STATS’ Data Supports FourFourTwo’s 100 Best Footballers of 2017

By:

It’s always been a conversation starter. Conversation has a tendency to escalate to debate. Debate at times gives way to argument.

And it goes 98 footballers deeper than that never-ending question: Messi or Ronaldo? In the FourFourTwo UK office, a friendship occasionally cools off for a day or two when compiling the 100 Best Players in the World.

The same thing happens across the Atlantic in the STATS headquarters as the list is revealed, so 2017 brings a fitting union of expert opinion and reliable data. FourFourTwo enlisted STATS this year to provide the leading football publication with analytical support for their much-anticipated annual list.

The 2017 version marks its 11th year, but it’s the first with STATS augmenting the collective opinion of FourFourTwo’s extensive worldwide staff of journalists. The list is being revealed from No. 100 to No. 1 over the course of the week, and it’s the culmination of a comprehensive global effort to provide an objective take on the calendar year’s performances.

“I would always say that it’s a starting point for conversation,” said FourFourTwo Global Digital Editor Gary Parkinson. “It’s a collective opinion. It’s a snapshot of the moment in time. There are a lot of moving parts to the 100 because it’s not just about the year that’s gone. There’s also an element of form is temporary; class is permanent.

“There are a lot of different ways to look at this, but it is certainly an entertaining and involving and engaging way to look at the year gone by, and it’s fascinating to look back through the history as well to look at years past. Hopefully we’re doing our job in analysing and educating, and STATS is now a part of it.”

This year, it began with a 19-year-old American at No. 100:

(Graphics by Olivier Maurel)

It’s the latest step for a partnership in which STATS has provided FourFourTwo with live data-driven insights during Premier League matches for the 2017/18 campaign.

“It’s very impressive – the level of service that STATS provides,” Parkinson said. “It easily meets the level of information we need to back up our judgement. It’s not purely a data-driven list, but we can use the data that STATS provides to make our case for why this guy should be No. 37 whereas the other is No. 38.”

Those numbers were revealed midweek with an Italian playing in Paris narrowly missing the top 30:

The value of the list doesn’t end at the fan engagement level. FourFourTwo isn’t shy about pointing out the occasional miss, but over a decade of the list shows there’s plenty to be gained from it in terms of forecasting football trends. At the list’s 10-year mark, FourFourTwo analysed itself, finding the 100 in recent seasons reflects the concentration of top talent moving more and more each year to top clubs. Or, from an international perspective, the list can help foretell how one nation might be about to experience an upturn in form while another is headed for a drought.

Evolving football trends or not, the list’s objective remains unchanged as it enters its second decade. Where certain awards in sport may heavily consider a given player’s collection of trophies, FourFourTwo tries to value the individual.

“It’s about the players within the teams,” Parkinson said. “Football is a team game, but this is an individual award that recognises that the player can be performing better than their team. So it’s not just about medals and even necessarily victories.

“A goalkeeper that’s playing really well on a struggling team might not win the most games, but he could be impressing. It’s about evaluating the players individually, and it’s not necessarily about the results of his team in the calendar year.”

FourFourTwo will announce its 2017 top 10 on Friday. Let the debates begin.

STATS Playing Styles Offers Growing MLS Market Opportunity for Richer Media Analysis

By:

There hasn’t been much lacking from the Seattle-Toronto MLS Cup rematch in terms of promotable storylines – except for true quantitative analysis to help contextualize them and show how the 2017 version could be more entertaining than 2016

MLS got it right this season when it elected to schedule playoff matches for midweek, and the structure conveniently delivered an opportune conclusion. It culminates in the favorable narrative of an MLS Cup rematch between Canada’s largest market and one of the league’s most popular clubs, and both sides arrive rich with subplot.

The historical storylines of this final go well beyond Seattle-Toronto Part II: It’s a dream matchup of the club striving for the first domestic treble in MLS history – Toronto – versus the club plenty anticipated might accomplish the feat a few years back. Seattle never reached that level of success with what many would consider its most talented rosters, but here it is trying to at once deny another club of that feat while itself becoming the third repeat champion in MLS history.

Not bad in terms of avenues on which to hype the Dec. 9 matinée in Toronto.

But there remains the question of how North American media can better engage soccer fans in a market that’s still relatively new to the sport and a sport that’s still conspicuously devoid of the kind of statistical tools to make that job easier. Moving MLS out of the weekend competition it faced with American football gave MLS more of an audience to work with, but how does the league keep that audience’s attention?

U.S. sports fans are used to quantitative analysis to augment or fuel the above storylines, and that’s an area where soccer has historically lagged.

STATS Playing Styles can help.

First, it’s of note to acknowledge the league’s recent gains thanks at least in part to its shift in viewership tactic. It’s been an autumn of headlines pointing out rating declines in the NFL, yet according to Sports Business Journal, regular-season MLS viewership across all networks was up four percent from 2016 and 41 percent from 2014. In Canada, TSN/CTV and TVA were up eight percent from 2016. Heading into its telecast of the weekend’s final, ESPN reported a 38 percent increase for its playoff audience. This coincides with MLS selling out advertising inventory for all of its U.S. network affiliations for the regular season, playoffs and MLS Cup.

No one’s getting burned, so it follows it might be time to double down and enrich media coverage to make the most of fan engagement.

It of course helps that the rematch of last year’s MLS Cup worked out as it did, but regardless of matchup or audience, the fluid nature of soccer has always been difficult to objectively analyze with relevant metrics on levels fans of other U.S. sports expect. STATS can now do just that.

For Seattle and Toronto, the on-pitch storyline painted by Playing Styles pits the league’s two most up-tempo teams.

Toronto has been one of the most dominant regular-season clubs in MLS history, and a win puts it in the discussion for the best MLS team of all-time. Let’s first glance at what that looks like in terms of style for the club that won the Supporters’ Shield – with a league-record point total, tied for the most wins (Seattle 2014) and tied for the second-most goals in league history – and the Canadian Championship (the U.S. Open Cup is typically associated with the treble, but being a Canadian club, Toronto doesn’t participate). Here are Toronto’s 2017 regular-season styles measured against league averages:

Toronto’s 2017 regular-season playing styles measured against MLS averages (0%). (Graphics by Stephan van Niekerk)

Playing Styles shows Toronto might not differentiate itself from the league average quite as much as one might expect given the level of dominance shown on the traditional table with team goals and points. However, notice its fast-tempo style is 58 percent above MLS average, which led the league. Fast tempo is defined as a possession in which the player releases the ball to a teammate in less than two seconds or the player dribbles at a high tempo, which is all objectively measurable thanks to STATS’ plentiful Tier 6 data.

The Reds’ MLS Cup counterpart followed in second for fast tempo:

Seattle’s 2017 regular-season playing styles measured against MLS averages (0%).

Seattle wasn’t quite as much of a fast-tempo side as Toronto, but it differentiated itself from the league in other ways – most notably build up (plus-30 percent) and sustained threat (+25), which combined with fast tempo, make up the typical profile of a dominant possession-based attacking side.

This is just the foundation of what Playing Styles is capable of displaying. It provides match-by-match breakdowns of team and individual possessions with their associated styles, which styles were present on shots and goals, how each player affects each style with offensive and defensive contributions, and much more. In short, it’s a tool that can complement an analyst’s keen eye and help get the point across with objective data rather than simply trusting otherwise unsubstantiated opinion.

As for the 2017 MLS Cup, an interesting consideration is how the above styles have carried over – or not carried over – into the postseason. Toronto scored three goals in four playoff matches, advancing to the final on a 3-2 aggregate spanning four matches against New York and Columbus, so it shouldn’t come as too much of a surprise that the most dominant club in the regular season played conservatively in the Eastern Conference semifinals and finals:

Toronto’s 2017 playoff playing styles measured against league averages (0%).

The Reds sat in more, relying on counter attack and direct play with increased frequency. Compared to the regular season, they displayed increases of 11 percent in counter attack and 22 percent in direct play.

The Sounders, conversely, went at playoff teams with far more dynamism. Their direct play dropped off, while the previously discussed possession-based styles flourished:

Seattle’s 2017 playoff playing styles measured against league averages (0%).

It cost them nothing on the back end as they cruised through four games with Vancouver and Houston on a 7-0 aggregate. So much of what’s been said about them in recent weeks has had to do with not giving anything up, and while the back line and goalkeeping deserve plenty of credit for compiling the longest shutout streak in MLS playoff history, Playing Styles gives more complex context to the success. It shows the Sounders’ midfield and attacking players certainly deserve some credit for the impressive run of clean sheets. They’ve made things quite a bit easier on their defense with that possession-based success, not to mention a high press that’s increased its presence by 28 percent from the regular season.

That’s how each club’s regular season compares to their postseason runs. Let’s now move onto the specifics of these teams actually playing and how we can reach deeper levels of matchup insight using Playing Styles.

It’s one thing for media to point out how Seattle hasn’t defeated Toronto from the run of play over the past two seasons. They played to a 1-1 draw in their 2016 regular-season meeting in Toronto, went goalless before Seattle won in penalty kicks in last year’s final, and Toronto won 1-0 in Seattle on May 6 this year as Jozy Altidore converted a 23rd-minute spot kick. This gives us some surface-level historical context. Take it for whatever it’s worth. It adds another level of analysis and insight to consider a match in detail.

Looking into the playing styles for each club in last year’s final, it’s easy to see that neither club asserted itself beyond the 8-0 discrepancy in shots on goal, which makes it seem like the Reds might have flat out dominated the match. Toronto, despite that advantage, was at -21 percent of the league average for sustained threat. Seattle, meanwhile, was -70 percent of the league average for fast tempo in that match, which works out to 200 percent less than its playoff run from this season.

With neither team asserting itself in possession-based styles, one might think counter attack and direct play came into play. Not so. Toronto was at -34 percent for counter and -30 for direct play. Seattle was -41 and -27. It was a very conservative match that fit the profile of a tight final.

