Stats at updated for all leagues. Below is an example of what you can do with the data. Highest defensive line 'so far' belongs to Man City and Arsenal, lowest - Man United and Everton. Highest press intensity (PPDA) - Leeds, Chelsea, Liverpool
Premier League game style evolution for the past decade
▪ passes from goal-kicks are shorter
▪ crosses are less frequent
▪ pass directness is lower
▪ less shots % from outside of penalty box
▪ more 'high turnovers'
* all metrics are per team per game
The best central midfielders in the Premier League for ball progression are KDB, Rodri, and Thiago. The best for ball winning are Gueye, Doucoure, and Palhinha. Above average in both categories: Thiago, Casemiro, Marc Roca
When Arsenal is in possession against opponents’ low block and Bukayo Saka receives the ball, a chance that within next 3 actions Arsenal will make a shot or deep completion equal to 11.5% (higher than Messi 20/21)
Arsenal currently leads as the best defensive team in the big-5 leagues. Their success stems from a combination of high press efficiency and the quality of their mid/low block when needed
What was Pep Guardiola’s plan to mitigate the impact of the early red card? Just ask Bernardo and Kevin to cover 2 different positions at the same time. Heroic performance
#MCIFUL
Arsenal's ball progression efficiency compared to the last season. While Rice is holding up high standards, the departure of Xhaka, positioned at the 78th percentile among PL central midfielders, has been succeeded by someone now at the 34th percentile
#PSGRMA
A role of deep-lying creator suits Leo Messi so well at this point of his career. He used to be a solution against small teams in La Liga who park the bus and we've seen the same tonight
Smart clubs make smart transfers and Kaoru Mitoma is a prime example. As soon as he gets enough minutes on the pitch, he will be the best ball carrier in the Premier League
Moises Caicedo's key strengths are remarkable ability to execute precise and reliable passes even in the face of intense pressure and exceptional performance in ground duels. He could potentially replace either Xhaka or Partey in Arsenal's starting eleven
I can't stop admiring Bernardo Silva. Look at ball-receiving heatmaps, he was everywhere on the pitch. My favourite is when Bernardo drops back into first line to break opponent's press. Amazing player, happy he scored tonight
#MCIRMA
Players with the most ball progressions into deep areas via pass cross or carry from the last season:
Mbappe (PSG)
Salah (LIV)
Júnior (RMA)
Messi (PSG)
Saka (ARS)
Odegaard (ARS)
Martinelli (ARS)
KDB (MCI)
Casemiro is in a top tier of La Liga central midfielders who good at both 'ball winning' and 'ball progression'. I understand age and contract duration doubt, but I can't believe fans say 'we need someone better on the ball' 🙄
What players do next after successful dribble?
Kingsley Coman attempts a shot or provides a key pass in 25.5% of cases.
2 out of 3 successful dribbles of Bryan Gil and Jadon Sancho end wither by possession loss or safe play without progression
Data at updated. A new addition is 'carries into penalty box' for players. At the top right corner of the graph Premier League best ball carriers: Martinelli, Salah, Sancho, Zaha, Sterling
* Saint-Maximin would be on top with enough played minutes
Bayer Leverkusen is one of the best counter-pressing teams in Europe. Their ability to swiftly regain possession high up the pitch immediately after losing it is notable, enabling them to maintain relentless pressure on opponents (1/2)
I've analyzed all goal-kicks played in various European leagues since 2019. The table shows outcomes of the following 15 events
If you play short, your sequence ends up on average 20 meters higher up the pitch. Also your chance to retain possession twice higher
The highest counter-pressing success rates in Premier League this season belong to Man United (👀), Man City and Arsenal. Chelsea rate is also pretty high, but they aren't good at converting these situations into shots or deep completions (y-axis)
This season we witnessed seven mid-season manager appointments in the Premier League. Six of them improved xG difference and Points Per Game of their new sides. Is the so-called “new manager bounce” a real thing? 🧵
1/ No team in the top-5 degrades the quality of their opponents' buildup more than Liverpool do. Liverpool's opponents are passing with the lowest pass accuracy this season
Even though Zinchenko's out-of-possession limitations are a bit exaggerated, I believe Riccardo Calafiori is a defensive upgrade who, in theory, can also progress the ball effectively
Among all transfers from Germany to England happened this season only Emmanuel Dennis matched his previous Bundesliga numbers (npxG+xA).
