Month: May 2018

The European season is over. The World Cup is still two weeks away. We have entered the Great Content Gap. But never fear. I shall step in to the breach and fill the yawning hole in your football media consumption diet with my first ever StatsBomb Mailbag. Let’s kick things off with a question from the boss. Where do you want the content side of StatsBomb to be one year from now? --Ted Knutson Because Ted’s the boss his question goes first, but because I run the website I get to refuse to answer his other question about what my most painful moment as an Everton supporter was. What’s interesting about trying to answer this, other than it being essentially an interview question that I’m now answering in public, is that StatsBomb has a lot of freedom in terms of direction. As I’ve written about before, the website exists because the greater StatsBomb data collection and consulting company exists. We are lucky to get to build on the data and not be beholden to selling ads to justify our content. So, if the purpose isn’t to chase clicks (and it isn’t) then how do you measure growth, and channel ambition. Well, one axis is simply the quality of the work. A year from now I want Statsbomb to be a place where readers know they’ll be getting great content, and writers know they have a home for smart ideas. Just because we aren’t selling ads doesn’t mean we’re indifferent to getting eyeballs. We want people to read StatsBomb, it’s just we want them to do it because they believe in our product and our content, not because we have a business model that depends on it. That doesn’t mean that great content can’t also make money. It wouldn’t make sense for StatsBomb to become a subscription site (since the entire point is to be a free place where people can read good work which is supported by the resources that the services StatsBomb actually do sell) but that doesn’t mean there aren’t other ways to grow a business. One particular idea that has always intrigued me is trying to make and sell a Baseball Prospectus style season preview, perhaps with periodic email newsletter updates over the course of the season for those who purchased it. It’s a good engine for developing a discrete marketable content product but also not cannibalizing the free content we’re committed to giving y’all. A year from now I hope that the content on the website continues to grow, evolve and improve (and maybe increases in volume) and that on top of that we are launching different types of content products, some of which are even intended to be profitable. Oh, and I also want a copy editor.     https://twitter.com/joe_fishfish/status/1001901722183327744 This was such a good question that Ted decided to answer it on Twitter before I could even write the mailbag. https://twitter.com/mixedknuts/status/1001908098297147392 The question and answer here get at something I think is often misunderstood about analytics and the nerds that do it. Frequently analysts will be skeptical of causal claims surrounding soft factors. Things like confidence, or succumbing to pressure in big moments, or any of the other clichés that swirl around the idea of mentality. But, crucially, that doesn’t mean that they don’t exist. Human beings are human. Sometimes they feel good and sometimes they feel bad. Sometimes they perform well. Sometimes they perform poorly. Sometimes my articles contain lots of typos and sometimes I write clean copy. For data driven people the skepticism isn’t that the underlying thing can't be true, it’s the assumption of causality. Analytics is a field that’s deeply aware of the idea of uncertainty and randomness. It’s not that a player can’t be clutch, it’s that a player who isn’t clutch in a big moment would look more or less the same to an observer as one who had a four-month-old baby that kept him up half the night, or one who was fighting off some bad pad thai or one who just plain old got unlucky. Claiming to know that it was specifically a weak mentality (or whatever you want to call it), as opposed to a weaker claim of simply acknowledging that as one of a number of possibilities is where the difficulties lie. Analysts are rightly quite reticent to make those links strongly given the relatively little data we have. And, quite frankly, historically we aren’t great at correctly predicting who is bad in a big moment and who isn’t. Players choke until they don’t. Arjen Robben was a failure in big moments until he scuffed a shot that found its way into the back of the net in a Champions League final. Then, he isn’t. It’s not that soft factors don’t exist, it’s that all too commonly the things that end up defining a player’s mentality are the things they don’t have control over as opposed to the things they do. It’s an important distinction. It defines how analysts look at a problem. Ted’s response is typical of good work. Be skeptical, but look for data. Try to ground the problem as much as possible. Accept both the reality that humans are humans, and that we are terrible at understanding the transmission mechanism from the human brain to performance on the field. Then, work to bridge that gap.   