Does Manchester City Really Have a Fatal Flaw?

Despite a declining defense, City might be more dangerous than ever before

Over the past two seasons, Manchester City solved soccer. It’s supposed to be a game of trade-offs: attack more aggressively, and you weaken the defense; pack more bodies behind the ball, and you blunt your attack. The same, in theory goes for the players you pick. They all have strengths and weaknesses that add to and take away from a given lineup’s playing profile. For all that, I love this definition from former Chelsea manager Gianluca Vialli: “Deconstruct tactics, and you’ll find that basically it’s a way to minimize a team’s weaknesses while maximizing its strengths.”

But since about August 2017, Pep Guardiola and his players have found a way to maximize their strengths -- and eliminate their weaknesses. They won a Premier League-record 100 points in the 2017-18 season and then followed that up with a performance that might’ve been even better: 98 points, despite a 97-point challenger that won the Champions League final, a third-place team that won the Europa League, a fourth-place team that lost in the Champions League final, and a fifth-place team that lost in the Europa League final. In both championship seasons they created the most expected goals and conceded the fewest, per FBRef. Over the past two years, they’ve been the best attacking team in England, and they’ve been the best defensive team, too.

That’s not supposed to happen, but thanks to a machine-like offensive rhythm that produced tap-in after tap-in, massive personnel investment that produced an incomparably deep roster,  a swarming high press, and an onslaught of strategically-timed tactical fouls, it did. That is, until now.

Their top-line performance was perfect, and trade-offs didn’t even really exist within the micro-aspects of City’s play, either. Part of City’s defensive dominance came from their ability to keep possession and defend effectively high up the field -- two aspects that tend to reinforce each other. But when teams play that aggressively, they have historically conceded high-quality chances. Great pressing teams become great defensive teams because of how few chances they concede, not because they’re necessarily adept at dealing with the chances that do occur. Per Stats Perform data, City allowed the fewest shots in the league last season (6.3, next-best was 8.2), and on average they won the ball back farther from their own goal than anyone else(48.2 meters, next-best was 44.8). I’ll let Statsbomb’s Grace Robertson take it from here:

Last season, City had an xG per shot conceded of 0.09, around league average. This isn’t how a side that presses the ball as high up the pitch as City would typically operate. The point of a high press is that you limit the number of shots your opponent is able to take. The risk, though, is that once the opposition breaks your press, you’re badly exposed, and they can carve out an excellent scoring opportunity. That this team could press opponents so hard to concede the fewest shots in the league while also making sure those shots were only of middling quality seemed like genuine magic.

Eight games into this season, City are still the kings of shot suppression (7.1) and pressing-height (48.7 meters), but the bottom has finally given out. Here’s Grace again:

And this is where we find the leak. City’s xG per shot conceded this season is 0.17, nearly double that of last year and the worst in the Premier League ... City concede the second fewest deep completions in the league and the second fewest opponent passes inside their own box. Teams still find it awfully difficult to get the ball up the pitch against Guardiola’s side and generate shots. But, once they do work it into dangerous areas, it seems to be easier than ever to turn those situations into good chances.

Despite facing a similar number of shots as the league-leading defensive unit of this past two years, City have conceded more xG (9.5) than seven other teams and more goals (9) than five others. They’re now in second place in the table, eight points back of Liverpool, who have won eight from eight. After City’s 2-0 loss to Wolverhampton, betting markets finally flipped and made Liverpool the favorites for the Premier League title. Given that their attack the best not just in the Premier League but in all of Europe by a significant margin, the defense seems like the obvious explanation for the sudden decline. Throw in new personnel at defensive midfield and fewer personnel at center back, and you can start to tell yourself a pretty compelling story.

Except, I’m not sure it’s true. Well, I’m sure part of it is true: Manchester City’s defense is currently closer to league-average than league-best. But I’m not sure Manchester City are any worse, on the whole, than they were last year. Their per-game goal differential this year is plus-2.25; last season, it was plus-1.89. The year before that, it was 2.07, which ... was the best goal-differential in Premier League history! Same goes for xG. This year?  It’s 1.86. Last year? It was 1.51. The year before? It was 1.55. In terms of scoring more goals and creating better chances than their opponents, City might be better than they’ve ever been.