A general rule for this direction of transfers is 38% decline. Not sure if this applies if a player moves to Man City
Press-resistance of central midfielders. To measure the ability to beat an opponent's pressure I used two metrics: 1 Pass completion under pressure. 2 End result of all sequences where a player was under pressure (i.e. xGBuildup per sequence)
Who is the most effective ball progressor via pass in Big-5 this season? Exequiel Palacios, Rodri, and Frankie de Jong make up the top 3, with Toni Kroos in a league of his own
I've processed 150k throw-ins from big-5 leagues trying to understand if fast throw-in makes any difference.
Looks like the most optimal preparation time lays within first 8 seconds. Next, every 2 additional seconds decreases 'success rate' by 2-5 percent
Pep on Grealish: “We didn’t sign him for the goals or assists… it’s about what he can produce for the other ones”.
When City is in possession against low block and Grealish receives the ball, a chance of making shot or deep completion within next 3 actions equal to 8.1%
There is only 2 forwards in Europe who equally great* at scoring goals, generating attacking threat and linking-up play
And guess what, both played yesterday against Chelsea
*'great' - 90+ percentile for all 3 metrics
** domestic competitions + UCL/UEL
Here is the Premier League "Team of the Season" based on ball progression effectiveness, calculated as "fields gained via pass + carry per possession lost"
Examining pass completion 'in buildup' for European teams and their opponents this season. Those in the top right corner prioritize patient attack buildup, denying the same opportunity to their opponents
When the game state is "winning" Liverpool and City continue possessing the ball and concede a low amount of threat. Barca and Bayern - high possession % and high volume of conceded threat. Man United, low possession % and high threat, Atletico Madrid - dark arts
When
#ManCity
starts possession from goal kick, on average they progress the ball as far as 59.5 meters from start. Which is 6 meters further than second-best team (
#Liverpool
) and 13 meters further than third-best (
#Chelsea
)
I've processed 50k corners and compared 2 strategies: short pass and cross into penbox.
As efficiency metric I used xG created and conceded within next 10 events after a corner.
Long corners just a bit more "risky" (chance to concede), but "reward" is higher (chance to create)
Last season, Liverpool and Manchester City were the two best PL teams at preventing opponents from generating threat and shots within the first 15 seconds of their possessions. However, things have changed this year
"We ask goalkeepers to play long balls because we believe it is safer". However, data does not confirm this assumption. In fact, teams tend to concede more expected threat within the next 10 actions if their keepers prefer long passes over short
Only Trent, Robertson and Cancelo were better than Zinchenko by threat creation and ball progression last season. Superb signing if happens
* calculated for Premier League fullbacks with 1000+ minutes, non-adjusted by team possession
#Arsenal
#PremierLeague
Average defence line height in Premier League last season. Big gap between first two and others. Happy to see Brighton fourth in the list
#PremierLeague
Roberto Firmino is the best threat creator from ‘zone 14’ among all Premier League strikers for the last 5 seasons. In addition to regular xT, I calculate xT-facilitated (or “assisted”). And guess what, Romelu Lukaku is by far the best facilitator
The most threatening PL players of the last season. Bruno Fernandes, Trippier, and Trent were almost twice as dangerous at passing compared to their teammates. Doku, Mbeumo, and Luis Diaz stood out at carrying the ball into the ‘danger zone’. Michael Olise excelled in both
After the restart, the top three Premier League teams are Brighton, Manchester City, and Arsenal, followed by Liverpool, Chelsea, and Manchester United
Threat creation and ball progression by
#PremierLeague
fullbacks (every dot is a single game)
🟡 Performance of the season belongs to Joao Cancelo against Wolves
🔵 Numbers-wise Reece James doesn't perform on top level consistently
🔴 Trent simply the best
Although I already suspected that Casemiro had a significant impact, I was still surprised to learn that Manchester United concedes 0.3 xThreat and 0.8 xG more without him. And it's not just about defence
I’ve developed a twitter bot which will be posting post-match reports for 'popular' teams
Works semi-automatically, i.e. might be some [significant] delays between final whistle and corresponding post
Here is how Middlesbrough - Tottenham report looks
Can't believe how much Arsenal fell apart after Zinchenko substitution. It wasn't because of his defensive skills, not at all, all the difference made his ability to retain possession
Comparing
#Arsenal