We could spend all day analyzing the specifics of why the 2016 final played out as it did. We’ll pick one and run with it in Playing Styles to show how things may open up this year. The tool allows us to dig down to individual roles, of which there are plenty of interesting ones to choose from on both of these teams.

Let’s consider the value of Clint Dempsey, who wasn’t around for last year’s playoffs but has three of Seattle’s seven 2017 playoff goals. It’s well known that Seattle won last year’s title despite not putting a shot on target in 90 minutes of regulation or 30 minutes of extra time. It wasn’t just the final. Without Dempsey, that’s just how Seattle played last postseason.

We already noted above how that’s been anything but the case this postseason and Seattle has exerted its dominance over the past four games, three of which Dempsey has played in. Dempsey’s offensive contribution percentage to the team output when on the pitch is 22 percent this season. Significant to say the least. It’s the second-highest rate among regulars on either Seattle or Toronto, sandwiched between Altidore (23) and Sebastian Giovinco (21). In the playoffs, Dempsey has upped that to 27 percent, which is greater than the two prolific Toronto attackers combined (25).

Dempsey missed the playoff opener in Vancouver, a scoreless draw in which the Sounders put one shot on frame with six total attempts. In their last three playoff matches without him, they managed a total of 15 shots. With him over the last three matches, Seattle has 40. And 13 have come from possessions characterized as crossing style, resulting in four goals. Dempsey scored two of those. It’s a style that was, like others, absent without Dempsey: The last three playoff matches without him furnished just three such shots off crosses.

This might be another away final for the Sounders, but that doesn’t mean they’re going to mirror the conservative style they showed in the previous one. From a media perspective, in this case, STATS Playing Styles offers a way to objectively show a U.S. audience how a soccer match might have more action.

Seems like something those with an interest in ratings could get behind.

Burnley or Watford: Which Premier League Climber’s Impressive Start is More Sustainable?

By:

How STATS Playing Styles Foretells Viable Continuity for One and the Potential of a Defensive Letdown for the Other

More than a third of the way through the Premier League season, the table’s top six come as a surprise to few. The same cannot be said for the clubs situated in seventh and eighth entering the midweek 14th round of fixtures.

Six months after finishing just above the drop and separated only on goal difference with 40 points each, Burnley and Watford find themselves in the top half of the table and at the heels of a few clubs that will be playing Champions League football into the knockout phase.

The Hornets have done it with a new manager, an identifiable change in style, and an influx of fresh talent into key roles. The Clarets have been patient. Sean Dyche was retained, and they trusted the players and style that made for a touch-and-go 2016-17 campaign would come around. Both have thus far worked out with eighth-place Watford five points clear of ninth-place Brighton & Hove Albion, so let’s make sense of their respective climbs and the contrasting methods with which they’ve earned results to determine which club has staying power in the top half.

Marco Silva arrived at Watford this summer with a bit of a mess on his hands after Walter Mazzarri lost 20 matches and posted a minus-28 goal difference that in most seasons would signal relegation. STATS Playing Styles reaffirms a lack of identity:

Watford’s 2016-17 playing styles measured against Premier League averages (0%). (Graphics by Stephan van Niekerk)

Just over three months into this season, Silva has been heavily tied to the Everton position, and it’s not difficult to see why given the makeover that’s taken place at Vicarage Road. Consider Watford’s 2017-18 styles through 13 matches. Note that they’ve become more of a possession-based attacking team with noteworthy gains in build up, sustained threat and fast tempo, but pay specific attention to the club’s counter attack:

Watford’s 2017-18 playing styles through 13 matches measured against Premier League averages (0%).

A season after being 20 percent below the league average, Watford’s counter-attacking style ranks second behind only Manchester City (plus-36 percent) and ahead of Arsenal (+21), Liverpool (+16), Tottenham (+11) and Manchester United (+9). But that’s only part of the story. Now consider the efficiency of their transition game. Watford have had 70 possessions on which counter attack accounts for at least 50 percent of the possession’s value, and they’ve led to seven goals. That’s after managing four all of last season on 135 such possessions.

That’s what a new manager and the right players to carry out a plan can do. Those right players have been the 20-year-old Richarlison and Abdoulaye Doucouré. The duo has accounted for nine of the club’s 22 goals, but their value goes deeper than that in Watford’s style. For example, STATS Playing Styles Player Focus shows Richarlison’s individual influence on the counter attack has been massive, accounting for 41 percent of the club’s counter-attack distance dribbled when he’s on the pitch. That adds up to 469.1 metres, which trails only Mohamed Salah (532.2) and Kevin De Bruyne (506.4) in the Premier League. In 2016-17, Watford didn’t have anyone in the top 30.

Here’s where sustainability comes in. The Hornets’ 22 goals for come in just above their 21.7 expected goals for (xF), which means they’re not getting lucky or scoring in unlikely situations. Their scoring has been reasonable when weighed against historical league averages, meaning they’re creating opportunities that should allow them to sustain that level of scoring, which is considerably higher than last season’s rate of 39 goals for in 38 matches with an xF of 45.6.

It’s been chaotic at times – Watford were a defensive rollercoaster for nearly two months with 18 goals conceded in seven matches – but that’s also part of what makes their season sustainable. Their 21 goals against are by no means pretty and are more than anyone presently higher than 15th on the table, but again, it’s not an unrealistic or unsustainable departure from what’s expected. Their 23.7 expected goals against (xA) signals they’ve allowed fewer goals than the chances they’ve allowed would foretell, but it’s not the kind of disparity that throws up warning signs beyond the kind of defensive problems they’re already well aware of with those 21 actual goals conceded.

And that’s where bad news might come in for their fellow Premier League climbers.

First, the good. Consider Burnley’s 2016-17 styles, and it’s clear this was a compact team that attacked directly:

Burnley’s 2016-17 playing styles measured against Premier League averages (0%).

Now consider this season through 13 matches, and little has changed – very, very little to the point that at first glance, it might look like the same web:

Burnley’s 2017-18 playing styles through 13 matches measured against Premier League averages (0%).

While Watford made changes, Burnley are showing there might be something to be said for sticking with a system that may be on the verge of working. Patience doesn’t often win out in football when results aren’t coming, but the Clarets stuck with Dyche, and Dyche stuck with many of his guys. Burnley lost Michael Keane to Everton, but they’ve settled on a deep-lying back four of Steven Ward, Ben Mee, James Tarkowski and Matthew Lowton, who have started all 13 matches together and been lauded as a disciplined and organised unit. Nick Pope has been strong in goal after Tom Heaton went down, posting five clean sheets in nine starts.

Burnley are seeing results, but it’d be irresponsible not to look deeper into where those results are coming from given their style is nearly identical to last season. The truth is they’re living more dangerously than the traditional table shows. The Clarets have managed to turn 12 goals for into 22 points in 13 matches, putting them a point back of sixth-place Liverpool. Those 12 goals are in line with their expected 12.9, so it’s not as if they’ve been unlucky and are creating chances they can count on developing into a higher rate of scoring as the sample size grows. In short, offensively, this is who they’re supposed to be, which doesn’t leave much room for error at the back.

What’s worse is they’re actually still giving up chances considered of higher quality than their 10 goals against imply. Their 23.0 xA signifies their defensive successes have more to do with keeping legitimate chances out of the back of the net than they do with keeping the opposition from generating chances in the first place. That should act as a major warning sign going forward in terms of sustaining that surface-level defensive success.

Recall that things started to fall off for Burnley around this time last season with five losses in six matches and 13 goals conceded. It may very well play out again this campaign.

That doesn’t mean Burnley aren’t the well-managed club making the most of comparatively meager resources that they’ve often portrayed as. They absolutely are, and they nearly turned that into another point at the weekend until a stoppage-time penalty allowed Arsenal to escape Turf Moor with three points.

It just means sustaining their top-half presence is going to take something particularly special, whereas Watford could be better suited for nipping at the big clubs’ heels.

Valverde’s Barcelona: Stylish No More? Think Again

By:

STATS Playing Styles and Ball Movement Points Dispel the Myth that Barcelona’s Signature Form has Greatly Suffered Under Ernesto Valverde and Show How Philippe Coutinho Fits as a January Addition

A mid-August Spanish Super Cup defeat seemed likely to foreshadow Barcelona spending another campaign looking up at Real Madrid. It’s been anything but, yet even an unexpected eight-point lead over the European title holders hasn’t shielded the Blaugrana from criticism centered around the waning of their dominant possession-based attacking style. First-year manager Ernesto Valverde has said he won’t apologise for style when winning. The numbers show he might be right not to, and his detractors may want to take a look at the objective data before firing more arrows.

To say the Catalans are still at their best would be irresponsible because their best was another level of greatness. Even an undefeated La Liga run into November can’t hide that. Sergio Busquets, one of the holdovers from the club’s peak seasons, acknowledges they’re not playing with that same elegance he grew used to. Luis Suárez probably isn’t made for the left side the way Neymar was. Ousmane Dembélé is injured. Andrés Iniesta is still a midfield genius, but one with 33-year-old legs. Lionel Messi’s role has become more central and withdrawn under Valverde after we grew used to seeing him work into the middle from the right side in the MSN years under Luis Enrique.

Given Barcelona’s staunch summer pursuit of Philippe Coutinho, it seems they might have seen much of it coming and tried to bring in a talented player that’d fit Valverde’s system. And while STATS Playing Styles shows how Liverpool’s attacking midfielder might be the right guy to return to that style, it also shows Busquets might not be giving his team quite enough credit in his assessment of their early-season form as it compares to their results.

Busquets is on record saying the Blaugrana haven’t played “brilliantly,” which is difficult to define. It may look different, but the results in terms of objective style haven’t changed quite as much after Neymar’s departure as the critics may indicate.

