Manchester City 2-2 Arsenal: The Debrief
Stats, and thoughts from Arsenal's draw away to Manchester City
It is such a bummer that this match was turned into a debate about refereeing again, because it deserved so much more. I am not going to go deep here on the decision but rather save that for the VAR/Contensious call tracking update that I update.
As I write this I haven’t updated it yet, but it might be there by the time that you read this.
I do want to stick to things that are bit more real here so let’s get straight into it.
Manchester City 2-2 Arsenal: The Debrief
Red cards will screw up Arsenal’s numbers for a long time
99 - Minutes (including stoppage time) Arsenal have played this season down a man, the most in the Premier League
20% - The percentage of the minutes that Arsenal have played this season where they have been down a man.
47 - Shots Arsenal have conceded while down a man
52% - The percentage of the shots that Arsenal have allowed this season that have come while down a man
3.1 - xG Arsenal have conceded while down a man
49% - The percentage of the expected goals that Arsenal have allowed this season that have come while down a man
This is going to be something that will need to be watched and taken into account when we are looking at Arsenal’s stats for a while. Over the course of the season this should be become a small enough part of their total minutes (and maybe even get some time playing a man up too) that it will come out as inconsequential. It might be closer to week 15 where the stats for Arsenal have a bigger meaning compared to the normal 8-10 and until then it will always need to come with an asterisk attached.
If you want to pick on Arsenal’s strategy while down a man, I could understand that if you were talking Brighton but against Manchester City I think that they did essentially what they needed to do.
Against Brighton they probably could have spent more time in possession rather than ceding almost all the ball to them (76% to 24% in that match). In that match Arsenal did have a couple nice breaks (nearly matched them on xG while down a man) and were able to release pressure more effectively while limiting the overall times they were broken down to just one time.
Against Manchester City it was total sell out for defense and maybe one of Gabriel Martinelli or Kai Havertz does something special to get the ball into the City half for a spell of relief. Arsenal perhaps could have left a man higher as an outlet vs playing the 5-4-0 or 6-3-0 formation but you can’t really deny that they frustrated Manchester City in attack even with some crazy numbers put up by Manchester City.
One Way Traffic, but made it pretty ineffective
415 - Passes attempted by Manchester City in the second half
60 - Passes attempted by Arsenal in the second half
323 - Passes attempted in the final third by Manchester City in the second half
32 - Passes attempted in the penalty area by Manchester City in the second half
21 - Passes attempted in the final third by Arsenal in the second half
In the second half with Arsenal defending a lead while down a man, they decided the best strategy was to turn the game into 54 minutes of attack vs defense practice, daring Manchester City to try and find a way through.
This created some funny looking stats and a very funny looking pass map for the second half.

This is in pretty stark contrast to the first half where Manchester City had slightly better stats but it was a much more even game, with Arsenal seeming to grow into the match after a blitz from Manchester City to start the match.
Possession - 68-32, in favor of Manchester City
Field Tilt - 82-18, in favor of Manchester City
Touches in the box - 14-6, in favor of Manchester City
Non-Shot xG - 0.5-0.3, in favor of Manchester City
Shots - 5-4, in favor of Manchester City
xG - 0.9-0.7, in favor of Manchester City
Looking at the graphics and they all I think help tell this story, City were the stronger team overall, but after the quick start where they were able to generate their two best shots of the match, Arsenal were starting to get a foothold. It looked set up for a second half with lots of possibility to showcase two very high quality teams, instead we got a sending off for what now appears to be a pretty lame technical violation.
In the second half these numbers no longer were close.
Possession - 93-7, in favor of Manchester City
Field Tilt - 97-3, in favor of Manchester City
Touches in the box - 41-3, in favor of Manchester City
Non-Shot xG - 2.0-0.3, in favor of Manchester City
Shots - 28-1, in favor of Manchester City
xG - 1.9-0.05, in favor of Manchester City
Even with the numbers stacked so far against them, Arsenal did a decent job employing the bend but don’t break defensive structure.