Ok, now it’s time to fill this particular precondition for doing a mailbag. https://twitter.com/Classlicity/status/1001898252084527106 Listen, the obvious answer to this question is Jose Mourinho. The man is a born troll. He’s a master at deploying the same arguments against adversaries that he ridicules as absurd when deployed against him. He says things that are technically correct while being wildly inaccurate if contextualized. He is brilliant at taking any argument and twisting it around until its fought on whatever grounds he wants to fight it on. When he’s feeling threatened he bursts into the football world’s mentions and refuses to leave. Of course he’d have burner accounts. Mourinho would have a gazillion sock puppets. He’s already the king of turning any fan base that rallies around him into an army of trolls repeating at ever ascending volume whatever dubious claim he’s making about why this week’s disappointing result wasn’t his fault. Jose Mourinho is the king of the trolls, burner accounts are par for that course.   https://twitter.com/kevinmccauley/status/1001899773362888704 Injuries and analytics are a combustible brew. There is definitely both a desire for, and a realistic possibility of, advancements through data which can help both spot injuries and prevent them. There’s also a lot of quackery. And, because we’re dealing with science and medicine and data, it’s even harder of non-experts to determine the difference between the two. On a general top down lever there are things teams can do to mitigate injury risks, and lots of the sports science industry is dedicated to developing injury prevention methods. Whether that’s monitoring work load (and having Arsene Wenger ignore that monitoring and play players who are in the “red zone” anyway) or identifying when movement mysteriously drops off, or advising on sleep cycles and eating habits, teams across sports certainly try and ring every last drop of an edge that they can get out of the medical side of data. How much any of this works is an open question though. And in a specific case, like Nabil Fekir’s surgically repaired knee, it’s even more questionable. There’s no way that from the outside anybody is going to know more about Fekir’s health than a good sports medicine team working with him personally. The strongest claim that you might be able to make is that data could eventually be used to raise the kind of red flags that might make you leery enough to want to get a second, third or fourth opinion.   https://twitter.com/GregorydSam/status/1001908460752162816 I founded my own short-lived blog in April of 2013, and started writing for Grantland in June of that year. I think what’s surprised me the most has been the distribution of work done in public vs private since then. In American sports a lot of work was done publicly before teams began to take it seriously. Baseball work was around for a long time, and it took a more widespread adoption by the public before teams began committing to using that work to gain an edge for themselves. In football teams, or companies that make their money consulting with teams, stepped into the analytics world much more quickly. The pipeline from talented blogger to super-secret team consultant happened faster than it did in American sports. It actually happened faster than public awareness of analytics happened. Despite all the usual silly battles over real football men, and air conditioned offices, owners have been fairly quick to hire analytics people. Now, that’s different than saying that whoever they hire as any real influence, or that team transfer policy, or management choices have reflected analytics best practices. Managing teams are tricky, effecting change is difficult, and doing analytics well is hard. Put all that together and you’ve got an environment where hiring people is the beginning of the process not the end of it. But, in general I’ve been surprised that the speed of teams doing at least some analytics works seems to have outstripped the speed with which work is viewed as mainstream by fans.   Have I been a big enough influence on your writing that you’re going to stretch this mailbag content into a two day thing? --Fake Bill Simmons You better believe it.