First consider Barcelona’s style from 2016-17 when Neymar played 30 league matches:

Barcelona’s 2016-17 playing styles measured against La Liga averages (0%). (Graphics by Stephan van Niekerk)

Now consider this season below. Barcelona have actually played slightly more fast-tempo football without Neymar when measured against La Liga averages. Busquets might be noticing the difference in particular areas, which makes sense given the pitch perspective he’s working with: As a holding midfielder, he’s responsible for plenty of build up. That, along with Barcelona’s sustained threat – two styles indicative of possession-based dominance – have dropped off. Their build-up style is down from +125 percent of league average last season to +103 this campaign. Their sustained threat is down from +88 to +62. They’re still working the ball forward through build up at La Liga-best rates, though it might not be the level of dominance Busquets grew used to:

Barcelona’s 2017-18 playing styles through 11 matches measured against La Liga averages (0%).

STATS Playing Styles also throws into question the assumption that Valverde has transformed the team as a whole back into the ball-recovery style of old under Pep Guardiola. They’re a top-six team in high press among La Liga teams this season, but that style wasn’t as absent last season under Luis Enrique as many have thought. It’s actually dropped some from +19 percent of league average to +15 under Valverde.

There are even deeper shifts that have come from the personnel changes of the summer, and some have been rather positive. We’ve used STATS Ball Movement Points in the past to show the now-quantifiable value of playmakers such as Kevin De Bruyne, and we’ll use it here to show how Valverde is getting the most out of two players who fell off some in Enrique’s last season.

BMP considers every involvement a player has in a possession to credit or discredit decisions with the ball and reward creativity. It’s what football minds could always see but never calculate. It goes beyond expected assists by looking at the full chain of passes, weighing the probability of that pass leading to a shot later in the play, and assigning a value between zero and one. Passing points generate expected shot points, so if a player generates one BMP, he has generated passes to lead to or defend one shot. It expresses the level of threat or wastefulness that can be attributed to a player. It’s broken down into categories of offensive and defensive as well as positive and negative (oBMP+, oBMP-, dBMP+, dBMP-) with net values telling the more conclusive story.

oBMP also applies to defenders, and it’s particularly useful with wing backs. Jordi Alba has experienced something of a rejuvenation this season, stating himself that he feels he’s been opened up on the left side to attack more with Neymar gone. Alba ranked 24th in La Liga oBMP last season (3.41). This campaign, his 1.31 mark through just nine matches is up to 12th and on the level of players such as Busquets, Marco Asensio and Luka Modrić.

One player he’s not ahead of is Ivan Rakitić, whose midfield play hasn’t advanced just because of Messi’s positional changes. It’s also been augmented with Nélson Semedo at right back as Barcelona’s long-sought replacement for Dani Alves. Midfielder Sergi Roberto had the job last season, which resulted in playing with three at the back much of the time.

Rakitić ranks fifth in La Liga oBMP (1.68) behind just Messi, Isco, Jonathan Viera and Toni Kroos. The initial reaction to that might be one of surprise since Rakitić isn’t scoring – he had eight goals last season and has none so far this campaign – but Barcelona have enough scorers. They pay the Croatian for his midfield process, and that’s what oBMP calculates. It’s been measurably better after he ranked 27th in oBMP last season at 3.19. He’s over half way to that mark in 11 games, and that’s supported in Playing Styles with Rakitić’s percentage of on-pitch contributions for his club.

First, 2016-17. The number at the base of each style indicates his rank on the Barcelona squad based on his percent contribution to a given style when he’s on the pitch:

Ivan Rakitić’s 2016-17 contributions to Barcelona’s playing styles. The number at the base of each style indicates his rank on the Barcelona squad based on his percent contribution to a given style when he’s on the pitch.

Notice he accounted for nine percent of Barcelona’s oBMP+, and his player rankings among his own club in possession-based styles of maintenance, build up, sustained threat and fast tempo are lacking. Compare it to this season, and he’s clearly got a more involved role in the midfield’s attack with an increased oBMP+ contribution:

Ivan Rakitić’s 2017-18 contributions to Barcelona’s playing styles through 11 matches. The number at the base of each style indicates his rank on the Barcelona squad based on his percent contribution to a given style when he’s on the pitch.

He now ranks second in on-pitch build-up involvement percentage behind only Iniesta a season after trailing 10 Barcelona players. He now ranks third in fast-tempo involvement behind Dembélé’s small sample size and Alba a season after trailing nine teammates.

But Rakitić’s offensive involvement only tells part of the story. Semedo isn’t mirroring Alba on the other side of the pitch as Alves did and Roberto did at times. Rather, he’s playing in a truer defensive role, leaving Rakitić responsible for more of the right-side attacking distribution and less of the defending. The result? The midfielder ranks 52rd in La Liga defensive Ball Movement Points (dBMP) this season (0.25) after placing 205th last campaign (0.23). Barcelona are leaning on someone else to break up attacks in important situations, and Rakitić is probably in more suitable surroundings to avoid giving the ball away in dangerous circumstances.

Do a player-focused comparison in Playing Styles, and that’s supported by a Semedo’s 10 percent on-pitch defensive contribution versus Roberto’s seven percent mark last season.

Let’s now look ahead to the transfer window. We’ve used Playing Styles here to show team form and deep-level individual production. Let’s combine the two to show how a player might be of use in a different system.

It’s easy to say Coutinho is a fit for Barcelona for surface-level reasons. He’s a player who can slot into that left-side attacking role, possibly freeing up Suárez to work more centrally, or Coutinho can fall into more of a midfield role to replace Iniesta’s still-busy but worn legs. However, especially in today’s transfer market with fees reaching incredible levels for players of Coutinho’s quality, clubs might need a bit more empirical evidence. That exists with Coutinho, and it goes beyond relying on the assumption that a move from a gegenpressing Jürgen Klopp system to Valverde’s is in some ways a lateral move.

Coutinho has only played in five of Liverpool’s 11 league matches this season, so we have nearly equal sample sizes of the Reds’ style with and without him. Consider first Liverpool’s matches without him:

Liverpool’s 2017-18 playing styles without Coutinho measured against Premier League averages (0%).

The five matches he’s played have all been starts in which he’s played 90 minutes twice and 79 minutes three times. Consider the Reds’ style here, noting the massive increase in sustained threat and fast tempo, and also the dip in direct play – all of which suit Barcelona:

Liverpool’s 2017-18 playing styles with Coutinho measured against Premier League averages (0%).

Ronaldinho has gone out of his way to state Coutinho would be a perfect fit for the 2005 Ballon d’Or winner’s former club, and the data shows he’s probably right about his fellow Brazilian No. 10. Coutinho can play at fast tempo and won’t be overwhelmed by a Barcelona attacking style often dictated by the likes of fleet-footed Messi or the overlapping Alba. What’s more is he might be the right player to retrieve what Barcelona have lost in terms of style this season because he’s clearly familiar with operating in tight spaces of sustained attacking threat like Iniesta. Notice even the ever-so-slight increase in Liverpool’s high press when Coutinho plays. That’s a system both Liverpool in recent seasons and the current Barcelona boss value.

It’s all supported by Coutinho’s involvements for Liverpool. He’s contributing 18 percent of their build up, which is higher than even Iniesta (15). He’s contributing 17 percent of their sustained threat, which at Barcelona trails only Messi (19). And he’s contributing 15 percent of their fast tempo, which would lead the Spanish giants. How effective has he been in doing so? His 17 percent oBMP+ contribution leads Liverpool, meaning he’s their most ambitious creator. That mark is bettered by one man on either club: Messi.

That’s a style that may suit Valverde. And Busquets might even call it brilliant.

We Thought We Might Find You Here

By:

Revisiting STATS’ Preseason Premier League Predictions by Using the Most Telling Metric in Football – Expected Goal Value – to Forecast Results

Arsenal head to the Etihad in an attempt to disrupt the Premier League leaders. Manchester United visit Stamford Bridge trying to further distance themselves from the English holders. Tottenham try to avoid a slipup against the division stragglers.

It’s going to be a big Sunday at the top of the Premier League table, and 10 matches in, it’s about time to revisit the predictive analytics that hinted we’d be sitting exactly here. The top five is surprising to few as the calendar switches to November, but we didn’t settle for a simple opinion piece when crafting a narrative for why this would happen. STATS gave the data-driven specifics of why that would be the case way back at the beginning of August.

First, a glance at the top of the table with traditional goals and points versus expected goals and points. We’ll explain plenty of this as we go.

Sortable 2017-18 Premier League Table

(Top Five Through 10 Matches)
TeamGoals ForExpected Goals For (xF)F-xF (+/-)Goals AgainstExpected Goals Against (xA)A-xA (+/-)Goal Diff.Expected Goal Diff. (xGD)GD-xGDPtsExpected Pts (xPts)Pts-xPts (+/-)
Manchester City3528.16.966.6-0.62921.57.52821.96.1
Manchester United2322.90.1410.9-6.91912.07.02318.05.0
Tottenham Hotspur1922.6-3.679.6-2.61213.1-1.12019.70.3
Chelsea1812.95.11013.1-3.18-0.28.21913.25.8
Arsenal1920.2-1.21310.32.769.9-3.91917.71.3

Manchester City

We’re starting at the top, but we’ll bury the lede some by beginning with City’s seldom-tested last line of defence and work our way forward.

We noted in August that City’s goalkeeping last year was part of the reason they finished so far back of Chelsea, and Pep Guardiola did something about it by adding Ederson. The Benfica import has been doing his job in the way last season’s keeping duo didn’t, even if he’s only been faced with 18 shots on target in 10 league matches. Ederson’s save plus-minus (calculated by subtracting expected saves from actual saves) is plus-0.5, which means he’s operating just above league averages. It’s not exactly on the level of Lukasz Fabianski (+6.4), Nick Pope (+5.1) David De Gea (+5.0) and Fraser Forster (+4.7), but he hasn’t been a leaky liability that’s canceling out the attractive work of City’s attack.

