is your answer.


Download Download

Visualising Attacking Momentum Throughout a Football Match

Football matches are a prime example of a complex system that changes dynamically throughout the 90 minutes of play. One team can dominate the match for a given period of time, controlling possession and playing in the opposition’s half of the pitch. But this domination of possession and territory is rarely constant for the entirety of the match, and is a fragile domination that often leaves the team vulnerable to counter-attacks by the opposition.

That is the reason why we have developed an Attacking Momentum plot that illustrates the dynamic flow of a match. The X Axis represents the time in the match, while Y Axis represents the direction in which each team is attacking. In the example below Manchester City are attacking towards the top of the chart and United towards the bottom, and the line in the middle being the half-way line. The coloured spikes represent where the team’s chain of possession ended on the length of the pitch, narrower peaks represent short chains of possession and similarly wider ones represent longer chains. The circles in the chart indicate shots by the team with the smaller coloured indicator circles indication whether the shot was on-target (green), off-target (red), hit the woodwork (yellow) and a star indicating that a goal was scored.

A great example of this plot’s effectiveness in illustrating the changes of momentum in a match is the recent Manchester derby, this was truly a cliche’  “game of two halves”.  City dominated the first half, creating plenty of goalscoring chances, dominating possession and playing in United’s half. This domination meant that United had no shots at all in the first half and ultimately went into the changing rooms at half-time 2-0 down. The second half was an entirely different story, Mourinho’s men came out in the second half with something to prove. They took control of the match, and scored 3 goals by the 70th minute. This change in the match dynamics is clearly visible, in the first half City had more of the momentum illustrated by the high volume of light-blue peaks which were replaced by red peaks during Manchester United’s dominant period of the game and also highlighted as the only period of the match in which United had any shots on goal.

We can use this plot to better understand Roma’s surprising elimination of Barcelona from the Champions League in the quarter-final stage.

The first match between ended in a 4-1 victory to Barca who dominated the match in terms of shots on goal and Attacking Momentum. As is visible in the above figure, and expected at home in the Nou Camp, Barcelona enjoyed longer chains of possession and limited Roma to only 6 shots from open play. Adding salt to their wounds, Barcelona’s first 2 goals in the match came from Roma own goals through De Rossi and Manolas. Luckily for them, Roma’s late goal from Dzeko in the 80th minute gave them a lifeline in the tie, and ultimately proved to be crucial as they knocked the Blaugrana out in the second leg.

The second leg, shown above, was a completely different story. Roma completed the turnaround coming back from 3 goals down and going through on away goals, through an early goal from Dezko as well as De Rossi and Manolas scoring again, this time for their own team. As can be seen in the plot of the match, Roma had the majority of the Attacking Momentum in the match, they had longer chains of possession than Barcelona and created many goalscoring opportunities. Interestingly, Barcelona were limited to only 9 shots on goal and the majority of these shots came early in the first half or late in the second half, meaning that Roma very capably stopped Barca from playing their usual style of football. This is very impressive considering how dominant Barcelona have been in their domestic league this season, looking extremely likely to win the title in La Liga.

This is another plot that allows us to see the narrative of a football match swiftly and easily. There is little statistical evidence that momentum does exist in football, but regular spectators of the game will know that the 90 minutes of a football match are dynamic and the two teams trade periods of domination in a match. Using these plots we can now visualise this better than before as the Attacking Momentum plot allows us to see changes in the flow of football matches over time.

Emri Dolev

Data Scientist at Twelve Football


Download Download

Visualising a Football Match in One Plot

During the 17/18 Premier League season, Expected Goals (xG) have received more spotlight and their usefulness has been hotly debated. xG’s greatest public platform is provided by the BBC’s popular Match of the Day program, in which after a given match’s highlights are shown, the respective teams’ xG for the match is presented at the bottom of the screen (along with other traditional summary statistics).

After Arsenal lost 3-1 at home to Manchester United back in December, Wenger defended his team’s performance and claimed that the xG stat gave a better representation of the match than the final scoreline did. “In this kind of game you have to be efficient. We had the chances to score but we didn’t take them. We have produced an excellent performance but we conceded three goals. We had at least 10 chances and we were not efficient enough in the box. That’s the reality.” Sky Sport’s Jeff Stelling then attacked Wenger, “He’s the first person I’ve ever heard take any notice of expected goals, which must be the most useless stat in the history of football! What does it tell you?”