  Images provided by the Press Association

At some point in the early 2000s, the football fan became a pundit. Can you really blame them? Technological advances meant that the most casual hobbyist could put up a blog that looked roughly as reputable as a specialist site and skim information from around the world. Distribution channels — social media, podcasts, and Fan TV-type video — multiplied. It’s a fun hustle for those who can make it work. This development hardly troubled traditional football media, which has always internalized the bare minimum from its challengers whilst fending them off. It did, however, change the language of football fandom. Your average fan now speaks in the sweeping, definitive language of a talking head, interpreting everything — who did what now; who should be signed; who should be fired, because someone should always be fired — through the prism of whether it helps a team. This leaves scant room for idle curiosity, amusement, or one’s own esoteric feelings. It is a view from nowhere that mainly sees transfer rumours and crises. The main difference between radio show callers and talking heads employed by those same stations is microphone quality. The football world’s dirty secret is that most everyone hates Paul Merson while also wanting to be him. All of this has been a boon to Merson et al. — actual pundits with institutional support. Centrist punditry establishes its authority in opposition to perspectives that can be characterized as radically unreasonable. Data analysts, then, are written off as nerds who cannot see the world beyond their spreadsheets. The idea of football as an artistic entertainment — the movement of bodies in space, like ballet, but with a ball — is derided as the fantasy of humanities graduates who never played the sport. These uncharitable conceptions present different schools of thought as antagonists, leaving them even more siloed, their only commonality being the way their niche-ness is used as proof of their invalidity The only winner is the pundit’s faux-everyman routine. This starting point hinders talk of conciliation between the statistical and aesthetic realms,. limiting it to a focus on lowering expectations.  It forces a counterargument that claims art is not the enemy of science, and vice versa, and stops there. That construction betrays a fundamental misunderstanding of soccer aesthetics and a curious lack of faith in how analytics can be used. In actuality, data and analytics can help us better discuss and appreciate soccer’s aesthetic dimension. /// Artistic criticism is overwhelmingly the work of considering what a piece evokes. Much of that is introspection, the hard work of thinking about what something made you feel. In turning that feeling into an argument, one then considers the mechanics of the work in question: what did it do to produce this outcome? This is where the language of artistic criticism varies from field to field. One might consider brushstrokes in painting or word choice in poetry or movement in ballet.  Using the language of the medium, a reviewer asks whether the work at hand evokes what it set out to evoke, and whether that’s of any value. Or, artlessly, was it good? This analytical work must build upon a shared understanding between writer and reader of what is actually being criticized. In many fields, this is the maligned but useful work of summarizing. It  helps the reader of criticism to know that Martin Amis’s The Pregnant Widow is actually about a bunch of childless twenty-somethings. Criticism can be 99 percent figurative but not 100. In some fields, though, such summations are not useful. The plot of Cinderella tells you next to nothing about Prokofiev’s ballet. Stop there and you’re left with exuberant metaphors about grace, elegance yet absolutely no sense of the actual show. Soccer has a similar problem. A scoreline, like the fact that (spoiler alert) Cinderella finishes the ballet with the prince, only tells you how the piece under consideration ends — not what previously happened. Plot summarization is an insufficient starting point in certain forms of artistic criticism, orthogonal to the endeavor being undertaken; you need a language of movement. Enter statistics. Football statistics are not marketed as a language of movement, but that is one of their functions. Before being spun into models, they are inventories of actions performed and positions occupied. The heat maps proliferating in your Twitter timeline and on this website are accounts of spatial relations. Most other systems for thinking