That was one of many moves City made this summer, and the spending spree turned plenty of people off to the idea of Guardiola’s supposed genius. It doesn’t take a football mastermind to hand out millions. But that’s a lazy way of looking at things. It doesn’t have as much to do with big spending as one might think. Rather, it has to do with some of the players who have been around performing to high standards, and for that Guardiola deserves at least partial credit.

We spoke in detail a few weeks ago about the once-hidden, now-quantifiable value Kevin De Bruyne brings as a creative force. But back in August, we pointed out that the Sergio Aguero of last season wasn’t exactly the Sergio Aguero we’re used to, despite reaching 20 goals in 31 league matches. He had a -4.1 scoring plus-minus (G-xG), meaning he should have scored four more goals than he did given the opportunities he had. Sure, we can say he was productive with those 20 goals, but it’s hard to say he was efficient. As was expected based on performance of past seasons, that efficiency returned, and he’s back to converting his chances at more impressive rate with a +1.2 mark in just seven matches.

He’s not the only one. In terms of goal plus-minus, Leroy Sane (+3.6) leads the Premier League, and Raheem Sterling (+2.6) is third a season after a -1.8 mark, the latter being one of the players we noted could be in for a turnaround back in August.

We also noted Manchester United’s finishing wasn’t going to be suppressed as it was in the second half of last season, and the Red Devils provided results right from the start.

Manchester United

If anyone tells you the Premier League has already been decided, at least make them expand on their argument beyond eye test and a five-point lead. Yes, City have looked like legitimate Champions League contenders, and yes, United have cooled in recent matches after petrifying back lines and keepers out of the gate.

It certainly doesn’t look good for anyone other than City, but what’s interesting to consider here is there may be more potential for Jose Mourinho’s side to close the gap than for Guardiola’s to expand it, and that logic goes beyond the obvious surface-level arguments such as Zlatan Ibrahimovic’s looming return.

We could talk about how Romelu Lukaku has scored seven goals in 10 matches for United yet still has a -1.1 plus-minus a season after being incredibly efficient at +9.9, but let’s skip right to the bigger picture of complete team performance. Their No. 9’s nice goal total with rather average finishing is a microcosm of the whole. United have a +0.1 goal plus-minus, meaning they’re right at their expected scoring for the season.

Their goals against plus-minus (A-xA), however, is at -6.9, meaning they’ve allowed far fewer goals than anticipated, thanks mostly to stellar work from De Gea. It works out to a goal difference discrepancy (GD-xGD) of seven goals – their actual goal difference is 19 while their expected is 12.0. That means they’re overachieving, so where’s the potential to close the gap?

City are actually overachieving a bit more. Their goal difference of 29 is 7.5 higher than expected. What it amounts to is a small amount of intrigue on the table. City’s 28 points are a considerable inflation from their 21.9 expected points. United’s are as well – 23 and 18.0 – but that would suggest the potential for at least a slight closing of the gap as the season goes on.

Here’s where we’ll go out on a limb with an intriguing deviation and state United’s defensive productivity might be sustainable because the sample size is greater than 10 matches. Mourinho’s side did this last season, allowing a league-low 29 goals with a 39.3 xA. Mourinho’s keeper did this last season, posting a +7.1 save plus-minus. Oh, he also did it in 2015-16 at +5.8. And 2014-15: +6.8. That’s a sample size to rely on. De Gea is not lucky. He is objectively better than most keepers, so United will continue to allow fewer goals than expected.

And then they went and added a central presence that Chelsea really could have used during this run of matches without N’Golo Kante. Nemanja Matić has further stabilized things for United, and that’s measurable with STATS Ball Movement Points, which we’ve used and explained in detail in a number of recent posts. Matić ranks 11th in offensive ball movement points (oBMP) among Premier League midfielder and 17th in defensive ball movement points (dBMP). He’s one of four midfielders to rank in the top 20 of each category.

For the sake of this point, the defensive number is the one to pay attention to – it means Matić makes responsible decisions with the ball in his own half – and that only helps the back line and standout keeper.

Bring in the surface-level changes possible such as a healthy Ibra working with Lukaku, and there’s at least some intrigue. But as we said in August, it’s going to take something special to deny Guardiola his first Premier League title.

It follows that it’ll take something truly exceptional for any London club to legitimately contend.

Tottenham

The love Harry Kane has received this season has spread well beyond the UK. It’s been immense on an international level with the 24-year-old being thrust into the Messi-Ronaldo discussion as a truly elite-level finisher. On the surface, that might be warranted given the sustained scoring run he’s been on. But you can read about that elsewhere while we focus on the predictive end.

Kane leads the Premier League with eight goals in nine matches, but the true performance numbers show he’s not necessarily distinguishing himself as the efficient finisher as he did last season. He finished the season with a +13.3 plus-minus, which led Europe’s top-five leagues. This season, he’s at -0.2.

This was the concern we noted entering the season for Spurs, and it’s played out on a team level. Sure, Kane has remained productive, but it’s hard to say the same for the team as a whole. In terms of scoring productivity, Tottenham has returned to Earth a season after scoring a league-best 86 goals. That happened because they were notably efficient at 17.6 goals above expectation.

That’s not all on Kane. As we pointed out before the season started, Heung-Min Son went wild last season with a +6.2 plus-minus, meaning his 14 goals were nearly twice that of what was realistic measured by league averages. Dele Alli was at +5.3. It simply wasn’t sustainable on a team level, and that’s played out with the two combining for four goals and a -1.9 differential in a collective 19 league matches.

As a team, their 19 goals are tied for fourth and their 22.6 expected goals for rank third. That means there’s room for improvement this season, which carries across to expected points and makes Spurs London’s best hope for title contention. Consider Spurs 19.7 expected points are ahead of United and 2.2 behind City, and the Manchester clubs may still feel occasional pressure from a side that wasn’t able to end its title drought with a statistical dream season last campaign.

Come spring, Tottenham may again have to settle for topping their North London rivals, though that’s far from guaranteed to happen.

Arsenal

The thrust of our warning for Arsenal in August was that strong goalkeeping could be as much a sign of weakness at one position as a show of strength at another. The Gunners’ +11.7 save plus-minus last season lead the Premier League, and this term they’ve shown how unsustainable that was.

That’s not to say Petr Cech hasn’t been bad this season. His -0.4 save plus-minus implies he’s made the saves expected of him, and less than one save separates him from Ederson, whom we spoke of favourably above. They just haven’t quite figured things out defensively, as is evident by those 13 goals against that put them behind Burnley, Newcastle, Southampton, Brighton & Hove and Swansea, while no one else in the top five has allowed more than 10.

But there’s a positive here. Notice above we said they haven’t quite figured it out. Last season, Arsenal conceded 44 times with a 55.1 xA. That works out to 1.45 expected goals against per game. This year, through 10 matches, their expected goals against is 10.3. Slide the decimal for your per-match average, which over the course of the season is a substantial improvement.

That signals that they’re at least not giving as many high-percentage scoring chances as they were last season. And for what it’s worth, it’s better than the title holders.

Chelsea

We began our Chelsea section in August by stating the following: “The Blues conceded 33 goals last season, which was third to Manchester United and Tottenham. That number matters because it was consistent with the club’s expected goals against (31.8). They didn’t have a keeper constantly bailing them out. Their system worked.”

What also matters is having the players on the pitch who make that happen, which hasn’t been the case and is evident by the fact that Antonio Conte has used different starting XIs in the club’s last 13 matches.

Kante hasn’t played since September, and Chelsea have conceded 11 times in the six matches across all competitions he’s missed after allowing five in their previous nine. Again, it certainly doesn’t help that Matić unfortunately now plays for the club they’re facing this weekend.

Add in backline red card suspensions and inconsistency with new faces, and the lack of stability results in a 13.1 xA – 41.2 percent of the way to last season’s xA through just 26.3 percent of their fixtures. It signals they might be lucky to have only conceded 10 times in league play.

The Blues’ expected goals for – which you can see and sort in the table above – is an even more alarming indicator, but don’t forget last season. Recall Conte made in-season changes that were a major part of Chelsea’s title run. But given the strength at the top and the aptitude the Manchester clubs are displaying, it’s going to take something incredible to swing the kind of comeback necessary for a third Chelsea title in four years.

We ended that season preview by saying three terms – productivity, stability and efficiency – are telling, and they’re more measurable than ever. Given how this has played out, we’ll add that they’re more predictive than ever, too.

Valencia in Transition

By:

Using STATS Playing Styles and Advanced Metrics to Show How Marcelino’s Valencia are Employing an Effective Counter Attack to Return to Spanish Relevance

Marcelino couldn’t contain his excitement and pulled up short with a hamstring problem after Simone Zaza scored a late winner against Real Sociedad in late September. The injury was of little concern to his club for at least a triad of reasons.

  1. Marcelino, 52, is Valencia’s manager, and no part of a professional football match has hinged on the performance of his legs since 1994.
  2. It was overshadowed by the fact that it gave Valencia yet another result to sway detractors amassed over the past two seasons back in their favour.
  3. This was hardly an exceptional occurrence. Marcelino, as a manager, once injured himself taking a seat for a presser.

It’s unsurprising to see Marcelino competing for the celebratory spotlight with his players in his first season with the club, though he’s said he recognises the need to tone down his touchline merriment. Given the past two campaigns, it’s of greater note that Valencia have had so many such opportunities through their first eight matches with the excitable man in their coach’s box.

Using STATS advanced metrics, we can show his players might need to reciprocate and go crazy for their new manager once in a while, even if he’s being particularly unpredictable when determining whether he’ll include each of them in his starting XI. Eight different lineups in as many fixtures might seem as erratic as the manager’s touchline fervor, but there appears to be something of a method behind it. We’ll get back to that in great detail on team and individual levels.

