Attack vs Defense: which is more important in the PL?

Scandi Red

Hates Music.
Joined
Sep 25, 2022
Messages
6,827
Obviously you want to have the best of both. And the Champion often does. But let's say that you were given the choice between two teams. Which of the following should you pick?

Team Attack
- The best attack in the league
- Not the best defense, but still comfortably top 4 level

Team Defense
- The best defense in the league
- Not the best attack, but still comfortably top 4 level

----

My gut-instinct says that Team Attack would be the better pick, but let's find out! Here are some simple ground rules:


1. I will start the count from the 2009-10 season and upwards. 15 seasons should give us a decent amount of data without going too far back. It also features a good collection of different teams, managers and styles.

2. To make things simple, the definition of "best attack" and "best defense" are simply most goals scored and least goals conceded.

3. If the Champion has scored the most goals but not conceded the fewest goals, then Team Attack gets 1 point. If the Champion has conceded the fewest goals but isn't the top scorer, then Team Defense gets 1 point. In every other case it's a draw and neither team gets a point.


Going by the rules above there are 7 seasons that will be filtered out because the Champion had both the best attack and defense (or neither). That leaves us with 8 seasons. So what does the score look like?

- drumroll -

Attack wins 6 - 2

It's also worth noting that there were a few draws that would have gone to Team Attack if the Champion had conceded just one more goal. In other words, the 6-2 scorelines actually flatters Team Defense a bit.

Now you might think that Team Attack also did quite well defensively in their 6 wins. But the average number goals conceded for the Champion is 32 and Team Attack only did better than this once (23 GA). Generally they did much worse (32, 34, 37, 37, 43).

Overall I think this shows that if you can only choose to be the best at one thing, choose Team Attack!

EDIT:
Or to put it a different way: if you have the best attack in the league then you have a 73.33% chance of winning the league. If you have the best defense then you have a 46.66% chance of winning the league.

--


BONUS
Here are our 3 best attacking and defending seasons post Fergie.

Best attack
1. 2021 - 73 goals scored, 2nd place
2. 2018 - 68 goals scored, 2nd place
3. 2020 - 66 goals scored, 3rd place

Best defense
1. 2018 - 28 goals conceded, 2nd place
2. 2017 - 29 goals conceded, 6th place
3. 2016 - 35 goals conceded, 5th place

So even under our sub-optimal circumstances it's quite clear attack trumps defense. And as you can see: 2 times we've delivered Champion level defensive numbers (in terms of goals conceded) and we've still finished outside the top 4!
 
Last edited:
Obviously you want to have the best of both. And the Champion often does. But let's say that you were given the choice between two teams. Which of the following should you pick?

Team Attack
- The best attack in the league
- Not the best defense, but still comfortably top 4 level

Team Defense
- The best defense in the league
- Not the best attack, but still comfortably top 4 level

----

My gut-instinct says that Team Attack would be the better pick, but let's find out! Here are some simple ground rules:


1. I will start the count from the 2009-10 season and upwards. 15 seasons should give us a decent amount of data without going too far back. It also features a good collection of different teams, managers and styles.

2. To make things simple, the definition of "best attack" and "best defense" are simply most goals scored and least goals conceded.

3. If the Champion has scored the most goals but not conceded the fewest goals, then Team Attack gets 1 point. If the Champion has conceded the fewest goals but isn't the top scorer, then Team Defense gets 1 point. In every other case it's a draw and neither team gets a point.


Going by the rules above there are 7 seasons that will be filtered out because the Champion had both the best attack and defense (or neither). That leaves us with 8 seasons. So what does the score look like?

- drumroll -

Attack wins 6 - 2

It's also worth noting that there were a few draws that would have gone to Team Attack if the Champion had conceded just one more goal. In other words, the 6-2 scorelines actually flatters Team Defense a bit.

Now you might think that Team Attack also did quite well defensively in their 6 wins. But the average number goals conceded for the Champion is 32 and Team Attack only did better than this once (23 GA). Generally they did much worse (32, 34, 37, 37, 43).

Overall I think this shows that if you can only choose to be the best at one thing, choose Team Attack!

--


BONUS
Here are our 3 best attacking and defending seasons post Fergie.