about how players occupy space — tactical maps with dots and arrows, lineup graphics, positional names and numbers — are approximate representations of how players should move as opposed to those actually performed in matches. Quants and aesthetes who consider football, like ballet, the study of how bodies move in space, can find common cause in the need for more precise spatial information. Expressionist football writing has regularly built on the foundations laid by prevailing forms tactical analysis. For a time in the early aughts, “false nine” was both a meaningful positional description and metaphysical proposition. This term, which was rooted in a specific observation about movement, birthed thousands of florid meditations about our place in the cosmos. That it can no longer serves either purpose speaks to the ways tactical and aesthetic discourses in football are intertwined. Other terms — the dreaded, ineffable likes of “gegenpressing,” “half spaces” and “between the lines” — followed a similar trajectory. Our appreciation of the subtler pleasures of midfielders who don’t rack up the goals owes a great deal to these concepts. It is the fate of all tactical concepts to eventually become platitudinous. For our aesthetic understanding of football to move forward, descriptions of what is happening on the pitch must also be progressing. The statistical view of football imbues these measures with little more than instrumental value. There are no extra points for flair. Sites like this one differentiate between all sorts of passes, touches, and shots, but those differences only matter insofar as they tell us whether an action helped bring about a goal. In this worldview, creativity and assists are largely synonymous; the former is not an end unto itself. . That is where the aesthetic point of view branches off from statistical analysis. Inefficiency is often more compelling to the aesthete. A touch with the outside of the boot may be profoundly unhelpful to the team, yet oddly compelling. (See also: nutmegs.) Waxing lyrical about strange-but-evocative movements still requires a clear sense of what is happening on the pitch, otherwise it risks becoming meaningless or ascribing qualities to players who do not possess them. The latter problem often manifests as physical, national, and perniciously racial determinism: all small midfielders are twinkle-toed magicians unless they’re English (all-action) or black (powerful); every tall forward is an oafish header merchant. The artistic analysis of football, however, should seek to choose adjectives and metaphors based on what is actually happening on the pitch. Data remains fallible and artistic criticism remains subjective — Charlie Adam's potshots will evoke different sentiments in different observers — but the statistical language of movement puts the undertaking on sounder footing. /// The economics of football media have allowed the schism between the artistic and statistical visions of the sport to persist. In order to survive, all but the most traditional of approaches to the beautiful game tend to be reduced to their most easily monetized forms. Stats, then, are the purview of scouting types, gamblers, and/or almighty nerds. (Hello, dear StatsBomb reader.) The vision of football as a balletic undertaking is monetized in the form of over-sized prints and print products that a certain type of middle-class fan can keep on display. (Disclosure: I’m an editor of Howler Magazine and would love to sell you a luxurious print product.) The difference between these camps is not actually substantive so much as they just don’t communicate. Add in a decade of parochialism and you have our current mess. We have all become quite good at articulating where the statistical and artistic projects diverge and failed to develop language around their considerable common ground. This has only benefited the traditionalist pundits who always sought to cast these camps as dueling irrational extremists. The solution to these problems needn’t be particularly radical. Data analysts do not have to make sense of Wayne McGregor’s Genus and relate it to matches. Analysts interested in artistic performance don’t have to all go out and take remedial R statistical programming lessons. (Both of those options are fun, mind you.) A marked improvement in how we discuss football could be achieved by building a shared understanding of what both projects seek to achieve and how they overlap. We all want to know what actually happens on the pitch, we just use that information towards different ends.

