Manchester City Analysis: Why the Leaky Defence Is Being Hidden by the Attack

 

Manchester City have won all five Premier League games and scored 13 goals. But Sunderland's 3.57 xG against them exposed something the league table cannot show. I have watched every City match this season, and the results look perfect while the performances do not. The question is not whether City can score. It is whether they can control games without relying on outscoring opponents every week.

 

Read More: Manchester City 2026/27 Season Preview & Premier League Predictions

 

Five Wins, But Are City Controlling Games?

 

Honestly, no. Not in the way their perfect record suggests. Winning five straight Premier League matches is not luck. It takes quality, consistency, and the ability to solve problems when games get difficult. City have shown all of that. But the way they have won tells a more complicated story.

 

On opening weekend, City needed goals in the 84th and 91st minutes to come back against Bournemouth. I watched that one live, and I kept thinking to myself, this does not look like a team in control. Then their 1-0 win over Coventry? Comfortable on paper, sure, but if you actually sat through it, you would know it was not. The newly promoted side created enough chances to make City sweat throughout.

 

The 1-0 victory over Manchester United was a different story though. After Phil Foden's red card in the 23rd minute, City played most of the derby with ten men. And you know what? They stayed compact, limited United's threat, and controlled the match without the ball. That performance proved Maresca's side can defend when required.

 

The problem is they have not consistently needed to.

 

Against Sunderland, City were dragged into chaos. They had to retake the lead three times in the first half. Sunderland repeatedly found space in dangerous areas. The result was spectacular. The performance was not.

 

A perfect record can hide the difference between winning matches and controlling them. City are doing the first extremely well. The second remains a work in progress.

 

What Did Sunderland's 3.57 xG Tell Us?

 

It told us that City's defensive problems are real. Sunderland created enough quality chances to score more than three goals, and that should worry any City fan. The most striking number from that game was not City's five goals. It was Sunderland's 3.57 expected goals.

 

I have followed Sunderland closely since they came back up, and this number genuinely surprised me. They are not a high-volume chance-creation team. Across their first 43 league matches since the start of last season, they exceeded 2.0 xG only twice. Their previous best was 2.44 against Newcastle. They had produced less than 1.0 xG in 22 of those 43 games.

 

So creating 3.57 xG against Manchester City was highly unusual.

 

What made it more concerning was the type of chances. These were not speculative efforts from outside the box. Brian Brobbey repeatedly found himself in dangerous central areas and scored three close-range goals. Sunderland penetrated City's defence often enough to create opportunities in the most dangerous zones.

 

Even when City led 4-3, the game remained alive. A Rúben Dias goal-line clearance prevented Sunderland from finding another equaliser.

 

Expected goals should not be read as saying Sunderland deserved to win. City scored five because they were more efficient in attack and deserved the three points. But xG tells us about the quality of chances a team allows. By that measure, City's defensive performance was concerning.

 

Sunderland did not merely score three goals. They created enough opportunities to score even more.

 

Do Four Clean Sheets Mean City's Defence Is Settled?

 

On the surface, yes. But the deeper numbers say otherwise. Four straight clean sheets suggest the defence had settled. But clean sheets only describe the outcome. They do not explain how many dangerous chances a team had to survive.

 

Across that four-match run, City conceded chances worth approximately 2.82 xG. Opponents created opportunities that would reasonably have been expected to produce around three goals. City conceded none.

 

Some of that is good defending. Some is goalkeeping. Some is poor finishing from opponents. A goal-line clearance or a missed one-on-one can completely change the statistical outcome of a match. City now have Gianluigi Donnarumma, a goalkeeper capable of providing exactly that kind of margin. But relying on goalkeeping and opposition finishing becomes dangerous when the same pattern repeats.

 

The Sunderland match was important not because it suddenly created a defensive problem. It was important because it made an existing concern impossible to ignore.

 

City are allowing too many high-value chances.

 

What Has Maresca Changed in City's Shape?

