What the statistics tell us: five lessons for England from Euro 2020

Success has many parents, failure is an orphan. The England men’s football team which has brought so much joy to a nation the past few weeks fell at the final hurdle. After defeat comes the post-mortem. And while for a few troglodytes this means carte blanche to unleash hatred and violence (and if you know what ‘troglodyte’ and ‘carte blanche’ mean you’re unlikely to be one of them) for the rest of us there is a chance to make an honest assessment of Gareth Southgate’s team.

Gareth Southgate’s achievements as England manager (a World Cup semi-final, a Nations League semi-final, and a Euro final) are unprecedented. Equally as impressive is how he has transcended the role and helped forge a team that once again feels connected to the fans.

For the first time since Glenn Hoddle at France ’98 England had a manager with a tactical plan and the conviction to drop and pick players based on that plan. And in reaching the final we have seven matches’ worth of statistics to look at. And in studying those data five key messages come out with some hint of how England can improve for next year’s World Cup.

England pass but don’t press and the tournament ended with a familiar pattern

At Euro 2020 England were a team who liked to pass the ball. Only Spain managed more sequences involving more than 10 passes. Yet when it comes to pressing the picture is mixed. Only Spain and Italy had more pressing sequences (passages of play where an opponent has the ball in your defensive third but through pressing within 3 or fewer passes the ball is back in their half). However, all three other semi-finalists and the Dutch, who were knocked out in the round of 16, managed more high turnovers (sequences in open play starting within 40 metres of the opponent's goal). In terms of passes per defensive action (PPDA), a measure of pressing, England fell very short. Essentially, the lower the number the more pressing a team is. England were a distant fourth amongst the semi-finalists. Spain had the lowest PPDA at Euro 2020, Denmark and Italy were seventh and eighth respectively. England had the nineteenth lowest PPDA at Euro 2020. Scotland with a PPDA of 15 was fourteenth.

However, under Gareth Southgate, a pattern is emerging against their toughest opponents. ‘First half good, second half not so good’ was a trope under Sven Goran-Erikkson but has repeated under Southgate. Against Italy England started brightly, scored in the second minute, finished the first half on a high, and then faded in the second. This was a familiar pattern, having also been seen in the 2018 World Cup semi-final against Croatia (scored in the 5th minute), and the Nations League matches against Spain (three first-half goals) and the Netherlands (scored in the 31st minute).

In all four matches, England scored first and mostly controlled the first half before ceding the second. In all four matches, England ended up being dominated in terms of possession and shots. Only against Spain did England avoid defeat and win despite being thoroughly outplayed, largely because they’d managed a 3–0 lead and just held on.

All of this suggests a worrying pattern for Southgate’s England. ‘Scoring too soon’ is a cliche in football but seems to be the norm for England in their biggest matches. Southgate plans meticulously yet it feels like the bigger matches slip away from him with changes made too late or not at all. Against Italy, Southgate did not make any substitutions until after England had conceded an equaliser.

England reliant on Kane and Sterling

Only John Stones (679) and Kalvin Phillips (665) played more minutes than Kane (649) and Sterling (641) at Euro 2020. The forwards had the most shots and highest expected goals ratings of any England player with Maguire and Mount coming distant thirds respectively. Against Italy neither player managed a single shot. By controlling Kane and Sterling Italy were able to nullify most of England’s threat.

Kane and Sterling were not at the end of movements; they also involved themselves in passing sequences. In total Sterling was involved in 24 passing sequences which finished with a shot and 6 which finished with a goal. Kane was involved in 21 sequences which ended with a shot and 7 which ended with a goal. The total expected goals for movements (total expected goal value of unique open play shot or goal-ending sequences that a player was involved in) involving Sterling was 7.08 and 6.38 for those involving Kane. Kalvin Phillips was a distant third.

With the next World Cup coming in less than 18 months this is concerning for Gareth Southgate. England’s reliance on Kane and Sterling is obvious. There are alternatives; Marcus Rashford is an obvious one. The question is whether Southgate will experiment by resting Kane and Sterling to offload pressure and find those alternatives.

Defenders the best form of attack

While England were reliant on Kane and Sterling Euro 2020 was a tournament where defenders were the best form of attack. Of the 14 players who created a chance for England at Euro 2020, 8 were defenders or defensive midfielders. The top ten average progress (the vertical distance the ball was carried) for England featured five defenders. Luke Shaw created two goals for England, Kalvin Phillips and Kieran Trippier created one each. Luke Shaw’s expected assist, a measure of the quality of a chance created, was the highest for England at 1.65.

If we look at the expected goals for passing sequences as already discussed Sterling and Kane were a distant first and second for England. But, incredibly, six defensive players were in the top ten.

Three of England’s ten goals were scored by defenders or defensive midfielders (Maguire, Shaw, and Henderson). Kane and Sterling scored the rest. Against Italy in the final Shaw scored and Maguire had a headed chance, making them more of an attacking threat than Kane or Sterling. Maguire’s penalty against Italy was the best of the shoot-out. ‘Shawberto Carlos’ has been a hero this summer and the inclusion of him and Maguire in the team of the tournament was well deserved.