The good folks here at StatsBomb are very excited about the Champions League final. We’re so excited that we wanted to make sure we got some of our cool new toys out in public view before the last match of the season kicks off. This involved making a compromise or two. The data isn’t quite done yet. Such is life in big data city. Who knew launching a technology company could be so hard? But, rather than deprive the people of what they want (and we are quite confident that what the people want is pictures with bright colors and lots of squiggly lines), we just decided to make do with some slightly incomplete data. Here, without further ado, are a bunch of cool pictures that help explain why Liverpool are so dang good. They just happened to be drawn from a slightly incomplete picture of 32 matches of Premier League data rather than waiting for our diligent data mice to finishing sewing the last six games in.  

Liverpool’s Defensive Pressure.

We’ve talked about a lot about how we record pressure in matches. Liverpool are a perfect example of how doing that helps paint a fuller data picture of what a team is actually doing on the pitch. Here’s what Liverpool look like as a defensive team without pressures included. It paints a pretty good picture. The degree to which Liverpool bother opponents deep in their own territory is apparent. Looking at that map it doesn’t get anything noticeably wrong. Now, let’s look at a map with pressures included. Yowser. This really makes clear how aggressively Liverpool are taking the game to opponents. Those on ball actions don’t come from nowhere. Liverpool contest everything across the field, and it’s those actions that force opponents back and into the kinds of mistakes that ultimately lead to turnovers and easy opportunities for Jurgen Klopp’s band of merry pressers to take advantage of. Now, let’s look at the same dynamic on an individual player level. Here’s Roberto Firmino’s defensive actions without pressures included. Firmino has rightfully garnered a reputation as a player willing to do tremendous amounts of defensive work for Klopp’s side, exactly the kind of high motor ball harassing forward a good pressing team needs. This particular chunk of data doesn’t disprove that, but it doesn’t exactly prove it either. It shows a player who is quite active in a fairly limited forward zone. Clearly he’s committed defensively, but the breadth of his commitment is still mostly something that data didn’t capture. And now to add in the pressure. Hey there Bobby! That’s a more accurate picture of the player we keep seeing on the pitch. Add in pressure events and Firmino’s defensive range and determination becomes much more apparent. All this pressing information is contained on Liverpool’s defensive radar too. The radar in particular shows Liverpool with a very high defensive pressure rating while only having a mediocre pass per defensive action score. That, in part, is because of the presence of all those pressures. Liverpool have a cohesive pressing unit that makes life difficult for opponents even when they aren’t tackling, intercepting or otherwise actively taking the ball away. The one possible fly in the ointment here is that Liverpool’s xG per shot conceded is below average. It’s tempting to look at that is a result of the tradeoffs a pressing team makes, a concession that when the press is broken opponents will get very good shots. It turns out that’s not the case for Liverpool, at least not this season. The result of all of Liverpool’s effective pressure is that the team simply doesn’t give up a lot of shots of any variety. They restrict opponents bad shots, and they also don’t concede very many golden chances either. Opponents simply have a very hard time creating anything against this Liverpool defense. Teams' average xG per shot isn't higher against Liverpool because they creati more good chances. They don't. Rather, it's because in addition to preventing good chances Liverpool also prevent a high number of bad ones.   Salah Days It will not surprise anybody with a pulse that Mohamed Salah is the central hub of Liverpool’s attack. The attack does almost everything well. They keep possession and pass the ball from back to front and take an avalanche of chances. The only place where they might get a slight demerit is the average shot quality of their chances, but that's largely a function of how often defenses sit deep and make themselves difficult to break down. Sometimes there’s not much to do but hammer away with mediocre chances until one finally goes in. The majority of the goal scoring thrust obviously comes from the front three of Salah, Firmino and Sadio Mane. And, of course, Salah in particular has had an absolutely unstoppable season. His scoring numbers crashing into the box from his right side wide forward position are enormous. It’s not just that he’s cutting off the right side and onto his stronger left foot to terrorize the defense, it’s that when he does that he consistently gets deep into the heart of the backline trying to stop him. Ineffective wide forwards end up cutting in and shooting from the edge of the 18. Salah, on the other hand, lives six to twelve yards out. Oh, in case you’re wondering why that one Salah goal he scored from the parking lot seems like a good chance. It’s because it actually looked like this. Despite Salah’s magnificent scoring season, Liverpool’s attack remains highly balanced. Looking at the distribution of the team's xGChain per 90 (among players with more than 900 minutes played because I'm mean and wanted to deny the departed Coutinho any glory) paints a picture of a team operating with Salah as the first among equals rather than a lone superstar. He might get to finish a disproportionate number of the moves, but Salah is standing on the shoulders of all of his teammates working to find good opportunities. So, that’s Liverpool. They’re a hard pressing, high scoring, beast of a unit. Now, we have more ways than ever to show how they do it.   Images provided by the Press Association

Buying players based on summer tournament performances is common practice. It's also a bad idea. Eye-catching performances under extremely specific conditions, over the course of only five or six matches, will often turn into overblown transfer fees and contracts that seem to last forever. That’s why it’s important to highlight lesser known players before the World Cup and take a deeper look who actually had impressive seasons and might end up breaking out in Russia for the world to see. Andrija Zivkovic is exactly that type of player.

Zivkovic plays for Serbia, a team with a solid talent pool. Thanks to a soft path to qualification, they competed with Ireland, Wales and Austria for the top two spots in their group, Serbia managed to earn a decent World Cup draw. While Brazil is likely to win Group E, second place is wide open. Serbia will only have to get by Switzerland and Costa Rica to advance.