First, a bit on why Valencia’s strong start matters.

It wasn’t as long ago as it may seem that clubs outside of Madrid and Barcelona won the Spanish top flight. The last were Valencia in 2003-04 and ’01-02, and before that Deportivo La Coruña in 1999-2000. Modest success hasn’t eluded Valencia since. European football had been an expected part of the gig at Mestalla until recently. But there’s no arguing the past two campaigns in which supporters have endured successive 12th-place finishes – their first in the bottom half since 1987-88.

Los Che now find themselves second in the table as one of three unbeatens in what’s arguably Europe’s top league. The other two – Barcelona and Atlético Madrid – have reached at least the quarterfinals of the Champions League for the past four years.

So how do Valencia find themselves back up in the early-season mix for direct Champions League qualification? It’s not for a lack of competition. Quite the opposite, in fact, and it could be argued Valencia’s early-season fixtures have been as demanding as any Spanish club. Four of their eight league matches have come against clubs playing European football. That injury-inducing match came away to Sociedad a week before the same 3-2 scoreline played out in less-exciting fashion at home against Athletic Bilbao, Sociedad’s Europa League contemporary. But those victories bury the lede of quality draws in Marcelino’s second and third matches with the club.

Valencia left the Bernabéu with a 2-2 result against the reigning European and Spanish champions after holding a lead into the 83rd minute, then followed the first international break with a scoreless home draw with Atlético.

Most recently was Los Che’s chaotic 6-3 win Sunday at Real Betis, who remain in the top half of the table.

So what’s changed from a season ago? A bit of everything. The manager, of course. Players. Players’ efficiency. Method – and that’s where we’ll begin by calling upon STATS Playing Styles before going more granular with advanced individual metrics.

Last season, Valencia went through four different managerial periods and three different bosses – Pako Ayestarán until Sept. 20, club ambassador and longtime centre back Voro González for the next eight days, Cesare Prandelli from Sept. 28-Dec. 30, and the ever-present stopgap Voro again until the end of the season. In terms of style, that unsurprisingly amounted to very little differentiating from La Liga averages:

Valencia’s 2016-17 playing styles measured against La Liga averages (0%).

They played more of a fast-tempo game than much of the league, but they didn’t sustain threat when doing so and were rather blasé in all other areas. What followed was a minus-9 goal difference for their worst mark since ’07-08 (-14) when Los Che finished 10th.

Through eight matches this term, there’s still not some overhaul of telling possession-based attacking styles that typically signify a dominant club – they rank 16th in possession at 45.3 percent, which is lower than last season (48.3 percent). It’s unsurprisingly a drastic departure from other top teams in the table. Barcelona are first (61.2 percent) and Real Madrid are second (60.6). But there is order to how Valencia score goals. It’s frequently about transition:

Valencia’s 2017-18 playing styles through eight La Liga matches measured against league averages (0%).

That plus-55 percent counter attacking style against the league average leads La Liga – yes, ahead of even counter masters Real Madrid (+37 in second). Among the top-five European leagues, only Benevento in Italy are countering at a higher percentage against their league average. Anyone familiar with the Italian table then jumps to a logical question: Why are Valencia succeeding and Benevento the clear-cut worst side in Serie A with eight losses and a minus-17 goal difference?

The answer is probably that counters ending with a striker tripping over the ball don’t mean much. The Serie A newcomers have had 59 possessions with a counter attacking value of at least 50 percent, and it’s amounted to two goals. And they spend far more time defending, as is evident by their overall style, so they’re not exactly creating chances in other tactically sound ways:

Benevento’s 2017-18 playing styles through eight Serie A matches measured against league averages (0%).

Valencia, meanwhile, are effective in their counter – more effective even than Real Madrid. Among Los Blancos 48 possessions on which their counter attacking style is of a value of at least 50 percent, they’ve scored once. Valencia have 53 such possessions and four goals after scoring eight goals off the counter all of last season.

All of this must originate somewhere, and that’s where counter attack regains come in. Last season, Valencia were 13th in La Liga in regains to begin a counter attack (154). That trailed leaders Real Madrid by 70, so nearly two per match. Their counter attack distance covered (8,705 metres) – made up of the total of counter attack distance carried and counter attack distance passed – ranked 14th. This season, they’re first in regains (53) and distance (3,171 metres) – more than a third of the way to reaching last season’s marks.

There are more reasons transition works for one club and doesn’t for another. To measure the efficacy of the correlating defensive and midfield play, we’ve got to get beyond simple sums. We showed last week how Kevin De Bruyne has been one of Europe’s most dominant offensive players, despite having a comparatively limited direct involvement with goals and assists. We did this with STATS’ ball movement points. BMP is a metric that considers every involvement a player has in a possession to credit or discredit decisions with the ball and reward creativity. It’s what football minds could always see but never quantify. It goes beyond expected assists by looking at the full chain of passes and weighs the probability of that pass leading to a shot later in the play. Passing points generate expected shot points, so if a player generates one BMP, he’s generated passes to lead to – or defend – one shot.

Yeah, that’s ambitious. So how is this done? The process is measured and assigned objective value using machine learning and massive amounts of historical league data to express the level of threat or wastefulness that can be attributed to a player. It’s broken down into categories of offensive and defensive as well as positive and negative with net values telling the more conclusive story.

There’s dBMP+, which measures how many created chances a defender prevents – breaking up attacks in important situations. There’s dBMP-, which measures liabilities in possession – giving the ball away in dangerous areas. Combine that for net dBMP. While Benevento sit in the bottom half of Italy with a 0.13 dBMP rating, Valencia (0.27) lead Spain. So we’d previously established with playing styles that Benevento are spending a lot of time defending, and dBMP helps us show that they’re not making great decisions with the ball when doing so. Valencia might not be the most attack-minded club in Spain, but they’re at least effective in their own half. That might not matter quite as much for ball-dominant clubs such as Barcelona. It absolutely does for sides that have to pick their attacking moments judiciously.

So on the pitch, who specifically is to reward for executing the system Marcelino seems to be implementing?

We’ll start with the sexy goal-scoring numbers from a striker mired in that special brand of Italian sorrow this time last year for his happenings with club and country.

Zaza scored six goals in 20 matches in his time with Valencia last season and has passed that already this season with seven and three winners. With six goals in his last four matches, he seems a healthy distance from his Euro 2016 penalty miss for Italy and his disappointing spell with West Ham United. The numbers back that up with the 26-year-old ranking in the top five in Europe’s top-five leagues in finishing with an expected goal differential of plus-3.5 among a pretty elite group a season after posting a minus-1.9 xGD. Notice than in Spain, he jumped this past weekend ahead of even a guy named Messi:

As we noted before with ball movement points, midfield play has a lot to do with Valencia’s success, and that’s true on an individual level as well. Bringing in on-loan Geoffrey Kondogbia from Inter Milan as a central presence might have displaced 20-year-old Carlos Soler some after the latter became a mainstay in the middle last season, but it seems to be working out for Marcelino. Kondogbia, who’s attracting attention from top Premier League clubs, ranks second among all midfielders in Europe’s top-five leagues in dBMP, and he’s one of three to really distinguish himself from the pack:

Valencia don’t quite make the same use of the corresponding playmaker guiding a dangerous attack at the other end. Their top-ranked player in oBMP among the top-five leagues is Dani Parejo tied for 29th, but when filtering that down to only La Liga, it’s good enough for fifth among a star-studded top 10. It’s rather impressive when considering the opportunities and surrounding creatives much of the rest of this list has to work off of:

(Graphics by Stephan van Niekerk)

Bored yet? OK, let’s talk goals again. We can’t forget about Rodrigo, who scored five goals in 19 La Liga matches last season and was an objectively mediocre finisher with a -0.6 xGD. He was with Spain as they wrapped up qualifying last week for reasons that go beyond David Villa nearing 36 years old. Rodrigo has scored in five straight matches for the club and also got one in his start against Albania on Oct. 6. Although none for Valencia have been match-winners, he has compiled early-season efficiency (+1.6 xGD) to show he’s not exactly feasting on scraps.

Finally, goalkeeping. Neto, who spent the past few seasons behind Gianluigi Buffon at Juventus, has a +2.1 expected save differential, which is calculated by subtracting expected saves from saves to show how a keeper is performing against league averages. That mark ranks sixth in the division and, you guessed it, is better than his former mentor Gigi (+0.7). It’s not quite the level of Pau López (+6.2), Jan Oblak (+6.0) and Guaita (+5.3), but the Valencia keeper is still going above and beyond from time to time. It’s also important here to consider that Valencia aren’t leaning on him to continually bail them out in an unsustainable way.

So Valencia have a manager pushing for some consistency in style, and he has players making it happen at various levels that we’re now equipped to properly measure. That’s what it takes to earn 18 points in Spain through eight matches, three of which have been draws and another three being one-goal victories. But this is La Liga, home to the two most dominant clubs in the world in recent years. Recall Real Madrid’s Spanish-record unbeaten run of 40 in all competitions ending in January. Barcelona notched 39 in 2015-16. Is it right for a club that’s won domestic titles of its own to make much of this just yet?

Given the circumstances of the past two seasons, it somehow feels right for Marcelino to stick with those celebrations.

The Daring Process of Kevin De Bruyne

By:

How STATS Ball Movement Points Show Manchester City’s Midfielder Has Been the Most Dangerous Creator in Europe This Season

Kevin De Bruyne’s 67th-minute strike in a 1-0 win at Stamford Bridge wasn’t just an aesthetically impressive example of fast-tempo football – it was arguably the most important goal to be scored in England so far this season. It distanced Manchester City from title holders Chelsea entering the international break and gave the 26-year-old his first Premier League goal of the campaign against the club that sent him to Wolfsburg.