to last year: Enhanced attack and defense quality, more effective pressing and counter-pressing, and improved control in the buildup. Progress across all phases
Counter-pressing(gegenpressing) doesn’t guarantee you protection from opponent's fast-breaks
Take Atlético and Brighton. Simeone’s side is ‘below average’ at counter-pressing, but much more successful at neutralizing fast-breaks than Brighton who prefer to gegenpress
Bournemouth’s opponents lose possession in the “danger zone” on average 32 times per game and complete 78% of their pass attempts. After 4 game weeks, Bournemouth’s press is the most effective at disrupting buildup. I’ve started adding 2024-25 stats to
#Liverpool
tonight had season lowest possession %, lowest field tilt %, lowest % of attacks reached final third, second-lowest threat created, second-lowest def. line height
Big W for Tuchel imo as he forced one of the best PL teams play unusual football
#CHELIV
Long throw-ins straight into the penalty box are more efficient than short passes
Results based on xG of shots generated by team or opponent within next 3/10/20 actions after throw-in attempt (made in final third)
In this game, Manchester United recorded their lowest pass completion 'in buildup' this season, hitting just 76%, marking the first time they've fallen below 80%
Unique pass clusters of central midfielders. Passes that a player attempts much more often than others on the same role
Toni "maestro" Kroos
Youri "quarterback" Tielemans
Declan "nothing special" Rice
* calculated for all live pass attempts made in big-5 last season
Martín Zubimendi ranked 23rd among central midfielders in the Premier League and La Liga for “ball progression effectiveness” last season. Both Mac Allister and Endo are higher on this list
A tale about Antonio Conte who switched from "Back 4" (in yellow) to "Back 3" (in blue) and then destroyed the Premier League in 2016/17. Game against Arsenal (in pink) has been pivotal in this context as introduced Marcos Alonso for the first time
#PremierLeague
#ChelseaFC
Andre Onana solved
#MUFC
problems in the first phase of buildup. Even teams like Arsenal think twice before jumping into a high press. Now they need a few more “Onanas” in the next phases of possession
There’s no right or wrong way to play football, but calling Real Madrid’s low block tonight a 'good defensive performance' seems driven by outcome bias. They allowed roughly 3 xG and conceded 40 entries into their danger zone
Standout performance by Aurélien Tchouaméni tonight. By far the best ball progressor on the pitch. According to my 'progression effectiveness' model, Tchouaméni is currently the second-best player of the tournament, just behind Toni Kroos
Brentford is the best team in Europe in terms of disrupting opponents buildups started by goalkicks
I calculate a metric which shows how far (in meters) teams manage to progress the ball from goalkick or prevent opponent progression
Arsenal’s numbers are peculiar: xGDiff, possession, and field tilt are all negative. The elephant in the room? Minutes played at 11v11. Exclude shots after red cards to either side, and Arsenal jumps to the top in xG conceded
Arsenal's recent performance closely resembles their pre-World Cup levels from last season. The graph illustrates entries into the 'danger zone' made by Arsenal and their opponents over the past three years
Points per game and schedule difficulty after seven games
- Liverpool: great results, easiest opponents
- Man City: great results, tough schedule
- Fulham: solid results, average opponents
- Wolves: poor results, harsh schedule
My experience with "SB360". First of all, I want to thank
@StatsBomb
for letting us play with the dataset.
For all completed passes on Euro-2020, I've calculated the number of bypassed (packed) players.
The picture demonstrates how Leo Bonucci "packs" 9 Belgians
Style of play at World Cup-2022 so far is the closest to what we see in top-5 leagues:
▪ % of shots outside of the penalty box is lower
▪ defensive lines are higher
▪ goal kicks are shorter
Going to monitor and check if numbers regress no "normal world cup level" later
Despite the decline, Liverpool is still the best Premier League side at disrupting opponents' buildup. The data at has been updated for the last time for the 22/23 season
Goalkeeper pass completion correlates very well with team pass completion. It is fine to have a low percentage if you play for Empoli. David De Gea doesn't look bad, he is pretty much aligned with the rest of the team. The biggest outlier in the Premier League is Ramsdale 😅
End of season is a perfect time to revisit old templates. Some of mine require updates but fine in general, others I will never use again, so it’s their last dance. Thread contains 10+ different visualizations for Premier League 21/22
Player contributions into breaking opponents’ "low blocks". When Neymar receives the ball, a chance that PSG will make a shot or deep completion within next 3 actions equal to 9.6% - the highest number among all PSG players [1/n]