They’ve got Kevin de Bruyne after he only played around 900 minutes last season, and now he’s got nearly double the assists (and expected assists) of anyone in Europe. Perhaps he’s skewed the balance of the team, but in terms of goals, the attack has improved more than the defense has declined. And that’s the whole point, isn’t it?  

Sure, but at the same time, it’s hard not to look at Liverpool and see their performance as an argument in favor of balance. They’re second in xG for and second in xG against. As I wrote on Friday, they’re the only team in Europe with a top-10 defense and a top-10 attack. Jurgen Klopp’s side may not have had the highs of City’s 8-0 over Watford; instead, they just have eight mostly ho-hum wins in eight games. The ceiling seems lower, but the floor seems higher. City’s attack is so powerful that it’s turned this guy and this group of players into a bunch of pragmatists:

Stopping a goal is worth more than scoring a goal. Think of it this way: If you don’t concede a goal, the worst your team can do is draw. If you score one or two or even three, every result remains on the table. That’s why it makes sense for clubs with fewer resources to invest in the defense rather than the attack; removing one goal from the GA tally creates more value for your team than adding one to the G column.

However, City and Liverpool are two of the richest teams in the world; they’re both trying to find edges, but neither one is looking to maximize a limited budget. At the uppermost echelon of the game, research suggests that the best teams are more likely to win games that feature more goals. It might not feel that way -- more goals suggests less control and a bigger downside -- but in the long run, it’s true. You, a reader of this newsletter, probably know this by now since I’ve cited it plenty of times, but we have new subscribers, so here’s Ben Marlow of the consultancy 21st Club:

The principal reason why variance exists is that football is a low-scoring sport – a goal is more influential on the outcome of a match than, say, a three-pointer in basketball or a penalty in rugby. Teams often lose games that, based on performance, they really deserved to win. In fact, our analysis suggests that, on average, the better team (that is the team that creates notably more dangerous chances) wins only around 64% of the time.

While the factors that cause variance may be outside of our control, the extent to which we are affected by them is not. The premise is simple: variance is the result of football’s low scoring nature, so it can be reduced by making the games higher scoring.

The data bears this out – our analysis suggests that the better team wins around 75% of the time in games with over 2.5 goals vs. 50% in games with fewer than 2.5 goals.

Those same patterns seem to show up at the top of the Premier League. Take a look at the last 10 teams to win the title, matched up with where they ranked in goals scored and conceded:

-Manchester City, 18-19: 1st attack, 2nd defense
-City, 17-18: 1st attack, 1st defense
-Chelsea, 16-17: 2nd attack, 3rd defense
-Leicester, 15-16: 3rd attack, 3rd defense
-Chelsea, 14-15: 2nd attack, 1st defense
-City, 13-14: 1st attack, 2nd defense
-Manchester United, 12-13: 1st attack, 5th defense
-City, 11-12: 1st attack, 1st defense
-United, 10-11: 1st attack, 3rd defense
-Chelsea, 09-10: 1st attack, 2nd defense

Seven of the last 10 Premier League winners have scored the most goals in the league, while the stingiest defense has lifted the trophy just twice over that stretch. No team has ranked as low as City’s defense currently does, but no team has ever scored as many goals as they do either.

Last year, Liverpool made up ground on City’s impeccable open-play performance with their own near-flawless set piece execution. Klopp’s team scored 22 goals from dead-ball situations while Guardiola’s side only put up 12. That’s not the explanation for this year’s gap, though, as they’re both tied-for-second in the PL so far with four tallies from set plays. And so, City sit behind Liverpool right now mainly because shit happens in soccer more than it does in any other sport. City dominated Tottenham to the tune of 3.0 xG to 0.2 ... and the game ended 2-2. They lost to Norwich, 3-2, despite an xG advantage of 2.5 to 1.6. Liverpool, meanwhile, took down Chelsea, 2-1, even though they created only 0.9 xG and conceded 1.5. They beat Southampton by the same scoreline despite an even xG (1.6 to 1.6). Flip those results to more closely match the performances, and you flip Liverpool and City’s places in the table.

Instead, reality. City sit eight points back with 30 matches to go. It’s the most exciting set up we could’ve asked for: the two-time defending champ, the indomitable machine, the more talented team trying to chase down the side that came within a point of their first title since the Premier League began, clearly flawed but perfect so far. City’s defense is as vulnerable as its been since Guardiola arrived, but the current tactics should still give them their best chance to make things interesting. After years of magic making, they’re finally doing what Vialli said.