That finish, as pretty as it was, isn’t the main reason we should praise De Bruyne’s efforts for the Premier League leaders.

Even in that six-match goalless stretch to start the season, you weren’t going to hear many supporters complaining about De Bruyne’s form the way one might if Sergio Aguero went through such a drought. Nor will you in the week following his inclusion in the Ballon d’Or 30-man list, from which fellow City creative David Silva is somehow once again conspicuously absent.

De Bruyne has been directly involved in four of Manchester City’s 22 Premier League goals. His goal – which featured the Messi-esque combination of build up and finishing inclusions from a pretty one-touch layoff to Gabriel Jesus before getting it right back on the run for a two-touch 20-yard left-footer to the upper corner – and three assists works out to an 18 percent involvement from the goal-and-assist perspective. It’s not a particularly high rate, and on a personal level, it’d be his lowest in three seasons with City.

But that’s precisely why KDB’s start to the 2017-18 season can act as one of the most relevant contemporary examples of why there need to be better metrics in football. Now settled into a deeper central position after the tinkering Pep Guardiola went through with his creatives in his first season in charge, De Bruyne is still generating all sorts of threat. He’s always been the type of creative midfielder who gains praise for his passing ability and his field awareness. He makes work difficult for defensive players with his precision, and his ball movement is rarely lacking ambition. It’s now measurable with ball movement points, which reward the process in a way traditional binary metrics such as successful passes fall short.

And, through seven matches, BMP shows the Belgian has evolved into the most dangerous playmaker in Europe with his globally overlooked club teammate not far behind.

First, a quick rundown on BMP, which we’ve used a few times before when discussing transfers and how key players are contributing to the top-five European leagues. BMP is a metric that considers every involvement a player has in a possession to credit or discredit decisions with the ball and reward creativity. It’s what football minds could always see but never quantify. It goes beyond expected assists by looking at the full chain of passes and weighs the probability of that pass leading to a shot later in the play. Passing points generate expected shot points, so if a player generates one BMP, he’s generated passes to lead to – or defend – one shot.

Yeah, that’s ambitious. So how is this done? The process is measured and assigned objective value using machine learning and massive amounts of historical league data to express the level of threat or wastefulness that can be attributed to a player. It’s broken down into categories of offensive and defensive as well as positive and negative (oBMP+, oBMP-, dBMP+ dBMP-) with net values telling the more conclusive story.

Got it? Good. Back to KDB.

First, the basics. In 2015-16, his seven goals and nine assists among Man City’s 71 goals amounted to 22.5 percent involvement. Last season, his six goals and 18 assists out of 80 Man City league tallies amounted to a 30 percent involvement. At that level, this season’s inclusion looks like a regression.

It’s not. While Guardiola experimented last season, De Bruyne’s 7.62 oBMP still managed to place second in England behind Mesut Özil (9.00). The season before, Manuel Pellegrini’s last, his 4.14 oBMP ranked 16th in England, and that came behind teammates Yaya Toure (4.82 in eighth), Silva (4.60, 10th) and Fernandinho (4.32, 13th). Özil (10.95) led then as well. For the sake of comparison, La Liga’s leaders last season were Lionel Messi (7.52) and Toni Kroos (5.97).

Onto the current term. It’s of course very early in the season and this is a tight pack, but now consider this season’s oBMP rankings across Europe’s top-five divisions in league play. Of the top 20, 11 are midfielders, and plenty don’t have the goals or assists to display their true value:

(Graphics by Stephan van Niekerk)

De Bruyne is leading, but he’s also the only player to separate himself from the next best player by any considerable sum. Project that 2.29 mark out over a 38-game season, and he’s looking at a 12.44 oBMP that cruises past his own marks from the past two seasons and even Özil’s. And right behind him is his teammate, who again hasn’t been given any love by the Ballon d’Or brass despite consistently creating on elite levels among the world class.

It follows that these players must have a pretty impressive oBMP+, meaning they are ambitious and effective with ball circulation in the attack – they find the channels and play a mean through ball, or they consistently deliver that low, bending cross that makes centre backs trip themselves. It also follows that they may have a considerable oBMP- because of the number of chances they have to craft opportunity, but they’ve got to limit that to exist as a leading creative player. For example, Alexis Sanchez ranked fourth in oBMP+ last season (10.66) but first in oBMP- (-5.01), so his wastefulness drops his net oBMP (5.65), which was not only a considerable margin behind teammate Özil but also Arsenal midfielder Granit Xhaka (5.92).

Now keep in mind BMP does not take into account finishing, which STATS quantifies by calculating expected goal differential (subtracting expected goals from actual goals). That’s where players such as Radamel Falcao (plus-5.8), Paulo Dybala (+4.7), Ciro Immobile (+3.5), again Messi (+3.4) and Mathew Leckie (+3.1) have distinguished themselves this season.

Maybe you can see where this is going. We’re going to write more in the coming weeks and take this a step further by quantifying the value of overall offensive contribution. But when you focus solely on the process of getting the ball to those finishing players in positions to succeed, right now there’s been no better creative than De Bruyne.

Styles of Their Own: How Deep Data Differentiates the Best of Europe’s Best

By:

Assessing the Leaders of La Liga, Bundesliga, Serie A, Premier League and Ligue 1 with STATS Playing Styles, Expected Goals and Saves, and Ball Movement Points

The tables will tell you the five leaders of Europe’s top leagues all have goal differences between plus-17 and 19 through six or seven matches, but the brand of dominance with which they’ve arrived there varies. With Champions League fixtures occupying midweek, let’s look into the tendencies of Barcelona, Manchester City, Paris Saint-Germain, Borussia Dortmund and Napoli in league play using STATS’ advanced metrics.

We’ve talked about playing styles and expected goals and saves plenty. Ball movement points less so. BMP is a nifty metric that rewards creativity, going beyond expected assists by looking at the full chain of passes and weighing the probability of that pass leading to a shot. It measures the process and assigns objective value using machine learning and massive amounts of historical league data to calculate the level of threat a player creates.

Unsurprisingly, a few of those players suit up at Camp Nou, though Europe’s most dangerous creatives so far this season operate elsewhere. Read on for the details.

La Liga – Barcelona

Barcelona’s 2017-18 playing styles through six La Liga matches measured against league averages (0%).

A matter of weeks ago, the departure of Neymar and a 5-1 aggregate loss to Real Madrid in the Spanish Super Cup sent the Catalans – or at least their supporters – into something of a panic. Take into account they didn’t land Philippe Coutinho and their summer splash Ousmane Dembele will be shelved for at least three months, and on the surface it seems it should be Barca and not Real Madrid seven points back in La Liga. Instead, the Blaugrana rolled Juventus in the Champions League and have a plus-18 La Liga goal difference.

Yes, Messi has nine league goals in six matches. Yes, he’s outscored 14 of the 20 La Liga clubs. Yes, his 22 shots on goal are nearly half of the team’s 45, and his total either matches or betters the team total of seven La Liga clubs. But Barcelona didn’t need him to net in a 3-0 win at Girona over the weekend. The success might have something to do with a relatively healthy Andrés Iniesta and impressive play out of fellow midfielder Ivan Rakitić, who leads La Liga in offensive ball movement points (oBMP) with a 1.10 rating. Messi ranks second (1.05) ahead of Real Madrid’s Toni Kroos (0.98).

Make no mistake that this isn’t the Barcelona midfield of 2009, but they are operating with impressive tempo to complement their typical sustained possession. It’s only six matches, but they’re so far playing faster this year than they did with Neymar. Their fast tempo playing style is up from 154 percent above league average last season to +212 so far this term. In typical Barca fashion, they played directly less than any La Liga club last season (-40 percent). That’s barely changed (-34). Anyone they replace Neymar with won’t match his on-pitch value, but it’s becoming clearer that he wasn’t Barcelona’s engine. We’ll see if that changes with more demanding matches on the horizon, which is the case for all five clubs discussed here.

Premier League – Manchester City

Manchester City’s 2017-18 playing styles through six Premier League matches measured against league averages (0%).

No surprise here. Pep Guardiola assures us that this hasn’t already turned into a two-team race with his Manchester rivals, but that might just be coach speak. It may have been between those two all along, as we wrote before the season started. That argument was based largely on expected goals and saves, and a lot of that is playing out as the numbers implied it might. Take note of Ederson, who’s made more saves than expected. His plus-1.5 expected save differential – calculated by subtracting expected saves from actual saves – ranks just ahead of Gianluigi Buffon.

Assess City’s playing styles through six matches, and you’ll see plenty of evidence of the kind of big-money dominance they’re going to be capable of in Guardiola’s second season. Their offensiveness has increased from 45 percent above league average to +81, while maintenance – possession in one’s own half – has slightly decreased (+42 percent to +39). Attacking possession styles of build up (+69 to +102), sustained threat (+48 to +77) and fast tempo (+64 to +172) have all jumped along with crossing (+18 to +50). It’s a dangerous team that’s made up of dangerous individuals.

The same case can be made using ball movement points, but the most noteworthy players in that category aren’t Man City newcomers. Kevin De Bruyne might not be scoring, but his oBMP (2.03) is considerably higher than any player in the top-five European leagues with teammate and fellow playmaker David Silva (1.53) ranking second. The only other duo anywhere near that level isn’t the previously mentioned Barca duo, though one of the players famously moved from Catalonia this summer.

Ligue 1 – Paris Saint-Germain

Paris Saint-Germain’s 2017-18 playing styles through seven Ligue 1 matches measured against league averages (0%).

Since he’s already come up in the sections above covering two teams he doesn’t play for, let’s get to Neymar. Heads would soon roll if any other club were listed here after the summer PSG had. Any Neymar-Edinson Cavani penalty rift – perceived or otherwise – isn’t impacting their dominance. But the oBMP duo teased above wasn’t them. It’s Thiago Motta (fifth: 1.35) and Neymar (seventh: 1.27) among the leaders of Europe’s top-five leagues. Kylian Mbappe is settling in and staying out of the goal-scoring diva antics, but let’s focus on playing style because Paris have, essentially overnight, started challenging Barcelona as Europe’s most ball-dominant attacking club.