Best attack
1. 2021 - 73 goals scored, 2nd place
2. 2018 - 68 goals scored, 2nd place
3. 2020 - 66 goals scored, 3rd place

Best defense
1. 2018 - 28 goals conceded, 2nd place
2. 2017 - 29 goals conceded, 6th place
3. 2016 - 35 goals conceded, 5th place

So even under our sub-optimal circumstances it's quite clear attack trumps defense. And as you can see: 2 times we've delivered Champion level defensive numbers (in terms of goals conceded) and we've still finished outside the top 4!

When i was doing my Masters in Psychology i took stats class. So one day i was bored enough to calculate correlations between GF /GA vs points. My sample was EPL results from at least 5 seasons. Funnily enough correlation between GA and points was stronger thus proving the old cliche that attack wins you games and defence wins you titles.
 
Especially in the last 10 years, with titles often won with 90+ points, you have to win incredibly consistently and that means scoring a lot of goals so you have a lot of results that are protected against unlucky breaks like opponents scoring worldies, goals off deflections, or getting a soft penalty call. Even the best defense in the world won't be able to protect every one goal win against that kind of stuff. So its not surprising that the title winner is more likely to be the team that scores the most goals.

On a deeper level, I think the true trick in today's game is to be able to attack in a way that also prevents chances for the opponent and to be able to defend in a way that also creates chances for you. Its all about controlling and optimizing transitions between attack and defense. That's what the best teams are really good at doing and it goes a lot beyond simply having the best attack or best defense in the league.
 
Not a real answer, but in both cases Team Defense needs a REALLY good defense and structure to compromise for a weak attack, whilst the attack also needs to be REALLY good in patterns and skill to compensate a weak defense.

It's funny and like something that goes in waves. During Simeone-Atleti peak (2013-2016) and some others before and to a lesser degree after, it seemed defense and discipline could win you everything, especially at their freak levels. But even they had basically converted wingers in Juanfran and Filipe during their best games in FB positions. But it did make me look at the game differently. There were some games, some even just a few years ago like one against Liverpool, where without almost any attacking threat, Atleti just looked SO dominant sometimes. Like an actual impenetrable wall. Beautiful in its own way. Most games it was still really grinding it out and sometimes unfairly so. But those few games made me love defending, while before I just saw well timed through balls, perfect crosses born out of crazy vision and wild curlers of goals.

In general many normally cannon fodder teams really rose in power during those times through discipline and the learned arts of defending and focus. It lead to many boring games, especially among National Teams who are of course way less gelled together than club ones. When it's not impenetrable, it's usually very very boring, though I respect the art of "suffering" it out, too. But one of the reasons I began to dislike Simeone's approach, even with the occasional more offensive approach, is the lack of adaptability, and quickly grabbing back to old ways.

Plus, nowadays it just doesn't seem as effective to aim for 1-0 while building a stronger defense than the opposition, or settle for defending a 0-0 (or 1-1 or something) when the few creative forces can't get their engines running, or solve the opposite defense's puzzle or something. I wouldn't say team attack brings more of a guarantee, but we are still in some high press type of dominance. Perhaps at the end of it, now that talents like Pedri and Yamal win more games for you than more of the powerhouse types do. Guess every generation or so it's a countertype that kicks off the next generation's dominance.

The bus parking and quick counters made a lot of possession redundant, then high press made untechnical defenders less reliable. Now press-resistant midfielders are the most sought after and we have Gravenberch and Pedri as some of the best controlling midfielders around today. Once more of those develop I can imagine more tactics that make dribblers/carriers in midfield redundant again. Perhaps ETH was simply ahead of his time by trying to remove the midfield last year :lol: it did actually work against Liverpool in the FA Cup I guess.

So it seems now it is still best to sweep the opposition off their feet in order to not let them do so to you, so I say team Attack?
 
Depending on your style of play, if you’re attacking more then naturally you won’t be defending as much. If you’re more defensive heavy then you won’t be attacking as much. So attacking benefits both areas where as defending only really benefits the back.

Now that I’ve read what I just wrote it feels like I’ve just typed what everyone already knew and didn’t really need to post it. I’ve written it now though and it’d be a shame to have wasted my time so here you go.
 
In these days of having such a terrible attack, it has to be attack attack attack!
 
For leagues? Attack

For tournaments/knockout style competitions? Defense.

Biggest reason Amorim is getting such crap results is because we are bottom third level as an attacking unit and our options. Which means, especially as a new manager, that you can coach good performances and still lose/draw due to being unable to capitalize.
 
It depends on the actual numbers?

You could score 20 more than everyone else while conceding 2 or 3 more goals more than the best defence.