English defenders at Euro 2020 were allowed to carry the ball, create chances, and were involved in the sequences with amongst the highest chances of scoring. This was not a tournament for excellent English attacking midfielders. This brings us to the next point…

Grealish, Foden, and Saka not the saviours we thought?

A lot of column inches and air time was given to the virtues of Jack Grealish, Phil Foden, and Bukayo Saka. In one of his more parochial utterances, Lee Dixon referred to Southgate introducing Jack Grealish as “giving the people what they want”. However, the picture from the statistics would suggest that Dixon and ‘the people’ were wrong.

Grealish carried the ball for an average of 13.2 metres. Only Marcus Rashford carried the ball more. In terms of average carry progress, Grealish was fifth in the squad. The Aston Villa man also created two goals: Sterling’s versus the Czech Republic and Kane’s versus Germany. However, for a side built around passing the ball, he managed only 71 passes, the 15th highest in the squad.

Despite creating two goals his expected assist value was only 0.87. If we look again at the goals he created we can see why. The chance Grealish created for Sterling against the Czechs had an expected goal rating of 0.46 meaning there was a 46% chance of scoring given the position of two defenders and the ‘keeper.

Against Germany Grealish chipped the ball to Kane who received the ball slightly below head height a few yards from goal. Options for Kane were limited, it was too low to conventionally head and too high to chest and control or volley.

Compare that to Luke Shaw’s pass to Raheem Sterling for England’s first goal against Germany. It’s across the ground and into the space that Sterling had found. As a result, Sterling had a 66% chance of scoring and a relatively simple finish.

Compare that to Grealish’s pass where there was a 47% chance of Kane scoring. In the end, it was Kane’s reactions and an imaginative header into the ground which turned the chance into a goal. A successful pass but not a reliable one. The expected goals for passing sequences Grealish was involved in was the 10th highest for the squad, a statistic which becomes even less impressive as we learn that sixteen players in total were involved in a shot or goal-ending movement, placing Grealish in the bottom half.

The data for Grealish suggest the reason for being a crowd favourite: people love to see a player run with the ball. But the end results were mixed. He has a reputation as the most fouled player in the Premier League but at the elite level, you need more than just an ability to run.

Phil Foden is coached at club level by Pep Guardiola. Yet he only managed 71 passes, the same as Grealish. He created two chances, one more than Grealish, but had zero assists and an expected assist of just 0.08. His sequence involvement expected assists was 0.96, the 13th highest of the England squad. If fans had hoped that Foden would bring some Guardiola football to England it didn’t come to pass.

The picture is much the same for Bukayo Saka. Thirteenth in the England squad for both passes (81) and expected assists (0.07) Saka created just one chance, zero goals, and was fifteenth for sequence involvement expected goals. In other words, despite the hype, Saka was not a creative attacking force for England. A lot was made of his pace but his average progress carry was the 9th highest, Harry Maguire was 8th. Ignoring his penalty miss, Saka’s final performance was non-existent, his only opportunity to run being cruelly ended by Chiellini.

It is unfair to lay this all at the feet of Grealish, Foden and Saka. As already shown, England were a team that used Kane, Sterling, and defensive players as their main creative sources. The more attacking midfielders were largely bypassed. Southgate needs to assess how players like Grealish, Foden, and Saka fit into this picture. Foden and Saka in particular are young and have time on their hands. This may have been a tournament too soon for the Arsenal youngster and hopefully, he will be protected over the season ahead. Foden will have more tuition from Guardiola and Southgate should use that experience.

With 223 passes between them, an average expected assist of 0.34 and less creative involvement than several defenders, Grealish, Foden and Saka were not the saviours they were made out to be. Grealish at Manchester City is a strange fit.

Pickford better than Donnarumma

Italian goalkeeper Gianluigi Donnarumma was named Player of the Tournament for Euro 2020 and took the ‘keeper spot in the Team of the Tournament showing talent and attitude belying his youth. However, by every metric Jordan Pickford was superior at Euro 2020. Despite playing fewer minutes (690 to the Italian’s 719) Pickford faced more shots (18 to 13). He conceded fewer goals despite having greater expected goals conceded (number of goals that a keeper was expected to concede, given the quality of the on-target shots he faced).

This means that in terms of goals prevented (number of goals that a goalkeeper was expected to concede compared to the number that they actually conceded, according to expected goals conceded) Donnarumma prevented -0.1 goals. This suggests he actually conceded more goals than he should while Pickford prevented 2. His goal prevention rate was more than double the Italian’s. Pickford faces a lot of scrutiny at club level and Euro 2020 was no different, especially regarding the goal he conceded against Denmark and his demeanour on the ball. But the statistics are clear. He was brilliant at Euro 2020. And that’s without mentioning two fine saves in the final penalty shoot-out including against ‘Mr 100%: Jorginho. And he was unlucky with Italy’s equaliser as the ball ricocheted off the post following an excellent save. By all means laud Donnarumma. Pickford was actually better.

http://JamesThomas.me Doctor and Educator. Podcast https://www.takeaurally.com Blog on Medical Education and History of Medicine https://mcdreeamiemusings.com