Initially it seemed like Zivkovic might play a bit part at best. Serbia’s squad is mostly composed of players in or past the prime years of their careers. The eleven most used players by now former manager Slavoljub Muslin during qualification, where they topped their group with 21points while only losing once, average the ripe ol’ age of 29. This will, therefore, be the last tournament for a sizable portion of this group.

The appointment of a new manager, Mladen Krstajić, after the end of the qualifiers led to the recent integration of a few more interesting, younger, talents into the side. That includes not only Zivkovic, but the incredibly hyped Sergej Milinkovic-Savic as well, who, despite having over 7000 Serie A minutes under his belt, only debuted for Serbia in a friendly in November.

Before this year Zivkovic might have been best known as a Football Manager legend. He moved from Partizan to Benfica on a free transfer in the summer of 2016. While his talent was clear every time he stepped on the pitch, he was never given a chance at getting consistent playing time until early 2018. An injury to Filip Krovinovic, who otherwise could’ve been a World Cup revelation himself, left the team with a hole in midfield and no option to fill it other than, as it turns out, their most talented player. Zivkovic, a left-footed 21 year old who spent most of his career playing from wide, starred throughout the second half of the campaign as the left-sided center-midfielder in a 4-3-3.

His role was all about freedom: to roam into either of the wide areas with or without the ball, to arrive in the box in support of the lone striker or to aid in build-up. Zivkovic became the team’s creative dynamo and the numbers reflect that in a way that his 3 goals and 4 assists don’t fully capture.

Zivkovic’s 2.2 key passes per 90 minutes ranks him third best in the league for “central” players with over 1000 minutes, and the fact that 1.7 of those key passes came from open play situations makes that all the more valuable. His assists were all in repeatable contexts and the sheer number of situations he created for his teammates are proof of that.

He’s the kind of player to show up well in all kinds of passing models, completing a rather large and diverse number of difficult passes. His freedom to roam means he’s often compelled to drop and aid in build-up from deeper, where he pulls off longer side-switching passes with ease.  He also moves wide where he executed 1.6 crosses per 90 – top 10 in the league – with a solid 26% accuracy.

Zivkovic is technically sublime. He passes the eye test with flying colors. His ability to use different parts of his foot: from the outside to the side and across the laces helps his passing range be as diverse as possible and gives him a ton of solutions even when space is limited. Often, he’ll delay a pass, making his own task more difficult, just so he can assure his teammate’s run will align with the timing of the ball.

Give him the ball in the final third and he’ll find a way to get his team up close and personal with the opposing goalkeeper. And, as much as the definition of “through-ball” leads to a snow-ball of questions for stats providers across this sport, the lists of players who play the most of them always seem to pick out guys on the upper echelon of technique. Zivkovic leads the league with 0.3 per 90.

His capacity to dribble is of major importance too. Beyond just beating players one against one and being able to unravel himself from tight situations –which he does, with two completed take-ons per 90 and around a 65% success rate, there's a lot of value in being able to carry the ball from a deep position all the way to the final third. That’s true even if the situation doesn't involve overcoming a player with any sort of YouTube-montage-worthy skill-move. He's sublime at it, capable of bursts of acceleration to create separation from his marker and put his ball-carrying abilities to use.

Zivkovic is by no means perfect.

He can still improve when it comes to his box arrivals and has some awareness issues when having his back towards goal – too often choosing against turning into space simply because he didn’t realize said space was there. But, his positional change has helped accentuate his strengths and hide his weaknesses.

Serbia used a 3-4-3 during the entirety of the qualification process, but over the last friendly breaks the new manager has taken out one of the center-backs to add a third midfielder in a 4-2-3-1 or 4-3-3 set-up depending on personnel. The line-up could well end-up accommodating Milinkovic-Savic ahead of Serbian mainstay Nemanja Matic and Crystal Palace’s Luka Milivojevic. In this set-up Zivkovic is a front-runner to wreak havoc off the bench – also because he can fill in any creative role, either from wide or central – and to potentially unlock defenses such as Costa Rica’s.