The simple answer to that question is that xG assigns a value to the chances of a shot resulting in a goal and Total xG is a measure of the quantity and quality of the chances which a team creates during 90 minutes, but it on its own (just like other traditional football stats) is limited. Traditional stats that are shown at the end of matches can describe how a match unfolded, but each number holds a limited amount of information. Greater possession, on its own, doesn’t mean that a team has won, similarly the number of Shots on Target or Passing Accuracy does not, these numbers need to be considered together to understand the context of a match. Yet, xG can be used as a complementary tool to better tell the story of a that match. Thus, we has developed a new chart which gives a better record of the match and portrays its narrative in a more valuable way.


Above we can see, on average, how many chances and of what quality every Premier League team has created this season. In general, teams that have created more chances of higher quality (in terms of the average xG of their shots) tend to be more successful. The current top 4 in the Premier League occupy this area in the chart, along with Chelsea and Arsenal who are 5th and 6th respectively.

We also want to clarify the context of a single match using xG and take a team’s defensive ability into account. To do this, in the figure below, each team is represented by a circle with the colours of their home jerseys, and its area represents the percentage of possession it had during the match. The team’s circle location is based on the number of shots it had during the match, and the average quality of these shots calculated as the average xG of their shots. Both of these can be compared to the league averages represented by the dashed lines in the chart.

Going back Wenger’s quote, we see that his argument is justified by the figure. Arsenal had the majority of effective possession—which is calculated as the portion of the time that the ball was in play and controlled by a given team—and created many more chances of above-average quality. United, on the other hand, had a below-average number of shots of high-quality. The difference between winning and losing in this match was due to a difference in efficiency in finishing.

Below is another big clash between teams from the top-6 in December last year. Manchester City and Tottenham had a relatively equal share of effective possession, yet City used it more effectively to create more scoring opportunities with relatively high probabilities of being converted into goals. Tottenham, on the other hand, created a very small amount of chances for themselves which usually did not endanger Ederson’s goal. The final scoreline reflected this, as City won 4-1 with Eriksen grabbing an injury-time consolation goal from just outside of the 16-yard box.

The dominant display of Manchester City in their victory over Chelsea can be explained using the chart of the match. Pep Guardiola’s philosophy has always been based around possession, in this match his team enjoyed close to 70% of it. This meant that Chelsea did not have enough of the ball under their control to build dangerous scoring chances for themselves, this is visible below as Chelsea had an extremely low number of shots in the match and on average they were of very poor quality. Pep’s side dominated the possession and used it to create a league-average number of chances, yet theses chances were of below-average quality – attributing to City scoring only one goal. Pep’s possession philosophy worked like he is always adamant that it will, as towards the later stages of the match the Chelsea players seemed tired of chasing the City players’ shadows, stopped pressing and generally looked as if they were looking forward to the final whistle and disinterested in trying to tie the game.

Expected Goals, much like other summary statistics is not very effective at telling the story of a match on its own and without context. These chart sets out to uncover as much information as possible from the xG stat and try to portray the context of a given match more effectively than has been done before visually.


Emri Dolev

Data Scientist at Twelve Football


Download Download

Team Possession Using Effective Match Time

A common way of estimating team possession over a match is by looking at the ratio of a team’s passes over the total number of passes. We at Twelve have recently developed a different algorithm for determining team possession by looking at the times the ball is actually being played. We call this effective time and base our calculations on possession chains.

Each action in the field can be represented as an event. A possession chain is then a sequence of events that happen during a match. The first chain of the match begins with the first touch of the ball and continues with each pass made by players of the same team. After two successful passes, the team is said to be in possession of the ball until their chain is broken. This can happen when either the opposition makes two consecutive passes and wins the ball back, or the game is interrupted by a stop event (ball goes out of play, an attempt on goal is made, etc.). When one chain ends another one starts and the process goes on until the final whistle. By looking at the start and end times of each possession chains, we determine the effective time of the match. There are occasions possession cannot be determined before a stopping event occurs. In these cases, chains do not get assigned to a team, but still contribute to the total effective time of the match.

Below you can find possession information about the recent match between Manchester United and Liverpool. From left to right, each team’s effective time has been split into defending (left), midfield (middle), and attacking (right) areas. On the top left corner, you can find the total effective time of the match in minutes. This unassigned time is shown as a percentage on the bottom right, but also features in the possession bar as a grey area in the middle.

From the diagram, we can see that Liverpool enjoyed much of the possession in the midfield and going forward, while Manchester United were very effective from very few minutes in the final third. Mourinho’s side showed great defensive discipline absorbing a lot of pressure from Liverpool, but while in possession of the ball, transitioned from defence to attack only for a few minutes in the entire game.


Download Download