Their build up (+86 percent in 2016-17 up to +165), sustained threat (+48 to +111) and fast tempo (+115 to +238) have all spiked this season, while their direct play (-46 to -62) has fallen off even more. The same goes for maintenance, which has fallen from +89 of the average to +59. That all points toward that front three and the accompanying midfield having the ball in attacking circumstances. They’re making dangerous decisions when they do: Motta, Neymar, Adrien Rabiot (1.15) and Marco Verratti (1.05) make up four of Ligue 1’s top six in oBMP.

This version of PSG dropped points for the first time over the weekend with a goalless draw at Montpellier. Neymar, of course, didn’t play. Assuming he’s quickly back and healthy for the months to come, it’s hard to imagine any scenario other than this dominant of a Paris side taking back the title from depleted holders Monaco.

Onto another potential changing of the guard.

Bundesliga – Borussia Dortmund

Borussia Dortmund’s 2017-18 playing styles through six Bundesliga matches measured against league averages (0%).

It’d be easy here to argue that Bayern Munich are leaving the door open for Dortmund to capture their first title since 2011-12, but that might be taking due credit away from the challengers. Of the five clubs discussed here, Roman Burki has provided the most valuable goalkeeping, ranking 12th among the top-five European league keepers in expected save differential at +2.8 between Inter Milan’s Samir Handanović (+2.8) and Manchester United’s David De Gea (+2.5).

Surprisingly, Sokratis Papastathopoulos’ 1.29 oBMP is the highest any defender in the top-five leagues and leads Bundesliga players of all positions. They’ve also had some efficient finishing. Pierre-Emerick Aubameyang’s +2.1 expected goal differential is third in the Bundesliga, while Maximilian Philipp’s +1.8 is fifth. It’s of course a slippery slope to lean on such individual efficiencies, but these cases aren’t anomalies. PEA posted a Bundesliga-best +6.1 xGD last season, while Philipp’s nine goals with Freiburg came in a relatively impressive +2.6 above his expected mark of 6.4.

In terms of playing style, Dortmund have gone away from the high press some (down from +26 percent last season to +7), while increasing significantly in build up (+58 to +125) and fast tempo (+67 to +174). Their counter attacking has also increased from +19 to +41, which puts their playing styles web almost eerily in line with this week’s Champions League opponents Real Madrid.

Give Dortmund credit for their start, but another dark-horse club has gathered even more hype across Europe.

Serie A – Napoli

Napoli’s 2017-18 playing styles through six Serie A matches measured against league averages (0%).

We’ve saved the most intriguing – and possibly most exciting relative to their league – for last. Napoli may be the outliers on this list because they haven’t won a title since a guy named Maradona was around 1989-90 and Juventus have won the Serie A every season dating to 2011-12. And unlike other clubs on this list, they haven’t compromised their style or spent untold millions to jump in front of Juve on goal difference through six matches. That comes as a relief to plenty of the football world that sees manager Maurizio Sarri as one of the key names in pushing the modern game forward. As was pointed out in a recent ESPN FC piece, Fabio Capello boasted Sarri as an innovator on the level of 1970s Ajax, 1980s AC Milan and 2000s Barcelona. Pep Guardiola had a hand in the last of those and called Napoli one of the three best clubs in Europe back in August when Man City drew them in the Champions League group stage.

Napoli are highlighted by a rather un-Italian attack that’s scored more league goals than any club in Europe’s top-five leagues, including that explosive PSG side that’s played an additional match. How, specifically, are they doing it with a player payroll that ranks fifth in Serie A? Pace and press are key parts. In fact, Napoli might play as fast as any club on the planet. They don’t sustain threat like other clubs on this list (+13 percent of Serie A average last season and +10 this season), but their fast tempo (+265 up from +231) and high press (+78 up from +44) are increasing from already head-turning numbers last season.

On an individual level, Sarri is well aware of Lorenzo Insigne’s value, and that’s supported by a 1.19 oBMP that trails only Juve’s Miralem Pjanić in Serie A. What’s scary – or enthralling – is Napoli have only recently found the No. 9 to head this monster up. Dries Mertens has six goals with a +2.3 xGD, and that’s not a matter of the Belgian finding a streak of luck to start the campaign. He scored 28 goals in 35 matches last season with Serie A’s second best xGD of +6.9. It has, with little doubt, something to do with the system.

Long Live the High Press: The Subtle Adaptation of Premier League Newcomers Huddersfield Town

By:

Assessing the Terriers’ First Three Premier League Matches in Comparison with Their 2016-17 Championship Season Using STATS Playing Styles, Ball Movement Points and High Press Regains

Before anyone gets too excited here, let’s acknowledge two things.

1. Huddersfield Town are still 35 match weeks away from being 2015-16 Leicester City, which is to say let’s not waste any more time with a serious comparison.

2. Huddersfield Town are still 35 match weeks away from being 2016-17 Hull City, who won two of their first three matches before managing 28 points in the next 35 games on their way to deserved relegation with a Premier League-worst minus-43 goal difference.

But lumping the Terriers in with Hull might be more irresponsible than any comparison with those fantastic Foxes. The unsexy truth is Huddersfield likely fall somewhere between, which hardly warrants this level of attention. What does deserve acknowledgement is the tactical competence of a relatively unknown club that spent the past 10 seasons split between second- and third-tier football.

The Terriers represent a community trapped in a triangle of Manchester, Leeds and Sheffield in northern England and play at the aptly inconspicuous grounds of 24,500-seat John Smith’s Stadium. They arrived last month seeming to have an idea of how they’d like to play in the top flight, which we’ll show below. More so, they already seem comfortable implementing it, despite it being in some ways a significant departure from a more possession-based Championship system of last season that earned them their first top-flight campaign in 45 years. In short, the influence of another German manager in the Premier League on Huddersfield boss David Wagner is already showing, which we’ll get back to.

First, the cautionary tale of Hull. Take a deep look back at their start last year, and you’ll see the Tigers got those initial six points with plenty of direct play and possession in their own half (maintenance). They didn’t counter, their high press was lowly and their possession-based styles in attack (build up, sustained threat, fast tempo) were essentially nonexistent:

Hull City’s 2016-17 playing styles through three matches measured against league averages (0%).

That held true over the course of the season with maintenance and direct play as the only two areas they operated above league averages. It’s hard to say that amounts to a system of note, and it caught up to them. Their 80 goals against were the most in the Premier League since Fulham conceded 85 in 2013-14, and it can’t even be attributed to bad luck. Their expected goals against was 83.3, so they actually conceded fewer goals than what would be expected under league average circumstances.

But the data-driven story of Huddersfield’s seven points coming out of the international break looks considerably different. On the surface, they’ve played what one could say amounts to an average or below-average schedule. That said, they’re one of two sides (Manchester United) that haven’t conceded. Of the 18 sides that haven’t conceded through three matches in the Premier League era, the average finishing position is 5.3. Only four of those sides have finished outside of the top eight. Manchester City ’07-08 and Portsmouth ’06-07 finished ninth, and Birmingham City ’03-04 finished 10th. But don’t consider Huddersfield in the clear just yet. Charlton Athletic didn’t concede through three matches in 1998-99 and were relegated after finishing 18th.

Wagner is approaching the small sample of success with appropriate caution.

“We are happy with our start (to the season) because of our results, the clean sheets and because of the performances we have put in,” Wagner, the Premier League Manger of the Month, told the club’s official website. “The players can see now that if they follow their ideas and their identity, even in the Premier League, they have a chance.”

What exactly that identity is has shifted some with the climb. Consider last year’s side, which finished fifth and won the playoff for the third promotion spot. Along with Reading and Fulham, they were one of the Championship’s dominant possession-based attacking sides, even if they worked in a high press (they did, and we’ll get to that):

Huddersfield Town’s overall 2016-17 playing styles measured against league averages (0%).

It’s known that Wagner makes fitness a priority, which seemed to translate to widespread tactical and situational success last season to compensate for a minus-2 goal differential. They scored one goal in their last five matches, yet somehow managed to win the playoff.

Now consider this season and the significant departure in possession-based attacking styles:

Huddersfield Town’s 2017-18 playing styles through three matches measured against league averages (0%).

That’s not something Wagner didn’t foresee entering the season.

“We changed sometimes in the last season the style of our game as well,” he said. “When we played Newcastle away, we played slightly different than in other games. So there will be some games in the Premier League as well where we have to maybe slightly change our style. But the idea and our identity will always be the same.”

Their high press style is 52 percent above league average through three matches, which ranks first, but they’re also effective in execution of regaining the ball while pressing. Huddersfield are tied for the league lead in high press regains with a familiar club:

2017-18 Premier League Playing Style Totals

TeamMABUSTFTDPPMDPPDCMCDCADCARHPR
Huddersfield
Town
39728821550130737740712.01316
Liverpool678479246104111826530881.81516
Bournemouth582364228138130596659466.9915
Chelsea60735428987111697153707.71315
Leicester City37317013821140505463615.61115
Tottenham
Hotspur
685663399204778010035949.91415
Everton5702311441121646047581018.31614
Manchester
City
594739462218511119322694.21213
Southampton57747730016391737437760.11313
Watford334198205561465665541029.91713
Arsenal62771035031273956426488.5911
Stoke City33929019185113646754968.11510
Swansea City556226131121135623760479.8810
Brighton and
Hove Albion
54223713869115606356449.699
Newcastle
United
30115916523127666243903.5148
West Bromwich
Albion
24014712433163586759725.9138
Crystal Palace375278166110144706736498.195
Manchester
United
58769741924193748936939.0165
West Ham
United
51320611271134464752412.165
Burnley42119014857151806657645.0104
MA=Maintenance Involvements; BU=Build Up Involvements; ST=Sustained Threat Involvements; FT=Fast Tempo Possessions; DPM=Direct Play Passes Made; DPD=Direct Play Passes Defended; CM=Crosses Made; CD=Crosses Defended; CAD=Counter Attack Distance (Metres); CAR=Counter Attack Regains; HRP=High Press Regains.