Zivkovic is under contract with Benfica until 2021, but the constant need to sell by the Portuguese side combined with the fact that the Serb is one of the few valuable assets that could bring in revenue during this window, will make this a very interesting case to follow. Serbia have an intriguing national team and it will be interesting to see how their younger talent will stand-out among a solid foundation of personnel. Brazil will still likely tear through their group, but the Eastern European side has enough about them to overcome Costa Rica and Switzerland and finish in second place if they play their cards right. And if Zivkovic impresses during the process and earns himself a big money move, it will be justified not simply because of his performances this summer, but based on everything he accomplished before it.

The concept of pressing has existed in football for decades but its profile has been increasingly raised over recent years due to its successful application by numerous teams. Jürgen Klopp and Pep Guardiola in particular have received acclaim across their careers, with pressing seen as a vital component of their success. There are numerous other recent examples, such as the rise of Atlético Madrid, Tottenham Hotspur and Napoli under Diego Simeone, Mauricio Pochettino and Maurizio Sarri respectively.

Alongside this rise, public analytics has sought to quantify pressing through various metrics. Perhaps the most notable and widely-used example was ‘passes per defensive action’ or PPDA, which was established by Colin Trainor and first came to prominence on this very website. Anecdotally, PPDA found its way inside clubs and serves as an example of public analytics penetrating the private confines of football. Various metrics have also examined pressing through the prism of ‘possessions’, which Michael Caley has put to effective use on numerous occasions. Over the past year, I sought to illustrate pressing by quantifying a team’s ability to disrupt pass completion. While this was built on some relatively complex numerical modelling, it did provide what I thought was a nice visual representation of the effectiveness of a team’s pressing.

While the above metrics and others have their merits, they tend to ignore that pressing can take several forms and are biased towards the outcome, rather than the actual process. The one public example that side-steps many of these problems is the incredible work by the Anfield Index team through their manual collection of Liverpool’s pressing over the past few seasons but this has understandably been limited to one team.

Step-forward the new pressure event data supplied by StatsBomb Services. This new data is an event that is triggered when a player is within a five-yard radius of an opponent in possession. The radius varies as errors by the opponent would prove more costly, with a maximum range of ten-yards that is usually associated with goalkeepers under pressure. As well as logging the players involved in the pressure event and its location, the duration of the event is also collected.

The data provides an opportunity to explore pressing in greater detail than ever before. Different teams use different triggers to instigate their press, which can now be isolated and quantified. Efficiency and success can be separated from the pressing process in a number of ways at both the team and player-level. Such tools can be used in team-evaluation, opposition scouting and player recruitment.

One such application of the new data is to explore gegenpressing or counter-pressing, which is the process where a team presses the opposition immediately after losing possession. The initial aim of counter-pressing is to disrupt the opponent’s counter-attack, which can be a significant danger during the transition phase from attack-to-defence when a team is more defensively-unstable. Ideally possession is quickly won back from the opponent, with some teams seeking to exploit such situations to attack quickly upon regaining possession. Five seconds is often used as a cut-off for the period where pressure on the opposition is most intensely applied during the counter-press.

The exciting new dimension provided by StatsBomb’s new pressure data is that the definition of counter-pressing you would find in a coaching manual can be directly drawn from the data i.e. a team applies pressure to their opponent following a change in possession. The frequency at which counter-pressing occurs can be quantified and then we can develop various metrics to examine the success or failure of this process. Furthermore, we can analyse counter-pressing at the player-level, which has been out-of-reach previously.

The figure below illustrates where on the pitch counter-pressing occurs based on data from 177 matches from the Premier League this past season. The pitch is split into six horizontal zones and is orientated so that the team out-of-possession is playing from left-to-right. The colouring on the pitch shows the proportion of open-play possessions starting in each zone where pressure is applied within five seconds of a new possession.

