It is the nature of the modern Premier League, in which 90 points has come to feel like a basic minimum to win the title, that analysis of how the league was won tends to focus on where it was lost. Manchester City’s excellence has become so relentless that the assumption is they will surpass 90 points. The question is less about anything they have done than whether other teams could have done anything to overhaul that total.

In that context, the mind goes back to the final day of March last season, and Arsenal’s trip to Manchester City. Liverpool had beaten Brighton earlier in the day, giving them a three-point lead over Arsenal and four over City, having played a game more. Arsenal were content to frustrate City, drawing 0-0.

Although City shaded the xG, Arsenal had two of the three shots on target in the game. They had maintained their advantage and knew if they won all nine of their remaining games, the only team who could finish above them were Liverpool.

In the immediate aftermath, Mikel Arteta praised his side’s resilience, pointing out City had scored in every home game for three years and stressing the need at times to “leave your ego aside and your ideology aside and do what you have to do”. Arsenal had lost on their eight previous visits to the Etihad, including a 4-1 mauling at a similar stage the previous season. In the immediate context, it seemed a good result.

As it turned out, Liverpool won only four of their nine remaining games, while Arsenal won eight. Crucially, though, they lost at home to Aston Villa and so City, by winning nine out of nine, lifted the title for a fourth successive season.

So while in the most immediate sense, Arsenal lost the title at home to Villa – or in the unexpected Christmas defeats against West Ham and Fulham – there’s also a nagging sense that they let an opportunity slide away at City.

It was a game they had under control. Could they have been a little bolder in the final 20 minutes? Would it have been worth gambling to try to put clear water between themselves and the champions? Might they even have been more likely to win their final nine games had they not had the pressure of knowing they had to do so?

There is, of course, no definitive answer. Had Arsenal opened up, that might have handed a chance to City, lifted them above Arsenal and had everybody condemning Arteta for his hubris. It’s not a case of right and wrong but, with hindsight, and given that at the time of the Arsenal game, City hadn’t won any of eight matches against the sides who would finish in the top six, might that have been an opportunity missed?

That’s not to say Arteta should have attacked from the off, it’s not to criticise an approach that meant Arsenal conceded five goals fewer than City, 12 goals fewer than Liverpool and 22 fewer than any other side last season. It’s to suggest that in one period in one game there was a possibility to attack opponents who seemed to have become flat.

It’s certainly not to suggest that Arteta should be gung-ho at City on Sunday. The issue is more of being able to sense the emotion of a game and take advantage of psychological shifts, a difficult and imprecise ability that is perhaps being lost amid the detail of data-informed planning.

City in March was the only away league game Arsenal have not won this year. They have not even been behind in an away league game since losing at Fulham on New Year’s Eve, that bizarre game when they took the lead, seemed comfortable and then lost their way, as they had at Liverpool and West Ham the previous season. This occasional habit leaves major doubts about their psychological capacity to win the league – and was perhaps reflected in the way Declan Rice’s red card so unsettled them at home to Brighton.

Yet in 11 away league games since that Fulham aberration, Arsenal have scored 31 goals and conceded just three. For somebody who is often portrayed as a sort of Guardiola-lite, Arteta has a very clear streak of José Mourinho.

And that perhaps makes sense: with so many sides practising guardiolista football, competitive advantage is to be found at the margins, in refusing to engage and going long to evade the press, in defending – something Jürgen Klopp predicted five years ago after a goalless draw between Liverpool and Bayern Munich. Even Pep Guardiola, with his back fours comprising only central defenders and direct play to Erling Haaland, isn’t a classic guardiolista any more.

The 28% possession Arsenal had at the Etihad was strikingly low, but they also had less of the ball in the second part of last season against Brighton, Tottenham and Manchester United. Against Spurs last week, they had just 37%. That was perhaps a response to the absence of Rice and Martin Ødegaard as much as to the opposition, but the job was well done, the sense of control far greater than it had been in the 2-0 win at Aston Villa three weeks earlier. Against Atalanta on Thursday, set up to defend, they had 46% possession but, as at Villa, needed an astonishing double save from David Raya to maintain their clean sheet.

City this season have been a curious mix. They have conceded the opening goal in both home league games, struggled in patches against Brentford and West Ham, and were pedestrian against Inter, but Haaland, even by his own stratospheric standards, has been in exceptional form.

Arsenal shut City out in both league games last season, but that will be harder to achieve without Ødegaard and, in the form Haaland is in, offering up even a half-chance could be enough to undo them.

Potential points deductions notwithstanding, this fixture already has the sense of a vital game. Avoid defeat and Arsenal can reflect on being in touch with three of their toughest away games already played. Lose, though, and a five-point lead for City might already look decisive.