28 - Shots for Manchester City in the second half
1.8 - Expected goals for Manchester City in the second half
0.06 - Expected goals per shot for Manchester City in the second half
1 - Big Chance, John Stones (90+8)
19.9 - The average distance for the shots that Manchester City took in the second half
16 - Shots from further than 20 yards in the second half
0 - Other shot besides the goal that was rated as 15% or higher
The volume of shots here is eye catching but I think the key numbers here is that they limited the true break through chances to basically just one, and that came on one of the last kicks of the game from a scramble after a corner kick shot.
Manchester City ended this game with a TON of final third possession but they just really didn’t create that much possession in the dangerous areas around the goal. They had 523 touches in the final third and just 17 (3%) inside of 15 yards of Arsenal’s goal.
It may not have been pretty or “the right way” but Arsenal’s defensive shape was no doubt effective at the main job of no matter what do not let Manchester City play through the Arsenal’s middle. Looking at the central locations that are often the spaces on the field where a team can open up the defensive block, Arsenal just did not allow much penetration. Looking at the graphic below, you can pretty clearly visualize where Arsenal played their compact 5-4 and 6-3 block in this match.
Some thoughts on timewasting
John Stones on Arsenal 'timewasting' tactics:
"They've done it for a few years now and we know what to expect. "You can call it 'clever' or 'dirty'. It upsets the rhythm for everyone. "They use it to their advantage and I thought we dealt with it really well."
68.5 - The minutes of total ball in play time in this match by my caluculations, the 7th longest ball in play time total this season in the Premier League
62 - The average ball in play time for a Premier League match this season
39.9 - the average time it took on Arsenal’s 12 goal kicks in this match, average in the Premier League this season is 30 seconds.
82 - The longest goal kick time for Arsenal in this match, the quickest was 12 seconds (which happened in the second half, surprisingly)
25 - Seconds that Manchester City took on goal kicks, they had just 3 opportunities in this match, and they were 33, 35, and 6 seconds. Two of the three were longer than the average goal kick time in the Premier League this season.
On the goal kicks, Arsenal were a bit longer than normal but were not far off what Manchester City have normally done this season. Manchester City were a bit faster (helped by one really fast restart).
20.8 - Arsenal’s average time to take each of their 5 throw ins.
9.6 - Manchester City’s average time to take their 18 throw ins. It was 12.2 seconds in the first half before they sped things up in the second half.
34.4 - The average time for Arsenal to restart the game after their 9 free kicks
41.1 - The average time for Manchester City to restart the game after their 18 free kicks
0 - Yellow cards for timewasting during free kicks, throw ins, and goal kicks for Arsenal.
There is no doubt that Arsenal tried to find little margins to try and take time off of the clock but it was ultimately pretty fruitless with this game having more ball in play time than the average Premier League game and Arsenal just not having the ball enough to take advantage of chances to delay. This was also a referee that was pretty intent to enforce things strictly but didn’t see any egregious time wasting that was worthy of a yellow card. This is not a new tactic and it is one where teams that have it go against them don’t like it, teams that are holding on will try and find as many opportunities to do this.
Manchester City have long been one of the teams that have exploited the dark arts to their advantage. They have been one of the most cynical teams at using tactical fouls to stop counter attacks, they will surround and pressure referees (even though that is supposed to be a yellow card but unevenly enforced), they will exaggerate contact to draw fouls, penalties, and cards (look at Rodri doing that in the opening seconds of the match and staying down for several minutes).
One of the very fair criticisms of Arsenal for a long time is that they were naïve in not exploiting this, and doing too much time where they were too sporting and letting teams take advantage of them when they did not return the favor. Arsenal tried to waste time in this match, but they didn’t have the opportunities to really disrupt things and in the end had all the time wasted added back and more compared to the average Premier League match.
Rice got a yellow card for time-wasting, no? 83rd minute after Havertz was fouled