The figure illustrates that pressure is most commonly applied on possessions starting in the midfield zones, with marginally more pressure in the opposition half. Possessions beginning in the highest zone up the pitch come under less pressure, which is likely driven by the lower density of players in this zone on average. Very few possessions actually begin in the deepest zone and a smaller proportion of them come under pressure quickly than those in midfield.

From a tactical perspective, pressing is generally reserved for areas outside of a team’s own defensive third. The exact boundary will vary but for the following analysis, I have only considered possessions starting higher up the pitch, as denoted by the counter-pressing line in the previous figure.

In the figures below, the proportion of possessions in the counter-pressing zones where pressure is applied within five seconds is referred to as the ‘counter-pressing fraction’. In the sample of matches from the Premier League this season, a little under half (0.47) of open-play possessions come under pressure from their opponent within five seconds. At the top of the counter-pressing rankings, we see Manchester City, Tottenham Hotspur and Liverpool, which is unsurprising given the reputations of their managers. At the bottom end of the scale, we find a collection of teams that have mostly been overseen by British managers who are more-known for a deep-defensive line.

On the right-hand figure above, the strong association between counter-pressing and possession is illustrated, with the two showing a high correlation coefficient of 0.86 in this aggregated sample. Interpreting causality here is somewhat problematic given the likely circular relationship between the two parameters; teams that dominate possession may have more energy to press intensively, leading to a greater counter-pressing fraction, which would lead to them winning possession back more quickly, which will potentially increase their possession share and so on. The correlation is weaker for individual matches (0.36), which hints at some greater complexity and is something that can be returned to at a later date.

Perhaps the most interesting finding in the above figures is Burnley’s high counter-pressing fraction. The majority of analysis on Burnley has focused on their defensive structure within their own box and how that affects their defensive performance in relation to expected goals (xG). The figure illustrates that Burnley employ a relatively aggressive counter-press, especially in relation to their possession share.

Examining Burnley’s counter-pressing game in more detail reveals that they counter-press 18 possessions per game, which is above average and only slightly lower than Manchester City. However, they only actually regain possession within five seconds 2.5 times per game, which falls short of what you might expect on average and falls below their counter-pressing peers. In terms of the ratio between their counter-pressing regains and total counter-pressing possessions, they sit 17^th on 14%.

Burnley’s counter-press is the fourth least-effective at limiting shots, with 13% of such possessions ending with them conceding a shot compared to the average rate of 10%. However, one thing in their favour is that these possessions are typically around the league average in terms of their length and speed of attack, which will allow Burnley to regain their vaunted defensive organisation prior to conceding such shots.

The more dominant discourse around pressing is as an attacking rather than defensive weapon, so narratives are often formed around teams that regularly win back the ball through pressing and use this to generate fast attacks e.g. Liverpool and Tottenham Hotspur. As a result, a team like Burnley who seemingly employ counter-pressing as a defence-first tactic to prevent counter-attacks and slow attacking progress may be overlooked.

Burnley’s manager, Sean Dyche, has typically been lumped-in with the tactical stylings of the perennially-employed British managers who aren’t generally associated with pressing tactics. Dyche was reportedly most impressed by the pressing game employed by Guardiola’s Barcelona and he has seemingly implemented some of these ideas at Burnley. He has instilled an approach that combines counter-pressing and a low-block with numbers behind the ball, which is a neat trick to pull-off; Diego Simeone and Atlético Madrid are perhaps the more apt comparison given such traits.

The above analysis illustrates the ability of StatsBomb’s new pressure event data to illuminate an important aspect of the modern game. Furthermore, it is able to do this in a manner that directly translates tactical principles, separating underlying process and outcome, which is a giant step-forward for analytics. It also led to an analysis discussing the similarity between Guardiola’s legendary Barcelona team and Sean Dyche’s Burnley, which was probably unexpected to say the least.

This is just a taster of what is possible with StatsBomb’s new data. There's more information in this presentation from the StatsBomb launch event and you can expect more analysis to appear over the summer and beyond.