 

The biggest change is how clearly City are structured once they have the ball. One of the most notable developments under Maresca has been the way Manchester City transform their shape in possession. Against Norwich in the Carabao Cup, City frequently moved from their nominal 4-2-3-1 into structures resembling a 3-1-6 or 2-2-5-1.

 

The full-backs moved into the half-spaces. Players like Rayan Aït-Nouri and Allan held the width. Ayyoub Bouaddi could drop into the defensive line to create a temporary back three, while the attacking midfielders occupied pockets between Norwich's midfield and defensive lines.

 

This created multiple problems for Norwich. When Bouaddi dropped into the back three, City could create a three-versus-two overload against the opposition's front line. When Norwich's wide players tried to follow City's inverted full-backs, spaces appeared elsewhere.

 

Rayan Cherki was particularly important because he could move between these zones. Rather than staying permanently between the lines, Cherki could drop deeper to connect midfield and attack. That allowed Rico Lewis and Nico O'Reilly to stretch Norwich's midfield structure.

 

The principle is clear. Maresca wants City to create numerical advantages across multiple zones rather than simply dominate possession for its own sake. But unlike a purely aggressive attacking system, his structure is designed to retain a rest defence, with three defenders and one or two midfielders capable of staying behind the ball. That gives City protection against transitions while allowing more players to attack.

 

Another change I have noticed is how important the wingers have become to Maresca's approach. By keeping them high and wide, City can create more one-versus-one situations and give their wingers space to attack the full-back. I also think this has a knock-on effect centrally: when opponents shift across to deal with the threat out wide, players like Cherki and Foden can find more room between the lines. For me, this makes City's wide play more than just a way to stretch the pitch; it has become a repeatable route to creating cut-backs and chances in the box.

 

But it also creates a potential problem. The risk is not simply that too many players are positioned in front of the ball, but that City's high defensive line and zonal pressing structure can leave midfielders and full-backs covering large distances. This is where the transition phase becomes particularly important. Rodri's value was never just tackles and interceptions. He controlled the game, protected the space in front of the defence, and helped City manage transitions. With Rodri no longer providing that control, maintaining the balance between Maresca's positional structure, high press and defensive security becomes even more important.

 

In my view, Maresca's City are not a complete departure from Guardiola. I still see the same principles of positional play and rest defence, but expressed differently. The key changes are a more structured positional setup, high and wide wingers, and a greater emphasis on creating repeatable one-v-one and cut-back situations.

 

What Do City's Crossing Numbers Reveal This Season?

 

Under Pep Guardiola, City's crossing numbers from Opta Analyst have gradually declined in recent seasons. In 2025/26, they averaged 15.3 crosses per game. This season, that number has risen to 18.4.

 

But the more interesting change is what happens after the ball is delivered.

 

 

The increase is significant. City are attempting more crosses, generating more shots from them, scoring more often from them, and creating twice as much xG from crossing situations.

 

The obvious explanation is Erling Haaland. He gives City something Guardiola's later teams had increasingly moved away from: a physically dominant central target capable of attacking deliveries from wide areas. Since joining City, Haaland has scored 17 Premier League goals from crosses, the most of any player during that period.

 

Maresca is not simply asking City to cross more. He has a striker who makes crossing a much more valuable weapon.

 

Antoine Semenyo has also become important in this system, ranking highly for crosses attempted while contributing goals, shots, big chances, and expected assists from those situations.

 

This is one reason City's attack may be even more difficult to defend than last season. They can build through the middle. They can isolate wingers. They can attack the half-spaces. And now they are increasingly comfortable putting the ball into the box for Haaland.

 

Why Haven't City's Defensive Problems Cost Them Points?

 

This attacking variety explains why City's defensive problems have not translated into dropped points.

 

Haaland has started the campaign with five league goals. Cherki has three goals and two assists, while Semenyo has two goals and three assists. Foden remains a major creative threat. Doku provides another one-versus-one option. Enzo Fernández offers additional attacking presence from midfield.