Liverpool, managed by Wagner’s mentor Jürgen Klopp and his well-known gegenpressing striving for immediate ball recovery, led that category last season with 199, or 5.2 per match. It’s interesting to at least note here that Hull’s 112 were next to last, and Leicester’s 161 in 2015-16 were five behind leading Manchester United. That might be oversimplifying things, but the only teams to finish in the top six the past two seasons without being above league average percentages in maintenance, build up, sustained threat and fast tempo are Leicester, Tottenham Hotspur and Southampton in 2015-16. Those clubs all operated above the league average high press percentage and ranked in the top seven in regains.

Huddersfield through three matches are just ahead of Liverpool’s per-match pace from last season, but that actually shouldn’t come as too much of a surprise. They led the Championship last season with 217 over 46 matches (4.7 per match).

So while there was a shift in style from the Championship, that’s not to say they didn’t employ a high press last year. They’re not attacking with the ball this season, but they are similarly disrupting. They’ve just had to go about it differently and will have to continue to do so because they’re realistically not going to be a possession-based attacking team anytime soon at this level. That probably means their forwards are doing a bit more work this campaign, but they might have brought in the right attacker to head that up.

They’ve succeeded thus far with much of the same team under the same manager. Ten of the 15 players Wagner has used in their three matches are holdovers from the Championship side, but some key players certainly joined the club over the summer.

There’s Steve Mounie for the obvious reason of scoring twice in his first three games with the club, but he’s also contributed four high press regains. That ranks tied for third in a small sample size, but Mounie was one of 12 players in Ligue 1 last season with double-digit goals and high press regains:

2016-17 Ligue 1

PlayerTeamGoalsHigh Press Regains
Edinson CavaniParis SG3510
Alexandre LacazetteLyon2810
Florian ThauvinMarseille1513
Steve MounieMontpellier1413
Nicolas de PrévilleLille1416
Benjamin MoukandjoLorient1310
Emiliano SalaNantes1215
Lucas MouraParis SG1214
Jimmy BriandGuingamp1218
Loïs DionyDijon1111
Ryad BoudebouzMontpellier1116
Valère GermainMonaco1010
Players with at least 10 goals and 10 high press regains.

He did it for a club that seemed to succeed with a high press. Montpellier operated just three percent above the league average but ranked fifth in regains.

Looking further into individual performance and beyond playing style to quantify Huddersfield’s success, there’s Jonas Lössl with his three clean sheets in his first three with the Terriers, which have come with the keeper outperforming the league average of expected saves. Subtract the former Mainz 05 man’s expected saves from his saves, and his differential comes out at plus-2.3, meaning he’s saved them at least two goals the average keeper wouldn’t. As STATS has written before, goalkeeping was a huge part of Leicester’s dream run.

But there are also holdovers from last year who are properly adapting. STATS has brought ball movement points into analysis, which is broken down into categories of offensive and defensive as well as positive and negative (oBMP+, oBMP-, dBMP+ dBMP-). These metrics use machine learning to assign an objective value to every involvement a player has in a possession to credit or discredit decisions with the ball, measuring how dangerous a player is with ball circulation by relating it to the probability of a shot happening later in that play. Passing points generate expected shot points, so if a player generates one BMP, he’s generated passes to lead to one shot.

Aaron Mooy’s 0.67 oBMP+ ranks 15th in the Premier League ahead of players like Dele Alli, Wayne Rooney and Alex Oxlade-Chamberlain, despite the midfielder having to exist in a very different system. That doesn’t tell the whole story because he, like the rest of the team, is having to defend more than last season. Yet his 0.30 defensive contribution, calculated by adding dBMP+ and expected goals defended, is also interesting to consider. It ranks between noteworthy central presences such as N’Golo Kante (0.35), who we all remember from Leicester, and Nemanja Matić (0.29).

They also only have one player in the top 50 of dBMP-, which is impressive for a club that’s spending a decent amount of time in their own half. Again, these are small sample sizes, but it’s a start.

That’s all those seven points are thus far for the Terriers – a start. So are Huddersfield ’17-18 going to be Leicester ’15-16, Hull ’16-17 or Klopp’s Liverpool?

The answer is probably none of the above. They’ve simply evolved from their former selves, but that might have staying power.

Gylfi Sigurdsson, the Costly Set-Piece Specialist? Maybe Not for Everton

By:

Using STATS’ Advanced Metrics to Show the Move Presents the Midfielder with a Chance to Prove He Can Be a Playmaker at a Club with a More Possession-Based Style

It’s true that two of Gylfi Sigurdsson’s nine goals last season came on free kicks and three came from the penalty spot. It’s true that of his 13 assists, eight came on set pieces. It’s true that Fernando Llorente is 6-foot-4 and Wayne Rooney is 5-foot-9. It’s also true that Sigurdsson is probably too far into his prime to ever be Mesut Özil.

You’ve read all of this elsewhere in the days after his move to Everton.

But calling Everton’s £45 million signing of Sigurdsson money spent on a set-piece specialist is a bit shortsighted. Dead-ball goals, assists and teammate physique aren’t necessarily the most well-rounded ways to measure whether Sigurdsson will fit in the run of play at Goodison Park. The rest of Swansea’s starting XI didn’t move with him, so it’s necessary to also consider club playing styles he’s more likely to be a part of under Ronald Koeman rather than the Swansea managerial mashup he’d experienced since rejoining the club in 2014-15. It’s also necessary to consider his own distribution aptitude.

First, the highlights. They’re there to suggest Sigurdsson in his mid-20s became the player he rarely was with Tottenham, and they go beyond his Europa League dreamer from 50 yards on Aug. 24 against Hajduk Split. If you haven’t already, do yourself a favour and watch it:

Against Manchester United on April 30 as Swansea scrapped for a much-needed away point to fight relegation, Sigurdsson sets up for a 24-yard free kick in the 79th minute while trailing 1-0, sees a defender leave David De Gea’s line, and immediately placed the ball in the back of the net in the precise position of the departed defender. The Swans got that point because of it.

Against Sunderland on May 13 as Swansea tried to secure safety, there’s Sigurdsson’s long ninth-minute free kick dropped into the box just out of reach of the charging keeper on the head of Llorente to give the Swans an initial lead in a 2-0 final.

From the run of play, there’s his two-touch heal pass to Martin Olsson in the 69th minute to set up an equaliser against Burnley on March 4.

Those are contributions that are obviously identifiable as valuable to anyone with football sense, and it’s now quantifiable. STATS measures players’ team point contribution, which factors in objective value to such plays, much like expected goals. Sigurdsson contributed 5.4 points last season, which was right at his expected points (xP) of 5.5.

What might be more telling here is while Sigurdsson’s point contribution ranked 31st in the Premier League, it accounted for 13.2 percent of his team’s points in matches he played. That wasn’t far behind some pretty impressive names among players who appeared in at least half of a single club’s fixtures, and it was ahead of notable others. It’s also interesting to consider how many of Everton’s transfer window signings were similarly impactful for their clubs last season.

So it’s very possible Swansea would not be a Premier League club this season had he not been there.

But the key counterpoint to his performance in the run of play might be that ball movement metrics don’t speak as well for Sigurdsson as certain elite attacking midfielders. STATS’ data-science team is able to leave behind often-misleading binary metrics of passes completed and assign objective value to distribution based on pass risk and reward.

Similarly, they’ve brought ball movement points into analysis, which is broken down into categories of offensive and defensive as well as positive and negative (oBMP+, oBMP-, dBMP+ dBMP-). These metrics use machine learning to assign an objective value to every involvement a player has in a possession to credit or discredit decisions with the ball, measuring how dangerous a player is with ball circulation by relating it to the probability of a shot happening later in that play. Passing points generate expected shot points, so if a player generates one BMP, he’s generated passes to lead to one shot.

Looking only at his positive offensive involvements, Sigurdsson is impressive, ranking eighth last season among a truly elite bunch of Premier League creatives.

But bring in his negative involvements, and his 3.4 net oBMP levels out considerably to 33rd. Comparatively, he’s nowhere near the top four of Özil (9.0), Kevin De Bruyne (7.6), David Silva (7.6) and Eden Hazard (7.2).

The leaders in those categories, however, at least have an opportunity to play for clubs that attack. Not only do they attack, they attack with possession. This is where things get interesting in evaluating the opportunity before Sigurdsson under Koeman. On one hand, the attacking systems those elite players are a part of makes it impressive that they’re able to limit their oBMP-, meaning they make relatively few adverse decisions with the ball while attacking. On the other, there’s the argument that Sigurdsson’s Swansea surroundings didn’t give him the opportunity to do so.

Everton are not Manchester City or Arsenal, but they also aren’t a club struggling to avoid the drop each April and May. Swansea have been in that defensive role in the table, and it’s reflected in their playing style.

According to STATS Playing Styles, which measure a club’s time spent in specific styles compared to league averages, Swansea operated far less frequently in possession-based styles such as build up (minus-21 percent), fast tempo (-15) and sustained threat (-10).

Everton, meanwhile, were at least slightly above the league averages in all three categories while also counter attacking more frequently.

That might not be a significant enough change to turn Sigurdsson into Everton’s version of Özil. But before concluding Sigurdsson is little more than a dead-ball wiz, let’s first give him a chance to be the playmaking midfielder at a club that operates at or above possession-based norms in the Premier League.

That opportunity could come with the new-look Toffees. It’s less of a long shot than a 50-yard dreamer.