Likewise, you could have by far the best defence, letting in fewer than 20 all season in a year when the next best defence lets in 35-40 while they only score 1 or 2 more than you.

In both those rather exreme cases I'd expect to win the league easily.

Then there's eveything in between. Sometimes the 2nd-4th best attack is close to the best attack in numbers, sometimes much further off. Same for defences. Then there's how much separation there is between the top 4 performing attacks/defences and everyone else which will vary from year to year too.

Edit: Fewer, not less!
 
Last edited:
Really depends where you are as a club. A squad/club with low morale, lack of confidence and one that has made a habit of losing or getting battered - I'd say prioritise defence. Rebuild a bit of confidence and professional pride, by just not losing even if wins are had to come by.

Most other scenarios, I'd prioritise attack - 3 pints will get you further up the table, than 1.
 
I also did another check.

The average number of goals scored by the Champion in the last 15 years is 90 goals. The average number of goals conceded by the Champion is 32.

Now, how many times has a team managed to match (or do better) than these numbers and still not won the league? In this case, the lower number is better:

Attack: 4
Defense: 7

It's worth noting that 3 out of those 4 attacking instances occurred between 2020 and 2024 (Pep, Klopp and Arteta). So goals not being a guarantee for success is a fairly recent phenomenon, even if it's still clearly more effective than not conceding.
 
Neither. Do a Liverpool trick of getting the refs to drop to thir knees a swallow whatever you put in front of them and your golden.
 
Oddly had this same musing earlier today. But I would say best attack. Goals win you games, first and foremost, and devastating attacks eventually earn the respect of most opponents and they won’t attack you anyway.
 
Look at the last 10 title winners. Every side has cracked more than 80 goals besides Leicester.

Our last title success, we conceded 43 goals, but scored 86.

If you're not scoring goals(2 per game at least) in the PL, it's pretty much impossible to win the title.

Liverpool almost won the league in 2013-2014 when they conceded 50.
 
I also did another check.

The average number of goals scored by the Champion in the last 15 years is 90 goals. The average number of goals conceded by the Champion is 32.

Now, how many times has a team managed to match (or do better) than these numbers and still not won the league? In this case, the lower number is better:

Attack: 4
Defense: 7

It's worth noting that 3 out of those 4 attacking instances occurred between 2020 and 2024 (Pep, Klopp and Arteta). So goals not being a guarantee for success is a fairly recent phenomenon, even if it's still clearly more effective than not conceding.
When everyone started playing out from the back and hosting goals galore.
 
Attack is the best form of defence.

The reason we are losing nearly every game right now is because Amorim is playing too much of a defensive team with 7 defensive players and 4 attacking players;sometimes as low as 3.

It's uneven and draws the pressure from the opposition directly on to us rather than pinning the pressure on the opposition themselves.

That's when the defence becomes important because the attack is pressuring the opposition but whenever that is not done, the defence can sort out the mess.

SAF said something similar

Attack makes you win games, but defence makes you win titles.

However, you can't win titles without winning games and for that reason Attack comes first to win a game, with the defence then providing protection to get the games result.
 
In the previous 15 full seasons, only once has the winner conceded more than 38 (1 per game avg.) and that team was the only team scoring 80+ that season, scoring 11 more than the second highest scoring team.

Only twice in the same timespan has the winner scored fewer than 76 (2 per game avg.). One of those conceded the fewest in its season, and the other was Leicester in a freakishly close and low-scoring season goal scoring-wise (all top 4 teams with GD of 29-34, scoring 65-71, conceding 35-41).

Once you get passed those two thresholds (38 or fewer conceded, 76 or more scored) it looks like winning is slightly more correlated to scoring prowess than to defensive strength.

I'd go with Team Attack.

https://i.imgur. com/dbI9jeM.png
 
It's attack. Don't think Rafa ever truly appreciated this. His main downfall in this league.
 
What I would say is that the "best attack" team will generally prevail in a league based tournament. But in a knockout competition, its more 50-50. This is just my off the cuff view but I am looking at France 2018, Italy 2020, Argentina 2022. Not to say that these teams couldn't attack but I would say defense and control were more of the qualities I would associate with them rather than attack.
 
It's attack. Don't think Rafa ever truly appreciated this. His main downfall in this league.
The one season you had a great attack under him is the one you challenged for the league. Unfortunately, United and their joint best defence beat you out