 

Even City's younger players have shown there is considerable depth behind the established stars. Against Norwich, 17-year-old Floyd Samba scored twice, while Allan contributed a goal and an assist. Bouaddi completed 98 of 103 passes and created two chances.

 

The significance of those performances goes beyond the Carabao Cup. They demonstrate that City have multiple ways of hurting opponents.

 

That matters because elite attacking depth can compensate for defensive instability. If one player is unavailable, another can create. If the central route is blocked, City can cross. If the opposition presses high, they can exploit the space behind. If they concede, they have enough attacking quality to respond.

 

The danger is that this can become a form of defensive insurance. City may begin to rely on the assumption that they can always score again. That is a dangerous habit over a 38-game league season.

 

What Did the Manchester Derby Show About City's Control?

 

It showed that City can still control a game when they have to. The question is whether they can do it when they are not forced to. After Foden's red card in the 23rd minute, City were pushed into a completely different type of match. Instead of chasing another goal, they had to defend. And they did it impressively. That is why the win at Old Trafford stands out as such an important counterexample.

 

I went back and watched the second half again, and United never really built sustained pressure. Two efforts hit the woodwork, but it took until the 70th minute for United to register their first shot on target after the red card.

 

Elliot Anderson was central to that defensive effort, covering 11.9 kilometres and recording 13 defensive contributions, including three tackles, three clearances, an interception, and six recoveries. Haaland eventually scored the winner. That performance proves City are capable of the control Maresca wants. The real test is whether they can reproduce it when circumstances do not force them into it.

 

Does History Favour Attack or Defence in Title Races?

 

The traditional football argument says that attack wins games, defence wins titles. Premier League history is not quite so simple.

 

When I look at the 28 completed Premier League seasons played with 20 teams from 1995/96 to 2022/23, I find that the team with the best attack won the title 19 times, compared with 14 times for the team with the best defence. I have included the table below to show where each champion ranked for goals scored and goals conceded.

 

Source: Premier League. Attack rank is based on goals scored; defence rank is based on goals conceded. = indicates a tie.

 

That suggests attacking superiority may have a stronger historical relationship with title success than the famous cliché implies. But there is an important caveat: many champions had both the best attack and the best defence. That is exactly why City's current situation is so interesting. Their defence does not necessarily have to be the best in the Premier League if their attack can maintain a sufficiently strong advantage.

 

The problem is that maintaining 2.6 goals per game over an entire season is a much bigger challenge than sustaining that level over just five matches.

 

GoalBible Conclusion: Man City Do Not Need More Goals. They Need More Control.

 

Manchester City have already shown that their attacking quality can win them difficult games. The bigger question is whether City can turn that attacking strength into consistent control.

 

That is what makes Maresca’s preference for 1-0 over 5-3 so significant. A 5-3 win is entertaining, but a controlled 1-0 victory may say more about a team’s ability to win a Premier League title.

 

City have started the season perfectly. The next step is making their performances as convincing as their results.

 

What the Data Says About the Next Big Test for Man City vs Liverpool?

 

The questions surrounding Manchester City’s perfect start naturally lead into their next major Premier League test against Liverpool. If City’s attack remains their biggest strength while defensive control remains the main concern, two betting markets closely align with the trends highlighted in the analysis.

 

Note: Odds correct as of 24 September 2026.

 

City have scored 13 goals in their opening five league games, while Sunderland's 3.57 xG showed that opponents can still create good chances against them. Liverpool should provide a tougher attacking test, making BTTS a natural fit with the pattern seen so far. City have plenty of quality to score, but their defence has not always looked equally secure. Among the odds compared here, 20Bet's 1.52 for BTTS Yes is the highest listed price.

 

Note: Odds correct as of 24 September 2026.

 

City's attacking numbers and increased use of crosses have made them a consistent goal threat, while their defensive issues can make matches more open. With Haaland leading the attack and Sunderland already exposing City's vulnerability, Over 2.5 goals follows the same logic. For this market, 20Bet's 1.58 for Over 2.5 Goals is the joint-highest available quote.